MA58L@=V
MA-:%X$-M<7J;Y%DZB].R,*S`=>`[4`5W63[&B_ ^]!L<2'!7XX.6+2+7(
M[TM5HA;Y:.UJ`&TP_6#;S72@C[[.16V]8UGO6/Y/KI[][0;3'_R4P#MIB=0D
MM'GA?PPMD`+IX%.\G0C$7[I"I=Y#&(W0]L#>-)![H7S;@C#;Q]SQ(ADW+*TS
MA;U>9"W[AM?+E#KYJ/[^BM-,$9>VWXS54NDQ0K!AUPT#G_0Z<6!46LY.V3D&
MW76+;2VIUYJLA/M).6-;&>69[YE]R\&!""1BWH0*F[&*WXO`V)0NCT7:J&Q(
M'\H"9XK=HUH3]?[T-M)S).[!9LR_!,!*/$PJ!!962T*+10$^JDOLGN``<<
M&JH+84W&M)OND-)''!-8PG6>+>:`$R[#IPZ3"AR6A)VV09/>QG$"N%UA0YGM
M5\*`%2F']#?'5LWTB!_=*2Y#^GP0UZ%E=:(8#C"9(PT:F]A":@C3KHR8\$8M
M-@?'B[DAI_7,OMI!_*^^"B@.6)B"/S1S9+JY/KZ+1U`P%9N\1MDD1RTNE/:((2B;M<$:21C'U,D(,Q4A@6^:$32M
M8+F=N^:4KT46+4XL?F)4/&DR$OV%]%3`53*,W^%@@DRZEE&Y*+HU;+B=BKT\
M?CL"3S##FE$%RP8Y?N>+/QX1OGP4",)J%?X>/#T[-3(TT^
M-$;33@]?'!E%7QY>/C>3CE\>A3^_>JHD/7Y^]MMI>'YT<7E^_/CRZ(F*\RR\
M/']U^EA)^O7)\<7ASR='2M+%[Z>/P^,S)>703GIY?GRFIUR29MU:G!QNZ+
MX_\Z^OGX\@*GBN<%))\N4`7GS]2&7>!6')\\,=-.?@DO'_^BI)P^.S][]?+"
M@#M[>71J)"%F'!V^,!(O_\MD,DK\?V<_AX_/3B_/ST[4\H>_'CT)CY]<*&D(
MX\GE,4)P_S\[/3LU<7
M"E,E(L3`)WH&+V'G/,59RN\7AR]?(AC2$6KRT8N3,XV9+"4\/SQ]=J2GGYW_
MCD@YNSQZ?'DL1R[)N[@X?':$AN;%A=[(BR-4\?.S'YX@6AO'LY_^'
M$&J,0(/JY/@"56+TRB&80@;ADZ.3RT,C$Z4=_D[8;&2\^$]H=*!4@DI/_15U
MDM[:EZBI>#"C)#D*+E'WFZ/MZ$5XBOXQ!R9.__7PY)4YYA"&_WQU9"6K#5!J
M_!G][_#"!$:I3X[-$8X2+QX?G@"P6#J<6I/N[.0D_.WH^-GS2Y/TH_]\=?PK
MFH&HG\V[T\#T^1N#'2T:`WZST_"I^\>FG-],-SBE>9C2*W#W1#/T1-.#Y]
M8B0].?I52WEZ=GX))R(9J"5>_&;!H9F`^/;DZ*E*S,OCXRWM5_BO2S,%B\&C
M2R.1O@II)9]=:*41][7!2W+11H^K\>OW+FO3H]_I>5^/C\]Y>75NK1Z?/PN'=@X[AX_D*7
M3EAZA6@8J=,""R-(:KUZ>62WY%\OG_7M9KQ\-H`2=]3RI.J?M6XF2;9X-08K
MD@@&$$G18$[.T.J@(__M[/R)D?3BY_#$6DU/_^OH7%-5@+E[`5!Z`9!Z\?S<
MI)4F:5"O`&RO[%:^(HTRTI0:9`M.L$9JM.HD/$&*D97XXL)..[52+HXNK30T
MXB_-P?WSQ6YX?/)RT`_/GCX=:`-#R_KY^)F9=_)R;P=G[>W8.0A8H4VV+RR-5
MVKQ`VX9+M!%@_%!S7J&.>HG*,!&M:9RGS\`,I"B@>6Y4C%F'--'?SO&"20C3
M<+U$R["9>'[T#"F&9@(@%B^>'^EJC[69NGAY^)M6`@V$PR?'6$,Z1_B`!=H)
M0)D7/CF\/#3&GY&C=2P>,6>O-&WH\O>7OGW,*U1S2/>F<*J&_^7/%_JO\/#Q
MX[-7IY=Z+^`,K)9>'AF);!-FI%Z>'VJCX>)WM+T[>XD3/@PA>\7C"V)A$'N-
M+9%\\A0M^T]/#I_A]\A[V]O;'"W+.WE",LW4XY_MI--+`!0)KL=P,H@9IUNX
M22+##C3!E*Q*>WI:>RQ`FP0;1*<&R+>;;=LQ2;JJ(0`G8UHJM
M@2XK+`NOPR*%VW$69'G3ZNXL4[POU!("SHA/@6&*,E?!7.\)36,C!BP^5!O3
M0#'S9&R'&%?S=0`0Y#J?6]',33AR@V-^G9!`&CQI3HM`*$%`TYA<@)!;2J4X
MFYZS[NX>.,]93910>QSM0O"%SB&[-86C,?*HD$,NX-X0^;$#X)H#7%?DQ]=P
M=[+`5^08K1#&=QS(1#EDHV$"K`,&4@B'8KZ*R5&;.)-S=MJ",H2?YBT@:A!8
M'AN`^2+1CA5C5TFSX,))"AL^-LTF*09@?DU)X;]C5TFS(,`5/C0F6?X6/O_C
M$+I1VR0?/69";H:28\0
M/.*L58A-$(;2(06GV->)O1ATHZ1S!\GT5/1_\C2]6C/TX9>#VX$]JYTM]WH0)3#AP5J
MS8@/>]HC]#SWUGP]BCQTB0IRMQ-\B;COX:?ZH7YQ]";V4=[O>QRV`>IOLJ+4
M+Q&Q\2>\`N$96:A%88<^[XQFY;%\5P)P6:*943G.9E&2@G3JE>@UJ,66H'%2
MO$]'EK!6(6YOHO1Z,;?5!BHI;[.WCO=Q8,^M/$-YBI<4?4@59^$Q$\WFTSBD
M;P_4"9DF(+B;HNE]-8JFIG,-<_(L?3)%@%$P
M,>=##9Y@8QF%01,
M>$Z34*-EQ*\D'L"7,\"M*`G4H(@WE)9FJG,/0O=HQ^/"JFI^^*JDO14>S4P>
M.QA,RNO,-3%(/D.;-Z._\$O)FGQ,IN/8CD_%%L/\_1P0$V_C][;264334I,=
M3#(P'$+?Q@WB;8C'.(2'J?E3B7$777D#4%+5%\@H,TBL%J87M=W8$@Z6AI01
M^#:5M@;0VV%(!LCM2K!!1%<\C_*XR\:W-UJ#+!1A!\EF948WJ"=Y$3%.!X[+
M>E:(D@9._EJYAA<$+%=]-&Z7_VM;*?6!E6W=[.B28>V6&G]7FWU#D6S
MKFVYZJ;UUYP_SO)+W`K21++S)@ETPP,MG_"\3U*LS1!:B[&
MATK%HS++WU.358*OXJ10A)'R)BGP]JIRJNB?VH*&1:WK)%J#S6L?9J;SVH>N
MO[1U@<-SA<8U#LM<_8[.+59VQV&?B,#3RR4G;$8+X7P#@=,@YT"()%17X,
M`EQ+@.N*_!@`H,VN?^"C%_.>]VB@KK,>$ZCN28]9#CQ;,8'J'1':)-4[\;'Z
M7#^CT=`N<_BB(0#.7G3)WO1@1&_U$BS[;*+Z^(R:GAGAE-;8BZ!E'3
M9J=9R%!O>&QD'=,D0)31.,7]A-7!11Z3HYF]@^HH0^L.$%W8YDJ
M0P2)5,6&M8Z7LHZ7LHZ7\C7%2QGLF#+`CI3"7G/`IT)W8Q()FJYV\SNRW`W5
M!`J$D[@.A1+1'S3EFJ=YI1[)B%;#/Y>1(.S"$Y@6J?63?"_;A<)+V;DDN<-K`Q^;9M#3V3311
MRQ.[)S//J7\QSS4+W@PI`L8YU:LVXG?Z2ZB*'4TDUW;HA88P23/$]C'^[Y`E
MT4WM&/^7"@><#!@PT-3#,6C2H0Y`>G9,#J!I#DO`0_QU?W?OS9#?(M*(VMN1
MMXDP->3LQZ2+)1+*/A95O+O@[N4A70\07E+`G$\Z/GJ&$`YEPEHS9>!#UD?D0#M=,8'B!15%\64;>*']L#C&-@?#MN2
M"#@I3)#/R?3M[>PZ^*KKL30CV,0F)P"1N1H:(TV4PX>64$G-/9%`AK(.9?7%
MA9R/T^J4:N502OZ^9L&*]=Y!+YHIC2D.;#[5IQHH[*%9#.T_$:WL;2GX)*[\6\91XL70M0R='ZW"CY0A&LWG7]#Q_%C.VW'//D&-
M^]T.<.PY7^2Q$N72YK6H1N7VZA6I=X3BO$@0BH_>%*6B=ALCJY&O"N*3=S(&
M\B0N()>,6I<^U3';T$]6[.>-NJ(BGCMV$0[J=0?/9>E?H@7JWM_7"M.&J4M0
M=;>*\%7L5VW7!V#[`&])J1]$37"^QW+L1BW7%WTG2O<_5OU-]FD>LQ^TDP1J
MI$VHJ-.Y.P3J=-!TO&GC*5B!B
M]U-4@:]2'O%+NO3GE?,%/"D=-<2&^%T1];W@T:"9P=>G@EGO"R-5D9F"-0B:
MHP$(>["99U:PWK.M]VSK/=M7LF=C/C66/+"=\KPFJ+LH*Z/48'XV%7R=^O0;=C&FZE,5Z6)3R*AU/(@($(A8CMF1I@)/%"0M>
M8^Y82']ZSD-7<+9H_\FIX&NVZCW[0V]O"&9@UU2DJY2C;!S_<."!&65Y/%[,
MYC_T/$#XVBM*^F%_R,GX@--1*TA6DD9E//;0WISTHLSFN$8?Z1CF-IIB&),J
MG#6G)*D'9#N58]_\^)D2Z3?19;(7L7*RF)?Y4'0E2DA8`B;GM\-C'*CZ\/+5
M!:`5E'F4%O0J;TB0VMM`BQ3!83E\_EADI:0@CV>T]G%RRU[\#;\.4<5H04@,?6A=NH`*M!N7CNP^N]XUP?5A3#&=7^(F88UX!\T^QO2^#$YV`)Y2_(0'TW"&V!+JWM'M'CYX#`]DR5I
M9E[?D0DOU8@B"A&R=9)X,K$\EH-CKF"NT)_%Z][@S9`H3I1I\%TLNAY>#068
MEHY&1787Y\[\"YIJV%M]`S]:C?((6M$E[799Y6-;(3'
MVW^=H:F,L1:-5VR/KXM\5%`MC1$Q68V(1C3`^AL?N:NR8_4NA0CNV+JTZ#YW
MQ(3FH[)ZOHE7/^%N;)&8YK3`_=DF?U;O7?:I#0!V=&)I:Y?X%0E7!8S1@%J+
M7?L]T4JS:C:PV?KW,=K:2CN];5QR170UMAGWZ;?J8*!?:T."?BZF=:1=21P[
M)CB00LSD4*>NR<.W]2<73^G3J[#83]E*Q\7C-NK-#Q9%O..CE+87HM4=*I;6JHU_]^TA=-:+?[GRR,G;I6J(;=/UFV^]516<=:U9"N
M:9UQZ3([ZI3JG5[#3-+&:&W2Y?XVP$*S4]OHT\X(;C:$FW=+'=M+&RWYNSK&
M;@^P'M5<7-KM&\?RXCF'4"N5JQ"MAP7MV>P&&RS(37>["TH-S!O/085:"7:M
MZ2HF:4]56T&O8HWRG&E8#:M=6[,1[CWXL&AH0(2F,%#8Z=Y.I,5\3L%[.]'>
M#D`1^%B[>@QC>?*\+QYB0YTXV1X\\AG>.NJQU:"/-J9YE(ZSF2/@1,%SK9".
M,?R$2I+2@^P8+"(=M0B,UZM21HS"H&9P?HH'!XBCE9EXY>$HLY_2AA#[J7HD
M2EBP.5%/9FE2;B<1W,KQ*<*H7+H42>/XVD@IXKF!"HF&D!_^ZB>)C$YY!UTA
MO+93JJBG.GY"(2L$.PRF@G%8P2,ZWHD)6%26&`65+I)>?IF-CR7-RY!7MTI]
MVA%X.>&L^>:9;082B]>;^M0T#4UC7&CGP`PBQG)CG@N<(H3ANV)QE;P>O`&%
MV11&+/+3Y5'/*E#_NS%JR6E<3-T$(];=1E-9"72<@H&',WMXMK5"M;HK$
M_+6SV#S*H]GK?4HE&U2LKW2)!91^]WKPQGEC//)ECIPY6/F7>/GTEEXRZKU2
M1J846"KA<'P0QW1BP]`AM)0Z8UFGM].K+L=\'$+U-Q?:8X\8J-4,2C\Q@ZI(
MU5@T^R0L^O>7Q*)"$FN+JRT7+2P*3L=<\V)?NZ4$KQDU&_*8TG!:I.4Q
M0\QO7Y$]QO*$3O(XUI!I6@))'SE@S'T$J2+B-P*,VP!JI@S18N2QOEC?!%C?
M!%C?!/B:;@*P]P0M62"\H3491_--\:LZ5SOD3M_&A>]J>;#="W9W]GS@W9
M[5+K0D=;T$A$"QR\/9JBU8R+U4V4H$IRDCF+TU*DU%TO5(*C*W)?V!7..=-)NKF5I")?@IX/XTH&,>AWEUUW]EW['K@UT
M5@=2C"J&6FS1<(1'JR&A1@+\+$!Z6R,@E+!*D9#7874Y,^`J`9>VK3Q)KP'`
MPH/7#CR/YL%"G'WF\7P:C9@7I,HII*_5H%:/?!7C"PJWKN?W:/G9VQ)-%MDH
M_&N*QBL8.'SVMJB"!L#)W4MW`9.>L;<*E&-U)6V116(8"/371WD*F"54,-K6#&TZ32Y(CO
MZN!_Q3,QBUDL_+W&<9K-JO!QXBBJ*<&EGJM0A$I*+:P.#O!:E&JLNFKYFYA%
MZQ$%3_QX=(N&H?"#TT1W.DZN$_;HGSZSQ_%H#L<$L&'QKAQ:XR:?K.;K&C4K
M]_U56PE-_H-R37EJE+L%>N633UGY=B.4:;,[M".
M\0Y#6\>I\[N1<]%WJ&C>VNY&93:[4M3]K8!7'(;D+TUQ8(AF5P4FLO!0Z7W)
MM3ZM>KT(+2:W@`].JVOT4*O6VX&V!?E\%I4C6!N>HX086E!Q0;0!*Q97F7Q$
M6J\=92"=P!E#5NK[N$Z]:)F]11I%D]NN9-I887GY*R!(\04?RP:W2==YA)]A
MTEZPTB$6*7XKVP;1EX]Y6=#-B0&D[THI4*-7V*S0_:6Q2=2SIUDTCFZO%8'.
M4EZ_$7(FGL8SU:ITX(AOT]?1$[<(U`^ZTXMF(@@KG%Z,O7U['B^PSXME2O"X
MO&@VG2I/$M)7NJ<(%+%J!?\0#5%]!Y&E.-6`$LCK0]19'TV%`X8&N^J)=<51
MV)9^)`UTXZHGPE4$.(L9YYP`::N>Q'Y$TOSGKLO18X_&UI?H3TXP\)<77(?Q3X4=((4W2
MN*!*@NPLO;?^6"2CMZI]CZ*8:T:^,BNC:8B5#(\NJQSNF*9%'"C1;/SH710:
M)S/*N0R.O\B/T/)K_J?-PA6/H/!7?>1U+^@-=BN"#]HQ!?'QA.E!X/13WG$4
M[EMG@*AIH]'\O33'CI$6;9N%BQS>CVDS>.3?U9!#EA3>*1(*E>*SZ!T>_2Z6
MVD%^%$Q%/%>V(!3UW)9(XWB:S&RI5N8XTN0(GUI85/9L+$6_`9&BBG2$YD.#
M.FII:TNRB[S/#G=(K2%M8TP]*+?JT.MV'L*:X1T>LS>Y?:!1\`W,:#E.(*0I
M=&MN*;S*!,LKJ*W%$V\-+5>@^BCA3M/$CQ*Q5,]3S`)K/Z6UG]+:3^EK\E,2
M$4L->:":@)B+*!)'JD*AF:2=X;1-X:4:/'-+L;!JG&6W<1T=!C)_6M1[R:^B
MU+%2@Y+8]J'!]:/MCJS>(<1-8SJB6M4F1&G@40*D3>C+@7\IL+C3:*FI>OQ`
M]^"MM;0WI+?AXEN38'9%#ND,>*P`5GI]K$!F_%P+\R,0IEZ,CM'EK\(=VI0;PO6F9OYODL\9C
MJNF@ZMC[4C.:W_8C]R)B;I%&CNAH+L[KX:":L8A4B)F$*Y3[0VBH&BWW&E`,
MDF`!`N^5ZKC.`M.N_L:KEIEBQZ'HHZS=OG,/8$D750`#>ZH&VS2$;45T,'U5
MV[ZZ..64&Q5S^$3CY3W92Y/BF[64A9GN2^
M5V\BH2-"%=0D@1I"ZXT2Q28T7V(7!A#E7?5-WBQ59\.M'R`+89OR$E*[J76Z
MZ5(3YWDFSG[0CW0QKZ`GZO8.5T>DZ\V0B'-T3NR2I
MPWA4JV8)HR94:JH=V4]]94%_<(YF:ON"3R'5W46*6.+;V_T9N[)6<=M;RX:G
M0,U,%2Y.-=A5U=_=9#7U>.L(K"5JU#V935I]9DD553M%U`+7T`-%_YKN5,'L
MFT_+G#@VD#.X+J*I8Y_O?1),]U903Q'V8(L"
ML0"9UB,%55]_$((5AP#Z/@A&Q4Z_"D8A>6_'!WPOV!_40&:\1?&GC`M$6(_O
MY>:O^V^&Y+6JB\OSX]-G_?#QVK(MYQ
M(-Y9%?&N`_'NJHCW'(CW5D6\[T"\ORKB`P?B`P]B;0#W#K9K#W?S0_/IP#]9
MT)SN>:<<$STB=J2Q0PT+;'?D\HON3+:@25`['C4PSEU(&PU$'`.Q"B79H'
M]0@L#!#P'H1HG^9!+(;G*`(^4-9D]).8#G#(.?J"(0T^Q[C&PL8YN#2*^J[\
M``GQ#>"#8\Y_E;E51Y';X>R1%^64'>D
MK#QF+4?1=Z$85*$81['G0K%?
MA6*?H]AWH3BH0G'`41P`*#YTE%#1:GD12QAC"N[+,:0$%7;[%#K=#H;JJC@X
M>.37"G=VO``88G^_R;*IGFE(R4M35UHO5U@N5U@M5U@L5U@K5U@JFZV4M%\Z
M<'>Y=HUM+8UMK8QM+8SK=;'INJA(L#87R'_\]_8_ULOC_X7E45LN+6;_L[EJO7^OU:[U^@2C6Z]='6K^4
M?5SP(.AIR]F^WP!IW!TVL__M[WJ42`3SR+I3W@H/>;@7`P+]5!(RR!_O>
M-IC27/5RQ7Z)X:@'NATP6<4\$NU5@>+I-,3#G@"Z3@J$3=)"@P*@(;"-._SN
M)IF2`ZGB-<]Z$WQ#IWSPW7>!E:$X3J+O_GV>.51'#D_ZT*G!DWYU6R"#)O1I
MA?JU.%FK=A/QQV5M3S;6!]9OKPL&'ZL+#"2#6EU2BQJSHHZ'C,%GTF'>W$'=
M[F2RZ%&_D1U+[?1J$<*\B_V]51M-#>Z[V/_CCPJ:U8=[]92G==4<[5J9*GE3
MOVX3;SO\ZP5__14XVS=^!Y"FS.E>JJHIML&M0-S178&+J%*-I$2N&%@V4H<1?
MY1P1E&C4X`^F"'^\:7I5@B*5*@9>&"8?&TMP/^C)W`\=_J]37:6CRK]E]XPJ
MU&YE%-3;-*OCJ+J\OL;+(53>1E,Z/@HYDG@R,)PTX6,.`]9)LKB^F=XD/=B]
M?U_AO^@G7':#HS4.WO$-DF1:)BD5%E%:AG-*)T&.QY#R2Y&)7*H8T7TH@5N2
M/"0F"G&?U,I%'QZ>%B_@%FEC#^:3,HQP557K*.[=OJ]W[0U\]7!IB/#3CA^B
M6X(Y_1I#BT]_B_P1G>Z\%TPAA;)U*443)%%0:E\78*.Y(<$DTJY//JDEW2--
M2B=#XL'\M818W=$W:#18:FESE783<\`VI,%O(?F,!C!CEBM_\-D-<+UW(9#!
MUS@'KO7AY[AN7SV.Z^%I>;@"XW1#<%E;:_\4VYMM3%Q$+@I%_6&]59BKXIJ[
M6W=#@4*:51=IZ`Y`;15'@/\D5Y"B;=*/YEY!0G=?;[\A=&X+_=#>$F%$O4I$
M/8JHIR&JJ8]$VUP7P7\UT4.B;9<.@G*Z@OQ^)?E]2GY?)=\T.:C(:6-I!5[$
M@S=*@\P-M8V14L$:Q8,^0)I5)<@P^.#0OR!Y`$UL\[/O1/M"50;^B"8XV-UV
MG:LZ=JVM157!-`SJGE7>"_K>VU!]"L$CR^[&][ZZV'ID8]^G6R24H>MM!ODI&`OU;6(S!O!5K4>\\I>U]#
M'TYS-'`0B!ZCN>+=.1Y5;C*-K@MY9Z_O:C;N\_Z^MPMIG^-K/X_\761REV30
M"W\Z;V4&OHID9SCZCKFH58,V^-@]*_PIQ1$6Q`D*01JBAO<-0<Y<^
M7:C`[>V8]^,^HQ[:V_E,^P@3MFPO^>KP5T&[<&]'O#^)9CJ[-VS,7T4.XC07
M+8$B#Q41YXNMB76,.,]9S'<5+4'1=96E@\^L'*=:EZUQ(GWP71:2D+:DITQI
MJ8VKM3(0/>1J*=103+Q:KJJQO$%8RW:L;K2E?RPRY=58I,>`2XD5%7"I.BP,
M=I)*\PU>KH_=&XTB..,PA$$*_QTQ2L.#HE%$9C]F[%+%^
M<8SS1WF^U]]%';./:+_09EK]-61>QLK32&CHE#?<@4@M0$Q+K)0\5.JI_K;4
MSL`P(!2PL8?8A)A%:(L:HMBO#]HOS5ID(E$1X6WJGP8L_N[?)Z0,.V:R;F5B
MG$&[^T6LIW_HF'\Y3YFXC4VWL&TH+?OK+\;;GP)YGE#=L`]`PQX\J-FP#P#;
MQ7F:!@F9C10`FS\]THE-(C+#EW!A4B@4SM@\W8KH+//
M'L*L`*-)"`7EMW_R,T!MDM,9H<+R4";F?$#"BB>SF8U2R+%SP*+QX'?K^:]L
M.AYEB[0D7,5=P8@W;'AXE!`Z@/0-TKK@N^!_-S9ZP7_\!QI&?]$_>OR//O]C
MT$7PY*\]GK3#_]@5>?L\Z9'`M"TR>P)K3Z#M#<1?.^R0;%L7`&Q)2S,2.`D_
MU!AL($#4F'Z?,YR/D0<]9=Q@GGPCFLA;@28*'8:$*0]^8@JX\%'"I?0R/3ZY
M.$66T*M]WHB%"IG%Y(\JZYX8O@1:VO:T]*[@`::N"A5%D
M#&W]X;;+/:8;MVG[5&HA2-JMU?HD+BE0I!4(8_"%1(IPE;`/DMV`*,&K`1
M1J.6X#]^#`Z(#=6(=D'YM<7#TDDK+$?)K=SZI73@ZD\W^!/]7P4>3W;O#=IZ
M?>C6J6NP>EW>['Y]4G8^.BG>[$%]2G<_-:7>[)WZ#=G[S!OBS=ZMW\[]+[N=
MWNR]^FPX^*K9X,W>YURRUHP?`B4J,I?A8I4WP94H6G)MP\LN6M;`O1I9;M1U
M%VL*M(!#4>#*@J[1"S4+TJZ)%LGU)4O5-C<1*F*\_#/253"^_R'9WP#[%G./
MH^!169#3*WJ]KKFYP5R@93Q<<'&B16Y0COCY@]4CU#:N"-V_CQ.[U,?0&@N0
MXKO;!;I7IT$JQ,1VS7?L;I!;R#4":A'5,A&T486F\`EC@.XY:=H%)/TZ:?3E
M3^I2:36MGK9'>>I4]7@_^'4\T3$>!8_"".U.=+!+K:M'?C#W$$\%Q`,&XZ0>
MY_MIQQ"""W)R90K<>2W(RX]*G3NO!6UIORYM&2M*;>L*>U\^'.8L><6
MQ=3NB`84-60;)]YH@=W&]U'","J*."_#292@-?9;`/;;+=\)\E:PN[V/9^G+
M\Z/+R]_#IZ].'U\>GYV&(>KE;>7P73#)-L6:=E54/S6(P@XSFM$5>P>970T-
MSBZ$:2`Q\?OGN$;N?D+MR:9+P<-_*%Z4S!U"%``,Y=*.3&Z$&50JCA[$O0,-
M=E85$EUDX/ROXP(?X$(/`'FW[P6AEN&+<%WP]D"HXN8L4!1N,''I@PTIM;LA@
M)"WY;=!J/I*[!FN#2]7'3%C[:!C'MFL?C?8I7?MHM+.=^$+:N?;1J,.&>CX:
M?'W0]B54<`M=B<%TB3R1@@E@L@&U=KY2:+$6(E2HS33H^5&Z9@K\L
M5UM#HV:*K!)Y0E%E*_9`G#$>MP:GA[#=8(^_@U-WQ-V1J,T!T-$QB`8AO8RM
M,93X)6/=HXJ/QK=)'&NSB5"SK.ZU"/%XP+`^Z&FZ[80\%4YX&218T=&I'J*!
MGKCJGR=?+&KY.3#N8^[K;K".*+4:78KM38@[")8>Y!M#LG#*0G@Q;@^Q9Z
MZD-M\^+P>K[)9C&YKH2F98PD[RW:B#X_>W'TK;;14KM'EK!<]GF6(BFJ_6**
MQ6@4%X6YAV9[`548\8]Y9M-7UVE`L_<%:A2.%7+Q.#PY>W9\&IX>OC@*7QS^
MB_LA&15C=W16GD@2P#[`T?3F_"$(GAE0'@9$!$=T2[H)7*\!&$*EQ)_OS00_J4/PI[>QM]W<@
M$G@K_!W,$<$$B5BRJ)?G=ZA[92]_1TINB6JV!$TXSR--O[%6//Y!G<_9XKB@
M@JUY.^[*7!@)?WR.=?JG7):"O@^.'-%)FVCRN8HW[B88C?L"#QZ&CF(?W%;/
MN?-^B_HI4G?^X*?Y'?Z[VN]2KZE5.6S?*]H!2+=-=_33E`R@C=_^[[=#HRW6
MLF(O:7K]4F7EB$U3I+>%MFF)?TP"$JRZW8Y79(V":D.1P':?*^*6G*AG+/I.
M\F5%@Y&&R6\TTD#K&HYJGFX+IGH.N3GW*@ZY):J*LVX)*!HC.JBA'4SB6OG0
MV\#D!6CC"+Q9?14`;9R+MTQ0!4`;9^9_-\45`&T*345X)TL*Z\3DWKA*DA37D"V]_)4@+*\K7SZ)*$/^)S7K'M-XQK7=,
MZQW3>L>TWC%])DSYF#LF96E4W!-`%VEI[+4MMFJ4NJI#;MJ(.!V'_,BN7B2L
MASP2UL,:D;#&W)&!0,M(6%JZ?LRMLGA1Q'DH'2-M`&GO-DSFLEV>(P>!'K6>
ML]7P;5[!>%YM"1=$/N#5VZ;PQBZ42D,\6A%0M5<]TK#Z%20-5*A(=H5-E245
M[ZKJDHVK`J0%E6F).BM!6E"@Y9E2`M*%"?C/)*D!94J<^Y<94@+:A57WC[
M*T%:4+*^?A95@@BE"UK/`/5+*0V6@>YAV2ZBD/X`*6,JO*XW5/@?57AQ5?@=
M45[!'D>5OD9-O8QT;S'(N\CC553/407V)@+]B$2/0,Y$BBL1Z#%D>_8LXR4$
M^P?5\@QR^@39WD!>+R"%K1YNFJ4:NOM8CCYU7'RJG7L`?QFED(.%'YR.F#YR
M/JH?C!$B%Q5Q;$G8E4>.D$.MYE@CJM6N8_&O\9Y"$+5JD#6)R+^'D'"M!UM;
MFXS7)N.UR7AM,EZ;C-
MQZSW,>M]S)>\C^$KHQ8<4`_,L9P+#&!+=MSE=+R6HH5AX[$-Z@978U9@Y9'J
MHA2&6/WIXX(\3.T(Q\9BQJF6:_P`H`GQR#ZX^B>"VI2Q%MA#S6/N>?,=>\IB
M&SAO(-H"TA?*[U&/S+)QC*,O!-N]_6WTD5(;V]L[Y(==+PG?]XZU;6,@#MU(
ME7L[M%)9Q=Y.G4JZ73C:'#ZN0H.@0;`)6=:(;*&U0XWPD<%,:`A]B<_80F-$QF5IB&K`#]/7=(`.P(CC]M"RG>'A/,"S@Q+"2*(9T2Z
M\3QEH'.&N(I".5B8C*\4)A;]<92\33A0W.M)3@H1*7=;8D1Y&:^L$]5U[?
MOZ\-*?A8[OY]Z]T@""./2<2>;%5@[,B:\E!=%](BK)#]HK?H%3-BT%ZWJ\3"
M9&C9$["HD`B[DL?728&P\;A`;CG]J&N>+.NO-6&LWV-1-$>K'CT%E>*)I@T!
M<"2`QB8X2X/`1].LB$UXG@@5(,N3"HP3(,"I!3G503_(%0X_>6L%M((EIA7Y
M!X=J$$M?(-^OK12X?\F.<`%SD+Y$W!MT7<]_BW!;V$.WT)YL94W4^YN-2O[B
MKQQ71N`EP5(Z+_2P2^8"ETW'XMVM>\&C7D7$3J6T*`F]X27!M-X2C[1S#6=+
M(Q9'@*I>^7B7*M&5982Q'2>[(<[#@
MM_/AJ,S&(%"S/MAR7GL[OAFGOS/%LNQO4_.IP&0J3@_DT*E885IBEG/M$2*%
M#:@?V8@ZT-WCH8#-)@_P:.&1R1SQO^P=@AGLQ6P_4T/X9#&T$9[NJF&//WAHX\\*RMYD>?"1I:K55`4P)0H8@4R\6F\KTT`0?/P1[:=Z$\H_
MYV:4;42_@3>B_$-;WF^6WY'R#]J9BEUI;0KJ;ECA*&O@ZL0_W[:-[U=X)Z%)
M46].6-]]35B!(W?Y#0C_7&'2ZLP;'<[]\@'^W',,?Q_`L5MO!RQ%M_.`1>.C
M=Z>LK0.>K;*$$R&P=7'Y,Z
M=UX;YU.?BG)W7ALG4)]CJ]QY;1PO?6DM=N>I$4"`4R./B"5Z&R\%.L.196AE
MF2N1N/+:D+EU:G'GM2%S5Z7`G=>&S/V8U+GSVI"YGXIR=UX;,O=S;)4[KPV9
M^Z6UV)TG9&YO.:^SM3J^5L?7ZOA:'?^"6K56Q_]V=1RM-5OD32"@-F'A6E4$
M&XA\^2V(XD:U^?-;$,NM4N//;T%,_ZW4^O-;$-V?56O\^2V(]"^JM?[\%L3]
M5\4-?[Y8#O1#CF4V#>M3A_4V9[W-66]S/A/U>+W-66]SUJ<.ZU.']:E#NY2O
M3QUDWOK40%@W7Q5U86_9"&OVW`\:3D#>*&:/HG6'Z
MHU$X")HG+@;QBVOT?E#E73#]RE"MVSKP'1VSQY(WOM@1*T2.L&[G.&KF(23<
M==>^H>,:).P^E'X7RB8';R>,\-M4UB'!A`JZQ#!IB'WSADG#BMLU]KV:ZILR
MKCLR>ECQ>GLQ;+G?IC6IR;8MJU>3(ZCT'^9#F)&-NEW337DCR++2[FN@(H8VUCIBT3EG?H'0C,Q
M2AKJ_U8PRJ;3J(S':#V9S:,\%BIR1XWT0F-L=3"A'7-IZ[#EC,5C8;%:J,XR
M[/S9@:]W0QL+OHR:L5UHJGD;VU30;>7*9*?!I/L!ZD7W!H>V471.#61BS/B+
M"A83KF+=*AGAIG;,O@@VHJV`,UA=K8-H""0R;JOSB/Y=]!#7-C?LC$T*W*7X
M@(+]JH)70S[<<25(+^M3#07D]U:0:JRE[6,PG+]T]T[@61)30U/*;#%463(9K?+F/AP%:RBM'S,"-H7C,K8?_X+'P!#[$U7M!K[?[J#+DC#EG4F66B'!XE-IX6NK4DD8S
MA5IHJ:HJ;47M8@32J&H4G[;?]2G2J@K-7A%*]*TN,+$HB0\>*%//'%7TDU?.
M93V4^C@=:R^X-`_$YH2H//T8+Q6IC0Z1BNB^P((Z5J*YM?!,HTM]D-GM/<]8
MIRYO=GM/,K9`BC>[O2<8/SZEWNSVGEO\Y`WQ9K?WK.+GWDYO=GM/)W[A;/!F
MJPQ7X\EB].\R6+K%;W[P=$+9#)]=8LN9Z['R"F:[9WT=+4`L^J)>'D
M<\-4Q7`M6[AQ`/<$HE47+@.1+[^%I:M1;?[\%E:O5JGQY[>P@OVMU/KS6UC&
M/JO6^/-;6,R^J-;Z\UM8T[XJ;OCSY;._;!71UC8JW#FX`)*KVP?5`0#V]7;M/\41$/\*P?#\GEDRS;[N!T==AB/[,DZM%&9/)>X?PHY$X3:(B
MV/C6+/ZM[H^B&I.T`W\Q@-"O>%1F^7O7B7_5V#*L3Q29,NZ8EXGXO;')*D#3
M3RV[A7/Y#`&LL,Q*6)2(P3/;U86==V/X:9*^U4U$>AXV`+TK#9.0,)I]`+#1
M_`*HE5G3)MDB'?,)-(O+B/^-GSDF8PHGDK>3G5,9"B7.#8NT-'_CP(ALO=/]
MBSD*]*#HXT:,VQTAA2C5=*K3T4K]2JP:51N6>>*/J"2/1]32X?&DINHY_0L(
MRA^&>72'CT)NU%%*P>F1AYW1[9IV.>24'NW3<)3/YMDP`3NR99
MP?_)0)-L$/@]S6LYFG-,U>[F'+*NP96/28=')4/7AJ.YALH/T9+3>:,:JR!:
M_PS;5071DH/[%]GR*HB67.&_4MY40?@MO'*A
M4+8-PM`+X5Y?7EU?7EU?7K7RUI=7UY=75[R\NMY8K#<6GUQI66\LOJQVK3<6
MZXW%E[BQ\`7_Y";'59<1'8\GNX4%I$%=WNP6UHWV2/%FM[!6_&V4>K-;6!P^
MEX9XLUM8"KZ0=GJS6Y#Z7P<;O-E"NJO'2?]GPGJNMV+KK=@G5_/66[$OJUWK
MK=AZ*_8E;L5\9MGU&<_ZC,=)W?J,9WW&LS[C69_QK#<6ZXW%YZ!JK3<6ZXW%
M>F/Q^6PLE#,>X=ZOF!.[:MP1?A^G(O[@/P,@^*"D`X<<5,$QN6JP0`UR;X=>
M,]CN=F302N!N1<7E"NUB"HGR0]O*XZ5H=YK^[*CQ,<4=%.MV"G2%@(=*,H#U
MR%`$Y8.?S`L)/")*3P,F,5#4`IZ@@M=9F>&(5JB3PSC/L]Q1*VDVO4T&`*`Y
MZ`YJ*SBLQ2!D]VRVU>`I]+]:V%8U7HMRI:@BFB7[]&&%[X2-<8QWYSK)S^YP_#6*UTNNJ:4S
M$1AYPPI^MM\E`2I9(#34D]NNALH66\-I3T'28TC`F-'69\7?=+/#D$@'4B+Q
MCX4G`H(OVW%`]>M]2F0I[9.A3<%<*H#PV$%\WD14@TVN._DP"PV472W&+)HS
M8SH?Z?QQX<'2GL&BRI\46W(5
M53_U%N#&QKC+>>$L@*AT+\#RWA]ESIW7
M@C'ADU'NSFO!4/!9MLJ=U\+V_XMKL3M/;.>!1WH!N4K_I;MY]]K#-WMNS9>B
MI%L_LA%TKIQL/VS$,5:_^_=E:`;[^V"EZBG.]QFX^FT%9&!6@"X8HM0?2,#>
MSK:\WZ_<5%<8`TQ=I*XYP*UM--8NJL,95"H82\O^&IOA,L7ZV`+KZEM
M5+Z<<)_K$=6OJ]F/J:DKN_4R6N/5G82.&S(M1K6)$X#A?N#!H%1?/`U![ON_M"/OU+]4.JIHY1].L
MB#4[)V^U9N\.!:!EY)0'4XX3`ZQZHG91/C#+O*KZ_C,88"LVSE9'``Z91]O#
M(Z!Y.>-O=)TF*PVN;&YE4RE'F)$<&KQVAZE:L.M!#B,DI%*D:^K[RIR0720C
.#:+Q]O\!-6U&P&H/`P!E
`
end
From alan@linuxcare.com.au Fri, 10 Nov 2000 11:36:56 +1100 (EST)
Date: Fri, 10 Nov 2000 11:36:56 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: abort in eliminate_regs compiling glob.c from glibc
On Thu, 9 Nov 2000, John David Anglin wrote:
> > Breakpoint 2, eliminate_regs_in_insn (insn=0x406a0ba0, replace=0)
> > at ../../gcc/reload1.c:2826
> > 2826 if (! insn_is_asm && icode < 0)
> > (gdb) p debug_rtx (insn)
> > (insn/s 2711 2709 2719 (set (reg:SI 6 %r6)
> > (reg:SI 28 %r28)) 69 {pre_ldw-4} (insn_list 2708 (insn_list:REG_DEP_ANTI 2696 (insn_list:REG_DEP_ANTI 2702 (insn_list:REG_DEP_ANTI 2697 (nil)))))
> > (expr_list:REG_DEAD (reg:SI 28 %r28)
> > (insn_list:REG_RETVAL 2708 (expr_list:REG_EQUAL (expr_list (use (mem:BLK (scratch) 0))
> > (expr_list (symbol_ref/v:SI ("@strlen"))
> > (expr_list (reg/v:SI 4 %r4)
> > (nil))))
> > (nil)))))
>
> The `use' arises from the `__pure__' attribute in the prototype for strlen:
>
> extern size_t strlen (__const char *__s) __attribute__ ((__pure__));
I don't see this problem using current puffin CVS hppa64-linux gcc. Was
this with your REG_ELIMINATE patch?
Alan Modra
--
Linuxcare. Support for the Revolution.
From dave@hiauly1.hia.nrc.ca Thu, 9 Nov 2000 21:50:49 -0500 (EST)
Date: Thu, 9 Nov 2000 21:50:49 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: abort in eliminate_regs compiling glob.c from glibc
> > > 2826 if (! insn_is_asm && icode < 0)
> > > (gdb) p debug_rtx (insn)
> > > (insn/s 2711 2709 2719 (set (reg:SI 6 %r6)
> > > (reg:SI 28 %r28)) 69 {pre_ldw-4} (insn_list 2708 (insn_list:REG_DEP_ANTI 2696 (insn_list:REG_DEP_ANTI 2702 (insn_list:REG_DEP_ANTI 2697 (nil)))))
> > > (expr_list:REG_DEAD (reg:SI 28 %r28)
> > > (insn_list:REG_RETVAL 2708 (expr_list:REG_EQUAL (expr_list (use (mem:BLK (scratch) 0))
> > > (expr_list (symbol_ref/v:SI ("@strlen"))
> > > (expr_list (reg/v:SI 4 %r4)
> > > (nil))))
> > > (nil)))))
> >
> > The `use' arises from the `__pure__' attribute in the prototype for strlen:
> >
> > extern size_t strlen (__const char *__s) __attribute__ ((__pure__));
>
> I don't see this problem using current puffin CVS hppa64-linux gcc. Was
> this with your REG_ELIMINATE patch?
No. Well actually I saw it first with the patch. I see this with the
32 bit compiler. The only elimination with the 32 bit compiler is the
default frame pointer elimination.
I just tried the hppa64-linux-gcc compiler with the source that I posted
and it didn't abort. There were lots of warnings about int to pointer
conversions though.
Make sure you compile with -O2 or -O3? Register elimination only occurs
at -O2 or above. I see the problem both with a i686-linux cross compiler
and a fairly recent native hpux compiler under hpux 10.20. The problem is
not present in 2.95.2 but it doesn't support the pure atribute.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From matthew@wil.cx Fri, 10 Nov 2000 09:49:07 +0000
Date: Fri, 10 Nov 2000 09:49:07 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: linux bame
On Thu, Nov 09, 2000 at 03:53:48PM -0700, Paul Bame wrote:
> Somebody never imported 2.4.0-test6, then I imported -test10 on the main
> vendor branch and now can't (easily) undo that to import test6 and THEN
> test10. This workaround sucks.
don't use vendor branches. didn't you talk to mang about this?
--
Revolutions do not require corporate support.
From matthew@wil.cx Fri, 10 Nov 2000 10:18:08 +0000
Date: Fri, 10 Nov 2000 10:18:08 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: [parisc-linux] tulip DMA mapping
On Thu, Nov 09, 2000 at 12:12:25PM -0800, Grant Grundler wrote:
> 0 is a valid pci_map_single() return value when the system has an IO MMU.
Oh dear. You can bet tulip won't be the only driver which assumes it
isn't a valid return value. Can't our IOMMU code be limited in such a
way that 0 is not a valid return value? Say, constrain all allocated
addresses to the top half of the device bus?
(um, just check me on this, map_single returns a device bus address,
not a processor bus address, right?)
> The system will panic before pci_map_single() will fail.
> The driver needs to remember some other way if a buffer was mapped or not.
> Or the Documentation/DMA-mapping.txt should be changed - ie add this
> to the interface definition and I can reserve the 1st mapping
> entry so no-one uses it.
we should probably have a BAD_DMA_ADDR define which that array should be
initialised to. it's a little late in 2.4 to go through and audit all
the drivers again.
> Should I be mailing Jeff Garzik directly?
> Or can someone who knows Jeff point this out to him?
i've cc'd jeff & dave miller on this.
--
Revolutions do not require corporate support.
From davem@redhat.com Fri, 10 Nov 2000 02:16:11 -0800
Date: Fri, 10 Nov 2000 02:16:11 -0800
From: David S. Miller davem@redhat.com
Subject: [parisc-linux] tulip DMA mapping
Date: Fri, 10 Nov 2000 10:18:08 +0000
From: Matthew Wilcox
> Should I be mailing Jeff Garzik directly?
> Or can someone who knows Jeff point this out to him?
i've cc'd jeff & dave miller on this.
In 2.4.x there is _NO_ error return from the PCI dma functions except
the consistent DMA mapping ones.
This was an explicit design decision, the dynamic mapping functions
should never fail, and if they do it is a hard error.
Therefore no drivers need to check for failure, as far as they are
concerned, there is no failure.
So what is the issue? In 2.5.x I'll add an error return facility
(BTW: -1 ie. 0xfffffff would probably work as an error value on all
platforms :-)
Later,
David S. Miller
davem@redhat.com
From rhirst@linuxcare.com Fri, 10 Nov 2000 10:55:16 +0000
Date: Fri, 10 Nov 2000 10:55:16 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] parisc-linux reaches uptimes in excess of one day!
Just for interest... My A180 has been up for 25 hours, the first
12 of which I had a couple of telnet sessions open running vi, gcc,
etc. and since then it has been looping doing kernel builds.
I also ran cvs to update its kernel source tree from pehc.
So far it has completed about 23 kernel builds with one hiccup:
do_page_fault() pid=2420 command='cpp0'
YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000001100000000000001011
r0-3 00000000 4013ac38 400e134f 00000004
r4-7 40138c38 ffffffb5 2002014b 00000001
r8-11 00000004 00000000 00000000 4001ada0
r12-15 4001ad7c 0001b300 2001f601 00000001
r16-19 00012000 00000023 00000020 40138c38
r20-23 20020100 4012a51c 00000000 9999999a
r24-27 2002016c 2002014c 00000004 00013d34
r28-31 00000000 4013a438 200201c0 40041757
sr0-4 00000000 00000011 00000000 00000011
sr4-8 00000011 00000011 00000011 00000011
IASQ: 00000011 00000011 IAOQ: 400e13b7 400e13bb
IIR: 0c601094 ISR: 00000011 IOR: 00000004
ORIG_R28: 40019450
Richard
From rhirst@linuxcare.com Fri, 10 Nov 2000 11:12:20 +0000
Date: Fri, 10 Nov 2000 11:12:20 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] tulip DMA mapping
I've quoted the whole of Grants message below, so you can see the
context. It looks like tulip is treating zero as meaning it
doesn't have anything to pci_unmap...
Grant Grundler wrote:
> Hi all,
> I see a "bug" in tulip's usage of mapping services.
> It's not the bug I was looking for unfortunately.
>
> In line 217 of drivers/net/tulip/interrupt.c:
>
> if (tp->tx_buffers[entry].mapping)
> pci_unmap_single(tp->pdev,
> tp->tx_buffers[entry].mapping,
> sizeof(tp->setup_frame),
> PCI_DMA_TODEVICE);
>
> 0 is a valid pci_map_single() return value when the system has an IO MMU.
>
> The system will panic before pci_map_single() will fail.
> The driver needs to remember some other way if a buffer was mapped or not.
> Or the Documentation/DMA-mapping.txt should be changed - ie add this
> to the interface definition and I can reserve the 1st mapping
> entry so no-one uses it.
Richard
On Fri, Nov 10, 2000 at 02:16:11AM -0800, David S. Miller wrote:
> Date: Fri, 10 Nov 2000 10:18:08 +0000
> From: Matthew Wilcox
>
> > Should I be mailing Jeff Garzik directly?
> > Or can someone who knows Jeff point this out to him?
>
> i've cc'd jeff & dave miller on this.
>
> In 2.4.x there is _NO_ error return from the PCI dma functions except
> the consistent DMA mapping ones.
>
> This was an explicit design decision, the dynamic mapping functions
> should never fail, and if they do it is a hard error.
>
> Therefore no drivers need to check for failure, as far as they are
> concerned, there is no failure.
>
> So what is the issue? In 2.5.x I'll add an error return facility
> (BTW: -1 ie. 0xfffffff would probably work as an error value on all
> platforms :-)
>
> Later,
> David S. Miller
> davem@redhat.com
>
> ---------------------------------------------------------------------------
> To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com with
> `unsubscribe' as the subject.
>
>
From davem@redhat.com Fri, 10 Nov 2000 03:26:48 -0800
Date: Fri, 10 Nov 2000 03:26:48 -0800
From: David S. Miller davem@redhat.com
Subject: [parisc-linux] tulip DMA mapping
Date: Fri, 10 Nov 2000 11:12:20 +0000
From: Richard Hirst
I've quoted the whole of Grants message below, so you can see the
context. It looks like tulip is treating zero as meaning it
doesn't have anything to pci_unmap...
Thank you for the clarification.
Jeff, this is in fact a bug, please fix :-)
Later,
David S. Miller
davem@redhat.com
From jgarzik@mandrakesoft.com Fri, 10 Nov 2000 09:30:41 -0500
Date: Fri, 10 Nov 2000 09:30:41 -0500
From: Jeff Garzik jgarzik@mandrakesoft.com
Subject: [parisc-linux] tulip DMA mapping
"David S. Miller" wrote:
>
> Date: Fri, 10 Nov 2000 11:12:20 +0000
> From: Richard Hirst
>
> I've quoted the whole of Grants message below, so you can see the
> context. It looks like tulip is treating zero as meaning it
> doesn't have anything to pci_unmap...
>
> Thank you for the clarification.
>
> Jeff, this is in fact a bug, please fix :-)
np. Like Matthew(?) hinted, this is not the only place that needs
fixing. I'll take care of it.
Jeff
--
Jeff Garzik |
Building 1024 | Would you like a Twinkie?
MandrakeSoft |
From bame@riverrock.org Fri, 10 Nov 2000 08:57:47 -0700
Date: Fri, 10 Nov 2000 08:57:47 -0700
From: bame@riverrock.org bame@riverrock.org
Subject: linux bame
= On Thu, Nov 09, 2000 at 03:53:48PM -0700, Paul Bame wrote:
= > Somebody never imported 2.4.0-test6, then I imported -test10 on the mai
n
= > vendor branch and now can't (easily) undo that to import test6 and THEN
= > test10. This workaround sucks.
=
= don't use vendor branches. didn't you talk to mang about this?
Um, I have no information to go on from your note. All the (successful)
merges I've done before have used the cookbook CVS merge method including
a vendor branch. Several (N-1?) of the palinux merges have been
accompanied by updating the vendor branch. And this merge is going
well despite the ugly workaround, or so it appears to me. Just
importing files to a vendor branch should have no effect on anything
else unless CVS has some horrible bug (RCS does not). Before I make
what is apparently a serious mistake ("don't use vendor branches" sounds
pretty serious) please enlighten me!
-P
From grundler@cup.hp.com Fri, 10 Nov 2000 08:29:25 -0800
Date: Fri, 10 Nov 2000 08:29:25 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] tulip DMA mapping
Mathew,
I think the situation is clarified. This is just FYI.
Matthew Wilcox wrote:
> On Thu, Nov 09, 2000 at 12:12:25PM -0800, Grant Grundler wrote:
> > 0 is a valid pci_map_single() return value when the system has an IO MMU.
>
> Oh dear. You can bet tulip won't be the only driver which assumes it
> isn't a valid return value.
Well, then:
1) the driver writer made a wrong assumption that the "spec"
does not support.
2) that wasn't the case here - 0 was used as a flag to indicate
a mapping had (or hadn't rather) been done - not that the
mapping call had failed.
> Can't our IOMMU code be limited in such a
> way that 0 is not a valid return value? Say, constrain all allocated
> addresses to the top half of the device bus?
It could pretty easily by reserving the dma_map[0] entry during
init time. Dave Miller already made it clear that's not desirable.
> (um, just check me on this, map_single returns a device bus address,
> not a processor bus address, right?)
Yes. It's the address a device must use to master DMA transactions.
pci_map_single() input parameters are "struct pci_dev *", virtual host
memory address, and buffer size. The return is a device specific I/O
DMA address - ie this mapping cannot be shared with other devices.
FWIW, "I/O DMA address" are called IOVAs in HPUX (I/O Virtual Address).
The HW guys prefer another name but this one stuck.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From bame@riverrock.org Fri, 10 Nov 2000 14:28:33 -0700
Date: Fri, 10 Nov 2000 14:28:33 -0700
From: bame@riverrock.org bame@riverrock.org
Subject: [parisc-linux] kernel merge
=
= We're getting ready to do a kernel merge (to -test10 I think). Anybody
= concerns before we get started?
=
I'm getting ready to commit the changes to merge us up to 2.4.0-test10 as
soon as I'm confident 64-bit kernels are OK (32-bit seems fine).
Here's a brief list of changes made (and yet to be made -- if anyone
has the time/energy) to our parisc code (does not include changes in
common code!). While there's plenty yet to do, I think we're
no worse off than before the merge. Some parts of our parisc code
are getting rather moldy compared to the other ports FYI.
Important tags:
LINUS_240_TEST6 Linus' 2.4.0-test6
LINUS_240_TEST10 Linus' 2.4.0-test10
LINUS_240_TEST10_PREMERGE Our tree before the -test10 merge
LINUS_240_TEST10_MERGED Our tree after the -test10 merge
------------
Lots of 'extern __inline__' turned into 'static __inline__' and there
are more to do (TODO).
Beginnings of several smp_mb*() macros -- implemented as no-ops in
the non-SMP case (asm/bitops.h, asm/system.h)
SET_PERSONALITY() macro in asm/elf.h needs updating (TODO).
fcntl.h acquired a 'struct fcntl64' used with 64-bit offsets, implementing
fcntl(fd, F_S/GETLK[W]64, ...). Several other locking-related #defines added.
asm/mmu_context.h:init_new_context() now returns int (always 0), not void.
Should our asm/bitops.h routines be re-coded in assembly? (TODO)
Added #define RLIMIT_LOCKS to asm/resource.h
Added #define CLOCKS_PER_SEC to asm/param.h (how did we miss this one?)
Our asm/string.h is behind the times. Fixing it might get rid of a
bunch of compiler warnings! (TODO)
Removed mktime from drivers/char/genrtc.c (it's in a header file now)
x86 made a bunch of changes to asm-i386/floppy.h -- should we? (TODO)
looks like maybe the get/put_user_ret() functions in asm/uaccess.h are
obsolete? (TODO)
We'll need to add the getdents64() and fcntl64() syscall glue. (TODO!!!)
Our arch/parisc/config.in is in BAD SHAPE (TODO)
arch/parisc/process.c: added new argsto do_fork() and copy_thread().
IA64 seems to use the copy_thread() arg.
arch/parisc/signal.c: minor change to common data structure
drivers/scsi sym53c8xx_comm.h #ifdef doesn't work on parisc -- eliminates
unmap_pci_mem() causing link error (TODO - rhirst)
From pjlahaie@linuxcare.com Fri, 10 Nov 2000 17:14:06 -0500
Date: Fri, 10 Nov 2000 17:14:06 -0500
From: Paul J.Y. Lahaie pjlahaie@linuxcare.com
Subject: [parisc-linux] Beta CD
--uXxzq0nDebZQVNAZ
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Hello fellow PA-RISCers,
An preliminary beta CD for PA/Linux has been uploaded to
puffin.external.hp.com. If people could try it and forward any complaints
or suggestions to me, it would be greatly appreciated. The URL for the
image is: ftp://puffin.external.hp.com/pub/parisc/cd-images/palinux-0.5.iso.gz
- Paul
--uXxzq0nDebZQVNAZ
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE6DHMu8ggPQthPCzcRAgb0AJ4jJs5VOnm+9aDXUbtwht7hxAa+UwCgihAo
nEE25pYQwBwCDuALcSdyCNw=
=bTuw
-----END PGP SIGNATURE-----
--uXxzq0nDebZQVNAZ--
From randolph@tausq.org Fri, 10 Nov 2000 19:18:47 -0700
Date: Fri, 10 Nov 2000 19:18:47 -0700
From: Randolph Chung randolph@tausq.org
Subject: [parisc-linux] kernel merge
> Our asm/string.h is behind the times. Fixing it might get rid of a
> bunch of compiler warnings! (TODO)
if this is just an issue of implementing the various parisc-optimized string
routines, i've been working on this, but don't have everything ready
yet.
randolph
--
@..@ http://www.TauSq.org/
(----)
( >__< )
^^ ~~ ^^
From crypt@ihug.co.nz Sun, 12 Nov 2000 01:47:56 +1300
Date: Sun, 12 Nov 2000 01:47:56 +1300
From: crypt@ihug.co.nz crypt@ihug.co.nz
Subject: [parisc-linux] HP 9000 e25
Considering how often this question comes up is there anyone out there
that is interested in tracking down the info to port linux to this class
of box.
Server boxes may be classed as boring to some but there seem to be a
large number of people who have at least one of them sitting about doing
nothing.
Joe.
--
=======================================================================
in real life: Joseph Skinner |There's no such thing as a wizard
email: crypt@ihug.co.nz |who minds his own business
Analyst/Programmer | - Berengis the Black
| Court Mage to the Earls Caeline
========================================================================
From snaketails@optushome.com.au Sat, 11 Nov 2000 23:57:43 +1100
Date: Sat, 11 Nov 2000 23:57:43 +1100
From: Rod Smart snaketails@optushome.com.au
Subject: [parisc-linux] PA-RISC newbie
Hi there.
I just purchased at a **VERY** cheap price, a HP Visualize C-180
RISC system from a CAD drafting company, this box included a 4Gb Hard
disk, a video card with unknown amount of memory (& daughter board,
assuming its memory upgrade) and system board RAM is fully loaded
(768Megs).
I would like to run this system on Linux, and found the PA-RISC web
site from the www.linux.org web site :o)
The system currently has HP-UX installed, but not having any
usernames or passwords, I cannot access the system from the prompt.
I have read how to NFS-boot the HPbox from the message board, but I
couldn't figgure out where to find the kernel sources (or how I would be
able to compile them on the HPbox anyway).
I downloaded the vmlinux kernel file from the ftp site and the
NFS-ROOT and installed both on my current Linux box (Pentium 133
pre-MMX) in the format of how it was layed out in ...
LINUX/PA-RISC NFSROOT HOWTO
In there it states that you have to get the latest linux-2.3 tree
from CVS, but which CVS server I'm curious about as the ones on
ftp.kernel.org don't have "parisc" in the /arch/ branch, so where do I
find the CVS "tree" that I need to use?
I seem to have the "STEP 2." section running, as from playing with
the system I had found my way into the PDC prompt and have played with
the "boot" function. I tried to boot the installation CD (on a drive I
have borrowed from work) of HP-UX and reinstall so I have access, as I
don't really care whats currently on the system..
I'm not confident enough on how to create and burn a bootable CD to
try and boot the box that way, I still think the vmlinux kernel I
downloaded from the PA-RISC website has something to do with my problems
(or the permissions)
I do know that the Hp is talking to the Linux box (apart from the
lack of logged info) I get a report in /var/log/secure and messages that
the HP box has attempted a bootp session, but the HP box reports
something along the lines that the booting code was not found.
So, if anyone has any ideas on how to help me, I would be
appreciated ;o)
Oh, what *is* the RAM modules in the Hp systems anyway (apart from
the obvious 64Mx72-pin SIMM)
Thank you for your time and help :o)
PS. I would prefer replies to my E-mail address as I generally
forget to surf the net reading mailing lists....
From andrew@neep.com.au Sat, 11 Nov 2000 23:12:55 +0800
Date: Sat, 11 Nov 2000 23:12:55 +0800
From: Andrew Shugg andrew@neep.com.au
Subject: [parisc-linux] HP 9000 e25
> Considering how often this question comes up is there anyone out there
> that is interested in tracking down the info to port linux to this class
> of box.
>
> Server boxes may be classed as boring to some but there seem to be a
> large number of people who have at least one of them sitting about doing
> nothing.
>
> Joe.
I don't perceive the problem to be a lack of willingness, or interest, but
has already been stated the older systems contain a lot of proprietry stuff
that isn't sufficiently documented for people to work on.
Maybe as the parisc port grows more popular and attracts more resources,
some enthusiastic people will get it happening. =)
Andrew.
--
Andrew Shugg http://www.neep.com.au/
"Just remember, Mr Fawlty, there's always someone worse off than yourself."
"Is there? Well I'd like to meet him. I could do with a good laugh."
From grundler@cup.hp.com Sat, 11 Nov 2000 15:29:14 -0800
Date: Sat, 11 Nov 2000 15:29:14 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] HP 9000 e25
Andrew Shugg wrote:
> > Considering how often this question comes up is there anyone out there
> > that is interested in tracking down the info to port linux to this class
> > of box.
> >
> > Server boxes may be classed as boring to some but there seem to be a
> > large number of people who have at least one of them sitting about doing
> > nothing.
> >
> > Joe.
>
> I don't perceive the problem to be a lack of willingness, or interest, but
> has already been stated the older systems contain a lot of proprietry stuff
> that isn't sufficiently documented for people to work on.
AFIAK, the chipsets for I/O devices in E25 have plenty of documentation.
The issue is someone has to clean them up and get a lawyer to approve
their publication. It's all about IP and avoiding lawsuits. This has
been discussed before. Any HP employees interested in doing this "on
their own time" can contact me and I'll help locate unpublished docs.
Note that PDC is the same for Nova-Class (EFGHI-class) boxes as for
the workstations for the most part. So someone could start by just
trying to boot those boxes and see how far it gets before dying.
> Maybe as the parisc port grows more popular and attracts more resources,
> some enthusiastic people will get it happening. =)
Having worked on the HPUX SCSI driver (scsi3 and scsi1) for E25 and
similar boxes, I question anyone's sanity who volunteers to write
drivers for SPIFI chips - even with full documentation. I would rather
give folks that interested in contributing a 715/33! (or my gosh /50's!)
Yes - I know folks who collect PDP's and keep them running in their
garage...'nuf said.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From grundler@cup.hp.com Sat, 11 Nov 2000 15:38:58 -0800
Date: Sat, 11 Nov 2000 15:38:58 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] PA-RISC newbie
Rod Smart wrote:
> Hi there.
>
> I just purchased at a **VERY** cheap price, a HP Visualize C-180
> RISC system from a CAD drafting company, this box included a 4Gb Hard
> disk, a video card with unknown amount of memory (& daughter board,
> assuming its memory upgrade) and system board RAM is fully loaded
> (768Megs).
>
> I would like to run this system on Linux, and found the PA-RISC web
> site from the www.linux.org web site :o)
>
> The system currently has HP-UX installed, but not having any
> usernames or passwords, I cannot access the system from the prompt.
Yes you can. You just need to know how.
Interrupt the boot process to get a "BOOT ADMIN" prompt (or whatever
it's called - Boot Console Handler).
Then "boot primary isl" (shortened to bo pri isl)
You should have an "ISL>" prompt now.
ISL> hpux -is
And you will boot to single user state with a shell.
mountall and you should have the regular tools too.
vi /etc/passwd to your hearts content.
"There is no such thing at security with out physical security".
..
> I'm not confident enough on how to create and burn a bootable CD to
> try and boot the box that way, I still think the vmlinux kernel I
> downloaded from the PA-RISC website has something to do with my problems
> (or the permissions)
>
You can dd the ISO image to a regular hard disk (ie /dev/sdc) and move
that disk over to the C180. Then boot from that disk.
From BCH, use "sea" to locate all boot devices.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From randolph@tausq.org Sat, 11 Nov 2000 17:04:29 -0700
Date: Sat, 11 Nov 2000 17:04:29 -0700
From: Randolph Chung randolph@tausq.org
Subject: [parisc-linux] glibc question
I'm working with bdale and taggart to try to get the Debian glibc
package to compile under hppa. One of the things I saw today while
playing with this is that the build dies because it tries to link in
bsd-setjmp.c and bsd-_setjmp.c, which both define setjmp. Can someone
enlighten me about why this may be happening?
This is built from the glibc-2.2 sources, which I understand has
all/most of the changes in pehc cvs.
randolph
--
@..@ http://www.TauSq.org/
(----)
( >__< )
^^ ~~ ^^
From rajak@purdue.edu Sat, 11 Nov 2000 22:58:57 -0500 (EST)
Date: Sat, 11 Nov 2000 22:58:57 -0500 (EST)
From: Brian Poole rajak@purdue.edu
Subject: [parisc-linux] Problems booting new beta CD
Hello,
My machine is a 715/80 with 88M of RAM, 1G HD, external 4x CD drive &
floppy.
Trying to get the new palinux CD working on my 715/80 here and not having
too much success. Did get the CD to actually boot, but after all the
normal Linux init messages (which were very nice to see btw,
congratulations on how far it has already gotten ;) it stops at 'Switching
from PDC console'.
Looking thru the FAQ I saw a somewhat similar question, in that it had
'pdc' in it & was booting from the CD (but the 0.1 CD, not the one I am
using), that said the kernel on the CD booted from/to the
serial console. Is this also true of the 0.5 CD? If so, what is necessary
to boot one of these from console? I've been fooling with it to no
success.. I have it automatically booting from the CD, but it doesn't
actually boot. I replug in the monitor & keyboard and find that it can't
boot without the keyboard ('Failed to initiliaze a keyboard\n SYSTEM
HALTED').. How am I supposed to boot to the console if I can't boot w/o a
keyboard? I imagine if it has a keyboard it will boot to the screen like a
Sun.
Thanks for the help,
-b
From grundler@cup.hp.com Sat, 11 Nov 2000 23:31:37 -0800
Date: Sat, 11 Nov 2000 23:31:37 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] Problems booting new beta CD
Brian Poole wrote:
...
> Looking thru the FAQ
Brian,
Thank you for first reading the FAQ!
> I saw a somewhat similar question, in that it had
> 'pdc' in it & was booting from the CD (but the 0.1 CD, not the one I am
> using), that said the kernel on the CD booted from/to the
> serial console. Is this also true of the 0.5 CD?
I think it is.
Is there a way via PALO to direct the console to FBCON or STICON or
whatever it's called now?
I had the impression *eventually* the boot console would be whatever PDC
says it is - ie graphics console for the 712.
> If so, what is necessary to boot one of these from console?
kernel with CONFIG_STI_CONSOLE I think....but that's not enabled by default.
It may not be possible to use the graphics console unless one builds
a "custom" kernel.
> I've been fooling with it to no
> success.. I have it automatically booting from the CD, but it doesn't
> actually boot. I replug in the monitor & keyboard and find that it can't
> boot without the keyboard ('Failed to initiliaze a keyboard\n SYSTEM
> HALTED').. How am I supposed to boot to the console if I can't boot w/o a
> keyboard? I imagine if it has a keyboard it will boot to the screen like a
> Sun.
AFAIK, 712's are the only parisc workstations what won't boot w/o keyboard.
All others will auto-switch to use serial console if the keyboard
is not connected.
Both or the above questions sounds like a FAQ.
Could someone who knows more update/add those to the FAQ?
thanks,
grant
ps. The most up-to-date FAQ is at parisc-linux.org/faq.html even
though that's not officially online yet.
>
> Thanks for the help,
>
> -b
>
>
>
> ---------------------------------------------------------------------------
> To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com with
> `unsubscribe' as the subject.
>
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From alan@linuxcare.com.au Sun, 12 Nov 2000 23:08:35 +1100 (EST)
Date: Sun, 12 Nov 2000 23:08:35 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] glibc question
On Sat, 11 Nov 2000, Randolph Chung wrote:
> I'm working with bdale and taggart to try to get the Debian glibc
> package to compile under hppa. One of the things I saw today while
> playing with this is that the build dies because it tries to link in
> bsd-setjmp.c and bsd-_setjmp.c, which both define setjmp. Can someone
> enlighten me about why this may be happening?
I think bsd-_setjmp.c should define _setjmp, not setjmp.
Alan Modra
--
Linuxcare. Support for the Revolution.
From andrew@neep.com.au Mon, 13 Nov 2000 00:40:15 +0800
Date: Mon, 13 Nov 2000 00:40:15 +0800
From: Andrew Shugg andrew@neep.com.au
Subject: [parisc-linux] HP 9000 e25
Grant Grundler said:
> AFIAK, the chipsets for I/O devices in E25 have plenty of
> documentation. The issue is someone has to clean them up and get a
> lawyer to approve their publication.
My bad, I should've said "public documentation". =)
*prods LC FAQ maintainers*
Andrew.
--
Andrew Shugg http://www.neep.com.au/
"Just remember, Mr Fawlty, there's always someone worse off than yourself."
"Is there? Well I'd like to meet him. I could do with a good laugh."
From randolph@tausq.org Sun, 12 Nov 2000 11:06:38 -0700
Date: Sun, 12 Nov 2000 11:06:38 -0700
From: Randolph Chung randolph@tausq.org
Subject: [parisc-linux] glibc build fails / bash bug
Bdale, taggart and I have been looking at trying to build glibc on hppa
from Debian's sources. What we saw was that it looks like a lot of the
syscalls were not being reocognized as such by one part of the build, so
it tries to build things from the sysdeps/generic directory and fails.
After a lot of digging, I *think* what is at fault is actually bash. It
looks like during the build, a shell script (make-syscalls.sh) parses
through syscalls.list to generate syscall stubs that are needed for the
build to happen correctly, but these are not being generated. What it
boils down to, I think, is this:
(on hppa - bdale's J5K)
=============================
tausq@j5k:/space/tausq $ bash --version
GNU bash, version 2.04.0(1)-release (hppa-unknown-linux-gnu)
Copyright 1999 Free Software Foundation, Inc.
tausq@j5k:/space/tausq $ dpkg -l |grep bash
ii bash 2.04-7 The GNU Bourne Again SHell
tausq@j5k:/space/tausq $ cat test.sh
#!/bin/sh
echo "1 2 3 4 5
a b c d e
" |
while read a b c d e; do
echo $a $b $c $d $e
done
tausq@j5k:/space/tausq $ /bin/bash test.sh
1 2 3 4 5
=============================
(on other archs, tested with i386 and sparc)
samwise[11:06] ~% bash --version
GNU bash, version 2.04.0(1)-release (i386-pc-linux-gnu)
Copyright 1999 Free Software Foundation, Inc.
samwise[11:06] ~% dpkg -l |grep bash
ii bash 2.04-7 The GNU Bourne Again SHell
samwise[11:06] ~% /bin/bash test.sh
1 2 3 4 5
a b c d e
This causes the parsing routines to die quite miserably....
Anyone feel like trying to fix this? :-)
randolph
--
@..@ http://www.TauSq.org/
(----)
( >__< )
^^ ~~ ^^
From pdbeal@louisville.edu Sun, 12 Nov 2000 14:27:28 -0500
Date: Sun, 12 Nov 2000 14:27:28 -0500
From: Phillip D. Beal pdbeal@louisville.edu
Subject: [parisc-linux] Beta CD
On Fri, Nov 10, 2000 at 05:14:06PM -0500, Paul J.Y. Lahaie wrote:
> Hello fellow PA-RISCers,
>
> An preliminary beta CD for PA/Linux has been uploaded to
> puffin.external.hp.com. If people could try it and forward any complaints
> or suggestions to me, it would be greatly appreciated. The URL for the
> image is: ftp://puffin.external.hp.com/pub/parisc/cd-images/palinux-0.5.iso.gz
>
> - Paul
The kernel dumps on an HP755. Actually, how do you make these CD's? I
know how to use mkisofs to crerat a filesystem CD, but how you make it
bootable with a kernel image?
Thanks,
--
Phillip Beal
Electrical and Computer Engineering
S+LUG Vice-President
From bame@riverrock.org Sun, 12 Nov 2000 13:27:32 -0700
Date: Sun, 12 Nov 2000 13:27:32 -0700
From: bame@riverrock.org bame@riverrock.org
Subject: [parisc-linux] glibc build fails / bash bug
= After a lot of digging, I *think* what is at fault is actually bash.
Bash has a bunch of 'set' options, some shell variables, and probably
some compile-time configure options, which affect its behavior, so I'd
compare all those. One possibility is that the bash configure script on
hppa configures bash to be more like Posix, since that's what hp-ux shell
users expect.
The construct in question:
echo "a b c
1 2 3" | while read x1 x2 x3
*depends* on the way echo breaks or doesn't break lines, and the way
read parses them. Often scripts like that also depend on whether
the shell actually makes a new subprocess for the 'while' or it doesn't,
because that determines whether variables set in the loop will still be
set on exit from the loop. While I didn't find a way to make the
construction fail, a safer method (when you get a choice) is to use 'set':
set -- a b c \
1 2 3
while [ $# != 0 ]
do
echo $1 $2 $3
shift 3
done
-Paul "too much shell programming" Bame
From bame@riverrock.org Sun, 12 Nov 2000 17:25:45 -0700
Date: Sun, 12 Nov 2000 17:25:45 -0700
From: bame@riverrock.org bame@riverrock.org
Subject: [parisc-linux] kernel merge
Helge noticed SCSI fails (B160L, 712, I think it's ok on c3k)
after the -test10 merge. We have MANY differences from Linus in the SCSI
area. Anybody want to take a look at this? FYI to get
the pre-merge kernel you can use LINUS_240_TEST10_PREMERGE. Beware
that tags are sticky in CVS (use cvs update -A to fix that).
-P
From catfish@alltel.net Sun, 12 Nov 2000 19:05:21 -0600
Date: Sun, 12 Nov 2000 19:05:21 -0600
From: catfish@alltel.net catfish@alltel.net
Subject: [parisc-linux] Project Dead????
Hello,
Its been so long since I've had an email I was curious if this has
become a Dead project?
Just curious.
Terry
--
catfish: icq #20116127
From bame@riverrock.org Sun, 12 Nov 2000 18:15:07 -0700
Date: Sun, 12 Nov 2000 18:15:07 -0700
From: bame@riverrock.org bame@riverrock.org
Subject: [parisc-linux] Project Dead????
= Hello,
= Its been so long since I've had an email I was curious if this has
= become a Dead project?
You might want to check the e-mail archives to see what you've been
missing. I don't know why you haven't been receiving anything. The
project is quite alive.
http://www.parisc-linux.org/lists.html
-P
From bame@riverrock.org Sun, 12 Nov 2000 19:20:50 -0700
Date: Sun, 12 Nov 2000 19:20:50 -0700
From: bame@riverrock.org bame@riverrock.org
Subject: [parisc-linux] kernel merge
=
= Helge noticed SCSI fails (B160L, 712, I think it's ok on c3k)
= after the -test10 merge.
Fixed.
From test6 to test10 the SCSI host registration method changed.
Updated sim700.c, lasi7xx.h, zalon7xx.h
-P
From grundler@cup.hp.com Sun, 12 Nov 2000 21:34:08 -0800
Date: Sun, 12 Nov 2000 21:34:08 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] Problems booting new beta CD
Brian Poole wrote:
> Quoting Grant Grundler (grundler@cup.hp.com) from 11 November 2000:
> > Is there a way via PALO to direct the console to FBCON or STICON or
> > whatever it's called now?
> >
> > I had the impression *eventually* the boot console would be whatever PDC
> > says it is - ie graphics console for the 712.
>
> Working on a 715 here.. notice you refer to 712s throughout, hopefully we are
> on the same page. Maybe 712s are identical to 715s, I don't have any idea,
> just noticed the irregularity.
For the most part yes. I had just assumed you were using a 712.
> I meant to say 'from serial console' or something of the sort..
Serial console should just work.
Shouldn't need an HIL keyboard connected though.
>
> > > I've been fooling with it to no
> > > success.. I have it automatically booting from the CD, but it doesn't
> > > actually boot. I replug in the monitor & keyboard and find that it can't
> > > boot without the keyboard ('Failed to initiliaze a keyboard\n SYSTEM
> > > HALTED').. How am I supposed to boot to the console if I can't boot w/o a
> > > keyboard? I imagine if it has a keyboard it will boot to the screen like
> a
> > > Sun.
> >
> > AFAIK, 712's are the only parisc workstations what won't boot w/o keyboard.
> > All others will auto-switch to use serial console if the keyboard
> > is not connected
>
> Hmmm.. so I have manually switch it to use serial console?
No. Just disconnect the keyboard before powering on the machine.
Should end up with console on the serial interface.
You may want to set the keyboard/console path to the serial port.
This can be done from "BOOT_ADMIN>" prompt.
The default kernel has "CONFIG_HIL=y". But since I don't test on
boxes with HIL interfaces, I have no idea if a keyboard is required
since the driver is enabled. It sounds like a bug if the 715
can't boot from serial console w/o HIL keyboard.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From dhazeghi@pacbell.net Sun, 12 Nov 2000 21:34:19 -0800
Date: Sun, 12 Nov 2000 21:34:19 -0800
From: dhazeghi@pacbell.net dhazeghi@pacbell.net
Subject: [parisc-linux] [Semi OT] SOM Linker
Hello,
I have been watching this project for some time and wanted to thank you
guys for all the great work so far. The recent announcement of a BETA CD
is highly encouraging.
However I would like to know what work if any has been done on the SOM
linker which HP released to the public last November(?). It seems that
as of right now, it has not been touched since February 14, and the FSF
binutils snapshots still do not have any SOM support for ld. Has there
been any movement in merging this in, or is anybody working on this?
Thanks.
Dara Hazeghi
From deller@gmx.de Mon, 13 Nov 2000 08:32:54 +0100
Date: Mon, 13 Nov 2000 08:32:54 +0100
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] kernel merge
On Monday 13 November 2000 03:20, bame@riverrock.org wrote:
> =
> = Helge noticed SCSI fails (B160L, 712, I think it's ok on c3k)
> = after the -test10 merge.
>
> Fixed.
>
> >From test6 to test10 the SCSI host registration method changed.
> Updated sim700.c, lasi7xx.h, zalon7xx.h
>
> -P
Thanks Paul,
sim700 (53c710) now works again, but the second built-in controller (ncr/sym
53c8xx) still doesn't.
Greetings,
Helge
From rhirst@linuxcare.com Mon, 13 Nov 2000 10:24:03 +0000
Date: Mon, 13 Nov 2000 10:24:03 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] BEWARE: Makefile changed from parisc to parisc64
Confused me for a while, anyway. If you cvs update your kernel
source and expect to build a 32 bit kernel, you need to edit the
top level Makefile to change the arch := line.
Richard
From rhirst@linuxcare.com Mon, 13 Nov 2000 12:13:49 +0000
Date: Mon, 13 Nov 2000 12:13:49 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] kernel merge
On Fri, Nov 10, 2000 at 02:28:33PM -0700, bame@riverrock.org wrote:
> drivers/scsi sym53c8xx_comm.h #ifdef doesn't work on parisc -- eliminates
> unmap_pci_mem() causing link error (TODO - rhirst)
Fixed. This relates to NVRAM that some PCI scsi cards have to hold
config settings over reboot. CONFIG_SCSI_NCR53C8XX_NVRAM_DETECT
is normally defined in sym53c8xx_defs.h to turn that code on. When
I implemented Zalon (FWD) support I guessed that the 53c720 h/w
wouldn't have NVRAM implemented the same way, and turned off
NVRAM detect.
I've also replaced a chunk of Zalon specific code that got lost in
the merge, so Zalon/FWD/53c720 support works again.
There is a problem remaining when using the driver as a module;
it looks like something is trying to printk() from a string in
the module after the module has been removed. Havn't tracked
that down yet.
Richard
From adevries@linuxcare.com Mon, 13 Nov 2000 13:50:24 -0500
Date: Mon, 13 Nov 2000 13:50:24 -0500
From: Alex deVries adevries@linuxcare.com
Subject: [parisc-linux] [Semi OT] SOM Linker
dhazeghi@pacbell.net wrote:
> However I would like to know what work if any has been done on the SOM
> linker which HP released to the public last November(?). It seems that
> as of right now, it has not been touched since February 14, and the FSF
> binutils snapshots still do not have any SOM support for ld. Has there
> been any movement in merging this in, or is anybody working on this?
The initial plan was to do our 32-bit userspace with SOM, worrying about
ELF32 much later in the game. But ELF32 development happened a lot
quicker than expected, and so nobody's really done much on the SOM
linker.
I suspect it'd be very hard to use the SOM linker code to incorporate it
into binutils, but I could be wrong.
What are you actually trying to do?
- Alex
--
Alex deVries, Principal Solutions Architect, The Puffins at Linuxcare
613.562.2759 tel
alex@linuxcare.com, http://www.linuxcare.com/
Linuxcare, Support for the revolution.
From deller@gmx.de Mon, 13 Nov 2000 23:54:29 +0100
Date: Mon, 13 Nov 2000 23:54:29 +0100
From: Helge Deller deller@gmx.de
Subject: sym53c8xx-driver (was: Re: [parisc-linux] kernel merge)
On Monday 13 November 2000 13:13, Richard Hirst wrote:
> On Fri, Nov 10, 2000 at 02:28:33PM -0700, bame@riverrock.org wrote:
> > drivers/scsi sym53c8xx_comm.h #ifdef doesn't work on parisc -- eliminates
> > unmap_pci_mem() causing link error (TODO - rhirst)
>
> Fixed. This relates to NVRAM that some PCI scsi cards have to hold
> config settings over reboot. CONFIG_SCSI_NCR53C8XX_NVRAM_DETECT
> is normally defined in sym53c8xx_defs.h to turn that code on. When
> I implemented Zalon (FWD) support I guessed that the 53c720 h/w
> wouldn't have NVRAM implemented the same way, and turned off
> NVRAM detect.
>
> I've also replaced a chunk of Zalon specific code that got lost in
> the merge, so Zalon/FWD/53c720 support works again.
>
> There is a problem remaining when using the driver as a module;
> it looks like something is trying to printk() from a string in
> the module after the module has been removed. Havn't tracked
> that down yet.
>
> Richard
Hi Richard,
I'm sorry to say that the sym53c8xx still doesn't work on my B160L (32
bit-Kernel). It looks like the controller isn't detected at all, while it
worked perfectly with 2.4.0-test6.
Would you mind to take a look at the code again ?
Thanks,
Helge.
From rhirst@linuxcare.com Mon, 13 Nov 2000 23:27:52 +0000
Date: Mon, 13 Nov 2000 23:27:52 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: sym53c8xx-driver (was: Re: [parisc-linux] kernel merge)
Hi Helge,
The problem I fixed related to the ncr53c8xx driver (which shares
code with sym53c8xx), and was to make 53c720 support work again.
sym53c8xx worked for me on my A180. Please can you try booting
with
sym53c8xx=verb:7,debug:0xffff
and send me the output?
Thanks,
Richard
On Mon, Nov 13, 2000 at 11:54:29PM +0100, Helge Deller wrote:
> On Monday 13 November 2000 13:13, Richard Hirst wrote:
> > On Fri, Nov 10, 2000 at 02:28:33PM -0700, bame@riverrock.org wrote:
> > > drivers/scsi sym53c8xx_comm.h #ifdef doesn't work on parisc -- eliminates
> > > unmap_pci_mem() causing link error (TODO - rhirst)
> >
> > Fixed. This relates to NVRAM that some PCI scsi cards have to hold
> > config settings over reboot. CONFIG_SCSI_NCR53C8XX_NVRAM_DETECT
> > is normally defined in sym53c8xx_defs.h to turn that code on. When
> > I implemented Zalon (FWD) support I guessed that the 53c720 h/w
> > wouldn't have NVRAM implemented the same way, and turned off
> > NVRAM detect.
> >
> > I've also replaced a chunk of Zalon specific code that got lost in
> > the merge, so Zalon/FWD/53c720 support works again.
> >
> > There is a problem remaining when using the driver as a module;
> > it looks like something is trying to printk() from a string in
> > the module after the module has been removed. Havn't tracked
> > that down yet.
> >
> > Richard
>
>
> Hi Richard,
>
> I'm sorry to say that the sym53c8xx still doesn't work on my B160L (32
> bit-Kernel). It looks like the controller isn't detected at all, while it
> worked perfectly with 2.4.0-test6.
> Would you mind to take a look at the code again ?
>
> Thanks,
> Helge.
>
From deller@gmx.de Tue, 14 Nov 2000 01:13:58 +0100
Date: Tue, 14 Nov 2000 01:13:58 +0100
From: Helge Deller deller@gmx.de
Subject: sym53c8xx-driver (was: Re: [parisc-linux] kernel merge)
--------------Boundary-00=_ABNZX5KGKOOFAA0B1BIM
Content-Type: text/plain;
charset="us-ascii"
Content-Transfer-Encoding: 8bit
Subject:
On Tuesday 14 November 2000 00:27, Richard Hirst wrote:
> Hi Helge,
> The problem I fixed related to the ncr53c8xx driver (which shares
> code with sym53c8xx), and was to make 53c720 support work again.
> sym53c8xx worked for me on my A180. Please can you try booting
> with
>
> sym53c8xx=verb:7,debug:0xffff
>
> and send me the output?
>
> Thanks,
> Richard
Hi Richard,
the output and the relevant part of .config is attached.
Greetings,
Helge
--------------Boundary-00=_ABNZX5KGKOOFAA0B1BIM
Content-Type: text/plain;
name="log1"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="log1"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--------------Boundary-00=_ABNZX5KGKOOFAA0B1BIM--
From rhirst@linuxcare.com Tue, 14 Nov 2000 10:17:04 +0000
Date: Tue, 14 Nov 2000 10:17:04 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: sym53c8xx-driver (was: Re: [parisc-linux] kernel merge)
Hi Helge,
On Tue, Nov 14, 2000 at 01:13:58AM +0100, Helge Deller wrote:
> On Tuesday 14 November 2000 00:27, Richard Hirst wrote:
> > Hi Helge,
> > The problem I fixed related to the ncr53c8xx driver (which shares
> > code with sym53c8xx), and was to make 53c720 support work again.
> > sym53c8xx worked for me on my A180. Please can you try booting
> > with
> >
> > sym53c8xx=verb:7,debug:0xffff
> >
> > and send me the output?
> >
> > Thanks,
> > Richard
>
>
> Hi Richard,
>
> the output and the relevant part of .config is attached.
>
> Greetings,
>
> Helge
Thanks. I agree it doesn't look like the driver is even seeing the
chip; I wonder if PCI support is broken...
> dino_bridge_init: IO_ADDR_EN hasn't been configured.
> kernel BUG at dino.c:646!
Does it usually say that on bootup?
Richard
From matthew@wil.cx Tue, 14 Nov 2000 10:29:36 +0000
Date: Tue, 14 Nov 2000 10:29:36 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: [parisc-linux] kernel merge
On Fri, Nov 10, 2000 at 02:28:33PM -0700, bame@riverrock.org wrote:
> fcntl.h acquired a 'struct fcntl64' used with 64-bit offsets, implementing
> fcntl(fd, F_S/GETLK[W]64, ...). Several other locking-related #defines added.
>
> We'll need to add the getdents64() and fcntl64() syscall glue. (TODO!!!)
actually, we don't. Linux/PA-RISC has sufficiently wide data types
already so we don't have the grot required in other ports to support
the appropriate wide data types.
> looks like maybe the get/put_user_ret() functions in asm/uaccess.h are
> obsolete? (TODO)
yes, they are. exterminate! exterminate!
--
Revolutions do not require corporate support.
From deller@gmx.de Tue, 14 Nov 2000 14:11:42 +0100 (MET)
Date: Tue, 14 Nov 2000 14:11:42 +0100 (MET)
From: Helge Deller deller@gmx.de
Subject: sym53c8xx-driver (was: Re: [parisc-linux] kernel merge)
> Hi Helge,
>
> On Tue, Nov 14, 2000 at 01:13:58AM +0100, Helge Deller wrote:
> > On Tuesday 14 November 2000 00:27, Richard Hirst wrote:
> > > Hi Helge,
> > > The problem I fixed related to the ncr53c8xx driver (which shares
> > > code with sym53c8xx), and was to make 53c720 support work again.
> > > sym53c8xx worked for me on my A180. Please can you try booting
> > > with
> > >
> > > sym53c8xx=verb:7,debug:0xffff
> > >
> > > and send me the output?
> > >
> > > Thanks,
> > > Richard
> >
> >
> > Hi Richard,
> >
> > the output and the relevant part of .config is attached.
> >
> > Greetings,
> >
> > Helge
>
> Thanks. I agree it doesn't look like the driver is even seeing the
> chip; I wonder if PCI support is broken...
>
> > dino_bridge_init: IO_ADDR_EN hasn't been configured.
> > kernel BUG at dino.c:646!
>
> Does it usually say that on bootup?
Yep.
Has always been there, but nevertheless the scsi-driver worked before.
FYI: The non-pci sim700-driver also didn't showed up before pb fixed it in
CVS with a few one-line-patches.
NB: Could maybe someone (ggg?) explain me the kernel-bug mentioned above ?
>
> Richard
>
Greetings,
Helge
--
Sent through GMX FreeMail - http://www.gmx.net
From grundler@cup.hp.com Tue, 14 Nov 2000 08:10:42 -0800
Date: Tue, 14 Nov 2000 08:10:42 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] dino maintainer?
Anyone interested in maintaining Dino?
I don't have time for it and all the docs are on the web.
One of the "TODO" things is to convert dino.c to use struct
pci_hba the same way lba_pci.c does....
Richard Hirst wrote:
...
> Thanks. I agree it doesn't look like the driver is even seeing the
> chip; I wonder if PCI support is broken...
>
> > dino_bridge_init: IO_ADDR_EN hasn't been configured.
> > kernel BUG at dino.c:646!
>
> Does it usually say that on bootup?
per Helge's request:
The bug is normal for card-mode Dino - not for Built-in Dino.
I think Helge has the GSC 100BT card which is a card-mode Dino on-board
with one (or two) Tulip(s) behind it.
The warning is a reminder one can NOT use MMIO accesses to those
PCI devices and *only* I/O Port space (eg inb/outb).
If someone wants to fix the warning so it's quiet for card-mode
devices...see is_card_dino(d) in dino_driver_callback() for an
example.
FYI - card-mode dino was used for several different networking
interfaces but not SCSI interfaces.
hope this helps,
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From bame@noam.fc.hp.com Tue, 14 Nov 2000 09:35:01 -0700
Date: Tue, 14 Nov 2000 09:35:01 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
I've attached an overview of differences between our CVS linux sources
following the -test10 merge and the upstream -test10 sources. This
document is also at http://puffin.external.hp.com/~bame/diff.html
The raw 'diff' output (now a bit out of date) is at
http://puffin.external.hp.com/~bame/diff.out
If anyone gets bored, this list is full of small (and not so small)
tasks which range from very simple (drivers/block/rd.c) to fairly
complex (scripts/*).
-P
palinux Vs. linux 2.4.0-test10
Here's a brief summary of the differences in common code between
palinux (tag: LINUS_240_TEST10_MERGED) and the upstream -test10
sources. The full diff output is in diff.out. NOTE this does not
include machine-depend differences
* Minor changes in several locations to support GSC
* Minor top-level Makefile hacks, though the CFLAGS one is quite
important.
* Lots of RCS $Revision$ differences in ACPI (a different 'cvs
import' would've eliminated these differences).
* drivers/block/rd.c: obsolete debug code for parisc64. Changed a
constant from 0xffffffffL to 0xffffffffUL because of a parisc64
gcc bug initializing longs. The repaired code is probably "more
correct" anyway.
* drivers/char/Config.in: changes to support LASI, parisc real-time
clock (CONFIG_GENRTC)
* drivers/char/Makefile: Config-related changes to support HP
keyboards and RTC
* drivers/char/console.c: looks like dead or dying experimental
parisc code -- probably should be removed. Also some
parisc-specific comments and format changes which should
disappear.
* drivers/char/serial.c: support for GSC and A500 serial
* drivers/net/Makefile,Space.c: mostly LASI LAN support
* drivers/net/eepro100.c: no clue about this one
* drivers/net/tulip/interrupt.c: workaround for a B180+busy-lan boot
problem -- probably should be sent upstream.
* drivers/net/tulip/tulip_core.c: required #ifdef for hppa, also
printk() changes which appear valid
* drivers/parport/Makefile: GSC
* drivers/parport/parport_gsc.c: New file for palinux -- GSC
parallel ports -- required
* drivers/pci/pci.c: eh? Grant?
* drivers/pci/setup-bus.c: function definition tweek -- Grant?
* drivers/scsi: Lots of changes here -- rhirst? See for yourself.
Basics: support LASI and Zalon scsi, changes to 53c8xx drivers,
rename sim7x0 to sim710
* drivers/sound: support for HP "Harmony" sound
* drivers/video: STI and HP FB video drivers (iodccon is probably
worthless)
* fs: add support for SOM executables
* fs/binfmt_elf.c,exec.c: changes for stack-grows-up?
* fs/nfs/read.c: probably unnecessary hack for broken parisc64 gcc
* fs/proc/array.c: ?? something with signals ??
* fs/stat.c: added __hppa__ to several #ifdefs
* include/linux/binfmts.h,fs.h,kernel.h,tty.h,udf_fs_sb.h:
unnecessary differences?
* include/linux/init.h: we use different section names -- why????,
probably some unnecessary other differences too
* include/linux/mm.h: VM_STACK_FLAGS difference -- jsm?
* include/linux/wait.h: parisc debugging -- should be removed
* init/main.c: KWDB and GSC support plus a bunch of stuff which
should probably go away.
* kernel/Makefile,dma.c,fork.c,printk.c: eh?
* kernel/module.c: possible parisc-needed changes
* kernel/signal.c: unknown signal-related differences
* lib/inflate.c: changed some constants to work around parisc64 gcc
bug
* mm/mprotect.c: ?
* scripts/*: MANY differences here. Looks like a combination of
things we hacked to fix configuration problems plus MAYBE not
updating these files from new Linux versions in the past. 'make
menuconfig' is significantly different from upstream. Even the
mkdep.c program is different.
From bame@noam.fc.hp.com Tue, 14 Nov 2000 10:02:06 -0700
Date: Tue, 14 Nov 2000 10:02:06 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] kernel merge
= On Fri, Nov 10, 2000 at 02:28:33PM -0700, bame@riverrock.org wrote:
= > fcntl.h acquired a 'struct fcntl64' used with 64-bit offsets, implementing
= > fcntl(fd, F_S/GETLK[W]64, ...). Several other locking-related #defines add
ed.
= >
= > We'll need to add the getdents64() and fcntl64() syscall glue. (TODO!!!)
=
= actually, we don't. Linux/PA-RISC has sufficiently wide data types
= already so we don't have the grot required in other ports to support
= the appropriate wide data types.
Ok that sounds great but I need a clue how to see it. struct flock
contains an off_t which is 32 bits on narrow (wide too at the moment,
but that will probably change). Also, struct dirent contains
a 32-bit inode and a __kernel_off_t offset (d_off), which is also
32 bits narrow. I did see the ... in our fcntl() syscall definition
so I'll go check glibc to see what's happening there.
I wouldn't be so picky except that I need to write
the 32/64 syscall translators for these soon! Thanks!
-P
From bame@noam.fc.hp.com Tue, 14 Nov 2000 10:14:51 -0700
Date: Tue, 14 Nov 2000 10:14:51 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] kernel merge
= On Fri, Nov 10, 2000 at 02:28:33PM -0700, bame@riverrock.org wrote:
= > fcntl.h acquired a 'struct fcntl64' used with 64-bit offsets, implementing
= > fcntl(fd, F_S/GETLK[W]64, ...). Several other locking-related #defines add
ed.
= >
= > We'll need to add the getdents64() and fcntl64() syscall glue. (TODO!!!)
=
= actually, we don't. Linux/PA-RISC has sufficiently wide data types
= already so we don't have the grot required in other ports to support
= the appropriate wide data types.
Oh, and won't we have to support these syscalls anyway, because user programs
will make them? I suppose we could #define them in libc headers.
= > looks like maybe the get/put_user_ret() functions in asm/uaccess.h are
= > obsolete? (TODO)
=
= yes, they are. exterminate! exterminate!
Done
From pdbeal@louisville.edu Tue, 14 Nov 2000 13:40:24 -0500
Date: Tue, 14 Nov 2000 13:40:24 -0500
From: Phillip D. Beal pdbeal@louisville.edu
Subject: [parisc-linux] 735/755 and Kernel test10..
Hey,
I've been working on the 735 and 755 that I has access to and so far
the systems booted the test6 kernel, but scrambled the console as soon
as init was run. So, I figured I'd try the new test10 kernel that you
have added. Both system boot, and then stop at this line:
branching to kernel entry point 0x00100000
Can't select default wide mode, PDC_PSW call does not work
What does the above actually mean? How can I remove the PDC_PSW call
from the kernel so it can boot? I have plans to test this new kernel
image on a 715 later today.
Thanks,
--
Phillip Beal
Electrical and Computer Engineering
S+LUG Vice-President
From grundler@cup.hp.com Tue, 14 Nov 2000 10:51:26 -0800
Date: Tue, 14 Nov 2000 10:51:26 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] 735/755 and Kernel test10..
"Phillip D. Beal" wrote:
> Hey,
>
> I've been working on the 735 and 755 that I has access to and so far
> the systems booted the test6 kernel, but scrambled the console as soon
> as init was run. So, I figured I'd try the new test10 kernel that you
> have added. Both system boot, and then stop at this line:
>
> branching to kernel entry point 0x00100000
> Can't select default wide mode, PDC_PSW call does not work
Did you build this kernel yourself?
If so, it sounds like you built a 64-bit kernel since that's the default.
You need to change the ARCH line in the linux/Makefile to read "parisc"
instead of "parisc64".
grant
>
> What does the above actually mean? How can I remove the PDC_PSW call
> from the kernel so it can boot? I have plans to test this new kernel
> image on a 715 later today.
>
> Thanks,
> --
> Phillip Beal
> Electrical and Computer Engineering
> S+LUG Vice-President
>
> ---------------------------------------------------------------------------
> To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com with
> `unsubscribe' as the subject.
>
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From grundler@cup.hp.com Tue, 14 Nov 2000 11:08:20 -0800
Date: Tue, 14 Nov 2000 11:08:20 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] 735/755 and Kernel test10..
Grant Grundler wrote:
> If so, it sounds like you built a 64-bit kernel since that's the default.
Correction. *was* the default.
default ARCH was changed last night to parisc.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From ian.zink@maryville.com Tue, 14 Nov 2000 13:20:05 -0600
Date: Tue, 14 Nov 2000 13:20:05 -0600
From: Ian Zink ian.zink@maryville.com
Subject: [parisc-linux] Palinux on a 712/60
I have a hp 712/60 I was trying to get loaded with Pa-linux. I'm using the
0.5 version of the Pa-linux cd. If I boot right off the CD it all loads
until it gets to Switching to PDC. At that point nothing happens. From I
have read the list, that is because the kernel is switching the text
console. However, the 712s don't have consoles. From what I have also read
the CD should work if you pass the kernel the parameter console=tty. So I
tried to "boot scsi.2.0 isl" from the boot_admin prompt. It gave me the
PALO ISL, but I could not choose which line I wanted to edit. Further, I
couldn't even type "b" to boot. I don't know if the isl is freezing or what
is taking place
What I am wondering is there a way to boot a 712/80 without having
to get cross-compiler gcc, compile the kernel, etc. Is there someway I could
add the ramdisk-sti.tgz on the ISO to make it work? If so, what would need
to be done?
Thanks,
Ian Zink, Systems Engineer
Maryville Technologies
540 Maryville Centre, Suite 300
St. Louis, MO 63141
636/519-4182
(FAX) 636/519-4141
ian.zink@maryville.com
www.maryville.com
From mang@subcarrier.org Tue, 14 Nov 2000 14:17:35 -0500
Date: Tue, 14 Nov 2000 14:17:35 -0500
From: Michael Ang mang@subcarrier.org
Subject: linux bame
bame@riverrock.org wrote:
>
> = On Thu, Nov 09, 2000 at 03:53:48PM -0700, Paul Bame wrote:
> = > Somebody never imported 2.4.0-test6, then I imported -test10 on the mai
> n
> = > vendor branch and now can't (easily) undo that to import test6 and THEN
> = > test10. This workaround sucks.
If the sources on the linus branch have been religiously tagged every
time they're updated, then reverting to a previous would have been
relatively painless. I'm not sure what "this workaround" was, but I
guess at this point test10 is merged so the point is moot.
> = don't use vendor branches. didn't you talk to mang about this?
>
> Um, I have no information to go on from your note. All the (successful)
> merges I've done before have used the cookbook CVS merge method including
> a vendor branch. Several (N-1?) of the palinux merges have been
> accompanied by updating the vendor branch. And this merge is going
> well despite the ugly workaround, or so it appears to me. Just
> importing files to a vendor branch should have no effect on anything
> else unless CVS has some horrible bug (RCS does not). Before I make
> what is apparently a serious mistake ("don't use vendor branches" sounds
> pretty serious) please enlighten me!
Vendor branches are evil. (When I say "vendor branch" I mean the
special kind of branch created by "cvs import".) When you check in to a
vendor branch your changes will also be seen on the trunk, *unless* that
file has been previously modified on the trunk. This is almost never
what you want and adds confusion during merging (when you least want
it). Tracking third-party sources using a normal branch, as we are
doing, is much simpler and gives you more control.
When I did the original import of Linus' sources I converted the vendor
branch to a normal branch using cvs admin magic. So none of the
annoyances of vendor branches should affect us, as long as any new files
are added on the linus branch using "cvs add", NOT "cvs import".
When you say you "I imported -test10 on the main vendor branch" I hope
you really mean that you used "cvs add" on the linus branch. From your
other messages, your tags looked good.
- Mike.
From bame@noam.fc.hp.com Tue, 14 Nov 2000 13:00:41 -0700
Date: Tue, 14 Nov 2000 13:00:41 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: linux bame
= bame@riverrock.org wrote:
= >
= > = On Thu, Nov 09, 2000 at 03:53:48PM -0700, Paul Bame wrote:
= > = > Somebody never imported 2.4.0-test6, then I imported -test10 on the
mai
= > n
= > = > vendor branch and now can't (easily) undo that to import test6 and
THEN
= > = > test10. This workaround sucks.
=
= If the sources on the linus branch have been religiously tagged every
= time they're updated, then reverting to a previous would have been
= relatively painless. I'm not sure what "this workaround" was, but I
= guess at this point test10 is merged so the point is moot.
Like the comment said, there was no copy of plain -test6 in CVS (that I
saw). Without -test6 in CVS it's much harder to use cvs diff to figure
out the right way to merge files when there are conflicts.
I didn't realize this until -test10 was already there, so I *then*
brought in -test6. They're in the wrong order on the 1.1.1 branch, so
the standard merge command 'cvs -jlinus:yesterday -jlinus:'
won't work next time -- explicit names will be required.
= Vendor branches are evil. (When I say "vendor branch" I mean the
= special kind of branch created by "cvs import".) When you check in to a
= vendor branch your changes will also be seen on the trunk, *unless* that
= file has been previously modified on the trunk.
Thanks for clarifying what "evil" means! That is pretty ugly indeed!
= This is almost never
= what you want and adds confusion during merging (when you least want
= it). Tracking third-party sources using a normal branch, as we are
= doing, is much simpler and gives you more control.
But I've seen no cook book for it. I'm guessing that instead of cvs import
you use:
cvs co -rlinus linux
cd linux
cvs commit (make note of new files from commit)
cvs add
cvs commit
cvs tag LINUS_NEW_REVISION
perhaps with provision for removing obsolete files too. I suppose that is
simpler than a single cvs import command from a certain perspective :-)
= When I did the original import of Linus' sources I converted the vendor
= branch to a normal branch using cvs admin magic. So none of the
= annoyances of vendor branches should affect us, as long as any new files
= are added on the linus branch using "cvs add", NOT "cvs import".
Have you a pointer to the magic or the knowledge to recreate it? I
had no idea there was a special RCS marking for the evil type of branch.
= When you say you "I imported -test10 on the main vendor branch" I hope
= you really mean that you used "cvs add" on the linus branch. From your
= other messages, your tags looked good.
I used "cvs import", and either your branch magic worked, or I finished the
merge before anybody randomly updated from cvs. Since import used 1.1.1,
which is the branch you "fixed", it seems possible that 'cvs import' might
be rendered harmless but I don't know that for sure.
-P
From pjlahaie@linuxcare.com Tue, 14 Nov 2000 15:43:00 -0500
Date: Tue, 14 Nov 2000 15:43:00 -0500
From: Paul J.Y. Lahaie pjlahaie@linuxcare.com
Subject: [parisc-linux] Beta CD
--SkvwRMAIpAhPCcCJ
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
This is to answer all the questions about sti console. Currently, in the
CVS tree, serial console and STI console cannot be turned on at the same time.
This means that a choice between serial/sti needs to be done at compile
time. At the time we cut the beta, it was decided that serial console was
more important than sti console. I am also working on resolving the problem
with STI/serial console and once I have a fix ready, I will make a kernel
image available. The current beta CD is also expecting a console on
ttyS0 and does not currently open a ttyx getty. When I have a kernel
that can decide the console at runtime, I will look into the beta CD
issues.
If you have any more questions or suggestions, feel free to email me or
post on this list. Thank you.
- Paul
--SkvwRMAIpAhPCcCJ
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE6EaPR8ggPQthPCzcRAhT8AKC8EU8yusyoEvPHxKAQsaM0vMGthwCgnbbC
Yz6ZWjq3q9B80bI+YRxc8xo=
=HhRZ
-----END PGP SIGNATURE-----
--SkvwRMAIpAhPCcCJ--
From raj@cerias.purdue.edu Tue, 14 Nov 2000 16:18:06 -0500
Date: Tue, 14 Nov 2000 16:18:06 -0500
From: Brian Poole raj@cerias.purdue.edu
Subject: [parisc-linux] Beta CD
Sounds like a plan to me. I don't suppose you know how to make a 715
boot to console? Pulling the keyboard and monitor cables off the back
just make it stop at boot with the unable to initiliaze keyboard error
I posted with earlier. I am assuming there is some sort of boot_admin
trickery necessary, but I am unaware of what it might entail and my poking
around has yielded nothing..
Any advice here would be much appreciated.
-b
Quoting Paul J.Y. Lahaie (pjlahaie@linuxcare.com) from 14 November 2000:
> with STI/serial console and once I have a fix ready, I will make a kernel
> image available. The current beta CD is also expecting a console on
..
> If you have any more questions or suggestions, feel free to email me or
> post on this list. Thank you.
From dave@hiauly1.hia.nrc.ca Tue, 14 Nov 2000 16:40:52 -0500 (EST)
Date: Tue, 14 Nov 2000 16:40:52 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: abort in eliminate_regs compiling glob.c from glibc
> > Breakpoint 2, eliminate_regs_in_insn (insn=0x406a0ba0, replace=0)
> > at ../../gcc/reload1.c:2826
> > 2826 if (! insn_is_asm && icode < 0)
> > (gdb) p debug_rtx (insn)
> > (insn/s 2711 2709 2719 (set (reg:SI 6 %r6)
> > (reg:SI 28 %r28)) 69 {pre_ldw-4} (insn_list 2708 (insn_list:REG_DEP_ANTI 2696 (insn_list:REG_DEP_ANTI 2702 (insn_list:REG_DEP_ANTI 2697 (nil)))))
> > (expr_list:REG_DEAD (reg:SI 28 %r28)
> > (insn_list:REG_RETVAL 2708 (expr_list:REG_EQUAL (expr_list (use (mem:BLK (scratch) 0))
> > (expr_list (symbol_ref/v:SI ("@strlen"))
> > (expr_list (reg/v:SI 4 %r4)
> > (nil))))
> > (nil)))))
>
> The `use' arises from the `__pure__' attribute in the prototype for strlen:
>
> extern size_t strlen (__const char *__s) __attribute__ ((__pure__));
Here is a patch to fix the abort in eliminate_regs when it encounters a USE.
As I understand the situation, there are three conditions needed to trigger
it:
1) A function that contains insns with an eliminable register.
2) The function must call __builtin_alloca to change the frame size
from its initial size.
3) After the call to __builtin_alloca, there must be a call to a
pure function.
With the enclosed patch, I can now build glibc for hppa-linux with -O3
optimisation.
Please review carefully because I will be the first to admit that I don't
understand why the use is there in the first place and all the details of
what eliminate_reg does.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
2000-11-14 John David Anglin
* reload1.c (eliminate_regs): Don't abort on MEM USEs.
--- reload1.c.orig Wed Sep 27 14:27:23 2000
+++ reload1.c Tue Nov 14 16:01:56 2000
@@ -2499,6 +2499,10 @@
return x;
case USE:
+ /* Handle insn_list USE that a call to a pure functions may generate. */
+ new = eliminate_regs (XEXP (x, 0), 0, insn);
+ if (GET_CODE (new) == MEM)
+ return XEXP (new, 0);
case CLOBBER:
case ASM_OPERANDS:
case SET:
From grundler@cup.hp.com Tue, 14 Nov 2000 14:32:00 -0800
Date: Tue, 14 Nov 2000 14:32:00 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] Palinux on a 712/60
Ian Zink wrote:
> I have a hp 712/60 I was trying to get loaded with Pa-linux. I'm using the
> 0.5 version of the Pa-linux cd. If I boot right off the CD it all loads
> until it gets to Switching to PDC. At that point nothing happens. From I
> have read the list, that is because the kernel is switching the text
> console. However, the 712s don't have consoles.
712s have consoles. They have two outputs which can be used as
console by linux. The STI consoles (VGA-like spigot) and serial.
Connect a serial cable 9600-8-n-1 and run minicom at the other end.
> From what I have also read
> the CD should work if you pass the kernel the parameter console=tty. So I
> tried to "boot scsi.2.0 isl" from the boot_admin prompt. It gave me the
> PALO ISL, but I could not choose which line I wanted to edit. Further, I
> couldn't even type "b" to boot. I don't know if the isl is freezing or what
> is taking place
That's a different problem...pb?
> What I am wondering is there a way to boot a 712/80 without having
> to get cross-compiler gcc, compile the kernel, etc.
The CD was intended to also work on the 712 even though we may not have
tested on it.
> Is there someway I could
> add the ramdisk-sti.tgz on the ISO to make it work? If so, what would need
> to be done?
no clue. anyone else?
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From mang@subcarrier.org Tue, 14 Nov 2000 17:31:22 -0500
Date: Tue, 14 Nov 2000 17:31:22 -0500
From: Michael Ang mang@subcarrier.org
Subject: [parisc-linux] tracking third-party sources (was Re: linux bame)
Paul Bame wrote:
>
> = bame@riverrock.org wrote:
> = >
> = > = On Thu, Nov 09, 2000 at 03:53:48PM -0700, Paul Bame wrote:
> = > = > Somebody never imported 2.4.0-test6, then I imported -test10 on the
> mai
> = > n
> = > = > vendor branch and now can't (easily) undo that to import test6 and
> THEN
> = > = > test10. This workaround sucks.
> =
> Like the comment said, there was no copy of plain -test6 in CVS (that I
> saw). Without -test6 in CVS it's much harder to use cvs diff to figure
> out the right way to merge files when there are conflicts.
> I didn't realize this until -test10 was already there, so I *then*
> brought in -test6. They're in the wrong order on the 1.1.1 branch, so
> the standard merge command 'cvs -jlinus:yesterday -jlinus:'
> won't work next time -- explicit names will be required.
The best thing to do is to import -test10 again and move the static tag
by re-tagging.
> = Tracking third-party sources using a normal branch, as we are
> = doing, is much simpler and gives you more control.
>
> But I've seen no cook book for it. I'm guessing that instead of cvs import
> you use:
> cvs co -rlinus linux
>
> cd linux
> cvs commit (make note of new files from commit)
> cvs add
> cvs commit
> cvs tag LINUS_NEW_REVISION
> perhaps with provision for removing obsolete files too. I suppose that is
> simpler than a single cvs import command from a certain perspective :-)
I had a good chat with Paul about this, and we worked out that using
"import" is marginally better.
This is what the add/remove method would look like:
cvs co -rlinux linux
cvs rm
cvs add
cvs commit
cvs tag LINUS_NEW_REVISION
Add the import method:
cd linux
cvs import linux linus LINUS_NEW_REVISION
cvs admin -b
> = When you say you "I imported -test10 on the main vendor branch" I hope
> = you really mean that you used "cvs add" on the linus branch. From your
> = other messages, your tags looked good.
>
> I used "cvs import", and either your branch magic worked, or I finished the
> merge before anybody randomly updated from cvs. Since import used 1.1.1,
> which is the branch you "fixed", it seems possible that 'cvs import' might
> be rendered harmless but I don't know that for sure.
Using "import" to bring in new files taints them with the vendor branch
badness. These files should be adjusted using "cvs admin -b". Note
that "cvs admin" works directly on files in the repository at low level
(without any revisioning of changes) and is thus to be avoided if at all
possible. Please don't run "cvs admin" if you (the collective "you")
don't know the consequences.
- Mike.
From ian.zink@maryville.com Tue, 14 Nov 2000 17:08:33 -0600
Date: Tue, 14 Nov 2000 17:08:33 -0600
From: Ian Zink ian.zink@maryville.com
Subject: [parisc-linux] Palinux on a 712/60
Thanks for the reply, Grant. However, the 712s do not have a serial console.
They do have a com port, but it does not work as a console, unfortunately.
So I need do need to get a STI enabled kernel. I dropped Paul Lahaie a note
to see if he could send me such a kernel so I do not have to create the
entire cross-platform development environment just to boot one of these
712s. After I get it, I plan on expanding the 0.5 iso and making a new one
using the STI console kernel.
Ian
From kailasr@webcash.com Tue, 14 Nov 2000 14:58:55 -0800
Date: Tue, 14 Nov 2000 14:58:55 -0800
From: Kailashnath V Rampure kailasr@webcash.com
Subject: [parisc-linux] Boot command
I have booted HP A class server through network. On that I have repartioned
the system and followed the steps on http://www.parisc-linux.org/install.html.
I have used the following command to initialize the hard disk. The kernel I
have downloaded is vmlinux-20001018. and cpied it to /mnt/boot/vmlinux
after mounting /dev/sda3. I have used the following command to initialize
the HDD.
palo -I -k /boot/vmlinux -b /boot/iplboot \ -c '3/boot/vmlinux TERM=linux
HOME=/ root=/dev/sda3' \ /dev/sda
--------------------------------
I get the following error:
--------------------------------
SOFT Boot.
palo ipl bame@noam Tue oct 31 14:18:02 MST 2000
0/vmlinux 2138614 bytes@ 0x1f78c000
0/palo-cmdline '3/boot/vmlinux TERM=linux HOME=/ root=/dev/sda3'
/dev/ida1 82 62 1030688
/dev/ida2 f0 1030750 24738
/dev/ida3 83 1055488 1030750
Kernel:partition 3 file /boot/vmlinux
ext2 block size 1024
ext2_mount(partition 3) returns 0
ext2_open(/boot/vmlinux) = -1
open /boot/vmlinux failed
-------------------------------------
Please suggest.
From grundler@cup.hp.com Tue, 14 Nov 2000 15:18:10 -0800
Date: Tue, 14 Nov 2000 15:18:10 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] Palinux on a 712/60
Ian Zink wrote:
> Thanks for the reply, Grant. However, the 712s do not have a serial console.
> They do have a com port, but it does not work as a console, unfortunately.
It does for linux. That's what the "Switching from PDC Console" is about.
Connect the serial cable to the com port and look at the output.
I've done it in the past and know it worked at least once.
I haven't tried it with the lasted ISO - no time to play w/712's now.
> So I need do need to get a STI enabled kernel. I dropped Paul Lahaie a note
> to see if he could send me such a kernel...
Paul just sent mail to the list indicating he's working on console issues.
Please let him work on it.
Trust me, he'll tell us when he's done. :^)
grant
From dhazeghi@pacbell.net Tue, 14 Nov 2000 18:17:39 -0800
Date: Tue, 14 Nov 2000 18:17:39 -0800
From: dhazeghi@pacbell.net dhazeghi@pacbell.net
Subject: [parisc-linux] [Semi OT] SOM Linker
Alex deVries wrote:
> dhazeghi@pacbell.net wrote:
> > However I would like to know what work if any has been done on the SOM
> > linker which HP released to the public last November(?). It seems that
> > as of right now, it has not been touched since February 14, and the FSF
> > binutils snapshots still do not have any SOM support for ld. Has there
> > been any movement in merging this in, or is anybody working on this?
>
> The initial plan was to do our 32-bit userspace with SOM, worrying about
> ELF32 much later in the game. But ELF32 development happened a lot
> quicker than expected, and so nobody's really done much on the SOM
> linker.
That's what it looked like...
>
>
> I suspect it'd be very hard to use the SOM linker code to incorporate it
> into binutils, but I could be wrong.
>
> What are you actually trying to do?
I would like to be able to set up a cross compilation environment for hpux and
32 bit PA-RISC. However without a functional cross linker, this is impossible
to do, and as binutils has not got one yet, I thought perhaps the one that HP
open-sourced might be some use. It would seem logical that with the sources
available, it shouldn't be too difficult to fix the broken bits and get a SOM
linker working in binutils, but that doesn't seem to have happened yet. Oh
well, thanks for the info...
Dara Hazeghi
From dave@hiauly1.hia.nrc.ca Tue, 14 Nov 2000 22:00:25 -0500 (EST)
Date: Tue, 14 Nov 2000 22:00:25 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: [parisc-linux] [Semi OT] SOM Linker
> I would like to be able to set up a cross compilation environment for hpux and
> 32 bit PA-RISC. However without a functional cross linker, this is impossible
> to do, and as binutils has not got one yet, I thought perhaps the one that HP
> open-sourced might be some use. It would seem logical that with the sources
> available, it shouldn't be too difficult to fix the broken bits and get a SOM
> linker working in binutils, but that doesn't seem to have happened yet. Oh
> well, thanks for the info...
You should be able to build cross compilation tools under hpux for hppa-linux.
First you should install native versions of binutils and gcc under hpux (this
assumes that you have the hpux C compiler and linker). The release version of
gcc (2.95.2) would be a good choice. The standard hpux linker works fine with
gcc/gas for C compilations. Then follow the directions at www.parisc-linux.org
for building the cross compilation tools and linux.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From dhazeghi@pacbell.net Tue, 14 Nov 2000 21:23:41 -0800
Date: Tue, 14 Nov 2000 21:23:41 -0800
From: dhazeghi@pacbell.net dhazeghi@pacbell.net
Subject: [parisc-linux] [Semi OT] SOM Linker
John David Anglin wrote:
> > I would like to be able to set up a cross compilation environment for hpux and
> > 32 bit PA-RISC. However without a functional cross linker, this is impossible
> > to do, and as binutils has not got one yet, I thought perhaps the one that HP
> > open-sourced might be some use. It would seem logical that with the sources
> > available, it shouldn't be too difficult to fix the broken bits and get a SOM
> > linker working in binutils, but that doesn't seem to have happened yet. Oh
> > well, thanks for the info...
>
> You should be able to build cross compilation tools under hpux for hppa-linux.
> First you should install native versions of binutils and gcc under hpux (this
> assumes that you have the hpux C compiler and linker). The release version of
> gcc (2.95.2) would be a good choice. The standard hpux linker works fine with
> gcc/gas for C compilations. Then follow the directions at www.parisc-linux.org
> for building the cross compilation tools and linux.
This is not precisely what I mean. What I should have said is that I want to create
a cross compiler --host=i686-linux --target=hppa-hpux. I asked this list, because
you folks did some work on the SOM linker, which is at the moment the missing piece
for a cross toolchain which targets hpux.
Thanks,
Dara
P.S. On a different note, is the recipe fully up to date? I tried following it a
few weeks ago, but glibc did not complete building successfully.
From Arnaud.ATOCH@oecd.org Wed, 15 Nov 2000 10:05:04 +0100
Date: Wed, 15 Nov 2000 10:05:04 +0100
From: Arnaud.ATOCH@oecd.org Arnaud.ATOCH@oecd.org
Subject: [parisc-linux] Palinux on a 712/60
Hi,
I got a couple of 715 and I'd love having access to an ISO image with STI
console kernel too.
-----Original Message-----
From: Ian Zink [mailto:ian.zink@maryville.com]
Sent: Wednesday, November 15, 2000 00:09
To: 'parisc-linux@thepuffingroup.com'
Subject: RE: [parisc-linux] Palinux on a 712/60
Thanks for the reply, Grant. However, the 712s do not have a serial console.
They do have a com port, but it does not work as a console, unfortunately.
So I need do need to get a STI enabled kernel. I dropped Paul Lahaie a note
to see if he could send me such a kernel so I do not have to create the
entire cross-platform development environment just to boot one of these
712s. After I get it, I plan on expanding the 0.5 iso and making a new one
using the STI console kernel.
Ian
---------------------------------------------------------------------------
To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com with
`unsubscribe' as the subject.
From rhirst@linuxcare.com Wed, 15 Nov 2000 09:58:35 +0000
Date: Wed, 15 Nov 2000 09:58:35 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] dino maintainer?
On Tue, Nov 14, 2000 at 08:10:42AM -0800, Grant Grundler wrote:
> The bug is normal for card-mode Dino - not for Built-in Dino.
> I think Helge has the GSC 100BT card which is a card-mode Dino on-board
> with one (or two) Tulip(s) behind it.
>
> The warning is a reminder one can NOT use MMIO accesses to those
> PCI devices and *only* I/O Port space (eg inb/outb).
>
> If someone wants to fix the warning so it's quiet for card-mode
> devices...see is_card_dino(d) in dino_driver_callback() for an
> example.
>
> FYI - card-mode dino was used for several different networking
> interfaces but not SCSI interfaces.
But Helge has problems with the sym53c8xx driver on a B160L. Is
that a PCI card driven via Dino? And if so, are you saying he needs
to build his kernel with CONFIG_SCSI_NCR53C8XX_IOMAPPED=y so it
doesn't try to use MMIO?
Helge, it might be worth trying to switch on CONFIG_SCSI_NCR53C8XX_IOMAPPED
anyway just to see what happens. Otherwise someone needs to start
adding printk debug to figure out what is happening. I can't do
that as I don't have a sym53c8xx pci card.
Richard
From andrew@neep.com.au Wed, 15 Nov 2000 18:10:13 +0800
Date: Wed, 15 Nov 2000 18:10:13 +0800
From: Andrew Shugg andrew@neep.com.au
Subject: [parisc-linux] Palinux on a 712/60
Arnaud.ATOCH@oecd.org said:
> Hi,
>
> I got a couple of 715 and I'd love having access to an ISO image with STI
> console kernel too.
I've never made an El Torito CDROM, is it possible to have multiple boot
images and a boot loader on a single disc? ie could the actual boot
image be a boot loader (PALO) which then points at one or more kernels
on the CDROM?
Andrew.
--
Andrew Shugg http://www.neep.com.au/
"Just remember, Mr Fawlty, there's always someone worse off than yourself."
"Is there? Well I'd like to meet him. I could do with a good laugh."
From marteaut@esiee.fr Wed, 15 Nov 2000 11:25:07 +0100
Date: Wed, 15 Nov 2000 11:25:07 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] Palinux on a 712/60
Hi Ian,
If you like to see a linux kernel booting on your 712, you can donwload
a fully operationnal root file system with the STI console and an extra
terminal with
the RS232 port at
http://www.esiee.fr/puffin
You have also all the info to make a bootable hard disk so try it and give
us feedback...
Bye,
Thomas Marteau
ESIEE Team
From rhirst@linuxcare.com Wed, 15 Nov 2000 11:12:20 +0000
Date: Wed, 15 Nov 2000 11:12:20 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] Beta CD
On Fri, Nov 10, 2000 at 05:14:06PM -0500, Paul J.Y. Lahaie wrote:
> Hello fellow PA-RISCers,
>
> An preliminary beta CD for PA/Linux has been uploaded to
> puffin.external.hp.com. If people could try it and forward any complaints
> or suggestions to me, it would be greatly appreciated. The URL for the
> image is: ftp://puffin.external.hp.com/pub/parisc/cd-images/palinux-0.5.iso.gz
I just borrowed a CDROM drive for my 715/75 so I could try
it on there [Actually this is the palinux-0.5.iso.gz that I
commented on, not the final version]. I wasn't successful:
Sometimes the boot hangs after "ASP version 1 at 0xf0800000 found",
waits a few seconds and then HPMCs.
The scsi driver always has serious problems with the CD drive
but does detect other devices on the bus.
The kernel always crashes after "Serial driver version 5.01...",
(if it gets that far) with
Data access rights fault in kernel: Code=26 regs=c5f9c940 (Addr=00000003)
YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000001000000000000001010
r0-3 00000000 c0217800 c011e9f8 c5ff64a0
r4-7 c5ff6200 c023ab20 00000008 f0823800
r8-11 00000003 00000007 c019134c c02b20a0
r12-15 fffffffc c023a800 c023a800 c02c31cc
r16-19 c023a800 c023a800 c02b2024 00000000
r20-23 c5ff6234 f0823807 f0823800 0000000a
r24-27 ffffffff c5ff64a0 c5ff6200 c0258000
r28-31 ffffffff 000002c0 c5f9cb80 c012dfc0
sr0-4 00000000 00000000 00000000 00000000
sr4-8 00000000 00000000 00000000 00000000
IASQ: 00000000 00000000 IAOQ: c011e710 c011e714
IIR: 0f881093 ISR: 00000000 IOR: 00000003
ORIG_R28: 00000058
I'll investigate further when I have the time.
Richard
From adevries@linuxcare.com Wed, 15 Nov 2000 11:50:27 -0500
Date: Wed, 15 Nov 2000 11:50:27 -0500
From: Alex deVries adevries@linuxcare.com
Subject: [parisc-linux] dino maintainer?
Richard Hirst wrote:
> On Tue, Nov 14, 2000 at 08:10:42AM -0800, Grant Grundler wrote:
> > FYI - card-mode dino was used for several different networking
> > interfaces but not SCSI interfaces.
>
> But Helge has problems with the sym53c8xx driver on a B160L. Is
> that a PCI card driven via Dino? And if so, are you saying he needs
> to build his kernel with CONFIG_SCSI_NCR53C8XX_IOMAPPED=y so it
> doesn't try to use MMIO?
Hang on a sec... what Grant's saying is that *card-mode* dino is never
used for SCSI controllers, but on the B160L it would probably be
*chip-mode* dino.
Does this mean that all GSC SCSI expansion cards are Zalon based?
So what Helge has isn't a PCI card specifically, Dino and the 53c8xx are
all on the motherboard.
Does Dino handle IO memory mapping differently for chip or card mode?
- Alex
--
Alex deVries, Principal Solutions Architect, The Puffins at Linuxcare
613.562.2759 tel
alex@linuxcare.com, http://www.linuxcare.com/
Linuxcare, Support for the revolution.
From grundler@cup.hp.com Wed, 15 Nov 2000 08:06:32 -0800
Date: Wed, 15 Nov 2000 08:06:32 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] dino maintainer?
Richard Hirst wrote:
> On Tue, Nov 14, 2000 at 08:10:42AM -0800, Grant Grundler wrote:
> > The bug is normal for card-mode Dino - not for Built-in Dino.
> > I think Helge has the GSC 100BT card which is a card-mode Dino on-board
> > with one (or two) Tulip(s) behind it.
Helge confirmed he has no such card.
I think the PDC simply isn't programming the Dino IO_ADDR_EN
since there are no PCI devices in his B160.
Helge's B160 has a old rev of Dino PCI host bus adapter chip.
It's possible to have "silent" data corruption caused by older
revs of dino - 3.0 and older.
The latest PDC Revisions (5.x and later) know this and won't permit
the system to be booted unless the only devices on the PCI bus are
known graphics interface cards.
> > The warning is a reminder one can NOT use MMIO accesses to those
> > PCI devices and *only* I/O Port space (eg inb/outb).
> >
> > If someone wants to fix the warning so it's quiet for card-mode
> > devices...see is_card_dino(d) in dino_driver_callback() for an
> > example.
This is still correct for card-mode dino.
> > FYI - card-mode dino was used for several different networking
> > interfaces but not SCSI interfaces.
>
> But Helge has problems with the sym53c8xx driver on a B160L. Is
> that a PCI card driven via Dino?
I doubt it now. If Helge could send richard "in io" output, I think
that would clarify what's in the B160.
> And if so, are you saying he needs
> to build his kernel with CONFIG_SCSI_NCR53C8XX_IOMAPPED=y so it
> doesn't try to use MMIO?
no. no SCSI was ever implement on a card-mode dino board.
No reason to since they already had Zalon or open slots.
grant
> Helge, it might be worth trying to switch on CONFIG_SCSI_NCR53C8XX_IOMAPPED
> anyway just to see what happens. Otherwise someone needs to start
> adding printk debug to figure out what is happening. I can't do
> that as I don't have a sym53c8xx pci card.
>
> Richard
From grundler@cup.hp.com Wed, 15 Nov 2000 08:17:41 -0800
Date: Wed, 15 Nov 2000 08:17:41 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] dino maintainer?
Alex deVries wrote:
> Hang on a sec... what Grant's saying is that *card-mode* dino is never
> used for SCSI controllers, but on the B160L it would probably be
> *chip-mode* dino.
No such thing - you probably mean "Bridge Mode" and that's what
the built-in Dino is using.
> Does this mean that all GSC SCSI expansion cards are Zalon based?
>
> So what Helge has isn't a PCI card specifically, Dino and the 53c8xx are
> all on the motherboard.
>
> Does Dino handle IO memory mapping differently for chip or card mode?
yes. yes. yes (conditional).
"IO memory mapping" is a confusing term. I'll assume you mean MMIO.
(Memory Mapped I/O)
MMIO space access is independent of I/O Port space access.
MMIO space access simple isn't available for card-mode Dino since
niether PDC nor the OS assigns host physical address space to the
card-mode Dino (that's what IO_ADDR_EN is for). PDC does this
for Bridge-mode dino (built-in) - but apperently only when it needs
to.
I/O Port space accesses are done the same way for both modes.
I/O Port space access is implemented by poking registers on Dino.
Read the Dino Spec (or source) for more details.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From dave@hiauly1.hia.nrc.ca Wed, 15 Nov 2000 11:23:05 -0500 (EST)
Date: Wed, 15 Nov 2000 11:23:05 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: [parisc-linux] [Semi OT] SOM Linker
> This is not precisely what I mean. What I should have said is that I want to create
> a cross compiler --host=i686-linux --target=hppa-hpux. I asked this list, because
> you folks did some work on the SOM linker, which is at the moment the missing piece
> for a cross toolchain which targets hpux.
This also may be possible. The first step would be to copy the hpux headers
to a i686-linux system and see if you can build the HP linker. The next
step would be to try to build the cross binutils tools. I know this
requires the hpux headers and likely some hacking would be required to
get it to build.
The linker is probably the hard part. There may be byte ordering issues
and bugs in what HP contributed.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From pdbeal@louisville.edu Wed, 15 Nov 2000 11:47:09 -0500
Date: Wed, 15 Nov 2000 11:47:09 -0500
From: Phillip D. Beal pdbeal@louisville.edu
Subject: [parisc-linux] Beta CD
On Fri, Nov 10, 2000 at 05:14:06PM -0500, Paul J.Y. Lahaie wrote:
> Hello fellow PA-RISCers,
>
> An preliminary beta CD for PA/Linux has been uploaded to
> puffin.external.hp.com. If people could try it and forward any complaints
> or suggestions to me, it would be greatly appreciated. The URL for the
> image is: ftp://puffin.external.hp.com/pub/parisc/cd-images/palinux-0.5.iso.gz
>
> - Paul
the CD worked great on the 715 I tried, and I actually didn't have a
serial console machine, so I used my Palm Pilot and a link cable. Still
worked like a charm. How did you get the kernel image into the boot
sector of the CD? I'd like to try to build some CD's of the kernels I'm
building to test in some mahcines that are not on the same network as
the machine that I'm building everything from.
And I don't mind posting my CD images somewhere if they work...but most
of my testing is on a 735 or 755.
Thanks,
--
Phillip Beal
Electrical and Computer Engineering
S+LUG Vice-President
From bame@noam.fc.hp.com Wed, 15 Nov 2000 10:08:22 -0700
Date: Wed, 15 Nov 2000 10:08:22 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] Help with posix_types.h
I'm reviewing the posix_types.h to figure out what's right for 64-bit
linux. I know others may have thought through this before so I'm hoping
for guidance. For those unfamiliar, the type names in posix_types.h
are like __kernel_dev_t and usually are used to define the corresponding
"normal" type (dev_t). These should be settled before the 32/64
syscall wrappers can be completed.
__kernel_ino_t: often 32 bits, currently 32 bits for parisc and som3
64-bit kernels (mips64 and ia64). 64 bits on alpha and sparc64.
Seems to me this ought to be 32 bits on parisc and parisc64, or
64 bits on both, since it's a function of file system sizes not
processor width. HPUX kernel seems to always use 32 bits, but
64-bit userspace uses 64 bits. I propose 32 bits on both, but
willy had selected 64 bits in parisc64 for some reason.
__kernel_off_t: seems to be 32 bits on 32-bit cpus, 64 on 64-bit ones.
This is supposedly the offset from a beginning of a file.
HPUX appears to use 64-bits *in the kernel* for both 32 and
64-bit kernels.
The obvious pattern is to make ours 32 on narrow, and 64 on
wide palinux so I guess I propose that, and that's the way
it was before I hacked on it too.
Should we consider switching to 64 bits on narrow
palinux since this is related to file systems, not CPUs.
Note there's also a __kernel_loff_t -> loff_t which appears
to be defined as 64 bits. I'm not sure how this does/should
interact with off_t (lseek vs lseek64 for example).
__kernel_suseconds_t: which becomes suseconds_t which is used in
struct timeval {
time_t tv_sec; /* seconds */
suseconds_t tv_usec; /* microseconds */
};
I'm confused why hpux, and other systems,
like for this to be 'long' (e.g., 64 bits on 64-bit processors)
since it seems like values over a million are probably rare
and 32 bits seems to be enough for most implementations. sparc64
chose 32 bits for this and I want to do the same for parisc64
because it will reduce the amount of syscall structure repackings.
Comments?
__kernel_daddr_t: which is used indirectly in struct solaris_x86_vtoc and
solaris_x86_slice which *might* be an on-disk data structures
(used with CONFIG_SOLARIS_X86_PARTITION). So this needs to
be 32 bits if that's the case (definition from sparc) and several
archs have it 64 bits!
On the other hand, HPUX's man page says 'daddr_t used for disk
addresses except in an inode [and partition table I would think]
on disk'. So it should probably be the same
type as off_t. FYI the only other place it's used is in
'struct mtio' -- used to talk to magnetic tape units.
I'm guessing the sparc struct should not be using this type,
and that we should define it the same as off_t. Comments?
From kailasr@webcash.com Wed, 15 Nov 2000 09:55:09 -0800
Date: Wed, 15 Nov 2000 09:55:09 -0800
From: Kailashnath V Rampure kailasr@webcash.com
Subject: [parisc-linux] Boot command
Yes Thomas the vmlinux is in /boot of 3rd partition as the other partitions
are swap and f0.
At 11:48 AM 11/15/00 +0100, Thomas Marteau wrote:
>----- Original Message -----
>From: Kailashnath V Rampure
>To: Paul Bame
>Cc: Grant Grundler ; Matt Taggart
>;
>Sent: Tuesday, November 14, 2000 11:58 PM
>Subject: Re: [parisc-linux] Boot command
>
>
> > I have booted HP A class server through network. On that I have
>repartioned
> > the system and followed the steps on
>http://www.parisc-linux.org/install.html.
> > I have used the following command to initialize the hard disk. The kernel
>I
> > have downloaded is vmlinux-20001018. and cpied it to /mnt/boot/vmlinux
> > after mounting /dev/sda3. I have used the following command to initialize
> > the HDD.
> >
> > palo -I -k /boot/vmlinux -b /boot/iplboot \ -c '3/boot/vmlinux TERM=linux
> > HOME=/ root=/dev/sda3' \ /dev/sda
>
>This means that vmlinux must be in /boot in your third partition on your
>disk.
>Is it true in your case?
>
> > --------------------------------
> > I get the following error:
> > --------------------------------
> > SOFT Boot.
> > palo ipl bame@noam Tue oct 31 14:18:02 MST 2000
> > 0/vmlinux 2138614 bytes@ 0x1f78c000
> > 0/palo-cmdline '3/boot/vmlinux TERM=linux HOME=/ root=/dev/sda3'
> > /dev/ida1 82 62 1030688
> > /dev/ida2 f0 1030750 24738
> > /dev/ida3 83 1055488 1030750
> > Kernel:partition 3 file /boot/vmlinux
> > ext2 block size 1024
> > ext2_mount(partition 3) returns 0
> > ext2_open(/boot/vmlinux) = -1
> > open /boot/vmlinux failed
> > -------------------------------------
> > Please suggest.
> >
> > --------------------------------------------------------------------------
>-
> > To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com
>with
> > `unsubscribe' as the subject.
> >
> >
From bame@noam.fc.hp.com Wed, 15 Nov 2000 11:05:14 -0700
Date: Wed, 15 Nov 2000 11:05:14 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] Beta CD
= How did you get the kernel image into the boot
= sector of the CD? I'd like to try to build some CD's of the kernels I'm
= building to test in some mahcines that are not on the same network as
= the machine that I'm building everything from.
The magic for making bootable CDs is documented in the PALO README.html
which you have on your system already since you're building kernels.
-P
From marteaut@esiee.fr Wed, 15 Nov 2000 19:10:56 +0100
Date: Wed, 15 Nov 2000 19:10:56 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] Boot command
Hi again,
You put the swap partition before the f0 partition and here it is the f0
partition in first position.
Also, your hard disk is ida instead of sda, what's that ?
Bye,
ESIEE Team
----- Original Message -----
From: Kailashnath V Rampure
To: Thomas Marteau
Cc:
Sent: Wednesday, November 15, 2000 6:55 PM
Subject: Re: [parisc-linux] Boot command
> Yes Thomas the vmlinux is in /boot of 3rd partition as the other
partitions
> are swap and f0.
>
> At 11:48 AM 11/15/00 +0100, Thomas Marteau wrote:
>
> >----- Original Message -----
> >From: Kailashnath V Rampure
> >To: Paul Bame
> >Cc: Grant Grundler ; Matt Taggart
> >;
> >Sent: Tuesday, November 14, 2000 11:58 PM
> >Subject: Re: [parisc-linux] Boot command
> >
> >
> > > I have booted HP A class server through network. On that I have
> >repartioned
> > > the system and followed the steps on
> >http://www.parisc-linux.org/install.html.
> > > I have used the following command to initialize the hard disk. The
kernel
> >I
> > > have downloaded is vmlinux-20001018. and cpied it to /mnt/boot/vmlinux
> > > after mounting /dev/sda3. I have used the following command to
initialize
> > > the HDD.
> > >
> > > palo -I -k /boot/vmlinux -b /boot/iplboot \ -c '3/boot/vmlinux
TERM=linux
> > > HOME=/ root=/dev/sda3' \ /dev/sda
> >
> >This means that vmlinux must be in /boot in your third partition on your
> >disk.
> >Is it true in your case?
> >
> > > --------------------------------
> > > I get the following error:
> > > --------------------------------
> > > SOFT Boot.
> > > palo ipl bame@noam Tue oct 31 14:18:02 MST 2000
> > > 0/vmlinux 2138614 bytes@ 0x1f78c000
> > > 0/palo-cmdline '3/boot/vmlinux TERM=linux HOME=/ root=/dev/sda3'
> > > /dev/ida1 82 62 1030688
> > > /dev/ida2 f0 1030750 24738
> > > /dev/ida3 83 1055488 1030750
> > > Kernel:partition 3 file /boot/vmlinux
> > > ext2 block size 1024
> > > ext2_mount(partition 3) returns 0
> > > ext2_open(/boot/vmlinux) = -1
> > > open /boot/vmlinux failed
From dave@hiauly1.hia.nrc.ca Wed, 15 Nov 2000 13:09:49 -0500 (EST)
Date: Wed, 15 Nov 2000 13:09:49 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: [parisc-linux] Help with posix_types.h
The largest disk currently available I believe is the 180GB Seagate Baracuda.
The size of drives is increasing about a factor of 2 per year. The
__kernel_off_t definitely needs to be 64 bits to handle large drives
in both 32 and 64 bit systems. Disk blocks are typically 512 or 1024 bytes.
Thus, drives may exceed 4GB disk blocks in 3-4 years. Inodes are variable
in size (8KB average). Thus, we are a little further away from exceeding
the 32 bit inode limit.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From bame@noam.fc.hp.com Wed, 15 Nov 2000 11:12:57 -0700
Date: Wed, 15 Nov 2000 11:12:57 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] Boot command
= You put the swap partition before the f0 partition and here it is the f0
= partition in first position.
The partition order doesn't matter so long as the f0 partition and
the partition containing your kernel (an ext2 partition) *end* within
the first 2Gb of the disk. See the PALO README.html.
= Also, your hard disk is ida instead of sda, what's that ?
PALO lists the devices as 'ida' instead of 'sda' or 'hda' since it
is using the firmware 'IODC' interface to talk to the boot device,
and it has no idea what type of boot device IODC is providing.
-P
From rhirst@linuxcare.com Wed, 15 Nov 2000 18:48:08 +0000
Date: Wed, 15 Nov 2000 18:48:08 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] Single-stepping
(Oops, CC-ed to the wrong list first time!)
Hi John,
I've been helping Alan Modra out with kernel changes to support
single stepping for gdb. Paul Bame suggested I bounced our ideas
off you in case you (or anyone else) had any comments. I havn't
actually committed my changes yet.
The basic approach is to use the recovery counter to generate
a trap every instruction. The scheme is complicated because a
suspended process may or may not return to user space via an RFI.
If it was suspended as a result of an interrupt then we can
simply set PSW bit R in the tasks saved registers and it will
get loaded by the RFI. On every task switch I set the
recovery counter to 0, just in case the new process is being
single-stepped.
If a process is suspended during a syscall, then there is no
RFI on the return path to userland, and we have to handle things
differently. I have changed the syscall return path such that
it loads the recovery counter with 3 before updating the PSW
with a value from the tasks saved registers. If that PSW has
the R bit set, then the count of 3 will generate a trap on the
first instruction following the branch back to user space.
Note that PSW wasn't previously restored on the syscall return
path.
To avoid further complications of interrupts during the three
instructions when the recovery counter is decrementing, whenever
we set the R bit, we also clear the I bit to disable interrupts.
Nullified instructions are handled by the controlling process
manually moving the childs IAOQ over the instruction without
actually setting it running, because the recovery counter isn't
decremented for nullified instructions.
I need to do some more testing before committing this, but would
welcome any comments on the basic approach taken, areas I have
mis-understood, or problems with it that might not yet have
occurred to me.
Thanks,
Richard
From andreas@bawue.de Thu, 16 Nov 2000 20:20:41 +0100
Date: Thu, 16 Nov 2000 20:20:41 +0100
From: Andreas Thienemann andreas@bawue.de
Subject: [parisc-linux] 250/1 doesn't boot but dumps stack
This is a multi-part message in MIME format.
--------------F7EF1B69F0A7C4C7DD54E27D
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Hi,
I recently got a D-Class HP900 250/1 (at least it says that on the label) and
tried to get the pa-risc port running on it.
So I got a CVS checkout and build the whole toolchain without any major
hassles.
The kernel, including NFS-ROOT, built also without any glitches.
But when I try to boot up this kernel it dumps stack right after initialising
the pty interfaces. (I already left out everything unnecessary, such as
parallel port support)
After that it complains about "Data acces rights fault in kernel: Code=26
regs=c7f98780 (Addr=000000004)" and some more data...
For the curious, the boot sequence is attached.
I hope someone can give me a clue what might be wrong...
Thanks,
Andreas
--------------F7EF1B69F0A7C4C7DD54E27D
Content-Type: text/plain; charset=us-ascii;
name="hp9000-capture"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="hp9000-capture"
Firmware Version 36.34
Duplex Console IO Dependent Code (IODC) revision 0
------------------------------------------------------------------------------
(c) Copyright 1995-1998, Hewlett-Packard Company, All rights reserved
------------------------------------------------------------------------------
Processor Speed State Coprocessor State Cache Size
--------- -------- --------------------- ----------------- ----------
0 101 MHz Active Functional 256 KB
Central Bus Speed (in MHz) : 101
Available memory (bytes) : 134217728
Good memory required (bytes): 15245312
Primary boot path: 8/16/5.5 (dec)
Alternate boot path: 8/16/5.2 (dec)
Console path: 8/16/4.0 (dec)
Keyboard path: 8/16/7.0 (dec)
Processor is booting from first available device.
To discontinue, press any key within 10 seconds.
Boot terminated.
------- Main Menu -------------------------------------------------------------
Command Description
------- -----------
BOot [PRI|ALT|] Boot from specified path
PAth [PRI|ALT|CON|KEY] [] Display or modify a path
SEArch [DIsplay|IPL] [] Search for boot devices
COnfiguration [] Access Configuration menu/commands
INformation [] Access Information menu/commands
SERvice [] Access Service menu/commands
DIsplay Redisplay the current menu
HElp [
I could not find one. I found some apache-doc etc.
Can any on point me where I can try.
Regards
Kailas
--=====================_108207293==_.ALT--
From bame@noam.fc.hp.com Wed, 15 Nov 2000 16:59:18 -0700
Date: Wed, 15 Nov 2000 16:59:18 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] Apache package.
= I tried to build apache_1.3.12 on HP a class server. But I have error. I
= have tried to check the site
= ftp://ftp.debian.org/debian/dists/unstable/main/binary-hppa/
= I could not find one. I found some apache-doc etc.
We are still working on some kernel features which are required to
support Apache (system 5 shared memory).
-P
From cary@cup.hp.com Wed, 15 Nov 2000 16:00:49 -0800
Date: Wed, 15 Nov 2000 16:00:49 -0800
From: Cary Coutant cary@cup.hp.com
Subject: [parisc-linux] Help with posix_types.h
>You have certainly reserved a thread register in the ABI, right? If
>not, do it ASAP.
On PA-RISC, the thread pointer is kept in the read-only control register
CR 27.
For user-thread implementations, we provided a kernel API to change the
contents of the register.
-cary
From drepper@redhat.com 15 Nov 2000 16:04:51 -0800
Date: 15 Nov 2000 16:04:51 -0800
From: Ulrich Drepper drepper@redhat.com
Subject: [parisc-linux] Help with posix_types.h
Cary Coutant writes:
> On PA-RISC, the thread pointer is kept in the read-only control register
> CR 27.
>
> For user-thread implementations, we provided a kernel API to change the
> contents of the register.
This works fine for a 1:1 thread implementation but adds significant
runtime hits for a m:n implementation. The latter is the direction we
are heading.
--
---------------. ,-. 1325 Chesapeake Terrace
Ulrich Drepper \ ,-------------------' \ Sunnyvale, CA 94089 USA
Red Hat `--' drepper at redhat.com `------------------------
From jsm@udlkern.fc.hp.com Thu, 16 Nov 2000 02:01:12 -0700 (MST)
Date: Thu, 16 Nov 2000 02:01:12 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: Single-stepping
> I've been helping Alan Modra out with kernel changes to support
> single stepping for gdb. Paul Bame suggested I bounced our ideas
> off you in case you (or anyone else) had any comments. I havn't
> actually committed my changes yet.
>
I've decided to respond to the whole list, since others are now
participating in the discussion.
> The basic approach is to use the recovery counter to generate
> a trap every instruction. The scheme is complicated because a
> suspended process may or may not return to user space via an RFI.
>
There is no easy way to do single stepping on parisc. So any single
stepping design will be complicated.
> If it was suspended as a result of an interrupt then we can
> simply set PSW bit R in the tasks saved registers and it will
> get loaded by the RFI. On every task switch I set the
> recovery counter to 0, just in case the new process is being
> single-stepped.
>
> If a process is suspended during a syscall, then there is no
> RFI on the return path to userland, and we have to handle things
> differently. I have changed the syscall return path such that
> it loads the recovery counter with 3 before updating the PSW
> with a value from the tasks saved registers. If that PSW has
> the R bit set, then the count of 3 will generate a trap on the
> first instruction following the branch back to user space.
> Note that PSW wasn't previously restored on the syscall return
> path.
>
Just to be clear, it is impossible to restore the entire PSW without
an RFI. So, I assume you are referring to the system mask subset of
the PSW that can be manipulated by the ssm,rsm, and mtsm instructions.
You mention restoring from the task's saved registers, but we currently
do not save the system mask during a syscall (because it should be the
same for all processes). Have you added code to do that also? If not,
you are restoring from whatever the state was at the last interruption.
Which in this case works (since the R bit state will be changed
by another process while the debugged process is suspended, this should
guarantee that the R bit state is up to date), but it seems a little ugly.
In my opinion, you should just be checking a bit in the ptrace flags
in the task structure, and setting the R bit with an ssm instruction
based on that.
> To avoid further complications of interrupts during the three
> instructions when the recovery counter is decrementing, whenever
> we set the R bit, we also clear the I bit to disable interrupts.
Yuck, but I agree that it would be messier to have to deal with this in
the interrupt handlers. Please make sure that a comment is added that
explains what you are doing, and clearly documents the dependency on the
number of remaining instructions before we return to user privilege level.
I assume you restore the I bit in the recovery counter trap handler. I
can think of alternative ways of doing this, but they are probably just as
ugly (e.g. one possibility would be to do an rfi to set the L bit).
>
> Nullified instructions are handled by the controlling process
> manually moving the childs IAOQ over the instruction without
> actually setting it running, because the recovery counter isn't
> decremented for nullified instructions.
Does this code properly handle branches in the delay slot of another
branch? (you need to make sure you are not advancing the queues by just
adding 4 to each element). One concern I have about this method is that
the userland debugger has to cooperate to make this design work, i.e. the
single stepping is not accomplished entirely within the kernel, so we
cannot easily change the design for single stepping at a later date.
I wonder if it is necessary to do this. So what if we don't stop on the
nullified instruction. Since it is nullified, it doesn't actually do
anything, so why does the user have to see it, i.e. just let the recovery
counter trap happen on the next truly executed instruction (i.e. the
debugger performs a "double step" in this case). Am I missing something
here?
>
> I need to do some more testing before committing this, but would
> welcome any comments on the basic approach taken, areas I have
> mis-understood, or problems with it that might not yet have
> occurred to me.
OK, well here are some issues that you didn't mention, so I don't
know whether or not you addressed them:
1) When single stepping over a syscall, when do you actually stop the
single stepping and execute the syscall? Hopefully you are not
allowing single stepping after the gate instruction on the gateway
page (and returning control to a non privileged debugging process).
The recovery counter trap should detect when the user code gets
to the gateway page.
2) Does your solution properly handle single stepping into and out of
a signal handler? Note that the debugger will trap the signal as part
of this process. Since the return is handled through a hidden syscall
you may not have to do anything special here.
Note that HP-UX does not use the recovery counter for single stepping. I
made a few phone calls to various engineers to find out what the design
process was, and why they chose the solution they did, but I could not
find anyone who knew. Looking at the code in HP-UX it looks like someone
implemented that code a long time ago, and some of the engineers who have
worked on it since don't understand it, because some of the comments added
since then clearly show a lack of understanding of what is really going
on.
Others on this list have mentioned that MPE does use the recovery counter
for single stepping. Of course, MPE is not a Unix clone, so just because
it could be done on MPE doesn't mean that the recovery counter can cover
all cases on Unix (e.g. I have no idea how signals and syscalls are
implemented on MPE). But since I have no idea why the recovery counter
was not used for HP-UX, I can't say it is the wrong way to go. I can't
think of anything that will definitely rule it out, I'm just a little
uncomfortable with the fact that HP-UX chose not to use it.
One advantage of the HP-UX method is that it completely encapsulates the
single stepping inside the kernel, so it can be changed if necessary,
without having to modify gdb (and having to worry about old versions of
gdb).
Anyway, for reference, HP-UX does single stepping by using a combination
of the taken branch trap, and loading the instruction queues such that the
front of the queue points to the next instruction to be single stepped and
the back of the queue points to the first of two break instructions on a
"break" page. It does NOT insert break instructions into the code, so it
does not adversely affect execution on a SMP machine. Note that we
already put a bunch of break instructions before the syscall entry point
on the gateway page, so it would be easy to use our gateway page for the
"break page". This way, if the single stepped instruction branches, a
taken branch trap will be taken (which is important in the case where the
branch nullifies its delay slot). Otherwise, the instruction will be
executed and then the break instruction at the known location on the
"break" page will be executed. If the single stepped instruction
nullifies the next instruction, the second break instruction on the
"break" page will be executed.
Note that this is the short explanation. It is not as simple as it sounds.
One major complication is that branches with links don't work properly
with the instruction queue magic, so the link register has to be updated
in the taken branch trap handler. Also branch externals won't update
the space of the space queue tail properly (again, that has to be fixed
in the taken branch handler). I can provide more details if the recovery
counter method doesn't work out.
Sincerely,
John Marvin
jsm@fc.hp.com
From marteaut@esiee.fr Thu, 16 Nov 2000 10:04:07 +0100
Date: Thu, 16 Nov 2000 10:04:07 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] nfsroot - 712/60 - RPC - help
Hi Jack,
> (c) Copyright 1990-1993, Hewlett-Packard Company.
> All rights reserved
>
> Press to stop boot sequence.
> Selecting a system to boot.
>
> Booting
> palo ipl jkp2866@redtower Wed Nov 15 12:44:21 CST 2000
> 0/vmlinux 1606438 bytes @ 0x6800
> 0/palo-cmdline '0/vmlinux HOME=/ TERM=LINUX root=/dev/nfs
nfsroot=192.68.1.11:/tftpboot/pancho '
Are you sure that the command line is rigth? Is your Server address is
192.-68-.1.11
Here is probably your mistake!
Bye,
ESIEE Team
Visit http://www.esiee.fr/puffin
> Kernel: partition 0 file /vmlinux
> ELF32 executable
>
> Entry 00100000 first 00100000 n 4
> Segment 0 load 00100000 size 1097900 mediaptr 0x1000
> Segment 1 load 0020e000 size 150520 mediaptr 0x10e000
> Segment 2 load 00234000 size 55900 mediaptr 0x133000
> Segment 3 load 00244000 size 8192 mediaptr 0x141000
> branching to kernel entry point 0x00100000
> Set default PSW W bit to 0
> PDC Console Initialized
> The 32-bit Kernel has started...
> Enabled FP coprocessor
> Free memory starts at: 0xc026e000
> start_parisc(0x504d6c,0x504d6c,0x0,0x0)
> PALO command line: 'HOME=/ TERM=LINUX root=/dev/nfs
nfsroot=192.68.1.11:/tftpboot/pancho '
> PALO initrd 0-0
> model 00006000 00000481 00000000 00000000 77564139 00000000 00000004
00000072 00000072
> vers 0000000a
> CPUID vers 0 rev 0
> Searching for devices in PDC firmware... processor hpa 0xfffbe000
> an older box...
> Found devices
From rhirst@linuxcare.com Thu, 16 Nov 2000 12:00:47 +0000
Date: Thu, 16 Nov 2000 12:00:47 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: Single-stepping
Hi John,
On Thu, Nov 16, 2000 at 02:01:12AM -0700, John Marvin wrote:
> Just to be clear, it is impossible to restore the entire PSW without
> an RFI. So, I assume you are referring to the system mask subset of
> the PSW that can be manipulated by the ssm,rsm, and mtsm instructions.
Yes, mtsm in this case.
> You mention restoring from the task's saved registers, but we currently
> do not save the system mask during a syscall (because it should be the
> same for all processes). Have you added code to do that also? If not,
Yes I have.
> you are restoring from whatever the state was at the last interruption.
> Which in this case works (since the R bit state will be changed
> by another process while the debugged process is suspended, this should
> guarantee that the R bit state is up to date), but it seems a little ugly.
> In my opinion, you should just be checking a bit in the ptrace flags
> in the task structure, and setting the R bit with an ssm instruction
> based on that.
Sounds better, I'll look in to it.
> > Nullified instructions are handled by the controlling process
> > manually moving the childs IAOQ over the instruction without
> > actually setting it running, because the recovery counter isn't
> > decremented for nullified instructions.
Sorry, I worded that very badly. The code that moves the childs
IAOQ on is in the kernel, invoked as a result of the controlling
process calling ptrace(PTRACE_SINGLESTEP...) when the childs N
bit is set.
> Does this code properly handle branches in the delay slot of another
> branch? (you need to make sure you are not advancing the queues by just
> adding 4 to each element). One concern I have about this method is that
Current code does
/* Nullified, just crank over the queue. */
task_regs(child)->iaoq[0] = task_regs(child)->iaoq[1];
task_regs(child)->iasq[0] = task_regs(child)->iasq[1];
task_regs(child)->iaoq[1] = task_regs(child)->iaoq[0] + 4;
Does that look right to you?
> I wonder if it is necessary to do this. So what if we don't stop on the
> nullified instruction. Since it is nullified, it doesn't actually do
> anything, so why does the user have to see it, i.e. just let the recovery
> counter trap happen on the next truly executed instruction (i.e. the
> debugger performs a "double step" in this case). Am I missing something
> here?
I don't see why we really need to stop on a nullified instruction, but
I'll wait for Alan to comment as he wrote this initially.
> 1) When single stepping over a syscall, when do you actually stop the
> single stepping and execute the syscall? Hopefully you are not
> allowing single stepping after the gate instruction on the gateway
> page (and returning control to a non privileged debugging process).
> The recovery counter trap should detect when the user code gets
> to the gateway page.
At the moment my test harness notes IAOQ=0x100 and stops single stepping,
but obviously the kernel needs to enforce that.
> 2) Does your solution properly handle single stepping into and out of
> a signal handler? Note that the debugger will trap the signal as part
> of this process. Since the return is handled through a hidden syscall
> you may not have to do anything special here.
Havn't looked at signal handling yet.
> Note that HP-UX does not use the recovery counter for single stepping. I
Thanks for the description of how HP-UX does it. I'll stick with
the recovery counter for now as it does seem to be basically working.
I'll also try to ensure that it is completely encapsulated within the kernel
so it is less painful to change later, if need be.
Thanks,
Richard
From rhirst@linuxcare.com Thu, 16 Nov 2000 12:09:57 +0000
Date: Thu, 16 Nov 2000 12:09:57 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] Single-stepping
On Wed, Nov 15, 2000 at 02:49:02PM -0500, John David Anglin wrote:
> > I've been helping Alan Modra out with kernel changes to support
> > single stepping for gdb. Paul Bame suggested I bounced our ideas
> > off you in case you (or anyone else) had any comments. I havn't
> > actually committed my changes yet.
> >
> > The basic approach is to use the recovery counter to generate
> > a trap every instruction. The scheme is complicated because a
> > suspended process may or may not return to user space via an RFI.
>
> I really don't know enough to comment on the implementation choice. Why
> did you decide on this approach as opposed to inserting breaks and
> enabling the taken branch branch trap (T)? It would appear that the recovery
> counter was intended to provide software recovery from hardware faults
> in fault tolerant systems. Possibly, Grant could comment on whether
> it is actually useful for this purpose.
Alan Modra made those early decisions, but I gather that he went
for the recovery counter because it at least appears to be rather
more straightforward.
Richard
From jsm@udlkern.fc.hp.com Thu, 16 Nov 2000 05:44:55 -0700 (MST)
Date: Thu, 16 Nov 2000 05:44:55 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: Single-stepping
Richard,
>
> Sorry, I worded that very badly. The code that moves the childs
> IAOQ on is in the kernel, invoked as a result of the controlling
> process calling ptrace(PTRACE_SINGLESTEP...) when the childs N
> bit is set.
>
Great.
> > Does this code properly handle branches in the delay slot of another
> > branch? (you need to make sure you are not advancing the queues by just
> > adding 4 to each element). One concern I have about this method is that
>
> Current code does
>
> /* Nullified, just crank over the queue. */
> task_regs(child)->iaoq[0] = task_regs(child)->iaoq[1];
> task_regs(child)->iasq[0] = task_regs(child)->iasq[1];
> task_regs(child)->iaoq[1] = task_regs(child)->iaoq[0] + 4;
>
> Does that look right to you?
>
Yes, that is the correct way to do it (I'll assume the duplicated line
is just a cut/paste error).
> > I wonder if it is necessary to do this. So what if we don't stop on the
> > nullified instruction. Since it is nullified, it doesn't actually do
> > anything, so why does the user have to see it, i.e. just let the recovery
> > counter trap happen on the next truly executed instruction (i.e. the
> > debugger performs a "double step" in this case). Am I missing something
> > here?
>
> I don't see why we really need to stop on a nullified instruction, but
> I'll wait for Alan to comment as he wrote this initially.
>
Given the above, i.e. that this is being handled in the kernel anyway, I
guess I don't really care which way this goes. Probably it is best to
do it whatever way gdb on hp-ux presents it.
> > 1) When single stepping over a syscall, when do you actually stop the
> > single stepping and execute the syscall? Hopefully you are not
> > allowing single stepping after the gate instruction on the gateway
> > page (and returning control to a non privileged debugging process).
> > The recovery counter trap should detect when the user code gets
> > to the gateway page.
>
> At the moment my test harness notes IAOQ=0x100 and stops single stepping,
> but obviously the kernel needs to enforce that.
>
You should also be checking the space. But yes, the kernel needs to enforce
this for security reasons. You should be able to do it in the recovery
counter trap handler (rather than having to test for it in the syscall
path, which affects all processes).
> > 2) Does your solution properly handle single stepping into and out of
> > a signal handler? Note that the debugger will trap the signal as part
> > of this process. Since the return is handled through a hidden syscall
> > you may not have to do anything special here.
>
> Havn't looked at signal handling yet.
>
I'm not sure that there is a real issue here or not. HP-UX has some code
for single stepping with respect to signal handlers, but I believe it may
only be necessary due to the saved state necessary as part of the iaoq
manipulation. Obviously you should test this case.
> > Note that HP-UX does not use the recovery counter for single stepping. I
>
> Thanks for the description of how HP-UX does it. I'll stick with
> the recovery counter for now as it does seem to be basically working.
> I'll also try to ensure that it is completely encapsulated within the kernel
> so it is less painful to change later, if need be.
>
Sounds ok with me. And as long as there are no corner cases, it probably
is the best solution, assuming we don't find another application for
the recovery counter.
John
From rhirst@linuxcare.com Thu, 16 Nov 2000 13:20:17 +0000
Date: Thu, 16 Nov 2000 13:20:17 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: Single-stepping
On Thu, Nov 16, 2000 at 05:44:55AM -0700, John Marvin wrote:
> > Current code does
> >
> > /* Nullified, just crank over the queue. */
> > task_regs(child)->iaoq[0] = task_regs(child)->iaoq[1];
> > task_regs(child)->iasq[0] = task_regs(child)->iasq[1];
> > task_regs(child)->iaoq[1] = task_regs(child)->iaoq[0] + 4;
> >
> > Does that look right to you?
>
> Yes, that is the correct way to do it (I'll assume the duplicated line
> is just a cut/paste error).
It's not duplicated (iaoq v. iasq).
> > At the moment my test harness notes IAOQ=0x100 and stops single stepping,
> > but obviously the kernel needs to enforce that.
> >
> You should also be checking the space. But yes, the kernel needs to enforce
> this for security reasons. You should be able to do it in the recovery
> counter trap handler (rather than having to test for it in the syscall
> path, which affects all processes).
I might come back to you on that when I've thought some more.
Thanks,
Richard
From ian.zink@maryville.com Thu, 16 Nov 2000 09:46:18 -0600
Date: Thu, 16 Nov 2000 09:46:18 -0600
From: Ian Zink ian.zink@maryville.com
Subject: [parisc-linux] I got the serial console to work...
After reading some posts, I went back to trying to use to a serial console
to no avail.. Then.. suddenly it hit me. Do'y. The cable I was using was
straight-through not Cross-over. Now I have the 712 all booted and running
Debian. Very cool, indeed. Thanks for all for your help. I think I will
still work on building a STI ISO for the fun of it :)
Thanks much,
Ian Zink, Systems Engineer
Maryville Technologies
540 Maryville Centre, Suite 300
St. Louis, MO 63141
636/519-4182
(FAX) 636/519-4141
ian.zink@maryville.com
www.maryville.com
From marteaut@esiee.fr Thu, 16 Nov 2000 18:18:58 +0100
Date: Thu, 16 Nov 2000 18:18:58 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] The config for 712 hp boxes
This is a multi-part message in MIME format.
------=_NextPart_000_00A5_01C04FF9.B01CC3A0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Hi all,
Here is the config file we use to compile a new kernel from the test10
sources. We have the console with the STI and an extra term on the serial
port.
Also, in this mail, you have the diff between our hp_keyb.c and the official
one but it gives you the ~ and the ' and others scancodes.
All this works for the 712 boxes. We were forced to reboot our box to test
the new kernel but before the uptime was 3 days and 4 hours during which we
ran dselect and did some compiling.
All the files inclosed are downloadable at
http://www.esiee.fr/~puffin/code/dl.html
Bye,
ESIEE Team
------=_NextPart_000_00A5_01C04FF9.B01CC3A0
Content-Type: application/octet-stream;
name="keyb.patch"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
filename="keyb.patch"
diff -urN linux2/drivers/char/hp_keyb.c linux/drivers/char/hp_keyb.c=0A=
--- linux2/drivers/char/hp_keyb.c Thu Nov 16 17:39:03 2000=0A=
+++ linux/drivers/char/hp_keyb.c Thu Nov 16 17:18:54 2000=0A=
@@ -70,12 +70,12 @@=0A=
#define K_F8 0x42=0A=
#define K_F9 0x43=0A=
#define K_F10 0x44=0A=
-#define K_F11 87=0A=
-#define K_F12 88=0A=
+#define K_F11 0x57=0A=
+#define K_F12 0x58=0A=
#define K_PRNT 0x54=0A=
-#define K_SCRL 70=0A=
-#define K_BRK 119=0A=
-#define K_AGR K_NONE=0A=
+#define K_SCRL 0x46=0A=
+#define K_BRK 0x77=0A=
+#define K_AGR 0x29=0A=
#define K_1 0x02=0A=
#define K_2 0x03=0A=
#define K_3 0x04=0A=
@@ -89,10 +89,10 @@=0A=
#define K_MINS 0x0c=0A=
#define K_EQLS 0x0d=0A=
#define K_BKSP 0x0e=0A=
-#define K_INS 110=0A=
-#define K_HOME 102=0A=
-#define K_PGUP 104=0A=
-#define K_NUML 69=0A=
+#define K_INS 0x6e=0A=
+#define K_HOME 0x66=0A=
+#define K_PGUP 0x68=0A=
+#define K_NUML 0x45=0A=
#define KP_SLH 0x62=0A=
#define KP_STR 0x37=0A=
#define KP_MNS 0x4a=0A=
@@ -111,13 +111,13 @@=0A=
#define K_RSBK 0x1b=0A=
#define K_ENTR 0x1c=0A=
#define K_DEL 111=0A=
-#define K_END 107=0A=
-#define K_PGDN 109=0A=
+#define K_END 0x6b=0A=
+#define K_PGDN 0x6d=0A=
#define KP_7 0x47=0A=
#define KP_8 0x48=0A=
#define KP_9 0x49=0A=
-#define KP_PLS 118=0A=
-#define K_CAPS 58=0A=
+#define KP_PLS 0x4e=0A=
+#define K_CAPS 0x3a=0A=
#define K_A 0x1e=0A=
#define K_S 0x1f=0A=
#define K_D 0x20=0A=
@@ -146,7 +146,7 @@=0A=
#define K_DOT 0x34=0A=
#define K_FSLH 0x35=0A=
#define K_RSFT 0x36=0A=
-#define K_UP 103 =0A=
+#define K_UP 0x67=0A=
#define KP_1 0x4f=0A=
#define KP_2 0x50=0A=
#define KP_3 0x51=0A=
@@ -156,11 +156,11 @@=0A=
#define K_SPCE 0x39=0A=
#define K_RALT 0x64=0A=
#define K_RCTL 0x61=0A=
-#define K_LEFT 105=0A=
-#define K_DOWN 108=0A=
-#define K_RGHT 106=0A=
-#define KP_0 82=0A=
-#define KP_DOT 83=0A=
+#define K_LEFT 0x69=0A=
+#define K_DOWN 0x6c=0A=
+#define K_RGHT 0x6a=0A=
+#define KP_0 0x52=0A=
+#define KP_DOT 0x53=0A=
=0A=
static unsigned char keycode_translate[256] =3D=0A=
{=0A=
@@ -198,7 +198,7 @@=0A=
=0A=
/* ----- the following code stolen from pc_keyb.c */=0A=
=0A=
-#ifdef CONFIG_MAGIC_SYSRQ=0A=
+=0A=
unsigned char pckbd_sysrq_xlate[128] =3D=0A=
"\000\0331234567890-=3D\177\t" /* 0x00 - 0x0f */=0A=
"qwertyuiop[]\r\000as" /* 0x10 - 0x1f */=0A=
@@ -207,7 +207,6 @@=0A=
"\206\207\210\211\212\000\000789-456+1" /* 0x40 - 0x4f */=0A=
"230\177\000\000\213\214\000\000\000\000\000\000\000\000\000\000" /* =
0x50 - 0x5f */=0A=
"\r\000/"; /* 0x60 - 0x6f */=0A=
-#endif=0A=
=0A=
/*=0A=
* Translation of escaped scancodes to keycodes.=0A=
------=_NextPart_000_00A5_01C04FF9.B01CC3A0
Content-Type: application/octet-stream;
name="config"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
filename="config"
#=0A=
# Automatically generated by make menuconfig: don't edit=0A=
#=0A=
CONFIG_PARISC=3Dy=0A=
# CONFIG_UID16 is not set=0A=
=0A=
#=0A=
# Code maturity level options=0A=
#=0A=
CONFIG_EXPERIMENTAL=3Dy=0A=
=0A=
#=0A=
# General options=0A=
#=0A=
# CONFIG_SMP is not set=0A=
# CONFIG_KWDB is not set=0A=
CONFIG_GSC=3Dy=0A=
CONFIG_IOMMU_CCIO=3Dy=0A=
CONFIG_GSC_LASI=3Dy=0A=
CONFIG_PCI=3Dy=0A=
CONFIG_GSC_DINO=3Dy=0A=
CONFIG_PCI_LBA=3Dy=0A=
CONFIG_IOSAPIC=3Dy=0A=
CONFIG_IOMMU_SBA=3Dy=0A=
=0A=
#=0A=
# Loadable module support=0A=
#=0A=
# CONFIG_MODULES is not set=0A=
=0A=
#=0A=
# General setup=0A=
#=0A=
CONFIG_NET=3Dy=0A=
# CONFIG_SYSVIPC is not set=0A=
# CONFIG_BSD_PROCESS_ACCT is not set=0A=
CONFIG_SYSCTL=3Dy=0A=
# CONFIG_BINFMT_SOM is not set=0A=
CONFIG_BINFMT_ELF=3Dy=0A=
# CONFIG_BINFMT_MISC is not set=0A=
# CONFIG_BINFMT_JAVA is not set=0A=
=0A=
#=0A=
# Parallel port support=0A=
#=0A=
CONFIG_PARPORT=3Dy=0A=
# CONFIG_PARPORT_PC is not set=0A=
# CONFIG_PARPORT_GSC is not set=0A=
# CONFIG_PARPORT_OTHER is not set=0A=
# CONFIG_PARPORT_1284 is not set=0A=
=0A=
#=0A=
# Block devices=0A=
#=0A=
# CONFIG_BLK_DEV_FD is not set=0A=
# CONFIG_BLK_DEV_XD is not set=0A=
# CONFIG_PARIDE is not set=0A=
# CONFIG_BLK_CPQ_DA is not set=0A=
# CONFIG_BLK_CPQ_CISS_DA is not set=0A=
# CONFIG_BLK_DEV_DAC960 is not set=0A=
# CONFIG_BLK_DEV_LOOP is not set=0A=
# CONFIG_BLK_DEV_NBD is not set=0A=
CONFIG_BLK_DEV_RAM=3Dy=0A=
CONFIG_BLK_DEV_RAM_SIZE=3D4096=0A=
CONFIG_BLK_DEV_INITRD=3Dy=0A=
=0A=
#=0A=
# Networking options=0A=
#=0A=
# CONFIG_PACKET is not set=0A=
# CONFIG_NETLINK is not set=0A=
# CONFIG_NETFILTER is not set=0A=
# CONFIG_FILTER is not set=0A=
# CONFIG_UNIX is not set=0A=
CONFIG_INET=3Dy=0A=
# CONFIG_IP_MULTICAST is not set=0A=
# CONFIG_IP_ADVANCED_ROUTER is not set=0A=
CONFIG_IP_PNP=3Dy=0A=
# CONFIG_IP_PNP_BOOTP is not set=0A=
# CONFIG_IP_PNP_RARP is not set=0A=
# CONFIG_NET_IPIP is not set=0A=
# CONFIG_NET_IPGRE is not set=0A=
# CONFIG_INET_ECN is not set=0A=
# CONFIG_SYN_COOKIES is not set=0A=
# CONFIG_IPV6 is not set=0A=
# CONFIG_KHTTPD is not set=0A=
# CONFIG_ATM is not set=0A=
# CONFIG_IPX is not set=0A=
# CONFIG_ATALK is not set=0A=
# CONFIG_DECNET is not set=0A=
# CONFIG_BRIDGE is not set=0A=
# CONFIG_X25 is not set=0A=
# CONFIG_LAPB is not set=0A=
# CONFIG_LLC is not set=0A=
# CONFIG_NET_DIVERT is not set=0A=
# CONFIG_ECONET is not set=0A=
# CONFIG_WAN_ROUTER is not set=0A=
# CONFIG_NET_FASTROUTE is not set=0A=
# CONFIG_NET_HW_FLOWCONTROL is not set=0A=
=0A=
#=0A=
# QoS and/or fair queueing=0A=
#=0A=
# CONFIG_NET_SCHED is not set=0A=
=0A=
#=0A=
# SCSI support=0A=
#=0A=
CONFIG_SCSI=3Dy=0A=
CONFIG_BLK_DEV_SD=3Dy=0A=
CONFIG_SD_EXTRA_DEVS=3D40=0A=
# CONFIG_CHR_DEV_ST is not set=0A=
# CONFIG_BLK_DEV_SR is not set=0A=
CONFIG_CHR_DEV_SG=3Dy=0A=
# CONFIG_SCSI_MULTI_LUN is not set=0A=
# CONFIG_SCSI_CONSTANTS is not set=0A=
=0A=
#=0A=
# SCSI low-level drivers=0A=
#=0A=
CONFIG_SCSI_LASI=3Dy=0A=
CONFIG_SCSI_ZALON=3Dy=0A=
CONFIG_SCSI_SYM53C8XX=3Dy=0A=
CONFIG_SCSI_NCR53C8XX_DEFAULT_TAGS=3D8=0A=
CONFIG_SCSI_NCR53C8XX_MAX_TAGS=3D32=0A=
CONFIG_SCSI_NCR53C8XX_SYNC=3D20=0A=
# CONFIG_SCSI_NCR53C8XX_PROFILE is not set=0A=
# CONFIG_SCSI_NCR53C8XX_IOMAPPED is not set=0A=
=0A=
#=0A=
# Network device support=0A=
#=0A=
CONFIG_NETDEVICES=3Dy=0A=
CONFIG_LASI_82596=3Dy=0A=
=0A=
#=0A=
# ARCnet devices=0A=
#=0A=
# CONFIG_ARCNET is not set=0A=
# CONFIG_DUMMY is not set=0A=
# CONFIG_BONDING is not set=0A=
# CONFIG_EQUALIZER is not set=0A=
# CONFIG_TUN is not set=0A=
# CONFIG_NET_SB1000 is not set=0A=
=0A=
#=0A=
# Ethernet (10 or 100Mbit)=0A=
#=0A=
CONFIG_NET_ETHERNET=3Dy=0A=
# CONFIG_NET_VENDOR_3COM is not set=0A=
# CONFIG_LANCE is not set=0A=
# CONFIG_NET_VENDOR_SMC is not set=0A=
# CONFIG_NET_VENDOR_RACAL is not set=0A=
# CONFIG_AT1700 is not set=0A=
# CONFIG_DEPCA is not set=0A=
# CONFIG_HP100 is not set=0A=
# CONFIG_NET_ISA is not set=0A=
CONFIG_NET_PCI=3Dy=0A=
# CONFIG_PCNET32 is not set=0A=
# CONFIG_ADAPTEC_STARFIRE is not set=0A=
# CONFIG_AC3200 is not set=0A=
# CONFIG_APRICOT is not set=0A=
# CONFIG_CS89x0 is not set=0A=
# CONFIG_DE4X5 is not set=0A=
CONFIG_TULIP=3Dy=0A=
# CONFIG_DGRS is not set=0A=
# CONFIG_DM9102 is not set=0A=
# CONFIG_EEPRO100 is not set=0A=
# CONFIG_LNE390 is not set=0A=
# CONFIG_NATSEMI is not set=0A=
# CONFIG_NE2K_PCI is not set=0A=
# CONFIG_NE3210 is not set=0A=
# CONFIG_ES3210 is not set=0A=
# CONFIG_RTL8129 is not set=0A=
# CONFIG_8139TOO is not set=0A=
# CONFIG_SIS900 is not set=0A=
# CONFIG_EPIC100 is not set=0A=
# CONFIG_SUNDANCE is not set=0A=
# CONFIG_TLAN is not set=0A=
# CONFIG_VIA_RHINE is not set=0A=
# CONFIG_WINBOND_840 is not set=0A=
# CONFIG_NET_POCKET is not set=0A=
=0A=
#=0A=
# Ethernet (1000 Mbit)=0A=
#=0A=
# CONFIG_ACENIC is not set=0A=
# CONFIG_HAMACHI is not set=0A=
# CONFIG_YELLOWFIN is not set=0A=
# CONFIG_SK98LIN is not set=0A=
# CONFIG_FDDI is not set=0A=
# CONFIG_HIPPI is not set=0A=
# CONFIG_PLIP is not set=0A=
# CONFIG_PPP is not set=0A=
# CONFIG_SLIP is not set=0A=
=0A=
#=0A=
# Wireless LAN (non-hamradio)=0A=
#=0A=
# CONFIG_NET_RADIO is not set=0A=
=0A=
#=0A=
# Token Ring devices=0A=
#=0A=
# CONFIG_TR is not set=0A=
# CONFIG_NET_FC is not set=0A=
# CONFIG_RCPCI is not set=0A=
# CONFIG_SHAPER is not set=0A=
=0A=
#=0A=
# Wan interfaces=0A=
#=0A=
# CONFIG_WAN is not set=0A=
=0A=
#=0A=
# Character devices=0A=
#=0A=
CONFIG_VT=3Dy=0A=
CONFIG_VT_CONSOLE=3Dy=0A=
CONFIG_GSC_PS2=3Dy=0A=
# CONFIG_HIL is not set=0A=
CONFIG_SERIAL=3Dy=0A=
# CONFIG_SERIAL_CONSOLE is not set=0A=
CONFIG_SERIAL_GSC=3Dy=0A=
# CONFIG_SERIAL_EXTENDED is not set=0A=
# CONFIG_SERIAL_NONSTANDARD is not set=0A=
CONFIG_UNIX98_PTYS=3Dy=0A=
CONFIG_UNIX98_PTY_COUNT=3D256=0A=
# CONFIG_PRINTER is not set=0A=
# CONFIG_PPDEV is not set=0A=
=0A=
#=0A=
# I2C support=0A=
#=0A=
# CONFIG_I2C is not set=0A=
=0A=
#=0A=
# Mice=0A=
#=0A=
# CONFIG_BUSMOUSE is not set=0A=
# CONFIG_MOUSE is not set=0A=
=0A=
#=0A=
# Joysticks=0A=
#=0A=
# CONFIG_JOYSTICK is not set=0A=
# CONFIG_QIC02_TAPE is not set=0A=
=0A=
#=0A=
# Watchdog Cards=0A=
#=0A=
# CONFIG_WATCHDOG is not set=0A=
CONFIG_GENRTC=3Dy=0A=
# CONFIG_INTEL_RNG is not set=0A=
# CONFIG_NVRAM is not set=0A=
# CONFIG_RTC is not set=0A=
# CONFIG_DTLK is not set=0A=
# CONFIG_R3964 is not set=0A=
# CONFIG_APPLICOM is not set=0A=
=0A=
#=0A=
# Ftape, the floppy tape device driver=0A=
#=0A=
# CONFIG_FTAPE is not set=0A=
# CONFIG_AGP is not set=0A=
# CONFIG_DRM is not set=0A=
=0A=
#=0A=
# File systems=0A=
#=0A=
# CONFIG_QUOTA is not set=0A=
# CONFIG_AUTOFS_FS is not set=0A=
# CONFIG_AUTOFS4_FS is not set=0A=
# CONFIG_ADFS_FS is not set=0A=
# CONFIG_ADFS_FS_RW is not set=0A=
# CONFIG_AFFS_FS is not set=0A=
# CONFIG_HFS_FS is not set=0A=
# CONFIG_BFS_FS is not set=0A=
# CONFIG_FAT_FS is not set=0A=
# CONFIG_MSDOS_FS is not set=0A=
# CONFIG_UMSDOS_FS is not set=0A=
# CONFIG_VFAT_FS is not set=0A=
# CONFIG_EFS_FS is not set=0A=
# CONFIG_JFFS_FS is not set=0A=
# CONFIG_CRAMFS is not set=0A=
# CONFIG_RAMFS is not set=0A=
CONFIG_ISO9660_FS=3Dy=0A=
# CONFIG_JOLIET is not set=0A=
# CONFIG_MINIX_FS is not set=0A=
# CONFIG_NTFS_FS is not set=0A=
# CONFIG_NTFS_RW is not set=0A=
# CONFIG_HPFS_FS is not set=0A=
CONFIG_PROC_FS=3Dy=0A=
# CONFIG_DEVFS_FS is not set=0A=
# CONFIG_DEVFS_MOUNT is not set=0A=
# CONFIG_DEVFS_DEBUG is not set=0A=
# CONFIG_DEVPTS_FS is not set=0A=
# CONFIG_QNX4FS_FS is not set=0A=
# CONFIG_QNX4FS_RW is not set=0A=
# CONFIG_ROMFS_FS is not set=0A=
CONFIG_EXT2_FS=3Dy=0A=
# CONFIG_SYSV_FS is not set=0A=
# CONFIG_SYSV_FS_WRITE is not set=0A=
# CONFIG_UDF_FS is not set=0A=
# CONFIG_UDF_RW is not set=0A=
# CONFIG_UFS_FS is not set=0A=
# CONFIG_UFS_FS_WRITE is not set=0A=
=0A=
#=0A=
# Network File Systems=0A=
#=0A=
# CONFIG_CODA_FS is not set=0A=
CONFIG_NFS_FS=3Dy=0A=
# CONFIG_NFS_V3 is not set=0A=
CONFIG_ROOT_NFS=3Dy=0A=
# CONFIG_NFSD is not set=0A=
# CONFIG_NFSD_V3 is not set=0A=
CONFIG_SUNRPC=3Dy=0A=
CONFIG_LOCKD=3Dy=0A=
# CONFIG_SMB_FS is not set=0A=
# CONFIG_NCP_FS is not set=0A=
# CONFIG_NCPFS_PACKET_SIGNING is not set=0A=
# CONFIG_NCPFS_IOCTL_LOCKING is not set=0A=
# CONFIG_NCPFS_STRONG is not set=0A=
# CONFIG_NCPFS_NFS_NS is not set=0A=
# CONFIG_NCPFS_OS2_NS is not set=0A=
# CONFIG_NCPFS_SMALLDOS is not set=0A=
# CONFIG_NCPFS_MOUNT_SUBDIR is not set=0A=
# CONFIG_NCPFS_NDS_DOMAINS is not set=0A=
# CONFIG_NCPFS_NLS is not set=0A=
# CONFIG_NCPFS_EXTRAS is not set=0A=
=0A=
#=0A=
# Partition Types=0A=
#=0A=
# CONFIG_PARTITION_ADVANCED is not set=0A=
CONFIG_MSDOS_PARTITION=3Dy=0A=
# CONFIG_NLS is not set=0A=
=0A=
#=0A=
# Sound Drivers=0A=
#=0A=
# CONFIG_SOUND is not set=0A=
=0A=
#=0A=
# Console drivers=0A=
#=0A=
=0A=
#=0A=
# Frame-buffer support=0A=
#=0A=
# CONFIG_FB is not set=0A=
CONFIG_STI_CONSOLE=3Dy=0A=
CONFIG_DUMMY_CONSOLE=3Dy=0A=
=0A=
#=0A=
# Kernel hacking=0A=
#=0A=
CONFIG_MAGIC_SYSRQ=3Dy=0A=
------=_NextPart_000_00A5_01C04FF9.B01CC3A0--
From grundler@cup.hp.com Thu, 16 Nov 2000 10:10:23 -0800
Date: Thu, 16 Nov 2000 10:10:23 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] The config for 712 hp boxes
"Thomas Marteau" wrote:
> Hi all,
>
> Here is the config file we use to compile a new kernel from the test10
> sources. We have the console with the STI and an extra term on the serial
> port.
Thomas - excellent - thanks!
> Also, in this mail, you have the diff between our hp_keyb.c and the official
> one but it gives you the ~ and the ' and others scancodes.
I noticed this mail was directed to me - but I can't take ownership
of this code. I have too many other things going on.
Helge - can you commit this fix too?
thanks,
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From deller@gmx.de Thu, 16 Nov 2000 19:25:49 +0100
Date: Thu, 16 Nov 2000 19:25:49 +0100
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] The config for 712 hp boxes
On Thursday 16 November 2000 19:10, Grant Grundler wrote:
> "Thomas Marteau" wrote:
> > Hi all,
> >
> > Here is the config file we use to compile a new kernel from the test10
> > sources. We have the console with the STI and an extra term on the serial
> > port.
>
> Thomas - excellent - thanks!
>
> > Also, in this mail, you have the diff between our hp_keyb.c and the
official
> > one but it gives you the ~ and the ' and others scancodes.
>
> I noticed this mail was directed to me - but I can't take ownership
> of this code. I have too many other things going on.
>
> Helge - can you commit this fix too?
Ok. I will test & evtl. commit it today.
>
> thanks,
> grant
>
> Grant Grundler
> Unix Systems Enablement Lab
> +1.408.447.7253
>
> ---------------------------------------------------------------------------
> To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com with
> `unsubscribe' as the subject.
From frank_rowand@mvista.com Thu, 16 Nov 2000 11:00:48 -0800
Date: Thu, 16 Nov 2000 11:00:48 -0800
From: Frank Rowand frank_rowand@mvista.com
Subject: Single-stepping
John Marvin wrote:
>
> Richard,
>
> >
> > Sorry, I worded that very badly. The code that moves the childs
> > IAOQ on is in the kernel, invoked as a result of the controlling
> > process calling ptrace(PTRACE_SINGLESTEP...) when the childs N
> > bit is set.
> >
>
> Great.
>
> > > Does this code properly handle branches in the delay slot of another
> > > branch? (you need to make sure you are not advancing the queues by just
> > > adding 4 to each element). One concern I have about this method is that
> >
> > Current code does
> >
> > /* Nullified, just crank over the queue. */
> > task_regs(child)->iaoq[0] = task_regs(child)->iaoq[1];
> > task_regs(child)->iasq[0] = task_regs(child)->iasq[1];
> > task_regs(child)->iaoq[1] = task_regs(child)->iaoq[0] + 4;
> >
> > Does that look right to you?
> >
>
> Yes, that is the correct way to do it (I'll assume the duplicated line
> is just a cut/paste error).
If iaoq[0] contains a branch, iaoq[1] is in the delay slot. The instruction
executed after iaoq[1] would then typically _not_ be iaoq[0] + 4 (the next
instruction would be the target of the branch at iaoq[0]).
> Sounds ok with me. And as long as there are no corner cases, it probably
> is the best solution, assuming we don't find another application for
> the recovery counter.
The recovery counter is very useful for performance measurement tools to
understand the cycles per instruction of a code path. (Using the recovery
counter for the debugger doesn't preclude using it for performance tools -
you just can't easily use it for both purposes at the same instant in time.)
> John
-Frank
--
Frank Rowand
MontaVista Software, Inc
From rhirst@linuxcare.com Thu, 16 Nov 2000 20:28:42 +0000
Date: Thu, 16 Nov 2000 20:28:42 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: Single-stepping
On Thu, Nov 16, 2000 at 11:00:48AM -0800, Frank Rowand wrote:
> John Marvin wrote:
> > > > Does this code properly handle branches in the delay slot of another
> > > > branch? (you need to make sure you are not advancing the queues by just
> > > > adding 4 to each element). One concern I have about this method is that
> > >
> > > Current code does
> > >
> > > /* Nullified, just crank over the queue. */
> > > task_regs(child)->iaoq[0] = task_regs(child)->iaoq[1];
> > > task_regs(child)->iasq[0] = task_regs(child)->iasq[1];
> > > task_regs(child)->iaoq[1] = task_regs(child)->iaoq[0] + 4;
> > >
> > > Does that look right to you?
> > >
> >
> > Yes, that is the correct way to do it (I'll assume the duplicated line
> > is just a cut/paste error).
>
> If iaoq[0] contains a branch, iaoq[1] is in the delay slot. The instruction
> executed after iaoq[1] would then typically _not_ be iaoq[0] + 4 (the next
> instruction would be the target of the branch at iaoq[0]).
But the above code is only executed if the current instruction is
nullified. In your example, the branch in iaoq[0] would be
nullified and therefore never taken.
Richard
From deller@gmx.de Thu, 16 Nov 2000 22:28:50 +0100
Date: Thu, 16 Nov 2000 22:28:50 +0100
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] The config for 712 hp boxes
On Thursday 16 November 2000 18:18, Thomas Marteau wrote:
> Hi all,
>
> [....]
> Also, in this mail, you have the diff between our hp_keyb.c and the official
> one but it gives you the ~ and the ' and others scancodes.
> [....]
Hi ESIEE-Team !
Thanks for your patch !
I committed your patch into CVS and just tweaked it a little for
CONFIG_MAGIC_SYSRQ. I hope this is ok for you ?
If your time permits, I would like to ask, if you maybe also want to look at
getting the CAPS-Lock-, Scroll-Lock- & Num-Lock-LED's working ?
Furthermore you mention in your project-history on your homepage something
about "keyboard and soundchip problems" (Oct. 24). I'm asking just out of
curiousity: Did you thought about programming the on-board sound-chips (which
I believe would be a hard job.) ?
> Bye,
> ESIEE Team
Best regards,
Helge Deller.
From taggart@carmen.fc.hp.com Thu, 16 Nov 2000 19:25:26 -0700
Date: Thu, 16 Nov 2000 19:25:26 -0700
From: Matt Taggart taggart@carmen.fc.hp.com
Subject: [parisc-linux] glibc merged to 2.2
Today Paul Bame and I merged/updated the puffin.external.hp.com glibc cvs to
glibc 2.2. After the merge I cleaned up the differences between our cvs and
2.2, mostly comment differences and formatting changes. With the exception of
the setjmp problem mentioned in,
http://puffin.external.hp.com/mailing-lists/parisc-linux/2000/11-Nov/0091.html
and the fact that we have/need newer "scripts/config.guess" and
"scripts/config.sub", we are completely sync'ed up with glibc 2.2.
--
Matt Taggart
taggart@fc.hp.com
From deller@gmx.de Fri, 17 Nov 2000 15:34:09 +0100
Date: Fri, 17 Nov 2000 15:34:09 +0100
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] The config for 712 hp boxes
[CC'ed the list, so maybe someone else can comment too....]
On Friday 17 November 2000 11:45, Thomas Marteau wrote:
> Hi Helge,
Hi,
>
>
> ----- Original Message -----
> From: Helge Deller
> To: Thomas Marteau
> Cc:
> Sent: Thursday, November 16, 2000 10:28 PM
> Subject: Re: [parisc-linux] The config for 712 hp boxes
>
>
> > On Thursday 16 November 2000 18:18, Thomas Marteau wrote:
> > > Hi all,
> > >
> > > [....]
> > > Also, in this mail, you have the diff between our hp_keyb.c and the
> official
> > > one but it gives you the ~ and the ' and others scancodes.
> > > [....]
> >
> > Hi ESIEE-Team !
> >
> > Thanks for your patch !
> >
> > I committed your patch into CVS and just tweaked it a little for
> > CONFIG_MAGIC_SYSRQ. I hope this is ok for you ?
> We didn't really understand why pckbd_sysrq_xlate was here for.
> If you can explain to us, it will be fine.
The best documentation you can find is in linux/Documentation/sysrq.txt.
The copy of pckbd_sysrq_xlate[] is a simple implementation of
keyboard-scancodes to chars, which is used by drivers/char/keyboard.c to call
handle_sysrq() from /drivers/char/sysrq.c.
This means, that you can press Alt-PrintScr (=> SysRQ) and a char at the
keyboard simultaniously and your machine for example syncs the disks, mounts
all discs read-only or reboots your machine immediately.
> > If your time permits, I would like to ask, if you maybe also want to look
> at
> > getting the CAPS-Lock-, Scroll-Lock- & Num-Lock-LED's working ?
> Normally, you can see yours leds switching on/off but we did not init them.
> So they are like boot admin switch them...
^^^^^^^ ^^^^^ ^^^
Sorry ?
> But we would like to know how and where we have to init them
> because we know how to. We have made test and it was working...
I'm not sure, but I think you should check the code in lasikbd_leds() in
hp_psaux.c.
The initialization should maybe be done in the same file in the function
lasi_ps2_register() before calling register_kbd_ops().
>
> > Furthermore you mention in your project-history on your homepage something
> > about "keyboard and soundchip problems" (Oct. 24). I'm asking just out of
> > curiousity: Did you thought about programming the on-board sound-chips
> (which
> > I believe would be a hard job.) ?
> Yes, it is in our internal todo list, if we have time to. Because we look at
> it with Thierry, we saw that, for LASI,
> you have an interface and then the chip to configure. But it is very
> interesting...
Yes it is, and it would be great if you worked on that too !
>
> Also, we want to rule the power leds but they are different if the box is
> 712 or B132 or ... So our question is how to implement
> this kind of initialsation and deinitialisation in a linux kernel and in a
> second time, how to do for differents kind of boxes.
As far as I know / have heard, there are only two types of LED's (but I may
be wrong here !). One is where the LED's are organized in a row on the front
panel (as my local machines here), and the other one, where you have
(LED/)LCD-Numbers on your front panel ?
They differ how to be accessed and where the controlling registers are
located. One method to distiguish them is maybe to place the special
initialisation code into drivers/gsc/lasi.c and asp.c, depending on the
machine and which method needs to be used.
Since I don't have any real documentation or knowledge of the programming of
the LEDs (beside of the PDC-calls I placed into setup.c), maybe someone else
can better comment on that or give some documentation ?
>
> Thanks for your answers, Helge
> Bye,
> ESIEE Team
Best regards,
Helge Deller.
From Arnaud.ATOCH@oecd.org Fri, 17 Nov 2000 16:30:42 +0100
Date: Fri, 17 Nov 2000 16:30:42 +0100
From: Arnaud.ATOCH@oecd.org Arnaud.ATOCH@oecd.org
Subject: [parisc-linux] Booting 712 with CD 0.5
Is the following output the normal behavior of Linux-PARisc on a 712/100 ?
kmem_create: Forcing size word alignment - nfs_fh
kernel BUG at pci-dma.c:391!
kernel BUG at pci-dma.c:400!
kernel BUG at pci-dma.c:391!
kernel BUG at pci-dma.c:400!
kernel BUG at pci-dma.c:391!
kernel BUG at pci-dma.c:400!
VFS: Disk change detected on device sr(11,0)
kernel BUG at pci-dma.c:391!
kernel BUG at pci-dma.c:400!
kernel BUG at pci-dma.c:391!
kernel BUG at pci-dma.c:400!
kernel BUG at pci-dma.c:391!
kernel BUG at pci-dma.c:400!
ISO 9660 Extensions: RRIP_1991A
VFS: Mounted root (iso9660 filesystem) readonly.
INIT: version 2.78 booting
INIT: Entering runlevel: 2
Setting up /tmp ramdisk4096+0 records in
4096+0 records out
mke2fs 1.19, 13-Jul-2000 for EXT2 FS 0.5b, 95/08/09
Filesystem label=
OS type: Linux
Block size=1024 (log=0)
Fragment size=1024 (log=0)
1024 inodes, 4096 blocks
0 blocks (0.00%) reserved for the super user
First data block=1
1 block group
8192 blocks per group, 8192 fragments per group
1024 inodes per group
Writing inode tables: done
Writing superblocks and filesystem accounting information: done
INIT: Id "T0" respawni
INIT: no more processe
From rhirst@linuxcare.com Fri, 17 Nov 2000 15:39:55 +0000
Date: Fri, 17 Nov 2000 15:39:55 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] SEGV signal handling bug (dynamic linking)
Hi,
Don't know if anyone expects this to work yet or not, but:
------------------------- cut -----------------------------
#include
#include
#include
#include
#include
#include
#include
#include
#include
char *mem;
void sig_handler(int sig)
{
int res;
printf("Trapped!!!\n");
res = mprotect(mem, 4096, PROT_READ|PROT_WRITE);
if (res < 0) {
perror("mprotect");
exit(1);
}
}
void install_handlers(void)
{
struct sigaction act;
memset(&act, 0, sizeof(act));
act.sa_handler = sig_handler;
sigaction(SIGSEGV, &act, NULL);
}
int main(int argc, char **argv)
{
int res;
mem = malloc(8192);
if (mem == NULL) {
perror("malloc");
exit(1);
}
mem = (char *)(((int)mem + 4095) & ~0x0fff);
res = mprotect(mem, 4096, PROT_READ);
if (res < 0) {
perror("mprotect");
exit(1);
}
install_handlers();
write(1, "Going\n", 6);
mem[24] = 17;
write(1, "Gone\n", 5);
return 0;
}
------------------------- cut -----------------------------
generates:
Going
Bus error
plus the following on the console:
do_page_fault() pid=167 command='ch'
YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000001001111111100001011
r0-3 00000000 fffff000 0000166f 00002944
r4-7 40138c38 2001fd8c 00002852 00000001
r8-11 00002862 0008b010 0009c290 0009cbf0
r12-15 00000000 00000000 0009cb50 00000000
r16-19 00000000 00000001 0000b71b 00000011
r20-23 00004000 40041fcc 40041fcc 00000008
r24-27 00000006 00001000 00000001 0000280c
r28-31 00000006 00000020 20020140 40041fd7
sr0-4 00000000 00000003 00000000 0000000a
sr4-8 0000000a 0000000a 0000000a 0000000a
IASQ: 0000000a 0000000a IAOQ: 0000167b 0000167f
IIR: 6293002e ISR: 0000000a IOR: 00004017
ORIG_R28: 00002880
!!die_if_kernel: ch(167): Unaligned data reference 28
YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000011001111111100001011
r0-3 00000000 fffff000 20020140 00002944
r4-7 40138c38 2001fd8c 00002852 00000001
r8-11 00002862 0008b010 0009c290 0009cbf0
r12-15 00000000 00000000 0009cb50 00000000
r16-19 00000000 00000001 0000b71b 00000000
r20-23 0000289f 40041fcc 40041fcc 00000008
r24-27 200201d0 20020150 0000000b 0000280c
r28-31 00000006 00000020 200203c0 40041fd7
sr0-4 00000000 00000003 00000000 0000000a
sr4-8 0000000a 0000000a 0000000a 0000000a
IASQ: 0000000a 0000000a IAOQ: 0000289b 0000289b
IIR: 0e801096 ISR: 0000000a IOR: 0000289f
ORIG_R28: 00002880
The first do_page_fault() is fine, it is the 'mem[24] = 17' line,
but the second isn't. The corresponding code is at the end of
.plt:
2898: 0e 80 10 96 ldw 0(sr0,r20),r22
289c: ea c0 c0 00 bv r0(r22)
28a0: 0e 88 10 95 ldw 4(sr0,r20),r21
28a4: ea 9f 1f dd b,l 2898 <__DTOR_END__+0x74>,r20
28a8: d6 80 1c 1e depwi 0,31,2,r20
28ac: 00 c0 ff ee # c0ffee
28b0: de ad be ef #deadbeef
However, if I make it statically linked, it works fine, giving:
Going
Trapped!!!
Gone
Richard
From bri@mojo.calyx.net Fri, 17 Nov 2000 10:49:22 -0500 (EST)
Date: Fri, 17 Nov 2000 10:49:22 -0500 (EST)
From: Brian S. Julin bri@mojo.calyx.net
Subject: [parisc-linux] debian archive beware libpam
JFYI to save other people some time. Before firing up
"apt-get upgrade" you'll be best off putting a hold on all the
libpam packages, as the 0.72-12 version on ftp.us.debian.org
fails on a bad symbol in libpam-misc, which locks you out
of the box except if you boot single.
--
Brian S. Julin
From grundler@cup.hp.com Fri, 17 Nov 2000 08:38:24 -0800
Date: Fri, 17 Nov 2000 08:38:24 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] Booting 712 with CD 0.5
Arnaud.ATOCH@oecd.org wrote:
> Is the following output the normal behavior of Linux-PARisc on a 712/100 ?
>
> kmem_create: Forcing size word alignment - nfs_fh
> kernel BUG at pci-dma.c:391!
> kernel BUG at pci-dma.c:400!
The bug is real. A driver is calling pci_map_single() and pci_unmap_single()
without specifying which direction the DMA is going. Cache flushing on 712
(PCXL) systems needs to know. I'll assume it's the SCSI driver since
it looks like we are talking to a SCSI CD-ROM.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From drepper@redhat.com 17 Nov 2000 09:09:10 -0800
Date: 17 Nov 2000 09:09:10 -0800
From: Ulrich Drepper drepper@redhat.com
Subject: [parisc-linux] SEGV signal handling bug (dynamic linking)
Richard Hirst writes:
> mem = malloc(8192);
> if (mem == NULL) {
> perror("malloc");
> exit(1);
> }
> mem = (char *)(((int)mem + 4095) & ~0x0fff);
> res = mprotect(mem, 4096, PROT_READ);
Read the Unix standard:
The behavior of this function is unspecified if the mapping was not
established by a call to mmap().
--
---------------. ,-. 1325 Chesapeake Terrace
Ulrich Drepper \ ,-------------------' \ Sunnyvale, CA 94089 USA
Red Hat `--' drepper at redhat.com `------------------------
From grundler@cup.hp.com Fri, 17 Nov 2000 09:09:04 -0800 (PST)
Date: Fri, 17 Nov 2000 09:09:04 -0800 (PST)
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] pci-dma.c BUG(391/400) PATCH
Hi all,
The following patch tries to point a finger at the offender
rather than just call attention to a problem in the service.
I don't have a PCXL/L2 machine setup right now.
Could someone test commit this patch for me?
thanks,
grant
grundler <505>cvs diff arch/parisc/kernel/pci-dma.c
Index: arch/parisc/kernel/pci-dma.c
===================================================================
RCS file: /home/cvs/parisc/linux/arch/parisc/kernel/pci-dma.c,v
retrieving revision 1.16
diff -u -p -r1.16 pci-dma.c
--- pci-dma.c 2000/11/08 06:20:27 1.16
+++ pci-dma.c 2000/11/17 17:04:22
@@ -387,8 +387,10 @@ static void pa11_dma_free_consistent (st
static dma_addr_t pa11_dma_map_single(struct pci_dev *dev, void *addr, size_t size, int direction)
{
- if (direction == PCI_DMA_NONE)
- BUG();
+ if (direction == PCI_DMA_NONE) {
+ printk(KERN_ERR "pa11_dma_map_single(PCI_DMA_NONE) called by %p\n", __builtin_return_address(0));
+ BUG();
+ }
flush_kernel_dcache_range((unsigned long) addr, size);
return virt_to_phys(addr);
@@ -396,8 +398,10 @@ static dma_addr_t pa11_dma_map_single(st
static void pa11_dma_unmap_single(struct pci_dev *dev, dma_addr_t dma_handle, size_t size, int direction)
{
- if (direction == PCI_DMA_NONE)
- BUG();
+ if (direction == PCI_DMA_NONE) {
+ printk(KERN_ERR "pa11_dma_unmap_single(PCI_DMA_NONE) called by %p\n", __builtin_return_address(0));
+ BUG();
+ }
if (direction == PCI_DMA_TODEVICE)
return;
From Arnaud.ATOCH@oecd.org Fri, 17 Nov 2000 18:10:52 +0100
Date: Fri, 17 Nov 2000 18:10:52 +0100
From: Arnaud.ATOCH@oecd.org Arnaud.ATOCH@oecd.org
Subject: [parisc-linux] Booting 712 with CD 0.5
Thanks for the enlightment.
Should I try a new CD-ROM or should I wait untill next ISO image is created
?
IS the shutdown also due to this bug ?
Setting up /tmp ramdisk4096+0 records in
4096+0 records out
mke2fs 1.19, 13-Jul-2000 for EXT2 FS 0.5b, 95/08/09
Filesystem label=
OS type: Linux
Block size=1024 (log=0)
Fragment size=1024 (log=0)
1024 inodes, 4096 blocks
0 blocks (0.00%) reserved for the super user
First data block=1
1 block group
8192 blocks per group, 8192 fragments per group
1024 inodes per group
Writing inode tables: done
Writing superblocks and filesystem accounting information: done
INIT: Id "T0" respawni
INIT: no more processe
-----Original Message-----
From: Grant Grundler [mailto:grundler@cup.hp.com]
Sent: Friday, November 17, 2000 17:38
To: Arnaud.ATOCH@oecd.org
Cc: parisc-linux@puffin.external.hp.com
Subject: Re: [parisc-linux] Booting 712 with CD 0.5
Arnaud.ATOCH@oecd.org wrote:
> Is the following output the normal behavior of Linux-PARisc on a 712/100 ?
>
> kmem_create: Forcing size word alignment - nfs_fh
> kernel BUG at pci-dma.c:391!
> kernel BUG at pci-dma.c:400!
The bug is real. A driver is calling pci_map_single() and pci_unmap_single()
without specifying which direction the DMA is going. Cache flushing on 712
(PCXL) systems needs to know. I'll assume it's the SCSI driver since
it looks like we are talking to a SCSI CD-ROM.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From rhirst@linuxcare.com Fri, 17 Nov 2000 17:38:18 +0000
Date: Fri, 17 Nov 2000 17:38:18 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] SEGV signal handling bug (dynamic linking)
On Fri, Nov 17, 2000 at 09:09:10AM -0800, Ulrich Drepper wrote:
> Richard Hirst writes:
>
> > mem = malloc(8192);
> > if (mem == NULL) {
> > perror("malloc");
> > exit(1);
> > }
> > mem = (char *)(((int)mem + 4095) & ~0x0fff);
> > res = mprotect(mem, 4096, PROT_READ);
>
> Read the Unix standard:
>
> The behavior of this function is unspecified if the mapping was not
> established by a call to mmap().
Yeh, but it works on m68k and i386, and works on hppa if statically
linked. And the code is in an example on the mprotect man page on
my Mandrake7 box.
Richard
From drepper@redhat.com 17 Nov 2000 10:06:21 -0800
Date: 17 Nov 2000 10:06:21 -0800
From: Ulrich Drepper drepper@redhat.com
Subject: [parisc-linux] SEGV signal handling bug (dynamic linking)
Richard Hirst writes:
> Yeh, but it works on m68k and i386, and works on hppa if statically
> linked. And the code is in an example on the mprotect man page on
> my Mandrake7 box.
Then shoot the guy who wrote the man page. It's wrong and will never
reliably work.
--
---------------. ,-. 1325 Chesapeake Terrace
Ulrich Drepper \ ,-------------------' \ Sunnyvale, CA 94089 USA
Red Hat `--' drepper at redhat.com `------------------------
From rhirst@linuxcare.com Fri, 17 Nov 2000 20:10:34 +0000
Date: Fri, 17 Nov 2000 20:10:34 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] SEGV signal handling bug (dynamic linking)
On Fri, Nov 17, 2000 at 09:09:10AM -0800, Ulrich Drepper wrote:
> Richard Hirst writes:
>
> > mem = malloc(8192);
> > if (mem == NULL) {
> > perror("malloc");
> > exit(1);
> > }
> > mem = (char *)(((int)mem + 4095) & ~0x0fff);
> > res = mprotect(mem, 4096, PROT_READ);
>
> Read the Unix standard:
>
> The behavior of this function is unspecified if the mapping was not
> established by a call to mmap().
Changed my prog to use mmap and get the same problem.
Richard
From grundler@cup.hp.com Fri, 17 Nov 2000 13:50:46 -0800
Date: Fri, 17 Nov 2000 13:50:46 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] PDC firmware revision FAQ update
Hi folks,
The issue of PDC/firmware revs came up locally and sounded like a FAQ.
I added
"10. How can I check if the PDC (firmware) revision is the latest?"
and what I knew/could find to our FAQ at:
http://www.parisc-linux.org/faq.html
The new FAQ entry should be visible to the world in the next hour or so.
**** WARNING ****
Firmware upgrades can *kill* your machine!
**** WARNING ****
Don't do it just because. Read the FAQ carefully. Take the
time to figure out why you might need the upgrade and expose
yourself to this risk.
grant
ps. please don't ask me why your favorite machine's PDC isn't listed.
I don't run the referenced sites.
From taggart@carmen.fc.hp.com Fri, 17 Nov 2000 18:26:19 -0700
Date: Fri, 17 Nov 2000 18:26:19 -0700
From: Matt Taggart taggart@carmen.fc.hp.com
Subject: [parisc-linux] New cross compiler
I have posted new i386/linux -> hppa/linux cross tools at,
ftp://puffin.external.hp.com/pub/parisc/binaries/LinuxX86/xc-20001117.tar.gz
it includes 32bit and 64bit cross-compilers and a static 32bit glibc 2.2. I
have also updated the include tarball,
ftp://puffin.external.hp.com/pub/parisc/src/include-20001117.tar.gz
it includes glibc 2.2 headers and the latest kernel headers.
--
Matt Taggart
taggart@fc.hp.com
From sandy@storm.ca Fri, 17 Nov 2000 22:54:58 -0500
Date: Fri, 17 Nov 2000 22:54:58 -0500
From: Sandy Harris sandy@storm.ca
Subject: [parisc-linux] Host for 712/60 compiles
A couple of us have 16 diskless 712/60s which we want to use for distributed
processing on easily parallelized tasks like factoring and perhaps crypto
cracking. A few questions arise.
Is PARISC Linux far enough along to be useful for that? We need no monitor or
console (except perhaps for initial debugging), no devices except ethernet,
and expect to run only one process per machine, but we need stability.
We'll need a host for cross-compiling. We have a 256 meg 715/100 with HP/UX
which we expect to use as the host for booting, handing out chunks of work,
storing results, etc. Should we compile there under HP/UX or would we get
better tools on one of our Intel Linux boxes?
Or is PARISC Linux far enough along we should put it on the 715 and get
native compilation?
From matthew@wil.cx Sat, 18 Nov 2000 07:08:12 +0000
Date: Sat, 18 Nov 2000 07:08:12 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: binutils taggart
On Wed, Nov 15, 2000 at 05:16:02PM -0700, Matt Taggart wrote:
> Modified files:
> . : config.guess config.sub
>
> Log message:
> New upstream versions from
> http://subversions.gnu.org/cgi-bin/cvsweb/config/
> Now config.guess returns the right thing for hppa-linux
so... we needed new versions anyway, even though we're not parisc-linux..?
:-) [not bitter, just amused]
--
Revolutions do not require corporate support.
From matthew@wil.cx Sat, 18 Nov 2000 07:24:54 +0000
Date: Sat, 18 Nov 2000 07:24:54 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: [parisc-linux] CVS linux Vs. -test10
On Tue, Nov 14, 2000 at 09:35:01AM -0700, Paul Bame wrote:
>
> I've attached an overview of differences between our CVS linux sources
ok, here's my memories.
> * Lots of RCS $Revision$ differences in ACPI (a different 'cvs
> import' would've eliminated these differences).
i assume someone's already done the appropriate cvs admin -ko magic
to fix this?
> * drivers/char/Config.in: changes to support LASI, parisc real-time
> clock (CONFIG_GENRTC)
istr this was `stolen' from the m68k port by sammy.
> * drivers/char/serial.c: support for GSC and A500 serial
if they're working, send to Ted and he'll do an update with Linus at
some point.
> * drivers/net/eepro100.c: no clue about this one
we were trying to get it to work for the Jan NYLWE show. i doubt we have
any changes of note. does anyone have an eepro in an hp?
> * drivers/video: STI and HP FB video drivers (iodccon is probably
> worthless)
agreed on iodccon. unless we're using it up until the point that one of
the more advanced consoles takes over? i don't think we are now.
> * fs/binfmt_elf.c,exec.c: changes for stack-grows-up?
Yes, that's what they're for.
> * fs/nfs/read.c: probably unnecessary hack for broken parisc64 gcc
we certainly don't want to send it upstream, but we don't want to take it
out until the bug is fixed. is fixing that bug on the webshite todo list?
> * include/linux/binfmts.h,fs.h,kernel.h,tty.h,udf_fs_sb.h:
> unnecessary differences?
mostly, yes.
> * include/linux/init.h: we use different section names -- why????,
> probably some unnecessary other differences too
because we use -ffunction-sections. text.init clashes with a function called
init, and the linker just merges it into the text segment. we need it to
be separate, so it became init.text. there's patches around to do this for
other architectures too. just bad naming choices initially.
> * include/linux/wait.h: parisc debugging -- should be removed
yeah, that can die now.
> * scripts/*: MANY differences here. Looks like a combination of
> things we hacked to fix configuration problems plus MAYBE not
> updating these files from new Linux versions in the past. 'make
> menuconfig' is significantly different from upstream. Even the
> mkdep.c program is different.
ok. mea extremely culpa here. i had an exclude file which included the
scripts/ directory. someone should now ditch our stuff, take what's in
-test10, hack it till it works for us and check it in.
--
Revolutions do not require corporate support.
From bzeeb+hplinux@zabbadoz.net Sun, 19 Nov 2000 13:12:25 +0100 (CET)
Date: Sun, 19 Nov 2000 13:12:25 +0100 (CET)
From: Bjoern A. Zeeb bzeeb+hplinux@zabbadoz.net
Subject: [parisc-linux] serial console ?
Hi,
I just started booting into a 715-100 using palinux-0.5.iso.
I did read multiple times that I need serial console. So I plugged in
serial cable I an normally using for x86 to x86 serial console on my
linux router but I did not get anything on both serial ports :-(
What information would one need to help ?
personal contact is prefered. I would then sum up all information I
got and post it to the list afterwards (if desired).
Have to say: I am 'new' to those workstation and had to hoover it
first after I got it ;)
I also could have a 735-99 with some kind of graphic accelerator if
that one would be better.
--
Bjoern A. Zeeb bzeeb at Zabbadoz dot NeT
56 69 73 69 74 http://www.zabbadoz.net/
From bzeeb+hplinux@zabbadoz.net Sun, 19 Nov 2000 17:06:11 +0100 (CET)
Date: Sun, 19 Nov 2000 17:06:11 +0100 (CET)
From: Bjoern A. Zeeb bzeeb+hplinux@zabbadoz.net
Subject: [parisc-linux] serial console ?
On Sun, 19 Nov 2000, Bjoern A. Zeeb wrote:
> I just started booting into a 715-100 using palinux-0.5.iso.
>
> I did read multiple times that I need serial console. So I plugged in
> serial cable I an normally using for x86 to x86 serial console on my
> linux router but I did not get anything on both serial ports :-(
Hi,
sorted out everything. serial cable was quite ok, but using cu I was
never able to reach IPL. Using minicom and everything was fine.
Had a great success afterwards:
installed from CD (palinux-0.5.iso)
Everthing worked at once out of the box :-)
openssh was enabled on default :-))
only one bad touch: please do not enable portmapper on default.
there are too much exploits out there these days.
Short: Great great great work !!!
--
Bjoern A. Zeeb bzeeb at Zabbadoz dot NeT
56 69 73 69 74 http://www.zabbadoz.net/
From dhazeghi@pacbell.net Sun, 19 Nov 2000 09:24:56 -0800
Date: Sun, 19 Nov 2000 09:24:56 -0800
From: dhazeghi@pacbell.net dhazeghi@pacbell.net
Subject: [parisc-linux] glibc-latest tarball broken
the glibc-latest tarball on pehc appears to be missing some important
subdirectories like scripts and sysdeps. Perhaps a legacy of the 2.2
merge?
Thanks,
Dara Hazeghi
From imatthae@grc.hp.com Sun, 19 Nov 2000 19:35:57 +0100
Date: Sun, 19 Nov 2000 19:35:57 +0100
From: Ingo Matthaes imatthae@grc.hp.com
Subject: [parisc-linux] A500 and glibc woes
finally we've got a A500 for palinux testing. Unfortunately
we have some problems with it.
A 32 bit kernel complains about a very old firmware and claims
"that machine will probably never run linux" But in fact, it
will :-)
First question: Did anyone ever succeeded with a 32bit kernel
on that hardware ?
A 64 bit Kernel boots fine and recognizes all available hardware
including the additional 100BT card. But it traps with the init
which comes with the latest nfsroot tarball. I built a new one
from the sourcesw of debians sysvlinux-2.78 and it does not trap
anymore, but the kernel told me about unimplemented 32 syscalls.
At this point I tried to build a 64 bit glibc in order to get
a crt1.o , which is needed to builf for a 64bit init.
Did anyone ever build a 64bit glibc out of the current cvs tree ?
Today I've got
ologue/epilogue insn
(insn 433 431 435 (set (reg:DI 26 %r26)
(reg:DI 2 %r2)) -1 (nil)
(nil))
../sysdeps/unix/sysv/linux/init-first.c:105: Unrecognizable insn:
(insn 440 438 441 (set (reg:DI 24 %r24)
(lo_sum:DI (reg:DI 1 %r1)
(symbol_ref:DI ("LP$-001")))) -1 (insn_list 437 (nil))
(expr_list:REG_DEAD (reg:DI 1 %r1)
(expr_list:REG_UNUSED (reg:DI 24 %r24)
(nil))))
../sysdeps/unix/sysv/linux/init-first.c:105: Internal compiler error in num_delay_slots, at insn-attrtab.c:2505
Thats the point I'm currently stucking.
BTW If someone involved into the port with access to the internal HP
Network needs access to that machine, please contact me.
Later
Ingo
--
Tel: ++49-2102-908-210 German Response Center
Fax: ++49-2102-907-934 Berliner Str. 111
Mailto: Ingo_Matthaes@hp.com 40880 Ratingen
Germany
HP Unix/Linux Competency Center
Network and High AvailabilitY
OpenPGP fingerprint = 4298E7785FFD65DC8950 14E9F17F8CB5B611AA4A
From alan@linuxcare.com.au Mon, 20 Nov 2000 14:03:16 +1100 (EST)
Date: Mon, 20 Nov 2000 14:03:16 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: Single-stepping
On Thu, 16 Nov 2000, John Marvin wrote:
> Anyway, for reference, HP-UX does single stepping by using a combination
> of the taken branch trap, and loading the instruction queues such that the
> front of the queue points to the next instruction to be single stepped and
> the back of the queue points to the first of two break instructions on a
> "break" page. It does NOT insert break instructions into the code, so it
> does not adversely affect execution on a SMP machine. Note that we
> already put a bunch of break instructions before the syscall entry point
> on the gateway page, so it would be easy to use our gateway page for the
> "break page". This way, if the single stepped instruction branches, a
> taken branch trap will be taken (which is important in the case where the
> branch nullifies its delay slot). Otherwise, the instruction will be
> executed and then the break instruction at the known location on the
> "break" page will be executed. If the single stepped instruction
> nullifies the next instruction, the second break instruction on the
> "break" page will be executed.
This is the path I started out on for hppa-linux, then hit the problem of
a branch that nullifies it's delay slot. At that point, I decided playing
with IAOQ_back wouldn't work as I missed the solution of enabling taken
branch traps. :-( If I'd seen this trick, then I would not have tried
using the recovery counter, and even now, it may be better to go back to
IAOQ fiddling. The recovery counter scheme has the disadvantage that
there's only one of them so you need to save/restore over task swaps or
introduce extra instructions in the syscall path - and be very careful.
> Note that this is the short explanation. It is not as simple as it sounds.
> One major complication is that branches with links don't work properly
> with the instruction queue magic, so the link register has to be updated
> in the taken branch trap handler. Also branch externals won't update
> the space of the space queue tail properly (again, that has to be fixed
> in the taken branch handler). I can provide more details if the recovery
> counter method doesn't work out.
I'm a little intrigued about these "complications". How can the link
register or space _not_ be updated properly? As far as I can see, the
only really tricky instruction to single-step is RFI - which shouldn't
ever occur in userspace, and which we'd just emulate if it was important.
Alan Modra
--
Linuxcare. Support for the Revolution.
From jsm@udlkern.fc.hp.com Sun, 19 Nov 2000 22:43:02 -0700 (MST)
Date: Sun, 19 Nov 2000 22:43:02 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: Single-stepping
> > Note that this is the short explanation. It is not as simple as it sounds.
> > One major complication is that branches with links don't work properly
> > with the instruction queue magic, so the link register has to be updated
> > in the taken branch trap handler. Also branch externals won't update
> > the space of the space queue tail properly (again, that has to be fixed
> > in the taken branch handler). I can provide more details if the recovery
> > counter method doesn't work out.
>
> I'm a little intrigued about these "complications". How can the link
> register or space _not_ be updated properly? As far as I can see, the
> only really tricky instruction to single-step is RFI - which shouldn't
> ever occur in userspace, and which we'd just emulate if it was important.
The problem is that the link register is set to IAOQ_Back + 4. and in
the case of ble, sr0 is set to IASQ_Back. Since we've played games with
the queues, IAOQ_Back and IASQ_Back are pointing at the break page, not
at the instruction following the branch.
The additional complication is that the taken branch trap traps at the
branch destination, not at the branch, so at the point of the trap you
don't know where you came from in order to fix the problem easily. So,
what HP-UX does is check each instruction before it executes it to see if
it is a branch, and if so, what the link register is (and that is all that
needs to be parsed, since we are not emulating the instruction). It then
stores the branch location, and also sets some branch state flags (e.g.
UBE for a branch external, and UBL for a branch with a link, both flags
being set for a ble instruction). Then in the taken branch handler you
have all the information you need to fix the queue. You also need
to check this saved state if a signal handler is invoked while single
stepping, so that the proper pc queue values can be saved in the signal
context.
John Marvin
jsm@fc.hp.com
From jsm@udlkern.fc.hp.com Sun, 19 Nov 2000 23:01:20 -0700 (MST)
Date: Sun, 19 Nov 2000 23:01:20 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: A500 and glibc woes
Ingo,
>
> finally we've got a A500 for palinux testing. Unfortunately
> we have some problems with it.
> A 32 bit kernel complains about a very old firmware and claims
> "that machine will probably never run linux" But in fact, it
> will :-)
Actually, there are no plans to ever support the A500 on a 32 bit
kernel. The A500 only has 64 bit firmware, so in order to support
it we would need to write a 32->64 bit firmware translation layer.
...
> A 64 bit Kernel boots fine and recognizes all available hardware
> including the additional 100BT card. But it traps with the init
> which comes with the latest nfsroot tarball. I built a new one
> from the sourcesw of debians sysvlinux-2.78 and it does not trap
> anymore, but the kernel told me about unimplemented 32 syscalls.
> At this point I tried to build a 64 bit glibc in order to get
> a crt1.o , which is needed to builf for a 64bit init.
>
Well, it might seem crazy, but our short term plans do not include
supporting a 64 bit user environment on the 64 bit kernel. Long term I
believe this will happen, but as you have already seen, work needs to be
done to support 64 bit glibc, shared libraries, etc.
So, in order to get the A500 to boot, we need to support the 32 bit user
environment on a 64 bit kernel. We've only recently gotten to the point
where the 64 bit kernel gets far enough to start running user space
applications. The problem is that we need to write translations for many
system calls, since type sizes (and the structures those types are in) are
different between 32 bit and 64 bit (the ugliest case is the ioctl system
call).
In summary, the A500 currently won't work, but since it is one of the
machines that HP has targetted for Linux support, you can be sure that
this will change fairly soon.
John Marvin
jsm@fc.hp.com
From grundler@cup.hp.com Sun, 19 Nov 2000 22:47:27 -0800
Date: Sun, 19 Nov 2000 22:47:27 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] A500 and glibc woes
Ingo Matthaes wrote:
> finally we've got a A500 for palinux testing. Unfortunately
> we have some problems with it.
Ingo,
uhmm...I'm not surprised. Let me help you understand the A500
and the direction we are going with support on the A500.
I'm one of several people working on A500 support.
A500 Features (marketing crud - you probably know this):
o 16GB of RAM
o 2-way 550 MHz PA8600
o 4 PCI-4X slots (3 of which are 1 slot per PCI bus)
o all in 2U rack mountable box.
Inside is (technical crud):
o PCX-W+
o Astro IOMMU (aka SBA. Provides cache-coherent DMA and virtual I/O DMA)
o Elroy PCI Host Bus Adapter (aka LBA)
o IOSAPIC interrupt controller (integrated in Elroy)
o PAT PDC (not legacy)
o ^B for service processor access (ie I can reset the machine remotely)
> A 32 bit kernel complains about a very old firmware and claims
> "that machine will probably never run linux" But in fact, it
> will :-)
That's right! :^)
> First question: Did anyone ever succeeded with a 32bit kernel
> on that hardware ?
No. Only on *64-bit* kernels. The reason is some of the PAT PDC
calls we need are *only* available in wide mode. We could have written
narrow<->wide mode translation routines but a 32-bit kernel ignores
the A500's best feature - 16GB of RAM. The issue is HP needs good
specweb99 numbers to sell these boxes and that requires GB of RAM.
The 512MB of RAM that 32-bit kernel currently supports (jsm tells me
3GB or so can be done) won't approach the perf levels which can be
acheived with 8 or 16GB. The HPUX specweb team (who just *beat* single
CPU TUX specweb99 numbers with HPUX on A500) told me. They have a clue.
You are welcome to write those translation routines and then
*remove* many the "#ifdef __LP64__" preprocessor directives in
arch/parisc/kernel/inventory.c and lba_pci.c. Send me the
patch and I'll test/review/commit the changes.
> A 64 bit Kernel boots fine and recognizes all available hardware
> including the additional 100BT card. But it traps with the init
> which comes with the latest nfsroot tarball.
There are issues/problems with PCI resource management and I have
some code waiting to be tested when I get in tomorrow.
But you can be very helpful with user space! Perhaps Paul Bame
can be more specific. But basically we don't have all the translation
routines in place for 32-bit applications invoking 64-bit kernel syscalls.
> I built a new one
> from the sourcesw of debians sysvlinux-2.78 and it does not trap
> anymore, but the kernel told me about unimplemented 32 syscalls.
Right.
> At this point I tried to build a 64 bit glibc in order to get
> a crt1.o , which is needed to builf for a 64bit init.
>
> Did anyone ever build a 64bit glibc out of the current cvs tree ?
I don't think so. Eventually we wanted to but haven't been able yet.
So this great that you are trying!
> Today I've got
...
Can someone one with a clue about toolchain look at that?
I'd be really impressed if someone got a 64-bit userspace built!
I can help get kernel working right but don't have a clue about the tools.
The 64-bit kernel can be booted on the following class of boxes
to about the same point as the A500:
o C160/180/200/240/360
o B2000/C3000/J5000/C3600/J5600/J6000
o some D/K/R-class machines
The key is the box must have PCX-U (PA8000), PCX-U+ (PA8200),
PCX-W (PA8500) or PCX-W+ (PA8600) processor.
Look in the HWDB (http://www.parisc-linux.org/hw.html) or
in /usr/sam/lib/mo/sched.models (HPUX 11.x) to determine which
processor you have.
> Thats the point I'm currently stucking.
>
> BTW If someone involved into the port with access to the internal HP
> Network needs access to that machine, please contact me.
Ditto. I can arrange access to my A500 as well. External access
to some limited number of people could be arranged if demand
is justifiable. But given the number of other boxes which can
run 64-bit kernel, I hope that's not necessary.
Thanks Ingo!
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From alan@linuxcare.com.au Mon, 20 Nov 2000 17:53:18 +1100 (EST)
Date: Mon, 20 Nov 2000 17:53:18 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: Single-stepping
On Sun, 19 Nov 2000, John Marvin wrote:
> > I'm a little intrigued about these "complications". How can the link
> > register or space _not_ be updated properly? As far as I can see, the
> > only really tricky instruction to single-step is RFI - which shouldn't
> > ever occur in userspace, and which we'd just emulate if it was important.
>
> The problem is that the link register is set to IAOQ_Back + 4. and in
> the case of ble, sr0 is set to IASQ_Back. Since we've played games with
> the queues, IAOQ_Back and IASQ_Back are pointing at the break page, not
> at the instruction following the branch.
Ah. That is a little nasty, especially given the effect on signal
handlers you mention below. Maybe using the recovery counter isn't such a
bad idea after all, especially since the added syscall and task switch
overhead can be quite small if the kernel only supports single-step by
one instruction.
> The additional complication is that the taken branch trap traps at the
> branch destination, not at the branch, so at the point of the trap you
> don't know where you came from in order to fix the problem easily. So,
> what HP-UX does is check each instruction before it executes it to see if
> it is a branch, and if so, what the link register is (and that is all that
> needs to be parsed, since we are not emulating the instruction). It then
> stores the branch location, and also sets some branch state flags (e.g.
> UBE for a branch external, and UBL for a branch with a link, both flags
> being set for a ble instruction). Then in the taken branch handler you
> have all the information you need to fix the queue. You also need
> to check this saved state if a signal handler is invoked while single
> stepping, so that the proper pc queue values can be saved in the signal
> context.
Another question for you and/or the list in general:
Why does struct pt_regs have an ipsw field? Seems like it currently is
unused.
Regards, Alan Modra
--
Linuxcare. Support for the Revolution.
From grundler@cup.hp.com Sun, 19 Nov 2000 23:05:26 -0800
Date: Sun, 19 Nov 2000 23:05:26 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] Host for 712/60 compiles
Sandy Harris wrote:
> A couple of us have 16 diskless 712/60s which we want to use for distributed
> processing on easily parallelized tasks like factoring and perhaps crypto
> cracking. A few questions arise.
>
> Is PARISC Linux far enough along to be useful for that? We need no monitor or
> console (except perhaps for initial debugging), no devices except ethernet,
> and expect to run only one process per machine, but we need stability.
It's not rock solid. On 712's, it should be pretty good though.
> We'll need a host for cross-compiling. We have a 256 meg 715/100 with HP/UX
> which we expect to use as the host for booting, handing out chunks of work,
> storing results, etc. Should we compile there under HP/UX or would we get
> better tools on one of our Intel Linux boxes?
I don't think anyone has tried to cross-compile parisc-linux on an HPUX
host in quite a while.
> Or is PARISC Linux far enough along we should put it on the 715 and get
> native compilation?
AFAIK, all of the debian packages on the ISO image are built natively.
But using a dual P700 linux box would be alot faster. :^)
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From sieler@allegro.com Sun, 19 Nov 2000 23:24:00 -0800 (PST)
Date: Sun, 19 Nov 2000 23:24:00 -0800 (PST)
From: Stan Sieler sieler@allegro.com
Subject: Single-stepping
Re:
> handlers you mention below. Maybe using the recovery counter isn't such a
quite true.
> bad idea after all, especially since the added syscall and task switch
> overhead can be quite small if the kernel only supports single-step by
> one instruction.
why the limit? We've used multi-instruction "single step" (oxymoron :)
for about 15 years on PA-RISC...no problems, efficient, and *very*
useful!
--
Stan Sieler sieler@allegro.com
www.allegro.com/sieler/wanted/index.html www.sieler.com
From grundler@cup.hp.com Sun, 19 Nov 2000 23:44:42 -0800
Date: Sun, 19 Nov 2000 23:44:42 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
Matthew Wilcox wrote:
> On Tue, Nov 14, 2000 at 09:35:01AM -0700, Paul Bame wrote:
> ok, here's my memories.
Thanks Matthew!
hehe...sounds like someone's getting older. :^)
...
> > * drivers/net/eepro100.c: no clue about this one
>
> we were trying to get it to work for the Jan NYLWE show.
I think I did that. IIRC, it's a one-line change to use I/O port
space since MMIO wasn't usable without more invasive changes.
> i doubt we have any changes of note. does anyone have an eepro in an hp?
I have picked nearly 30 i82557/i82558 PCI cards from scrap yard.
And maybe a few i82559 even. Did you need one (or two)? :^)
FWIW, this is the card/driver which I think was causing misaligned
data reference traps. I never had a chance to followup with this though.
At the time, I thought it would be *really* fun to show this working
to HP's marketing team...
> > * fs/nfs/read.c: probably unnecessary hack for broken parisc64 gcc
>
> we certainly don't want to send it upstream, but we don't want to take it
> out until the bug is fixed. is fixing that bug on the webshite todo list?
I don't think so. It's possible it's already fixed.
Relevant CVS log entry:
| revision 1.5
| date: 2000/07/18 03:15:25; author: dhd; state: Exp; lines: +5 -0
| ARGH! When I put in an assertion, NFS stops breaking randomly. I
| suspect this is a compiler or binutils problem but I can't see any
| clear differences in the generated code. I'll debug it later...
This sounds like the hppa64 bug we saw with %r29 getting trashed.
But I don't think David was working on hppa64 kernel at the time.
I can test 32-bit NFS Root tomorrow w/o assertion if no one else
beats me to it.
> > * include/linux/init.h: we use different section names -- why????,
> > probably some unnecessary other differences too
>
> because we use -ffunction-sections. text.init clashes with a function
> called init, and the linker just merges it into the text segment. we
> need it to be separate, so it became init.text. there's patches around
> to do this for other architectures too. just bad naming choices initially.
We need to resolve this in order to merge upstream.
Matthew, any advice on how we should proceed?
Or would be easier for you pester Alan Cox and just get it fixed?
> > * include/linux/wait.h: parisc debugging -- should be removed
>
> yeah, that can die now.
I'd be happy to fix this by clobbering the current version with what's in
linux-2.4.0-test10. But what is the "right" way to revert changes we've made
so this doesn't show up in next merge?
I'll need to know this in order to revert the fs/nfs/read.c change as well.
thanks,
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From milang@tal.org Mon, 20 Nov 2000 10:57:36 +0200
Date: Mon, 20 Nov 2000 10:57:36 +0200
From: Kaj-Michael Lang milang@tal.org
Subject: [parisc-linux] Palinux on a 712/60
> Thanks for the reply, Grant. However, the 712s do not have a serial
console.
Yes it does.. it's just undocumented. Unfortunately I don't have the
information here, but
I have succesfully changed the console to serial on one of my 712.
Kaj-Michael Lang
milang@tal.org
From alan@linuxcare.com.au Mon, 20 Nov 2000 20:05:58 +1100 (EST)
Date: Mon, 20 Nov 2000 20:05:58 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: Single-stepping
On Sun, 19 Nov 2000, Stan Sieler wrote:
> > bad idea after all, especially since the added syscall and task switch
> > overhead can be quite small if the kernel only supports single-step by
> > one instruction.
>
> why the limit? We've used multi-instruction "single step" (oxymoron :)
> for about 15 years on PA-RISC...no problems, efficient, and *very*
> useful!
Because you would then need to save and restore cr0 on task switches (or
only allow one task to be single-stepped at a time). That's four
instructions and two extra memory accesses per task switch. Which might
not seem very much, but at some point somebody will no doubt start caring
about pa-linux performance. For a single-step by one, you can simply set
cr0 to zero on a task switch, and possibly avoid touching cr0 on a task
switch at all with careful attention to various trap handlers.
Here's the idea. The tail of syscall_restore (64-bit stuff pruned) will
look like the following, with the first three instructions being the added
code to support single-step (and also wide/narrow switching for 64-bit)
ldi 3,%r20
mtctl $r20,%cr0 /* recovery counter, ptrace single-step */
LDREG TASK_PT_PSW(%r1),%r20
mtctl %r1,%cr30 /* intrhandler okay. */
mfsp %sr3,%r1 /* Get users space id */
mtsp %r1,%sr4 /* Restore sr4 */
mtsp %r1,%sr5 /* Restore sr5 */
mtsp %r1,%sr6 /* Restore sr6 */
depi 3,31,2,%r31 /* ensure return to user mode. */
mtsm %r20 /* restore irq state */
mfctl %cr27,%r20
be 0(%sr3,%r31) /* return to user space */
mtsp %r1,%sr7 /* Restore sr7 */
ptrace will fiddle with TASK_PT_PSW, setting the R bit and clearing the I
bit to enable the recovery counter - which will start counting down at the
mtsm instruction above, and reach zero on the user-space instruction, so
we'll trap after executing one user-space instruction. The task-switch
nonsense is to handle the case where we page-fault on the instruction and
switch to another task also doing single-stepping. You want to ensure cr0
is zero when we finally get back to the original task.
Now it might turn out that having extra instructions in the syscall path
is worse than extra code and memory accesses on task switch. If that
turns out to be true, then you'll probably get your multi-step ptrace. :-)
Alan Modra
--
Linuxcare. Support for the Revolution.
From M_Wild@tunstall.co.uk Mon, 20 Nov 2000 10:21:32 -0000
Date: Mon, 20 Nov 2000 10:21:32 -0000
From: Mark Wild M_Wild@tunstall.co.uk
Subject: [parisc-linux] Upgrading memory on a 710
Hi,
I have acquired some 8MB SIMMs suitable for my old 710 workstation and
wish to utilise them. I don't have any docs for the 710 and was wondering
if anyone could point me in the direction of some. I've looked on the HP
website but could only find some for 712s, 720, 730 etc. Alternatively, if
anyone knows whether any motherboard links need to be changed, whether
SIMMS need to be added in pairs / quads and which memory configurations
are allowed I would be grateful.
Thanks,
Mark Wild
--
From matthew@wil.cx Mon, 20 Nov 2000 10:50:57 +0000
Date: Mon, 20 Nov 2000 10:50:57 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: [parisc-linux] A500 and glibc woes
On Sun, Nov 19, 2000 at 07:35:57PM +0100, Ingo Matthaes wrote:
> finally we've got a A500 for palinux testing. Unfortunately
> we have some problems with it.
> A 32 bit kernel complains about a very old firmware and claims
> "that machine will probably never run linux" But in fact, it
> will :-)
> First question: Did anyone ever succeeded with a 32bit kernel
> on that hardware ?
It's not intended to work; I doubt anyone has tried. You should build
a 64 bit kernel.
> A 64 bit Kernel boots fine and recognizes all available hardware
> including the additional 100BT card. But it traps with the init
> which comes with the latest nfsroot tarball. I built a new one
> from the sourcesw of debians sysvlinux-2.78 and it does not trap
> anymore, but the kernel told me about unimplemented 32 syscalls.
> At this point I tried to build a 64 bit glibc in order to get
> a crt1.o , which is needed to builf for a 64bit init.
Don't try a 64-bit userland yet. What you need to do is implement
some of the 32-bit syscalls.
--
Revolutions do not require corporate support.
From matthew@wil.cx Mon, 20 Nov 2000 11:17:26 +0000
Date: Mon, 20 Nov 2000 11:17:26 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: [parisc-linux] CVS linux Vs. -test10
On Sun, Nov 19, 2000 at 11:44:42PM -0800, Grant Grundler wrote:
> Matthew Wilcox wrote:
> > On Tue, Nov 14, 2000 at 09:35:01AM -0700, Paul Bame wrote:
> > ok, here's my memories.
>
> Thanks Matthew!
>
> hehe...sounds like someone's getting older. :^)
... when i were a lad, all this were fields! Our dad used to kill us
every morning, we'd get up half an hour before we went to bed and walk
uphill both to and from school...
> ...
> > > * drivers/net/eepro100.c: no clue about this one
> >
> > we were trying to get it to work for the Jan NYLWE show.
>
> I think I did that. IIRC, it's a one-line change to use I/O port
> space since MMIO wasn't usable without more invasive changes.
sounds right. MMIO should work now though, right?
> > i doubt we have any changes of note. does anyone have an eepro in an hp?
>
> I have picked nearly 30 i82557/i82558 PCI cards from scrap yard.
> And maybe a few i82559 even. Did you need one (or two)? :^)
Heh, I only have a 712 right now :-)
> FWIW, this is the card/driver which I think was causing misaligned
> data reference traps. I never had a chance to followup with this though.
> At the time, I thought it would be *really* fun to show this working
> to HP's marketing team...
Oh yes, I remember that now. Tulip always does a copy (well, it doesn't _have_
to, but we tell it to, just like the Alpha does).
> > > * fs/nfs/read.c: probably unnecessary hack for broken parisc64 gcc
> >
> > we certainly don't want to send it upstream, but we don't want to take it
> > out until the bug is fixed. is fixing that bug on the webshite todo list?
>
> I don't think so. It's possible it's already fixed.
>
> Relevant CVS log entry:
> | revision 1.5
> | date: 2000/07/18 03:15:25; author: dhd; state: Exp; lines: +5 -0
> | ARGH! When I put in an assertion, NFS stops breaking randomly. I
> | suspect this is a compiler or binutils problem but I can't see any
> | clear differences in the generated code. I'll debug it later...
>
> This sounds like the hppa64 bug we saw with %r29 getting trashed.
> But I don't think David was working on hppa64 kernel at the time.
> I can test 32-bit NFS Root tomorrow w/o assertion if no one else
> beats me to it.
it was definitely a 32-bit kernel at the time. It might be the same bug,
but I'm not sure.
> > > * include/linux/init.h: we use different section names -- why????,
> > > probably some unnecessary other differences too
> >
> > because we use -ffunction-sections. text.init clashes with a function
> > called init, and the linker just merges it into the text segment. we
> > need it to be separate, so it became init.text. there's patches around
> > to do this for other architectures too. just bad naming choices initially.
>
> We need to resolve this in order to merge upstream.
> Matthew, any advice on how we should proceed?
> Or would be easier for you pester Alan Cox and just get it fixed?
Hm. Alan's not hacking on 2.4, last I heard. I might pester Linus and
see if we can change that. It's a mechanical change so he might not be
averse to it at this point. Bear in mind we don't need to do a complete
merge at this point -- most architectures have a separate patch to apply
on top of Linus' tree.
> > > * include/linux/wait.h: parisc debugging -- should be removed
> >
> > yeah, that can die now.
>
> I'd be happy to fix this by clobbering the current version with what's in
> linux-2.4.0-test10. But what is the "right" way to revert changes we've made
> so this doesn't show up in next merge?
I don't know that there's an official way to do this. I always changed
the file to its previous state and then committed it. There are a number of
ways of doing it; perhaps the cleanest is:
cvs diff -r1.4 -r1.5 fs/nfs/read.c >../read.c.diff
(then check the read.c.diff file)
patch -p1 <../read.c.diff
rm ../read.c.diff
or you can just delete the lines yourself. Use diff to make sure there
aren't any silly cosmetic changes (eg whitespace).
--
Revolutions do not require corporate support.
From grundler@cup.hp.com Mon, 20 Nov 2000 09:34:31 -0800
Date: Mon, 20 Nov 2000 09:34:31 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
Matthew Wilcox wrote:
...
> sounds right. MMIO should work now though, right?
It would have worked then too.
Changing it to use MMIO space would require disabling I/O Port space
by defining inb/outb locally to use gsc_readb/writeb.
My problem with doing that was the semantics are slightly different.
I/O Port space is non-postable and MMIO space is. I wasn't confident
a driver written to use I/O port space would work right under MMIO
though some certainly do.
...
> it was definitely a 32-bit kernel at the time. It might be the same bug,
> but I'm not sure.
Can't be. AP (%r29) is a 64-bit only construct.
dhd also just confirmed it was 32-bit kernel.
I'll try it now.
...
> > We need to resolve this in order to merge upstream.
> > Matthew, any advice on how we should proceed?
> > Or would be easier for you pester Alan Cox and just get it fixed?
>
> Hm. Alan's not hacking on 2.4, last I heard. I might pester Linus and
> see if we can change that. It's a mechanical change so he might not be
> averse to it at this point. Bear in mind we don't need to do a complete
> merge at this point -- most architectures have a separate patch to apply
> on top of Linus' tree.
Ok. What's the first step to getting arch/parisc* and include/asm-parisc*
into Linus's tree?
I had dinner with Bdale Garbee last night and one of two things he made
clear was we need to unfork from debian and linus's tree in order to move
forward. All our CVS branches need to become obsolete or "local sandboxes"
of the respective upstream partners. Feeding kernel bits upstream will
bring a new level of visibility (and *HELP*) to the parisc-linux port.
I totally agree with Bdale. I understand alot of work still needs to
happen in our tree (eg though sba_iommu.c works, it's current form sucks)
But pushing bits upstream to linus will not preclude us from doing that work.
I also find it odd that glibc is merged upstream *before* the kernel is.
For the record, the second issue bdale made clear was we need "boot
floppies" debian package working. We don't need more ISO images (no
offense to pjlahaie for his good work). "Boot floppies" is a pre-requisite
to becoming part of the next debian release. Given I still don't have
a clue how to build a debian package and I can still contribute alot
in other areas, it doesn't make sense for me to do it myself.
> > I'd be happy to fix this by clobbering the current version with what's in
> > linux-2.4.0-test10. But what is the "right" way to revert changes we've made
> > so this doesn't show up in next merge?
>
> I don't know that there's an official way to do this. I always changed
> the file to its previous state and then committed it. There are a number of
> ways of doing it; perhaps the cleanest is:
>
> cvs diff -r1.4 -r1.5 fs/nfs/read.c >../read.c.diff
> (then check the read.c.diff file)
> patch -p1 <../read.c.diff
> rm ../read.c.diff
>
> or you can just delete the lines yourself. Use diff to make sure there
> aren't any silly cosmetic changes (eg whitespace).
The part you described above is the easy part - np.
I'm worried about labels and tracking how we "name" the releases.
Mang or other CVS ninja's care to comment?
thanks,
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From rhirst@linuxcare.com Mon, 20 Nov 2000 17:58:38 +0000
Date: Mon, 20 Nov 2000 17:58:38 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] signal handling problems (32 bit kernel)
Hi,
Following on from my problems with SEGV last week, I've come up with
a few signal handling issues:
=============================================================
parisc:signal.c
/* Possibly bogus. */
#warning XXX FIXME probably bogus -PB
/* I think this is bogus -- it'll cause the first instn of the
* signal handler to be executed twice! Better might be to
* set iaoq[0] to one of the NOPs in the trampoline. -PB
*/
regs->iaoq[0] = (unsigned long) haddr | 3;
regs->iaoq[1] = (unsigned long) haddr | 3;
This causes real problems if the signal handler is a dynamic object
which has not yet been resolved; in that case the signal handler
points at the b,l instr in the following at the end of the .plt:
2700: 0e 80 10 96 ldw 0(sr0,r20),r22
2704: ea c0 c0 00 bv r0(r22)
2708: 0e 88 10 95 ldw 4(sr0,r20),r21
270c: ea 9f 1f dd b,l 2700 <__DTOR_END__+0x54>,r20
2710: d6 80 1c 1e depwi 0,31,2,r20
typically that results in a misalligned access on the first ldw
as r20 is screwed.
I've fixed that by setting iaoq[1] = iaoq[0]+4. Is that OK, or
should we use the NOP approach in the comment above?
=============================================================
If a program gets a SEGV, fixes the cause of it in its signal
handler, and returns, then the faulting instruction is not rerun.
I think that's wrong (differs from m68k anyway).
=============================================================
Set the following program running:
#include
#include
#include
#include
int v[8];
void (* fp)(int);
void sig_handler(int sig)
{
}
void poke(int i)
{
v[0] = i; v[1] = i; v[2] = i; v[3] = i;
v[4] = i; v[5] = i; v[6] = i; v[7] = i;
}
int main()
{
int j, i = 0;
struct sigaction act;
memset(&act, 0, sizeof(act));
act.sa_handler = sig_handler;
sigaction(SIGUSR1, &act, NULL);
fp = sig_handler;
fp(0);
printf("I am %d\n", getpid());
while (1) {
poke (++i);
j = v[7];
if (j != i)
printf("Wah!\n");
}
}
and then in another terminal do 'kill -USR1 '. The program
either goes 'Wah!', gives a SEGV, or works. That seems to be because
%r1 is corrupted while processing the signal. The signal handler ends
with a syscall (rt_sigreturn_wrapper), and %r1, at least, is not saved
and restored over the syscall. %r31 also appears to get corrupted, as
it is used in the final branch of the syscall return.
Richard
From sieler@allegro.com Mon, 20 Nov 2000 10:47:38 -0800 (PST)
Date: Mon, 20 Nov 2000 10:47:38 -0800 (PST)
From: Stan Sieler sieler@allegro.com
Subject: Single-stepping
Re:
> Because you would then need to save and restore cr0 on task switches (or
> only allow one task to be single-stepped at a time). That's four
> instructions and two extra memory accesses per task switch. Which might
> not seem very much, but at some point somebody will no doubt start caring
> about pa-linux performance.
And it still won't seem like much, then!
Non-memory-access instructions are cheap. An extra memory reference (from
something probably already in cache) and two extra instructions
would probably cost less than an hour per CPU over the next 10 *years*,
assuming 10 years of 1000 task switches per second on a slow 100 MHz CPU.
Of course, at the cost of an extra non-memory-referencing instruction or so,
you could say "at switch-to-task time: if PSW R-bit set, then load the saved
CR0 from memory and move it to CR0", saving one memory reference 99.99999%
of the time, resuling in an average of only one memory reference
per task switch normally.
I haven't look at interrupt handling / system calls closely, but I
hope there aren't other false savings. (E.g., failure to save/restore
the PID check flag ... sure, user processes *now* probably never have
pid checking disabled, but that's a very useful feature to have
available (with proper security controls, of course).)
(Yes, I'm one of the very few who use that feature on MPE/iX ... carefully,
of course :)
--
Stan Sieler sieler@allegro.com
www.allegro.com/sieler/wanted/index.html www.sieler.com
From grundler@cup.hp.com Mon, 20 Nov 2000 11:23:40 -0800
Date: Mon, 20 Nov 2000 11:23:40 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
Matthew Wilcox wrote:
...
> > * fs/nfs/read.c: probably unnecessary hack for broken parisc64 gcc
>
> we certainly don't want to send it upstream, but we don't want to take it
> out until the bug is fixed. is fixing that bug on the webshite todo list?
It's fixed. I was able to NFS boot my A180 and install palinux-0.5.iso
bits on the SCSI disk. I've reverted the change to the LINUS_240_TEST10
version.
> > * include/linux/wait.h: parisc debugging -- should be removed
>
> yeah, that can die now.
Is anyone else already doing this?
thanks,
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From herrold@owlriver.com Mon, 20 Nov 2000 15:40:24 -0500 (EST)
Date: Mon, 20 Nov 2000 15:40:24 -0500 (EST)
From: herrold herrold@owlriver.com
Subject: Serial vs. STI console selector was: Re: [parisc-linux] Beta CD
On Tue, 14 Nov 2000, Paul J.Y. Lahaie wrote:
> This is to answer all the questions about sti console. Currently, in the
> CVS tree, serial console and STI console cannot be turned on at the same time.
>
> This means that a choice between serial/sti needs to be done at compile
> time.
> If you have any more questions or suggestions, feel free to email me or
> post on this list. Thank you.
... Query --
Background: Back in the early 70's with a HP 8090 >, IBM 1401, IBM
1620, and IBM 360/40, we faced such issues (Don't ask my age,
please) -- The solution was to probe for, and look for 'sense
switches' in unusual state -- that is -- Some condition which we
could examine, and yet not hose up the boot proces on the host. --
CapsLock set on a keyboard driver -- DSR on a serial line, a
'console sense switch' normally kept off but turned on ... so on.
Sort of like working through a qualitative analysis flowchart in
non-organic chemistry.
Is this possible, to allow support of both, with a single kernel?
-- Russ
From grundler@cup.hp.com Mon, 20 Nov 2000 12:59:41 -0800
Date: Mon, 20 Nov 2000 12:59:41 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: Serial vs. STI console selector was: Re: [parisc-linux] Beta CD
"Russ" herrold wrote:
> Is this possible, to allow support of both, with a single kernel?
Yes. I think the issues are code not stomping on each other.
And it's easy than you might think - PDC can tell us what the
primary console is.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From grundler@cup.hp.com Mon, 20 Nov 2000 17:19:03 -0800 (PST)
Date: Mon, 20 Nov 2000 17:19:03 -0800 (PST)
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] 712 ISL and palo
Hi Paul (Bame),
I tried booting 712 with "bo scsi.4.0 isl" with the
palinux-0.5.iso image on a regular hard disk.
I wanted to edit the "root" parameter so it would
be /dev/sdb (also had a disk at scsi.0.0) instead of
/dev/scd0.
palo wouldn't accept any ps/2 keyboard input.
Don't know if this is a palo or PDC bug.
Any ideas?
thanks,
grant
From alan@linuxcare.com.au Tue, 21 Nov 2000 12:55:42 +1100 (EST)
Date: Tue, 21 Nov 2000 12:55:42 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: A DT_INIT/DT_FINI glibc patch.
On Thu, 16 Nov 2000, H . J . Lu wrote:
> While we are on this subject, I looked at the hppa code. It seems to
> have the same bug. I have an impression that HP never really used
> DT_INIT/DT_FINI. They use DT_INIT_ARRAY/DT_FINI_ARRAY. I don't know
> if we should fix hppa also.
Yes to all of these comments. elf32-hppa stole some ideas from ia64,
which is why the bug is common.
Fortunately, I think hppa ld.so etc. can be fixed in a way such that
current (broken ABI) binaries will still work.
Alan Modra
--
Linuxcare. Support for the Revolution.
From taggart@carmen.fc.hp.com Mon, 20 Nov 2000 20:12:33 -0700
Date: Mon, 20 Nov 2000 20:12:33 -0700
From: Matt Taggart taggart@carmen.fc.hp.com
Subject: [parisc-linux] glibc-latest tarball broken
dhazeghi@pacbell.net writes...
> the glibc-latest tarball on pehc appears to be missing some important
> subdirectories like scripts and sysdeps. Perhaps a legacy of the 2.2
> merge?
Disk space problem. I made some room and generated new tarballs.
Thanks for pointing out the problem.
--
Matt Taggart
taggart@fc.hp.com
From alan@linuxcare.com.au Tue, 21 Nov 2000 14:22:53 +1100 (EST)
Date: Tue, 21 Nov 2000 14:22:53 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: A DT_INIT/DT_FINI glibc patch.
On Mon, 20 Nov 2000, Grant Grundler wrote:
> I'm afraid we will break binary compatibility for elf32 hppa-linux again.
> I only raise this in case there is really a "right" way to fix the
> DT_INIT/DT_FINI problem.
The fix involves pruning out all the special handling in bfd/elf32-hppa.c
for DT_INIT,DT_FINI, and making some small changes to glibc. H.J. Lu has
already made the necessary framework changes, and all we need to do is
something like
#define DL_FUNCTION_ADDRESS(map, addr) _dl_start_address (map, addr)
#define DL_DT_INIT_ADDRESS(map, addr) \
((addr) & 2 ? (addr) : DL_FUNCTION_ADDRESS (map, addr))
with similar defines for DT_FINI. The test for (addr) & 2 is the
fairly innocuous fudge to support old binaries.
Another glibc issue (which is why I'm sending this back to the list) is
that there has been quite some discussion on the binutils list over the
use of the EI_OSABI field. The conclusion being that it isn't really
correct to use this field to discern the intendend execution platform.
It's purpose is really to tell ELF tools that a file contains fields and
values that need to be interpreted in a non-standard way. Since both
elf32-hppa and elf64-hppa follow the standards, I'm inclined to think that
the gnu tools should set EI_OSABI to zero for both HPUX and Linux targets.
Instead the .note.ABI-tag section should be examined to determine the
target, which sadly isn't set correctly at the moment. :-(
Alan Modra
--
Linuxcare. Support for the Revolution.
From bame@riverrock.org Mon, 20 Nov 2000 21:02:40 -0700
Date: Mon, 20 Nov 2000 21:02:40 -0700
From: bame@riverrock.org bame@riverrock.org
Subject: [parisc-linux] 712 ISL and palo
= palo wouldn't accept any ps/2 keyboard input.
= Don't know if this is a palo or PDC bug.
= Any ideas?
PALO bug. I didn't test the PS/2 case but I think I fixed it.
-P
From alan@linuxcare.com.au Tue, 21 Nov 2000 15:38:57 +1100 (EST)
Date: Tue, 21 Nov 2000 15:38:57 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: A DT_INIT/DT_FINI glibc patch.
On Tue, 21 Nov 2000, I wrote:
> Instead the .note.ABI-tag section should be examined to determine the
> target, which sadly isn't set correctly at the moment. :-(
Actually, it is set correctly. A comment in csu/abi-note.S was wrong.
--
Linuxcare. Support for the Revolution.
From jsm@udlkern.fc.hp.com Mon, 20 Nov 2000 22:47:41 -0700 (MST)
Date: Mon, 20 Nov 2000 22:47:41 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: [parisc-linux] Use of the EI_OSABI field
> Another glibc issue (which is why I'm sending this back to the list) is
> that there has been quite some discussion on the binutils list over the
> use of the EI_OSABI field. The conclusion being that it isn't really
> correct to use this field to discern the intendend execution platform.
> It's purpose is really to tell ELF tools that a file contains fields and
> values that need to be interpreted in a non-standard way. Since both
> elf32-hppa and elf64-hppa follow the standards, I'm inclined to think that
> the gnu tools should set EI_OSABI to zero for both HPUX and Linux targets.
I don't read the binutils mailing list, so I checked the archive. In my
opinion I didn't see a discussion, I saw H.J. Lu repeatedly asserting
this "fact". He may or may not be correct about original intention, but
the implementation so far has been that EI_OSABI is used to tell what the
target platform is. That is what FreeBSD is doing, that is what HP-UX is
doing, and that is what the IA-64 Unix System V Application Binary
Interface specifies:
http://developer.intel.com/design/IA-64/Downloads/24537002.pdf
Note that the second paragraph in section 1.1 of that specification
includes the following:
This document is the result of consensus among operating system
vendors intending to provide UNIX and UNIX workalike operating
systems on the IA-64 architecture. The vendors participating in
this effort include Intel, Sun Microsystems, SCO, IBM, SGI,
Cygnus Solutions, VA Linux Systems, HP, and Compaq.
So, If the specification is wrong according to H.J. Lu, why did both
Cygnus and VA Linux Systems not object to this:
Section 4.1.1.4 Operating System Identification
The e_ident[EI_OSABI] value identifies the operating system and
ABI to which the object is targeted, as listed in Table 4-1.
Table 4-1 lists the various ELFOSABI_* fields, e.g. ELFOSABI_HPUX,
ELFOSABI_NETBSD, ELFOSABI_LINUX, etc.
Note also, that the current mechanism for us to be able to differentiate
elf executables in the Linux kernel is the machine dependent
elf_check_arch() macro, whose only argument is a pointer to a
elf32_hdr or elf64_hdr. I think it would be ugly to try to get the
.note.ABI-tag section first for every exec of a new binary in order
to determine what the target ABI is.
In my opinion, unless H.J. Lu can get the IA-64 ABI changed, it is too
late to assert his view of what that field was supposed to be. Please
leave the EI_OSABI field set to ELFOSABI_LINUX until a better consensus
on this issue is reached.
John Marvin
jsm@fc.hp.com
From alan@linuxcare.com.au Tue, 21 Nov 2000 18:05:36 +1100 (EST)
Date: Tue, 21 Nov 2000 18:05:36 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] signal handling problems (32 bit kernel)
On Mon, 20 Nov 2000, Richard Hirst wrote:
> #warning XXX FIXME probably bogus -PB
> /* I think this is bogus -- it'll cause the first instn of the
> * signal handler to be executed twice! Better might be to
Definitely bogus, as with quite a lot of iaoq manipulation in signal.c
> I've fixed that by setting iaoq[1] = iaoq[0]+4. Is that OK, or
Sounds like the right thing to do. I reckon everywhere ioaq is fudged
from/to r31 should have this sort of mapping. ie. it should be
regs->gr[31] = regs->iaoq[0];
in sys_rt_sigreturn, and
err |= __put_user(regs->gr[31], &sc->sc_iaoq[0]);
err |= __put_user(regs->gr[31] + 4, &sc->sc_iaoq[1]);
in setup_sigcontext, and so on. I'm guessing the origial author of this
code didn't know which of iaoq[0] and ioaq[1] was ioaq_front. :)
> and then in another terminal do 'kill -USR1 '. The program
> either goes 'Wah!', gives a SEGV, or works. That seems to be because
> %r1 is corrupted while processing the signal. The signal handler ends
> with a syscall (rt_sigreturn_wrapper), and %r1, at least, is not saved
> and restored over the syscall. %r31 also appears to get corrupted, as
> it is used in the final branch of the syscall return.
I don't really understand what is going on here, but it seems wrong to me
that setup_rt_frame should be touching regs->iaoq at all when in_syscall.
Hmm, and in that case why all the other gr[31] to/from ioaq[] fudgery?
Alan
--
Linuxcare. Support for the Revolution.
From rhirst@linuxcare.com Tue, 21 Nov 2000 09:34:42 +0000
Date: Tue, 21 Nov 2000 09:34:42 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] signal handling problems (32 bit kernel)
On Tue, Nov 21, 2000 at 06:05:36PM +1100, Alan Modra wrote:
> On Mon, 20 Nov 2000, Richard Hirst wrote:
>
> > #warning XXX FIXME probably bogus -PB
> > /* I think this is bogus -- it'll cause the first instn of the
> > * signal handler to be executed twice! Better might be to
>
> Definitely bogus, as with quite a lot of iaoq manipulation in signal.c
As another example, if a process gets a signal as it is about to
execute the instr in the delay slot of a branch, it forgets that it
was supposed to be branching on return from the signal handler. Try
compiling the following and sending it a SIGUSR1:
#include
#include
#include
#include
void sig_handler(int sig)
{
}
int main()
{
struct sigaction act;
int i = 1;
memset(&act, 0, sizeof(act));
act.sa_handler = sig_handler;
sigaction(SIGUSR1, &act, NULL);
printf("I am %d\n", getpid());
while (i++)
;
printf("Escaped, i=%d!\n", i);
return 0;
}
Oh, you have to run it with "LD_BIND_NOW=1 " to avoid one of
the other problems.
Time to try and work out what signal.c is really trying to do, I guess.
Richard
From alan@linuxcare.com.au Tue, 21 Nov 2000 22:26:14 +1100 (EST)
Date: Tue, 21 Nov 2000 22:26:14 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] signal handling problems (32 bit kernel)
On Tue, 21 Nov 2000, Richard Hirst wrote:
> Time to try and work out what signal.c is really trying to do, I guess.
Maybe you should start by considering all the possible states a task can
be in when a signal is delivered, in regards to saved registers and stack
layout. It would be a _very_ good idea to formalize this once you've
sorted it out by splitting up struct pt_regs appropriately. ie. as other
architectures do, into struct pt_regs and struct switch_stack. Actually,
parisc could go one further and have three structures, one corresponding
to registers saved on syscall entry (new pt_regs), one corresponding to
macro callee_save (switch_stack), and one corresponding more or less to
macro save_specials. Quite a bit of work, but IMO well worth doing.
Alan Modra
--
Linuxcare. Support for the Revolution.
From matthew@wil.cx Tue, 21 Nov 2000 11:34:32 +0000
Date: Tue, 21 Nov 2000 11:34:32 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: [parisc-linux] CVS linux Vs. -test10
On Mon, Nov 20, 2000 at 09:34:31AM -0800, Grant Grundler wrote:
> Ok. What's the first step to getting arch/parisc* and include/asm-parisc*
> into Linus's tree?
Someone (probably me) sends him a patch. He told me at the Toronto
show that he was quite happy to apply anything that only touched those
two directories. (oh, and drivers/gsc wouldn't be a problem either).
Can I just check that no-one wants to rename drivers/gsc again? :-)
> I had dinner with Bdale Garbee last night and one of two things he made
> clear was we need to unfork from debian and linus's tree in order to move
> forward. All our CVS branches need to become obsolete or "local sandboxes"
> of the respective upstream partners. Feeding kernel bits upstream will
> bring a new level of visibility (and *HELP*) to the parisc-linux port.
that's true. last time we discussed this several people were unhappy
with the idea of sending our current work to Linus. Is anyone unhappy
with doing this now?
> I also find it odd that glibc is merged upstream *before* the kernel is.
glibc is more portable :-)
> The part you described above is the easy part - np.
> I'm worried about labels and tracking how we "name" the releases.
> Mang or other CVS ninja's care to comment?
don't tag it. just commit it. tags are laid down at big events, not
when you fix bugs or undo changes.
--
Revolutions do not require corporate support.
From jsm@udlkern.fc.hp.com Tue, 21 Nov 2000 06:11:29 -0700 (MST)
Date: Tue, 21 Nov 2000 06:11:29 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: CVS linux Vs. -test10
> > I had dinner with Bdale Garbee last night and one of two things he made
> > clear was we need to unfork from debian and linus's tree in order to move
> > forward. All our CVS branches need to become obsolete or "local sandboxes"
> > of the respective upstream partners. Feeding kernel bits upstream will
> > bring a new level of visibility (and *HELP*) to the parisc-linux port.
>
> that's true. last time we discussed this several people were unhappy
> with the idea of sending our current work to Linus. Is anyone unhappy
> with doing this now?
>
Are we suggesting that we populate include/asm-parisc* and arch/parisc*
WITHOUT the machine independent changes in the rest of the tree?
If that is the case, I guess I don't object, although I would want
to make sure that Linus knew the state of the code, and that it would
not work without a patch containing changes to the machine independent
part, and that followup patches to these branches are likely to be
huge. We should also add a README in arch/parisc that explains the
above (and where to get patches, how to access CVS, etc.). We also
need someone to set up an automatic nightly? patch generator.
I certainly don't want to try to get the machine independent changes in
yet, since we still have some major issues to fix/clean there, and I doubt
Linus would want them at this time anyway.
John Marvin
jsm@fc.hp.com
From rhirst@linuxcare.com Tue, 21 Nov 2000 16:54:42 +0000
Date: Tue, 21 Nov 2000 16:54:42 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] signal handling problems (32 bit kernel)
On Tue, Nov 21, 2000 at 06:05:36PM +1100, Alan Modra wrote:
> > and then in another terminal do 'kill -USR1 '. The program
> > either goes 'Wah!', gives a SEGV, or works. That seems to be because
> > %r1 is corrupted while processing the signal. The signal handler ends
> > with a syscall (rt_sigreturn_wrapper), and %r1, at least, is not saved
> > and restored over the syscall. %r31 also appears to get corrupted, as
> > it is used in the final branch of the syscall return.
>
> I don't really understand what is going on here, but it seems wrong to me
> that setup_rt_frame should be touching regs->iaoq at all when in_syscall.
Problem is that whenever a signal handler is invoked, it is followed
by a sys_rt_sigreturn syscall (invoked via the trampoline code), before
returning to original usercode. I can imagine that this might work if
the process in question was blocked on a syscall, but not if it was
suspended due to an interrupt. In either case the path back to the
original user code is the syscall return path, and that obviously cannot
put things back as they were if the process was interrupted.
I think the answer is for syscall_do_signal to save enough in the
pt_regs such that sys_rt_sigreturn_wrapper can return to userland
via an RFI (like intr_restore, but remembering to trace the syscall
exit if necessary) regardless of the process state when the signal
occurred.
Richard
From taggart@carmen.fc.hp.com Tue, 21 Nov 2000 10:20:35 -0700
Date: Tue, 21 Nov 2000 10:20:35 -0700
From: Matt Taggart taggart@carmen.fc.hp.com
Subject: [parisc-linux] kernel BUG at sched.c:692!
I arrived this morning to discover both my B160 and my C3000 were printing the
following as fast as they could,
Scheduling in interrupt
kernel BUG at sched.c:692!
Looking at linux/kernel/sched.c here's the relevant section,
690 scheduling_in_interrupt:
691 printk("Scheduling in interrupt\n");
692 BUG();
693 return;
scheduling_in_interrupt is called from line 516. Here's some context,
500 asmlinkage void schedule(void)
501 {
502 struct schedule_data * sched_data;
503 struct task_struct *prev, *next, *p;
504 struct list_head *tmp;
505 int this_cpu, c;
506
507 if (!current->active_mm) BUG();
508 if (tq_scheduler)
509 goto handle_tq_scheduler;
510 tq_scheduler_back:
511
512 prev = current;
513 this_cpu = prev->processor;
514
515 if (in_interrupt())
516 goto scheduling_in_interrupt;
517
518 release_kernel_lock(prev, this_cpu);
I thought it was odd that both systems chose to freak out last night when I
have never seen this problem on either of them. The B160 is running a slightly
newer kernel and had been up for about 3 days as of last night. The C3K had
been up for 4-5 days. When I left last night the B160 was doing a build
although the it looks like the build died before this problem occurred. The
C3K was idle.
Any ideas?
Thanks,
--
Matt Taggart
taggart@fc.hp.com
From robert.reilly@gecapital.com Tue, 21 Nov 2000 18:26:41 GMT
Date: Tue, 21 Nov 2000 18:26:41 GMT
From: Robert Reilly robert.reilly@gecapital.com
Subject: [parisc-linux] cd .5
Hi All,
I have HP9000/d212 I was able to boot off the cdrom image no problem,
however when I get the login prompt I am only able to type in two
characters and the screen redraws itself and prompts me for a login. I am
using a HP700 serial terminal on ttyS0.I briefly scanned the mailing list
but did not find anything. Any help would be appreciated.
--Robert Reilly
GE Capital Card Service
UNIX Administrator
robert.reilly@gecapital.com
From matthew@wil.cx Tue, 21 Nov 2000 17:38:57 +0000
Date: Tue, 21 Nov 2000 17:38:57 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: CVS linux Vs. -test10
On Tue, Nov 21, 2000 at 06:11:29AM -0700, John Marvin wrote:
> Are we suggesting that we populate include/asm-parisc* and arch/parisc*
> WITHOUT the machine independent changes in the rest of the tree?
Yes.
> If that is the case, I guess I don't object, although I would want
> to make sure that Linus knew the state of the code, and that it would
> not work without a patch containing changes to the machine independent
> part, and that followup patches to these branches are likely to be
> huge. We should also add a README in arch/parisc that explains the
> above (and where to get patches, how to access CVS, etc.). We also
> need someone to set up an automatic nightly? patch generator.
Agreed. The patch generation is not a big deal to arrange.
> I certainly don't want to try to get the machine independent changes in
> yet, since we still have some major issues to fix/clean there, and I doubt
> Linus would want them at this time anyway.
Certainly none of the stack direction changes. Some of the MI stuff is
arguably stuff we could put in.
--
Revolutions do not require corporate support.
From Pirvulescu_Alexandru@telemobil.ro Tue, 21 Nov 2000 20:49:57 +0200
Date: Tue, 21 Nov 2000 20:49:57 +0200
From: Alexandru Pirvulescu Pirvulescu_Alexandru@telemobil.ro
Subject: [parisc-linux] case LEDs
Hello
Is there any possibility to activate the case LEDs? I mean the heartbeat and
the network activity
I have a A180 machine
Alex
From grundler@cup.hp.com Tue, 21 Nov 2000 10:55:06 -0800
Date: Tue, 21 Nov 2000 10:55:06 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] case LEDs
"Alexandru Pirvulescu" wrote:
> Hello
>
> Is there any possibility to activate the case LEDs? I mean the heartbeat and
> the network activity. I have a A180 machine.
Yes.
I was already asked this weekend to dig up technical info on LED
and soft power control. I guess this is my reminder to do that. :^)
grant
From grundler@cup.hp.com Tue, 21 Nov 2000 10:59:24 -0800
Date: Tue, 21 Nov 2000 10:59:24 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] cd .5
Robert Reilly wrote:
> Hi All,
> I have HP9000/d212 I was able to boot off the cdrom image no problem,
> however when I get the login prompt I am only able to type in two
> characters and the screen redraws itself and prompts me for a login. I am
> using a HP700 serial terminal on ttyS0. I briefly scanned the mailing list
> but did not find anything. Any help would be appreciated.
Robert,
I would suspent the terminal isn't configured correctly for the
serial connection. Is it set up for 9600 baud, 8 data bits, no parity,
1 stop bit?
I'm pretty sure you have the baud rate correct since you get a login
prompt. I'm not sure all the other things must be correct to get
that far.
I'm not familiar with "HP700 serial terminal". But all the HP terminals
I've worked with in the past also support vt100 emulation mode. You
probably want to set that too.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From kailasr@webcash.com Tue, 21 Nov 2000 12:30:12 -0800
Date: Tue, 21 Nov 2000 12:30:12 -0800
From: Kailashnath V Rampure kailasr@webcash.com
Subject: [parisc-linux] Apache package.
Hi Paul,
I have installed Apache + mod_ssl + mod_perl on HP A Class server.
I need help on to change the server looking fort bootpd server and put the
IP address on the system. Also where can I find the ftp client for linux .
Regards
Kalas
At 04:59 PM 11/15/00 -0700, Paul Bame wrote:
>= I tried to build apache_1.3.12 on HP a class server. But I have error. I
>= have tried to check the site
>= ftp://ftp.debian.org/debian/dists/unstable/main/binary-hppa/
>= I could not find one. I found some apache-doc etc.
>
>We are still working on some kernel features which are required to
>support Apache (system 5 shared memory).
>
> -P
From grundler@cup.hp.com Tue, 21 Nov 2000 13:24:31 -0800
Date: Tue, 21 Nov 2000 13:24:31 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
Matthew Wilcox wrote:
...
> Someone (probably me) sends him a patch. He told me at the Toronto
> show that he was quite happy to apply anything that only touched those
> two directories. (oh, and drivers/gsc wouldn't be a problem either).
> Can I just check that no-one wants to rename drivers/gsc again? :-)
Hi Mathew,
I don't and it's a good question.
I would like a few files moved:
arch/parisc/kernel/ccio-dma.c -> drivers/gsc/ccio-dma.c
arch/parisc/kernel/ccio-rm-dma.c -> drivers/gsc/ccio-rm-dma.c
ccio will *always* be associated with a GSC bus since that's
the secondary bus. And ccio supports devices below dino.c which
already lives in drivers/gsc.
arch/parisc/kernel/lba_pci.c -> drivers/ropes/lba_pci.c
arch/parisc/kernel/sba_iommu.c -> drivers/ropes/sba_iommu.c
arch/parisc/kernel/iosapic.c -> drivers/ropes/iosapic.c
lba/sba code is equivalent to dino/ccio code for another set
of platforms. And long term, I'm certain iosapic.c does not
belong under arch/parisc. I can do this move if there are no
major objections.
Any reason why we couldn't do these moves *after* you submit a patch?
FWIW, here are issues I see with merging IA64 iosapic code with mine:
o iosapic "discovery" (I invented register_iosapic() interface for parisc)
o parisc PDC calls (initialization)
o interrupt policy decisions (eg EOI generation and picking a CPU)
o I don't have time to do it in the near future.
Folks working on IA64 stuff inside HP need to think about:
(a) if they want to do the merge any time soon
(b) which iosapic.c they want to start with
(c) where the final version should live
thanks,
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From pjlahaie@linuxcare.com Tue, 21 Nov 2000 16:41:48 -0500
Date: Tue, 21 Nov 2000 16:41:48 -0500
From: Paul J.Y. Lahaie pjlahaie@linuxcare.com
Subject: [parisc-linux] Fun build problems
--PW0Eas8rCkcu1VkF
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
Hello everyone,
Over the last couple of days, I've been trying to get the debian
boot-floppies building (I know some changes will be necessary) and as
it stands now, several packages are missing. In the process of trying
to compile/install these packages, I've run into some difficulties.
Here are some of them:
apt -- Package on pehc and .iso do not work. Missing lots of helper apps.
Seeing as apt was broken, I decided to download the woody version of apt so
I can get a newer version and hopefully skip some steps in building the
above packages. gcc 2.96 and apt-0.3.19 don't get along, so I tried picking
it up from cvs (supposed to have been fixed). Follow the directions on
cvs.debian.org
$ cvs -d :pserver:anonymous@cvs.debian.org:/cvs/APT login
(Logging in to anonymous@cvs.debian.org)
CVS password:=20
cvs login: authorization failed: server cvs.debian.org rejected access
Anyone know if there are other instructions other than what cvs.debian.org
says? If so, feel free to email me
rsync -- ldd crashes on dpkg-shlibdepends stage
Can I just manually put the required dependancies in?
autoconf -- installinfo spins
When installing the autoconf package (to compile dpkg) installinfo spins.
And after several minutes still doesn't return. It's taking up 100% cpu
at the time.
If I abort this, autoconf seems to be installed but cannot generate the
dpkg configure properly, some macros are missing
AM_CONDITIONAL(HAVE_CPLUSPLUS, test "$CXX" !=3D "")
Is in the ./configure script.
info -- Reinstall spins
If I try to reinstall info, it spins on the uninstall. 100% cpu.
If anyone has any solutions for any of the above problems, I am all
ears (eyes?).
- Paul
--PW0Eas8rCkcu1VkF
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE6Guwc8ggPQthPCzcRAgkGAJ9TzClXOg009RWO/v09ONBNnxJcbgCfcWcX
gj26osR06BqyJ0O+/Q83csk=
=YHRW
-----END PGP SIGNATURE-----
--PW0Eas8rCkcu1VkF--
From alan@linuxcare.com.au Wed, 22 Nov 2000 09:50:05 +1100 (EST)
Date: Wed, 22 Nov 2000 09:50:05 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] Use of the EI_OSABI field
cc to binutils because John makes some salient points.
On Mon, 20 Nov 2000, John Marvin wrote:
> > Another glibc issue (which is why I'm sending this back to the list) is
> > that there has been quite some discussion on the binutils list over the
> > use of the EI_OSABI field. The conclusion being that it isn't really
> > correct to use this field to discern the intendend execution platform.
> > It's purpose is really to tell ELF tools that a file contains fields and
> > values that need to be interpreted in a non-standard way. Since both
> > elf32-hppa and elf64-hppa follow the standards, I'm inclined to think that
> > the gnu tools should set EI_OSABI to zero for both HPUX and Linux targets.
>
> I don't read the binutils mailing list, so I checked the archive. In my
> opinion I didn't see a discussion, I saw H.J. Lu repeatedly asserting
> this "fact". He may or may not be correct about original intention, but
> the implementation so far has been that EI_OSABI is used to tell what the
> target platform is. That is what FreeBSD is doing, that is what HP-UX is
> doing, and that is what the IA-64 Unix System V Application Binary
> Interface specifies:
>
> http://developer.intel.com/design/IA-64/Downloads/24537002.pdf
>
> Note that the second paragraph in section 1.1 of that specification
> includes the following:
>
> This document is the result of consensus among operating system
> vendors intending to provide UNIX and UNIX workalike operating
> systems on the IA-64 architecture. The vendors participating in
> this effort include Intel, Sun Microsystems, SCO, IBM, SGI,
> Cygnus Solutions, VA Linux Systems, HP, and Compaq.
>
> So, If the specification is wrong according to H.J. Lu, why did both
> Cygnus and VA Linux Systems not object to this:
>
> Section 4.1.1.4 Operating System Identification
>
> The e_ident[EI_OSABI] value identifies the operating system and
> ABI to which the object is targeted, as listed in Table 4-1.
>
> Table 4-1 lists the various ELFOSABI_* fields, e.g. ELFOSABI_HPUX,
> ELFOSABI_NETBSD, ELFOSABI_LINUX, etc.
>
> Note also, that the current mechanism for us to be able to differentiate
> elf executables in the Linux kernel is the machine dependent
> elf_check_arch() macro, whose only argument is a pointer to a
> elf32_hdr or elf64_hdr. I think it would be ugly to try to get the
> .note.ABI-tag section first for every exec of a new binary in order
> to determine what the target ABI is.
>
> In my opinion, unless H.J. Lu can get the IA-64 ABI changed, it is too
> late to assert his view of what that field was supposed to be. Please
> leave the EI_OSABI field set to ELFOSABI_LINUX until a better consensus
> on this issue is reached.
>
> John Marvin
> jsm@fc.hp.com
I'm happy enough to leave things as they are in puffin CVS, but these
changes haven't been merged back to sourceware yet, partly because I had
some doubts myself as to whether setting EI_OSABI was correct. H.J. Lu
wasn't the only one arguing that EI_OSABI should be left at zero; Ulrich
Drepper also was quite vehement against changing sourceware FreeBSD
binutils.
BTW, it's not too hard to check .note.ABI-tag. The linker arranges for a
PT_NOTE program header entry to point to it, and the section itself is
virtually guaranteed to be read in with the header as it's placed right
after the header along with .interp.
Alan Modra
--
Linuxcare. Support for the Revolution.
From drepper@redhat.com 21 Nov 2000 15:27:19 -0800
Date: 21 Nov 2000 15:27:19 -0800
From: Ulrich Drepper drepper@redhat.com
Subject: [parisc-linux] Use of the EI_OSABI field
Alan Modra writes:
> Ulrich Drepper also was quite vehement against changing sourceware
> FreeBSD binutils.
I've never said anything about any *BSD, why should I? The *BSD
people wanted to change the Linux binutils.
Anyway, the ABI value is zero unless you implement ELF extensions.
--
---------------. ,-. 1325 Chesapeake Terrace
Ulrich Drepper \ ,-------------------' \ Sunnyvale, CA 94089 USA
Red Hat `--' drepper at redhat.com `------------------------
From alan@linuxcare.com.au Wed, 22 Nov 2000 11:13:29 +1100 (EST)
Date: Wed, 22 Nov 2000 11:13:29 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] Use of the EI_OSABI field
On 21 Nov 2000, Ulrich Drepper wrote:
> Alan Modra writes:
>
> > Ulrich Drepper also was quite vehement against changing sourceware
> > FreeBSD binutils.
>
> I've never said anything about any *BSD, why should I? The *BSD
> people wanted to change the Linux binutils.
Sorry, I stated that badly.
> Anyway, the ABI value is zero unless you implement ELF extensions.
Exactly what is an "ELF extension"? Anything outside gABI or
"gABI + psABI"? Handly the latter, as it seems to me that a processor
specific ABI can specify extensions. There's also the awkward possibility
that a psABI may specify an extension that is later incorporated into a
new revision of the gABI (eg. hpux and DT_INIT_ARRAY) Does that mean that
if a new revision of the gABI completely incorporates all previous
extensions, that EI_OSABI should become zero?
Yes, I'm arguing that "No ELF extensions => EI_OSABI == 0" is not
necessarily true, but I'm _not_ arguing that changing x86 Linux binutils
is wise. Historical usage is important. If we were to change x86 Linux
binutils to set EI_OSABI, then that can only be after a considerable
period of time to allow code such as glibc to accept a new branding.
Alan Modra
--
Linuxcare. Support for the Revolution.
From drepper@redhat.com 21 Nov 2000 16:31:39 -0800
Date: 21 Nov 2000 16:31:39 -0800
From: Ulrich Drepper drepper@redhat.com
Subject: [parisc-linux] Use of the EI_OSABI field
Alan Modra writes:
> Exactly what is an "ELF extension"? Anything outside gABI or
> "gABI + psABI"?
Anything beyond the psABI that cannot be handled by the rules in the
gABI.
--
---------------. ,-. 1325 Chesapeake Terrace
Ulrich Drepper \ ,-------------------' \ Sunnyvale, CA 94089 USA
Red Hat `--' drepper at redhat.com `------------------------
From hjl@valinux.com Tue, 21 Nov 2000 16:53:27 -0800
Date: Tue, 21 Nov 2000 16:53:27 -0800
From: H . J . Lu hjl@valinux.com
Subject: [parisc-linux] Use of the EI_OSABI field
On Wed, Nov 22, 2000 at 11:13:29AM +1100, Alan Modra wrote:
>
> > Anyway, the ABI value is zero unless you implement ELF extensions.
>
> Exactly what is an "ELF extension"? Anything outside gABI or
ELF extension is something whose interpretation is not documented in
neither gABI nor psABI. HP's old ELF is a good example since it didn't
implement DT_INIT/DT_FINI, but added DT_INIT_ARRARY/DT_FINI_ARRAY. A
normal ELF tool will have a hard time to interpret those fields and
their values.
H.J.
From matthew@wil.cx Wed, 22 Nov 2000 00:53:09 +0000
Date: Wed, 22 Nov 2000 00:53:09 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: [parisc-linux] CVS linux Vs. -test10
On Tue, Nov 21, 2000 at 01:24:31PM -0800, Grant Grundler wrote:
> I would like a few files moved:
>
> arch/parisc/kernel/ccio-dma.c -> drivers/gsc/ccio-dma.c
> arch/parisc/kernel/ccio-rm-dma.c -> drivers/gsc/ccio-rm-dma.c
> arch/parisc/kernel/lba_pci.c -> drivers/ropes/lba_pci.c
> arch/parisc/kernel/sba_iommu.c -> drivers/ropes/sba_iommu.c
> arch/parisc/kernel/iosapic.c -> drivers/ropes/iosapic.c
> Any reason why we couldn't do these moves *after* you submit a patch?
Better to get our house in order before we patchbomb Linus, I think.
Renames are hard enough in CVS; renames in diff -u format are a real
stinker :-)
--
Revolutions do not require corporate support.
From adevries@linuxcare.com Tue, 21 Nov 2000 21:27:51 -0500
Date: Tue, 21 Nov 2000 21:27:51 -0500
From: Alex deVries adevries@linuxcare.com
Subject: [parisc-linux] case LEDs
Grant Grundler wrote:
> Yes.
> I was already asked this weekend to dig up technical info on LED
> and soft power control. I guess this is my reminder to do that. :^)
Isn't there a PDC call (pdc_chassis?) to do this? Or is the heartbeat
LED done by hardware?
- Alex
--
Alex deVries, Principal Solutions Architect, The Puffins at Linuxcare
613.562.2759 tel
alex@linuxcare.com, http://www.linuxcare.com/
Linuxcare, Support for the revolution.
From randolph@tausq.org Tue, 21 Nov 2000 18:50:24 -0700
Date: Tue, 21 Nov 2000 18:50:24 -0700
From: Randolph Chung randolph@tausq.org
Subject: [parisc-linux] Fun build problems
--9zSXsLTf0vkW971A
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
> Seeing as apt was broken, I decided to download the woody version of apt =
so
> I can get a newer version and hopefully skip some steps in building the
> above packages. gcc 2.96 and apt-0.3.19 don't get along, so I tried pick=
ing
> it up from cvs (supposed to have been fixed). Follow the directions on
> cvs.debian.org
>=20
> $ cvs -d :pserver:anonymous@cvs.debian.org:/cvs/APT login
> (Logging in to anonymous@cvs.debian.org)
> CVS password:=20
> cvs login: authorization failed: server cvs.debian.org rejected access
Try:
cvs -d:pserver:anonymous@cvs.debian.org:/cvs/deity login
(empty password)
It should work. if not let me know and i can mail you a tarball. You
probably want to get the aliencode branch, which has hppa patches. Let
me know if you run into any problems.
> rsync -- ldd crashes on dpkg-shlibdepends stage
> Can I just manually put the required dependancies in?
Yes, or pick up the dpkg-architecture and dpkg-shlibdeps scripts from
http://puffin.external.hp.com/~tausq/, or wait for the next version of
dpkg to get built for hppa (it doesn't use ldd anymore)
> autoconf -- installinfo spins
>=20
> When installing the autoconf package (to compile dpkg) installinfo spins.
> And after several minutes still doesn't return. It's taking up 100% cpu
> at the time.
hmmm.. didn't see this. I had built a slightly older version of dpkg
successfully.
randolph (tausq@debian.org)
--=20
@..@ http://www.TauSq.org/
(----)
( >__< )
^^ ~~ ^^
--9zSXsLTf0vkW971A
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.1 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE6GyZgULspdC1Zp9IRAozyAKC/Df1fexltMjFlcwpeYlD+IIWEigCgiGyT
YKOjLA0BQzMo7qOj8jC1BTI=
=whh1
-----END PGP SIGNATURE-----
--9zSXsLTf0vkW971A--
From hjl@valinux.com Tue, 21 Nov 2000 19:11:29 -0800
Date: Tue, 21 Nov 2000 19:11:29 -0800
From: H . J . Lu hjl@valinux.com
Subject: [parisc-linux] Use of the EI_OSABI field
On Wed, Nov 22, 2000 at 02:03:07PM +1100, Alan Modra wrote:
> On Tue, 21 Nov 2000, H . J . Lu wrote:
>
> > On Wed, Nov 22, 2000 at 11:13:29AM +1100, Alan Modra wrote:
> > >
> > > > Anyway, the ABI value is zero unless you implement ELF extensions.
> > >
> > > Exactly what is an "ELF extension"? Anything outside gABI or
> >
> > ELF extension is something whose interpretation is not documented in
> > neither gABI nor psABI. HP's old ELF is a good example since it didn't
> > implement DT_INIT/DT_FINI, but added DT_INIT_ARRARY/DT_FINI_ARRAY. A
> > normal ELF tool will have a hard time to interpret those fields and
> > their values.
>
> Neither you nor Ulrich have responded to John's point that the IA64 psABI
> states
> "The e_ident[EI_OSABI] value identifies the operating system and ABI to
> which the object is targeted"
>
> Granted, this is only in a processor specific ABI, but it does flatly
> contradict your assertions that the purpose of EI_OSABI is to flag the
> presense of ELF extensions.
>
When I was at the IA64 ABI meeting yesterday, we were looking at the
EI_OSABI issue. Everyone at the IA64 ABI meeting agreed that what
Ulrich and I had said was the correct understanding.
"The e_ident[EI_OSABI] value identifies the operating system and ABI
to which the object is targeted" just means if the e_ident[EI_OSABI]
value is not ELFOSABI_NONE, you have to look somewhere else in addition
to gABI and IA64 psABI to interpret the ELF fields/values specific to
that OS. Otherwise, gABI and IA64 psABI are sufficient.
--
H.J. Lu (hjl@valinux.com)
From drepper@redhat.com 21 Nov 2000 19:18:21 -0800
Date: 21 Nov 2000 19:18:21 -0800
From: Ulrich Drepper drepper@redhat.com
Subject: [parisc-linux] Use of the EI_OSABI field
Alan Modra writes:
> "The e_ident[EI_OSABI] value identifies the operating system and ABI to
> which the object is targeted"
>
> Granted, this is only in a processor specific ABI, but it does flatly
> contradict your assertions that the purpose of EI_OSABI is to flag the
> presense of ELF extensions.
The meaning of the field changed over time. Or better said, some
people initially understood it differently.
--
---------------. ,-. 1325 Chesapeake Terrace
Ulrich Drepper \ ,-------------------' \ Sunnyvale, CA 94089 USA
Red Hat `--' drepper at redhat.com `------------------------
From alan@linuxcare.com.au Wed, 22 Nov 2000 14:03:07 +1100 (EST)
Date: Wed, 22 Nov 2000 14:03:07 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] Use of the EI_OSABI field
On Tue, 21 Nov 2000, H . J . Lu wrote:
> On Wed, Nov 22, 2000 at 11:13:29AM +1100, Alan Modra wrote:
> >
> > > Anyway, the ABI value is zero unless you implement ELF extensions.
> >
> > Exactly what is an "ELF extension"? Anything outside gABI or
>
> ELF extension is something whose interpretation is not documented in
> neither gABI nor psABI. HP's old ELF is a good example since it didn't
> implement DT_INIT/DT_FINI, but added DT_INIT_ARRARY/DT_FINI_ARRAY. A
> normal ELF tool will have a hard time to interpret those fields and
> their values.
Neither you nor Ulrich have responded to John's point that the IA64 psABI
states
"The e_ident[EI_OSABI] value identifies the operating system and ABI to
which the object is targeted"
Granted, this is only in a processor specific ABI, but it does flatly
contradict your assertions that the purpose of EI_OSABI is to flag the
presense of ELF extensions.
Alan Modra
--
Linuxcare. Support for the Revolution.
From rbradetich@uswest.net Tue, 21 Nov 2000 22:54:11 -0800
Date: Tue, 21 Nov 2000 22:54:11 -0800
From: Ryan Bradetich rbradetich@uswest.net
Subject: [parisc-linux] CVS linux Vs. -test10
Grant Grundler wrote:
> Matthew Wilcox wrote:
> ...
> > Someone (probably me) sends him a patch. He told me at the Toronto
> > show that he was quite happy to apply anything that only touched those
> > two directories. (oh, and drivers/gsc wouldn't be a problem either).
> > Can I just check that no-one wants to rename drivers/gsc again? :-)
>
> Hi Mathew,
> I don't and it's a good question.
> I would like a few files moved:
>
> arch/parisc/kernel/ccio-dma.c -> drivers/gsc/ccio-dma.c
> arch/parisc/kernel/ccio-rm-dma.c -> drivers/gsc/ccio-rm-dma.c
Grant,
Do you really want to merget the ccio-rm-dma.c file into Linus's tree?
It is just a reference file used to construct the real ccio-dma.c file ... I
don't believe it is referenced anywhere.
I'll double check this in the morning.
- Ryan
> ccio will *always* be associated with a GSC bus since that's
> the secondary bus. And ccio supports devices below dino.c which
> already lives in drivers/gsc.
>
> arch/parisc/kernel/lba_pci.c -> drivers/ropes/lba_pci.c
> arch/parisc/kernel/sba_iommu.c -> drivers/ropes/sba_iommu.c
> arch/parisc/kernel/iosapic.c -> drivers/ropes/iosapic.c
>
> lba/sba code is equivalent to dino/ccio code for another set
> of platforms. And long term, I'm certain iosapic.c does not
> belong under arch/parisc. I can do this move if there are no
> major objections.
>
> Any reason why we couldn't do these moves *after* you submit a patch?
>
> FWIW, here are issues I see with merging IA64 iosapic code with mine:
> o iosapic "discovery" (I invented register_iosapic() interface for parisc)
> o parisc PDC calls (initialization)
> o interrupt policy decisions (eg EOI generation and picking a CPU)
> o I don't have time to do it in the near future.
>
> Folks working on IA64 stuff inside HP need to think about:
> (a) if they want to do the merge any time soon
> (b) which iosapic.c they want to start with
> (c) where the final version should live
>
> thanks,
> grant
>
> Grant Grundler
> Unix Systems Enablement Lab
> +1.408.447.7253
>
> ---------------------------------------------------------------------------
> To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com with
> `unsubscribe' as the subject.
From jsm@udlkern.fc.hp.com Tue, 21 Nov 2000 23:50:03 -0700 (MST)
Date: Tue, 21 Nov 2000 23:50:03 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
> Better to get our house in order before we patchbomb Linus, I think.
> Renames are hard enough in CVS; renames in diff -u format are a real
> stinker :-)
In that case, we need to do some cleanup first. I've been lobbying for
the removal of the almost empty arch/parisc/real directory, and its few
remaining valid files moved to the kernel directory. There are also a
fair number of dead files. Every file that is not currently involved in
the build should be removed, unless a good case for it remaining can be
made. If the reason to keep it is not a long term reason, then that file
should not be sent to Linus (It sounds like it is a lot easier to add
files rather than remove/rename them).
If there are any files that are currently in use, but which we know
will eventually be removed, perhaps we should consider what to do with
that file (although I don't know of any files in this category at the
moment).
John Marvin
jsm@fc.hp.com
From grundler@cup.hp.com Tue, 21 Nov 2000 23:18:16 -0800
Date: Tue, 21 Nov 2000 23:18:16 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
Ryan Bradetich wrote:
> Do you really want to merget the ccio-rm-dma.c file into Linus's tree?
> It is just a reference file used to construct the real ccio-dma.c file ... I
> don't believe it is referenced anywhere.
Hi Ryan,
Yes I do. It is supposed to work for ccio+PCX-W platforms (eg C360).
Keeping it arround will document the pro/con's of that approach and
give folks who have time (and the right machine) something to experiment
with instead of writing it from scratch. If someone finds an application
it's good for (short transactions with low latency requirements perhaps),
it's worth having around.
It's not referenced because I didn't add a CONFIG_CCIO_RM_IOMMU flag
or ccio_rm_init() call to drivers/gsc/gsc.c:gsc_init(). You are welcome
add this CONFIG flag by hacking arch/parisc/config.in and defconfig.
If you do, please add rules which only allow one or the other
CONFIG_CCIO* option to be enabled.
thanks,
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From Pirvulescu_Alexandru@telemobil.ro Wed, 22 Nov 2000 09:36:58 +0200
Date: Wed, 22 Nov 2000 09:36:58 +0200
From: Alexandru Pirvulescu Pirvulescu_Alexandru@telemobil.ro
Subject: [parisc-linux] case LEDs
I think that the heartbeat LED is software because it starts with the kernel
boot and if the kernel stops the LED stops
blinking too. Is better to implement a software monitoring tool because you
have to watch the software for hanging.
Alex
PS. There is a function in the kernel source in
linux/arch/parisc/kernel/process.c named machine_heartbeat(). Any connection
with the heartbeat from the chassis?
----- Original Message -----
From: "Alex deVries"
To: "Grant Grundler"
Cc: "Alexandru Pirvulescu" ;
Sent: Wednesday, November 22, 2000 4:27 AM
Subject: Re: [parisc-linux] case LEDs
> Grant Grundler wrote:
> > Yes.
> > I was already asked this weekend to dig up technical info on LED
> > and soft power control. I guess this is my reminder to do that. :^)
>
> Isn't there a PDC call (pdc_chassis?) to do this? Or is the heartbeat
> LED done by hardware?
>
> - Alex
>
> --
> Alex deVries, Principal Solutions Architect, The Puffins at Linuxcare
> 613.562.2759 tel
> alex@linuxcare.com, http://www.linuxcare.com/
> Linuxcare, Support for the revolution.
>
>
From bvalkema@knowhowww.nl Wed, 22 Nov 2000 06:32:51 +0100
Date: Wed, 22 Nov 2000 06:32:51 +0100
From: Bas Valkema bvalkema@knowhowww.nl
Subject: [parisc-linux] case LEDs
Alex deVries wrote:
> Grant Grundler wrote:
> > Yes.
> > I was already asked this weekend to dig up technical info on LED
> > and soft power control. I guess this is my reminder to do that. :^)
>
> Isn't there a PDC call (pdc_chassis?) to do this? Or is the heartbeat
> LED done by hardware?
>
> - Alex
A couple of months ago, I asked the same question, got answer to look in the
mkLinux sources. I did, and I think it was a register (outb(0xblabla);).
Wrote a driver (and for WAX, got a Intel Flash 32 EISA running), can't
release it now, because of some copyright issues...
Bas
From grundler@cup.hp.com Tue, 21 Nov 2000 23:56:35 -0800
Date: Tue, 21 Nov 2000 23:56:35 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
John Marvin wrote:
> In that case, we need to do some cleanup first.
John,
I want a task list which leads us to a submital to linus.
That's why I listed the specific files I wanted moved.
Can you come up with (or ask someone else to come up) with a list of
files which meet your criteria?
All of your criteria sounds reasonable to me.
But I don't have a feel of which files meet your criteria.
If someone makes the task list, I'm happy to help with items
and verify the result works.
thanks,
grant
From jsm@udlkern.fc.hp.com Wed, 22 Nov 2000 01:11:57 -0700 (MST)
Date: Wed, 22 Nov 2000 01:11:57 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
> Ryan Bradetich wrote:
> > Do you really want to merget the ccio-rm-dma.c file into Linus's tree?
> > It is just a reference file used to construct the real ccio-dma.c file ... I
> > don't believe it is referenced anywhere.
>
> Hi Ryan,
> Yes I do. It is supposed to work for ccio+PCX-W platforms (eg C360).
> Keeping it arround will document the pro/con's of that approach and
> give folks who have time (and the right machine) something to experiment
> with instead of writing it from scratch. If someone finds an application
> it's good for (short transactions with low latency requirements perhaps),
> it's worth having around.
>
> It's not referenced because I didn't add a CONFIG_CCIO_RM_IOMMU flag
> or ccio_rm_init() call to drivers/gsc/gsc.c:gsc_init(). You are welcome
> add this CONFIG flag by hacking arch/parisc/config.in and defconfig.
> If you do, please add rules which only allow one or the other
> CONFIG_CCIO* option to be enabled.
>
Well, personally i'd vote to get rid of it. It works for ONE machine only,
and MAY have an advantage in some small case. But if we keep it, lets make
sure that it is real clear that it should NOT be the default choice.
It should be marked CONFIG_EXPERIMENTAL, and the text associated with it
should clearly show that it works on a C360 only. If possible, it should
also be made clear that ccio-dma.c works for C360, so people who have
C360's don't think they have to choose ccio-rm-dma.c.
Grant, I hope you are prepared to answer the parisc-linux mailing list
questions this is going to generate once parisc-linux starts becoming more
visible. Another FAQ entry perhaps? :-)
John Marvin
jsm@fc.hp.com
From jsm@udlkern.fc.hp.com Wed, 22 Nov 2000 01:52:04 -0700 (MST)
Date: Wed, 22 Nov 2000 01:52:04 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: [parisc-linux] Use of the EI_OSABI field
>
> BTW, it's not too hard to check .note.ABI-tag. The linker arranges for a
> PT_NOTE program header entry to point to it, and the section itself is
> virtually guaranteed to be read in with the header as it's placed right
> after the header along with .interp.
I didn't say it was difficult, I said it was ugly. It means another
parisc only change to the machine independent file fs/binfmt_elf.c,
since the hook provided will not allow this check. Without a change,
binfmt_elf.c won't be smart enough to differentiate a binary produced
by Gnu binutils for HP-UX and a binary produced by Gnu binutils for
Linux, so it will claim both, and then blow up later, rather than
not claiming the HP-UX binary and allowing it to be claimed by
an arch dependent binary handler further down the list.
And binfmt_elf.c does NOT read the program headers in the same read, so
another read would have to be done (the data should be found in in cache
rather than going to disk for it). Since we now need both the program
headers and a section header to determine whether or not we should claim
the file AND binfmt_elf.c also wants to look at those headers after
the file is claimed, a small redesign is probably in order (rather than
re-reading the headers). I'm not sure whether or not Linus would buy
that.
So, I guess I'll pursue the interpreter field instead, since that is
what sparc is doing (i.e. they have their own sparc only code in
binfmt_elf.c). Since that will be an easier sell. I need to do more
research here. I suspect that statically linked binaries are not going
to allow this solution to work though.
John Marvin
jsm@fc.hp.com
From alan@linuxcare.com.au Wed, 22 Nov 2000 23:28:45 +1100 (EST)
Date: Wed, 22 Nov 2000 23:28:45 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] binutils update
I've just committed a change to binutils that cures an ELF spec violation.
We now no longer create plabels for DT_INIT and DT_FINI and instead rely
on the dynamic linker to create them for us. This means you need an
updated glibc, that I committed a little earlier today, to create the
plabel function pointers if you update your binutils.
You have been warned...
Alan Modra
--
Linuxcare. Support for the Revolution.
From bruno_vidal@hpfrcu03.france.hp.com Wed, 22 Nov 2000 15:14:18 +0100
Date: Wed, 22 Nov 2000 15:14:18 +0100
From: Bruno Vidal bruno_vidal@hpfrcu03.france.hp.com
Subject: [parisc-linux] Crash dump module
Hi
Good news, I've finished to recompile the entire chain on hpux 11.00 64bits.
I've recompile my own kernel and booted a small 712/60. It work pretty well.
Now I'm going to start the work about crash dump. So, In my mind, I'll start
from the linux/kernel/panic.c function and I'll add a function dumpsys.c
in the linux/arch/paric directory -> is it okay for you ?
It will depend on the CONFIG_PARISC var (or do you prefer adding a CONFIG_DUMPSYS var ?)
thanks.
--
Vidal Bruno, (770-4271)
SSD-HA Team, HP-UX & LINUX Support
bruno_vidal@admin.france.hp.com
From randolph@tausq.org Wed, 22 Nov 2000 07:44:24 -0700
Date: Wed, 22 Nov 2000 07:44:24 -0700
From: Randolph Chung randolph@tausq.org
Subject: [parisc-linux] Crash dump module
> It will depend on the CONFIG_PARISC var (or do you prefer adding a CONFIG_DUMPSYS var ?)
I think there was a consensus to use __hppa__ instead of CONFIG_PARISC
...
randolph
--
@..@ http://www.TauSq.org/
(----)
( >__< )
^^ ~~ ^^
From mike_clapper@hp.com Wed, 22 Nov 2000 07:54:24 -0800
Date: Wed, 22 Nov 2000 07:54:24 -0800
From: CLAPPER,MIKE (HP-USA,ex1) mike_clapper@hp.com
Subject: [parisc-linux] B132L boot hang
Hello Everyone,
I download the cross-compilers and build a lifimage on Redhat 6.1. I set
up nfsroot on a S712 running hpux 10.20. With this I am able to boot a
B132L running pdc 6.1 -
almost. I get a hang right after -
VFS: Mounted root (nfs filesystem) readonly
Warning: unable to open an initial console
I have a 700/96 on serial port 1 as the console. I have hpux on a disc in
the unit I'm trying to boot - from HPUX I have verified I can nfs mount the
filesys I'm using for NFSROOT.
With linux in this condition the machine will respond to ping. I will also
react to a telnet - i get the telnet banner then "connection closed by
foreign host". FTP gets "connection
refused" - so it' partially alive.....
Is there a way to keyword search the developers archive? I vaguely recall
seeing the 'initial console' warning, but couldn't find it browsing the
developers archive
Thanks,
Mike
**********************************************
Mike Clapper
North American Crisis Management Team
Hewlett Packard
(405) 948-4715
**********************************************
From bame@noam.fc.hp.com Wed, 22 Nov 2000 09:02:03 -0700
Date: Wed, 22 Nov 2000 09:02:03 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
= I've been lobbying for
= the removal of the almost empty arch/parisc/real directory, and its few
= remaining valid files moved to the kernel directory.
Done.
arch/parisc/real
R.I.P.
From grundler@cup.hp.com Wed, 22 Nov 2000 08:27:22 -0800
Date: Wed, 22 Nov 2000 08:27:22 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] Crash dump module
Bruno Vidal wrote:
> Hi
> Good news, I've finished to recompile the entire chain on hpux 11.00 64bits.
> I've recompile my own kernel and booted a small 712/60. It work pretty well.
> Now I'm going to start the work about crash dump. So, In my mind, I'll start
> from the linux/kernel/panic.c function and I'll add a function dumpsys.c
> in the linux/arch/paric directory -> is it okay for you ?
YES!
> It will depend on the CONFIG_PARISC var
> (or do you prefer adding a CONFIG_DUMPSYS var ?)
As Randolph pointed out CONFIG_PARISC is deprecated.
Use "ifdef __hppa__" for changes in *common* (ie not arch/parisc)
files. They should not be needed for anything in arch/parisc
or linux/include/asm-parisc.
Adding a CONFIG_DUMPSYS is a good idea.
Look in linux/arch/parisc/config.in, linux/arch/parisc/defconfig,
and the various Makefiles for how CONFIG_* flags work.
thanks Bruno!
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From grundler@cup.hp.com Wed, 22 Nov 2000 09:27:20 -0800
Date: Wed, 22 Nov 2000 09:27:20 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] B132L boot hang
"CLAPPER,MIKE (HP-USA,ex1)" wrote:
>
>
> Hello Everyone,
>
> I download the cross-compilers and build a lifimage on Redhat 6.1. I set
> up nfsroot on a S712 running hpux 10.20. With this I am able to boot a
> B132L running pdc 6.1 -
> almost. I get a hang right after -
>
> VFS: Mounted root (nfs filesystem) readonly
> Warning: unable to open an initial console
I think I need more output. Based on ioscan output, the B132L
has two serial ports:
8/16/4 - off of LASI
8/20/2 - of WAX (EISA Bus Adapter)
(GSCtoPCI is at 8/0)
I don't think the one off of WAX is supported right now.
Are you using the default .config produced by "make oldconfig"?
It's also possible the LASI support is broken on B132L and I would
expect some console output indicating what's wrong.
> I have a 700/96 on serial port 1 as the console. I have hpux on a disc in
> the unit I'm trying to boot - from HPUX I have verified I can nfs mount the
> filesys I'm using for NFSROOT.
NFS root mounted fine. Console is the problem.
...
> Is there a way to keyword search the developers archive? I vaguely recall
> seeing the 'initial console' warning, but couldn't find it browsing the
> developers archive
Yes - http://puffin.external.hp.com/cgi-bin/mailgrep
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From cary@cup.hp.com Wed, 22 Nov 2000 10:06:05 -0800
Date: Wed, 22 Nov 2000 10:06:05 -0800
From: Cary Coutant cary@cup.hp.com
Subject: [parisc-linux] Use of the EI_OSABI field
As the original author of the proposal to add the EI_OSABI field to the
ELF format, let me try to clarify the intent of this field. The
authoritative definition of this field is found in SCO's gABI document,
which is still the official specification for the ELF format (this is
probably posted somewhere on SCO's web site). Here's what it says:
Byte e_ident[EI_OSABI] identifies the operating system and
ABI to which the object is targeted. Some fields in other ELF
structures have flags and values that have operating system
and/or ABI specific meanings; the interpretation of those
fields is determined by the value of this byte. The value of
this byte must be interpreted differently for each machine.
That is, each value for the e_machine field determines a set
of values for the EI_OSABI byte. Values are assigned by the
ABI processor supplement for each machine. If the processor
supplement does not specify a set of values, the value 0
shall be used and indicates unspecified.
The first sentence is still a bit misleading, and is an artifact of the
original proposal. Originally, the field was intended to identify the
target ABI (hence the name of the field). As we started discussing a
common Unix ABI for IA-64, however, it became clear that this field
wouldn't serve that purpose, but it was still needed to identify the set
of platform-specific ELF extensions that are used by the object file.
There are a number of fields in the ELF format for which ranges of values
or a set of flag bits are reserved for vendor-specific use (e.g.,
SHT_LOOS through SHT_HIOS for vendor-specific section types, and
SHF_MASKOS for vendor-specific section attributes). If an object file
uses any of these values or flag bits, the consumer of the file must
consult the EI_OSABI field to determine what those values or flags mean.
It works just like the e_machine field does for attaching meaning to
processor-specific values and flags.
The intent is that any ABI-conforming implementation will be able to
execute an ABI-conforming binary, even if it uses certain vendor-specific
features. In many cases, those vendor-specific features are hints for a
particular OS that can be ignored by other implementations. Where this is
not the case, and a vendor-specific feature must be understood by the
system in order to process the file correctly, we have a couple of
alternatives.
For section types and flags that a linker must understand, we have the
SHF_OS_NONCONFORMING flag -- if set, and a linker doesn't understand a
particular section type or flag, it must reject the file.
For executables that will execute only on a particular implementation, we
must use an alternate interpreter (PT_INTERP) or bind to
implementation-specific shared libraries. An ABI-conforming binary will
use the interpreter specified in the psABI spec, and will use only system
libraries specified there.
For statically-bound programs, I'm afraid we don't have a clear solution.
We took the general approach that such programs are not ABI-conforming in
the first place, and can use any mechanism they might choose to verify
that they are executing on the appropriate implementation.
-cary
From grundler@cup.hp.com Wed, 22 Nov 2000 11:55:38 -0800
Date: Wed, 22 Nov 2000 11:55:38 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
John Marvin wrote:
> Grant Grundler wrote:
> > Yes I do. It is supposed to work for ccio+PCX-W platforms (eg C360).
...
> Well, personally i'd vote to get rid of it. It works for ONE machine only,
> and MAY have an advantage in some small case.
And so does the OB600 mouse driver I rewrote/published.
AFAIK, it only works on OB600s.
I was originally thinking D/K/R-class boxes had PCX-W upgrades but
AFAICT, they don't.
> But if we keep it, lets make
> sure that it is real clear that it should NOT be the default choice.
>
> It should be marked CONFIG_EXPERIMENTAL, and the text associated with it
> should clearly show that it works on a C360 only. If possible, it should
> also be made clear that ccio-dma.c works for C360, so people who have
> C360's don't think they have to choose ccio-rm-dma.c.
IIRC, Comments in the headers of both ccio files make those issues clear.
I'm not sure where else that needs to be documented.
> Grant, I hope you are prepared to answer the parisc-linux mailing list
> questions this is going to generate once parisc-linux starts becoming more
> visible. Another FAQ entry perhaps? :-)
Ryan owns it. He's responsible for documenting it and adding FAQ questions.
He can choose to delete ccio-rm-dma.c as well.
:^)
I think it'd be a waste to throw it away before someone figures
out that it's really not useful - even if just for C360.
thanks,
grant
From mike_clapper@hp.com Wed, 22 Nov 2000 12:05:23 -0800
Date: Wed, 22 Nov 2000 12:05:23 -0800
From: CLAPPER,MIKE (HP-USA,ex1) mike_clapper@hp.com
Subject: [parisc-linux] B132L boot hang
Thanks Grant,
I verified the console is off 8/16/4.0 and that it is the LASI port. I also
downloaded the newest cross-compiler and source code. After rebuilding the
palo lifimage I get the same hang in the same place. Since I was cabled to
a dumb terminal I was not able to capture the boot output. I will find the
right cable to hook up to my pc so I can capture the output and send it to
you - perhaps an error occurred earlier in the boot that I overlooked.
Thanks for the help.
Mike
-----Original Message-----
From: Grant Grundler [mailto:grundler@cup.hp.com]
Sent: Wednesday, November 22, 2000 11:27 AM
To: CLAPPER,MIKE (HP-USA,ex1)
Cc: 'parisc-linux@thepuffingroup.com'
Subject: Re: [parisc-linux] B132L boot hang
"CLAPPER,MIKE (HP-USA,ex1)" wrote:
>
>
> Hello Everyone,
>
> I download the cross-compilers and build a lifimage on Redhat 6.1. I set
> up nfsroot on a S712 running hpux 10.20. With this I am able to boot a
> B132L running pdc 6.1 -
> almost. I get a hang right after -
>
> VFS: Mounted root (nfs filesystem) readonly
> Warning: unable to open an initial console
I think I need more output. Based on ioscan output, the B132L
has two serial ports:
8/16/4 - off of LASI
8/20/2 - of WAX (EISA Bus Adapter)
(GSCtoPCI is at 8/0)
I don't think the one off of WAX is supported right now.
Are you using the default .config produced by "make oldconfig"?
It's also possible the LASI support is broken on B132L and I would
expect some console output indicating what's wrong.
> I have a 700/96 on serial port 1 as the console. I have hpux on a disc in
> the unit I'm trying to boot - from HPUX I have verified I can nfs mount
the
> filesys I'm using for NFSROOT.
NFS root mounted fine. Console is the problem.
...
> Is there a way to keyword search the developers archive? I vaguely recall
> seeing the 'initial console' warning, but couldn't find it browsing the
> developers archive
Yes - http://puffin.external.hp.com/cgi-bin/mailgrep
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
---------------------------------------------------------------------------
To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com with
`unsubscribe' as the subject.
From kirkb@chrome.rose.hp.com Wed, 22 Nov 2000 12:10:10 PST
Date: Wed, 22 Nov 2000 12:10:10 PST
From: Kirk Bresniker kirkb@chrome.rose.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
Grant Grundler wrote:
| I was originally thinking D/K/R-class boxes had PCX-W upgrades but
| AFAICT, they don't.
|
The C-class upgrade to PCX-W was a one-off. Upgrades for similar enterprise
servers were not designed.
KMB
--
+============================================================+
| Kirk Bresniker (916) 748-2393 |
| 8000 Foothills Blvd |
| Roseville, CA 95747-5649 |
| kirkb@rose.hp.com |
From grundler@cup.hp.com Wed, 22 Nov 2000 12:11:23 -0800
Date: Wed, 22 Nov 2000 12:11:23 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
Grant Grundler wrote:
...
> arch/parisc/kernel/ccio-dma.c -> drivers/gsc/ccio-dma.c
> arch/parisc/kernel/ccio-rm-dma.c -> drivers/gsc/ccio-rm-dma.c
>
> ccio will *always* be associated with a GSC bus since that's
> the secondary bus. And ccio supports devices below dino.c which
> already lives in drivers/gsc.
Ryan - moving/keeping these files is up to you.
I was just sharing what I thought was "right".
Apologies for not making that clear earlier.
> arch/parisc/kernel/lba_pci.c -> drivers/ropes/lba_pci.c
> arch/parisc/kernel/sba_iommu.c -> drivers/ropes/sba_iommu.c
> arch/parisc/kernel/iosapic.c -> drivers/ropes/iosapic.c
I've talked to one of the folks working on IA64-linux.
They are interested in merging iosapic code but haven't even
looked at the parisc version I wrote. We talked a bit about the
issues and it doesn't sound like it's going to happen anytime soon.
In any case, iosapic.c doesn't belong under "drivers/ropes".
So none of this needs to move in the forseeable future.
It can all stay in arch/parisc/kernel.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From deller@gmx.de Wed, 22 Nov 2000 21:43:40 +0100
Date: Wed, 22 Nov 2000 21:43:40 +0100
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] B132L boot hang
> I think I need more output. Based on ioscan output, the B132L
> has two serial ports:
> 8/16/4 - off of LASI
> 8/20/2 - of WAX (EISA Bus Adapter)
>
> (GSCtoPCI is at 8/0)
>
> I don't think the one off of WAX is supported right now.
It should be detected & supported (at least it is on my B160L).
But it is not used as initial console since it normally gets assigned as
ttyS1 while LASI-serial gets ttyS0.
Greetings,
Helge.
From bame@noam.fc.hp.com Wed, 22 Nov 2000 16:03:12 -0700
Date: Wed, 22 Nov 2000 16:03:12 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] 64-bit progress
32-bit syscalls on 64-bit kernel are to the point where a few things
work, and signals appear to be working (I didn't implement *all* the
signal-related syscalls yet). I'll be continuing to produce syscall
wrappers for a while...
If you try to boot the standard NFS root with wide kernel, you'll want to
mv sbin/init sbin/init-real
then compile and install this program as sbin/init:
int main()
{
char *argv[2];
argv[0] = "/sbin/init-real";
argv[1] = 0;
execv(argv[0], argv);
}
I never felt comfortable I found the point where sys_execve figures out it
needs to pack the argv vector differently for narrow user apps (locally I
also am using PER_LINUX32 in binfmt_elf32.c) which causes the initial
exec(/sbin/init) to be passed incomprehensible arguments. This program
is a little temporary workaround.
-Paul Bame
From alan@linuxcare.com.au Thu, 23 Nov 2000 11:18:20 +1100 (EST)
Date: Thu, 23 Nov 2000 11:18:20 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: glibc
On Wed, 22 Nov 2000, Matt Taggart wrote:
> Hi Alan,
>
> First a timezone question... your mail headers have you in +11 and Ft. Collins
> is -7 so you're 6 hours behind + 1 day relative to us? So as I send this it's
> 14:05 here and 8:05 there?
Hi Matt,
I'm actually on +10:30 in Adelaide, but posting from a Linuxcare machine
in Canberra which is on +11.
>[snip segfaults]
> I am building native using dhd's binutils/gcc/glibc debs, using source checked
> out just after the glibc 2.2 merge. Do you think the merge / ld.so changes you
> just made would help this problem at all?
Quite likely. This fix
* sysdeps/hppa/dl-machine.h (ELF_MACHINE_START_ADDRESS): Define.
is fairly crucial, as without it %dp will not be set correctly. I should
have mentioned this when posting to the list about the binutils change.
Oh, and as of this email, I haven't actually run anything linked to the
new glibc. A little remiss, I know, but I've been deep in the gdb port,
and if I spend time tinkering with glibc, binutils and gcc (which I like),
gdb (which I don't like) will never be finished. ;-)
Alan
--
Linuxcare. Support for the Revolution.
From grundler@cup.hp.com Wed, 22 Nov 2000 18:24:52 -0800 (PST)
Date: Wed, 22 Nov 2000 18:24:52 -0800 (PST)
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] new SBA IOMMU version committed
Hi all,
I've committed my "second generation" I/O MMU code for
C3K/J5k boxes. It's only tested for 32-bit.
I'll be testing 64-bit shortly.
This code makes "perfect" use of the I/O pdir resource map
and we shouldn't see any panics due to out_of_resource.
grant
From grundler@cup.hp.com Wed, 22 Nov 2000 18:47:06 -0800 (PST)
Date: Wed, 22 Nov 2000 18:47:06 -0800 (PST)
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] CONFIG_DMB_TRAP diff
Hello again (last one until Monday - I promise),
With Lamont's wisdom, I implemented support for Date Memory Break trap.
This enables the kernel programmer to capture the evil code which
stomps on other people's "private" data. Only works for stores
through virtual addresses. gsc_writeX() and DMA will still
bypass this mechanism.
pb, dhd, (or some equivalent deity), could you review/commit this code?
Or tell me it's ok to commit?
I've touched:
arch/parisc/config.in
arch/parisc/kernel/entry.S
arch/parisc/mm/kmap.c
include/asm-parisc/pgtable.h
thanks,
grant
Index: arch/parisc/config.in
===================================================================
RCS file: /home/cvs/parisc/linux/arch/parisc/config.in,v
retrieving revision 1.25
diff -u -p -r1.25 config.in
--- config.in 2000/10/20 18:28:26 1.25
+++ config.in 2000/11/23 02:18:23
@@ -16,8 +16,12 @@ endmenu
mainmenu_option next_comment
comment 'General options'
-# bool 'Symmetric multi-processing support' CONFIG_SMP
-define_bool CONFIG_SMP n
+bool 'Symmetric multi-processing support' CONFIG_SMP
+# define_bool CONFIG_SMP n
+
+# One needs to tweak dmb_trap_11 code in entry.S to match.
+# Not tested for 64-bit kernel.
+bool 'Debug support for Data Memory Break Trap' CONFIG_DMB_TRAP
bool 'Kernel Debugger support' CONFIG_KWDB
# define_bool CONFIG_KWDB n
Index: arch/parisc/kernel/entry.S
===================================================================
RCS file: /home/cvs/parisc/linux/arch/parisc/kernel/entry.S,v
retrieving revision 1.53
diff -u -p -r1.53 entry.S
--- entry.S 2000/11/22 16:51:33 1.53
+++ entry.S 2000/11/23 02:18:23
@@ -23,6 +23,7 @@
*/
#include
+#include
/* the following is the setup i think we should follow:
* whenever the CPU is interruptible, the following has to be true:
@@ -349,7 +350,39 @@
.endm
#endif
+#ifdef CONFIG_DMB_TRAP
/*
+ ** Data Memory Bit trap interruption handler (parisc 1.1)
+ **
+ ** This is a debugging aid. Use it when you think someone else
+ ** is stepping on your memory. It only catches *virtual*
+ ** accesses. gsc_writeX() functions disable virtual translation
+ ** (D-bit) and will happily scribble whatever physical address
+ ** is passed in.
+ **
+ ** Here's how to use it:
+ ** 1) Call iterate_pages() from your init routine like this:
+ ** iterate_pages( my_private_mem, private_mem_size,
+ ** set_data_memory_break, 0);
+ ** 2) substitute your functions for your_function1 (or 2) in
+ ** dmb_trap_11 code below.
+ **
+ ** Thanks to Lamont Jones for telling me how to do this.
+ ** - ggg 1/22/2000
+ */
+ .macro dmb_11 code
+
+ mfctl %isr,spc
+ b dmb_trap_11
+ mfctl %ior,va
+
+ .align 32
+ .endm
+#else
+#define dmb_11 def
+#endif
+
+ /*
* dirty bit trap interruption handler (parisc 2.0)
*/
@@ -448,7 +481,7 @@ fault_vector_11:
naitlb_11 16
nadtlb_11 17
def 18
- def 19
+ dmb_11 19
dbit_11 20
def 21
def 22
@@ -467,7 +500,6 @@ fault_vector_11:
.import handle_interruption,code
.import handle_real_interruption,code
.import do_irq_mask,code
- .import parisc_stopkernel,code
.import cpu_irq_region,data
/*
@@ -903,11 +935,15 @@ dtlb_miss_11:
dep pte,8,7,prot
extru,= pte,_PAGE_NO_CACHE_BIT,1,r0
- depi 1,12,1,prot
+ depi 1,12,1,prot /* U-bit */
extru,= pte,_PAGE_USER_BIT,1,r0
depi 7,11,3,prot /* Set for user space (1 rsvd for read) */
extru,= pte,_PAGE_GATEWAY_BIT,1,r0
depi 0,11,2,prot /* If Gateway, Set PL2 to 0 */
+#ifdef CONFIG_DMB_TRAP
+ extru,= pte,_PAGE_DMB_BIT,1,r0
+ depi 1,4,1,prot /* B-bit */
+#endif
/* Get rid of prot bits and convert to page addr for idtlba */
@@ -1300,6 +1336,30 @@ dbit_trap_11:
rfir
nop
+
+#ifdef CONFIG_DMB_TRAP
+ .import your_function1,code
+ .import your_function2,code
+
+dmb_trap_11:
+ mfctl pcsq,t0 /* get space */
+ comb,<>,n t0,%r0,dmb_rfi /* not kernel - bail */
+
+ mfctl pcoq,t0 /* get offset */
+ ldil L%dmb_ok_function1, t1
+ dep %r0, 31, 12, t0
+ comb,=,n t0,t1,dmb_rfi /* it's ours - bail */
+
+ ldil L%dmb_ok_function2, t1
+ comb,<>,n t0,t1,intr_save /* not ours - panic */
+
+dmb_rfi:
+ mfctl ipsw,t0 /* Set PSW X-bit - just continue */
+ depi 1,11,1,t0 /* Set X-bit */
+ mtctl t0, ipsw
+ rfir
+ nop
+#endif
dbit_trap_20:
mfctl %cr25,ptp /* Assume user space trap */
Index: arch/parisc/mm/kmap.c
===================================================================
RCS file: /home/cvs/parisc/linux/arch/parisc/mm/kmap.c,v
retrieving revision 1.3
diff -u -p -r1.3 kmap.c
--- kmap.c 2000/05/05 18:05:47 1.3
+++ kmap.c 2000/11/23 02:18:23
@@ -43,7 +43,16 @@ static void unmap_cached_pte(pte_t * pte
}
#endif
+#ifdef CONFIG_DMB_TRAP
/* These two routines should probably check a few things... */
+void set_data_memory_break(pte_t * pte, unsigned long arg)
+{
+ pte_val(*pte) |= _PAGE_DMB;
+}
+
+#endif
+
+/* These two routines should probably check a few things... */
static void set_uncached(pte_t * pte, unsigned long arg)
{
pte_val(*pte) |= _PAGE_NO_CACHE;
@@ -106,7 +115,10 @@ static inline void iterate_pmd(pgd_t * d
} while (address < end);
}
-static void iterate_pages(unsigned long address, unsigned long size,
+#ifndef CONFIG_DMB_TRAP
+static
+#endif
+void iterate_pages(unsigned long address, unsigned long size,
pte_iterator_t op, unsigned long arg)
{
pgd_t *dir;
Index: include/asm-parisc/pgtable.h
===================================================================
RCS file: /home/cvs/parisc/linux/include/asm-parisc/pgtable.h,v
retrieving revision 1.29
diff -u -p -r1.29 pgtable.h
--- pgtable.h 2000/11/10 21:44:44 1.29
+++ pgtable.h 2000/11/23 02:18:23
@@ -109,6 +109,10 @@ extern void *vmalloc_start;
#define _PAGE_USER 0x400 /* Software: User accessable page */
#define _PAGE_USER_BIT 21 /* Needs to agree with _PAGE_USER above */
/* 0x800 still available */
+#ifdef CONFIG_DMB_TRAP
+#define _PAGE_DMB 0x800 /* Data Memory Break Trap */
+#define _PAGE_DMB_BIT 20 /* Data Memory Break Trap */
+#endif
#ifdef __ASSEMBLY__
#define _PGB_(x) (1 << (63 - (x)))
From jsm@udlkern.fc.hp.com Wed, 22 Nov 2000 22:50:14 -0700 (MST)
Date: Wed, 22 Nov 2000 22:50:14 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: [parisc-linux] CONFIG_DMB_TRAP diff
>
> Hello again (last one until Monday - I promise),
>
> With Lamont's wisdom, I implemented support for Date Memory Break trap.
>
> This enables the kernel programmer to capture the evil code which
> stomps on other people's "private" data. Only works for stores
> through virtual addresses. gsc_writeX() and DMA will still
> bypass this mechanism.
>
> pb, dhd, (or some equivalent deity), could you review/commit this code?
> Or tell me it's ok to commit?
>
> I've touched:
> arch/parisc/config.in
> arch/parisc/kernel/entry.S
> arch/parisc/mm/kmap.c
> include/asm-parisc/pgtable.h
>
Please don't. This solution is way more complicated than it should be.
Here are the problems with it:
1) As I had already mentioned in a previous discussion, the
pte's already reserve the location for the B bit (data memory
break trap) and the dtlb miss handlers already move the entire
group of bits that include this bit in one operation. So no
change is necessary to the dtlb miss handlers to specially
set that bit and incur extra instructions in the tlb miss handlers,
and no extra bits need to be allocated in the pte. Instead of
adding a new definition (e.g. 0x800, which is our last available bit)
use the one that is already reserved for it: 0x010.
2) There is no reason to add a special data memory break trap
handler. The general trap handler is more than sufficient for
this case. handle_interruption will be called if a data memory
break trap is encountered. Just add a new case for the list
of traps, and handle the trap in C. You can set the X bit by
simply setting it in the saved ipsw (in gr[0]) and it will be
set upon return from the trap, no muss, no fuss.
Note that the above also applies for the page reference trap. The T bit
is also already supported (0x040 in the pte) by the dtlb miss handlers.
Note that the reason I reserved these bits is because it would actually
take MORE code in the dtlb miss handlers to NOT support those bits and use
them for something else.
Another helpful hint for those wanting to use this feature. If you are
tracking corruptions that span multiple pages, then just setting the B bit
on each page may be all you need. But, when I've used the data memory
break trap for corruption tracking, typically I've wanted to track a
corruption that was happening to a particular variable, i.e. a 4 byte
quantity, and lots of other variables were being legitimately written on
the same page, so you wind up with thousands of data memory break traps,
where only one may be the one that is corrupting the location you are
interested in. But, all is not lost, the solution is still fairly simple.
The data memory break trap provides a valid iir, isr and ior. So once you
get the trap, a custom data memory break handler (which can be written
with a few lines of C in handle_interruption), simply uses iir, isr and
ior to check if the access was to the specific byte or bytes you are
interested in. I've simply used isr/ior to check for writes within the
word I was interested in. That may be enough for most cases. The
information you are missing from isr/ior is the size of the write
transaction. To get this you would need to parse the instruction stored
in iir (code to determine the size of a store from the instruction in the
iir will be necessary when an unaligned fault handler is written).
John Marvin
jsm@fc.hp.com
From jsm@udlkern.fc.hp.com Thu, 23 Nov 2000 01:29:55 -0700 (MST)
Date: Thu, 23 Nov 2000 01:29:55 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: [parisc-linux] CONFIG_DMB_TRAP diff
Alan Modra wrote:
>
> gdb will eventually want to use this trap too.
>
Cool! However, data memory break traps for user translations are a little
tougher. There are some small modifications in the machine dependent code
that I can make to make sure the B bit stays set during most VM operations
on a page. However, since the machine independent part of the VM system
doesn't know anything about this bit (nor should it), it won't be preserved
if the page is paged out (and subsequently paged back in). This could
be fixed fairly easily with some changes to the machine independent code,
but I don't think that would be appropriate.
Some potential solutions (each has its problems):
1) Just document the fact that the feature may not work on systems with
low memory. It's a parisc only feature, so perhaps we could live with that.
2) Lock the page in memory (using the mlock interface) when we set the B
bit on a page.
Just some thoughts on the subject.
John Marvin
jsm@fc.hp.com
From pjlahaie@linuxcare.com Fri, 24 Nov 2000 16:56:16 -0500
Date: Fri, 24 Nov 2000 16:56:16 -0500
From: Paul J.Y. Lahaie pjlahaie@linuxcare.com
Subject: [parisc-linux] boot-floppies dependancies
--GRPZ8SYKNexpdSJ7
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
I would like to know if anyone has built any of the following packages
that are needed for the boot-floppies.
Currently, the missing deps are:
cspsfonts
man-db
tetex-bin
recode
cslatex
libpaperg
tetex-extra
libnewt-dev
libgd1g-dev
gawk
libi18n-langtags-perl
dpkg-awk
debiandoc-sgml
I'm currently trying to get a working apt, since the one of pehc and the .iso
seem to be broken (no xfer methods). I'll let everyone know when a working
copy will be available.
- Paul
--GRPZ8SYKNexpdSJ7
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE6HuP/8ggPQthPCzcRAnyKAJ0fyraVXEvlI0yQLsli7FlnUUAPdwCfSyv/
YgAxOcVMDgWfHBc7lneuc1Y=
=tScM
-----END PGP SIGNATURE-----
--GRPZ8SYKNexpdSJ7--
From doneill@linuxcare.com Fri, 24 Nov 2000 17:01:00 -0500 (EST)
Date: Fri, 24 Nov 2000 17:01:00 -0500 (EST)
From: Dave O'Neill doneill@linuxcare.com
Subject: [parisc-linux] strace?
So, I've heard rumours that somewhere there's a working strace for
parisc.... Can anyone point me in the right direction?
Dave
--
Dave O'Neill, Senior Linux Consultant, Linuxcare, Inc.
desk: (613) 562-9949 fax: (613) 562-9700 cell: (613) 223-0219
doneill@linuxcare.com http://www.linuxcare.com/
From obrien@FreeBSD.org Sat, 25 Nov 2000 12:22:11 -0800
Date: Sat, 25 Nov 2000 12:22:11 -0800
From: David O'Brien obrien@FreeBSD.org
Subject: [parisc-linux] Use of the EI_OSABI field
On Tue, Nov 21, 2000 at 03:27:19PM -0800, Ulrich Drepper wrote:
> Alan Modra writes:
>
> > Ulrich Drepper also was quite vehement against changing sourceware
> > FreeBSD binutils.
>
> I've never said anything about any *BSD, why should I? The *BSD
> people wanted to change the Linux binutils.
No (don't put words in my mouth Ulrich as you'll be wrong 99% of the
time). I wanted the Sourceware Binutils to set the EI_OSABI to
"ELFOSABI_LINUX" when targeting Linux.
--
-- David (obrien@FreeBSD.org)
From obrien@FreeBSD.org Sat, 25 Nov 2000 12:28:40 -0800
Date: Sat, 25 Nov 2000 12:28:40 -0800
From: David O'Brien obrien@FreeBSD.org
Subject: [parisc-linux] Use of the EI_OSABI field
On Tue, Nov 21, 2000 at 07:11:29PM -0800, H . J . Lu wrote:
> When I was at the IA64 ABI meeting yesterday, we were looking at the
> EI_OSABI issue. Everyone at the IA64 ABI meeting agreed that what
> Ulrich and I had said was the correct understanding.
>
> "The e_ident[EI_OSABI] value identifies the operating system and ABI
> to which the object is targeted" just means if the e_ident[EI_OSABI]
Then is this *extremely* misleading text going to be changed???
"the operating system and" needs to be deleted in the _official_
_published_ specification.
--
-- David (obrien@FreeBSD.org)
From obrien@FreeBSD.org Sat, 25 Nov 2000 12:31:08 -0800
Date: Sat, 25 Nov 2000 12:31:08 -0800
From: David O'Brien obrien@FreeBSD.org
Subject: [parisc-linux] Use of the EI_OSABI field
On Tue, Nov 21, 2000 at 07:18:21PM -0800, Ulrich Drepper wrote:
> Alan Modra writes:
> > "The e_ident[EI_OSABI] value identifies the operating system and ABI to
> > which the object is targeted"
> >
> > Granted, this is only in a processor specific ABI, but it does flatly
> > contradict your assertions that the purpose of EI_OSABI is to flag the
> > presense of ELF extensions.
>
> The meaning of the field changed over time. Or better said, some
> people initially understood it differently.
Then lets get the _official_ wording changed. Obviously my
interpretation wasn't unreasonable. And the world does not need to have
you as a premadona keeper of the [unwritten] rules of the land.
--
-- David (obrien@FreeBSD.org)
From hjl@valinux.com Sat, 25 Nov 2000 12:33:40 -0800
Date: Sat, 25 Nov 2000 12:33:40 -0800
From: H . J . Lu hjl@valinux.com
Subject: [parisc-linux] Use of the EI_OSABI field
On Sat, Nov 25, 2000 at 12:28:40PM -0800, David O'Brien wrote:
> On Tue, Nov 21, 2000 at 07:11:29PM -0800, H . J . Lu wrote:
> > When I was at the IA64 ABI meeting yesterday, we were looking at the
> > EI_OSABI issue. Everyone at the IA64 ABI meeting agreed that what
> > Ulrich and I had said was the correct understanding.
> >
> > "The e_ident[EI_OSABI] value identifies the operating system and ABI
> > to which the object is targeted" just means if the e_ident[EI_OSABI]
>
>
> Then is this *extremely* misleading text going to be changed???
> "the operating system and" needs to be deleted in the _official_
> _published_ specification.
>
I will bring this issue up to ia64 psABI and gABI.
--
H.J. Lu (hjl@valinux.com)
From alan@lxorguk.ukuu.org.uk Sat, 25 Nov 2000 20:57:05 +0000 (GMT)
Date: Sat, 25 Nov 2000 20:57:05 +0000 (GMT)
From: Alan Cox alan@lxorguk.ukuu.org.uk
Subject: [parisc-linux] RPM and hppa
I've uploaded my first block of RPM packages to
ftp.linux.org.uk/pub/linux/alan/HPPA along with a tar ball of built RPM
tools for anyone who wants to play.
In doing so I've hit a few problems with the 0.5 iso (no suprises there since
its not actually meant to work reliably)
- g++ explodes trying to build groff after allocating about 400Mb of RAM.
Building -O0 works
- the configure script for procmail tries to find the largest argument set
that works (by searching). It crashes the kernel in doing so 8)
- ldd is causing page faults in ld.so (kernel logged ones) and dying with
segv. Fortunately it outputs the library list first
- The linker appears to have a problem when resolving symbols between three
shared objects while doing a shared object link.
[Example is rpm:
rpmlib is linked dynamically with -ldb3 -ldb
The linker emits messages about symbols being static and should be
built -fPIC. If you dump the libraries they are -fPIC.
It looks as if its resolving a symbol between two shared libraries
and making a static resolution that then blows up when the third
library gets involved
]
- The kernel shows occasional page cache corruption. This actually is quite
possibly generic test6 bugs
On the whole the toolset is working remarkably well. I've built over 100
source package sets so far including things like ncurses and most stuff just
builds or hits portability problems (eg gmp wants to use HP format asm,
zlib wants to be a non PIC library for performance but the HP tools dont allow
it)
Alan
From alan@linuxcare.com.au Sun, 26 Nov 2000 23:01:50 +1100 (EST)
Date: Sun, 26 Nov 2000 23:01:50 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] RPM and hppa
On Sat, 25 Nov 2000, Alan Cox wrote:
> - g++ explodes trying to build groff after allocating about 400Mb of RAM.
> Building -O0 works
Yeah, this is a known issue. I see on the gcc list that rth and others
have been working on gcc fixes recently that should address the problem.
I don't have the time/ability to fix it myself.
> - the configure script for procmail tries to find the largest argument set
> that works (by searching). It crashes the kernel in doing so 8)
>
> - ldd is causing page faults in ld.so (kernel logged ones) and dying with
> segv. Fortunately it outputs the library list first
How old are your glibc and binutils? I made some changes late October
that should have fixed this problem. See
http://puffin.external.hp.com/mailing-lists/parisc-linux/2000/10-Oct/0146.html
Does "readelf -d" on your hppa-linux ld.so show you have a DT_TEXTREL tag?
> - The linker appears to have a problem when resolving symbols between three
> shared objects while doing a shared object link.
>
> [Example is rpm:
>
> rpmlib is linked dynamically with -ldb3 -ldb
>
> The linker emits messages about symbols being static and should be
> built -fPIC. If you dump the libraries they are -fPIC.
>
> It looks as if its resolving a symbol between two shared libraries
> and making a static resolution that then blows up when the third
> library gets involved
>
> ]
Could you put all the objects involved in the link up for ftp
somewhere? I've just built rpm-2.5.6 without seeing this problem, but I'm
sure my setup is different to yours...
Regards, Alan Modra
--
Linuxcare. Support for the Revolution.
From alan@lxorguk.ukuu.org.uk Sun, 26 Nov 2000 12:12:21 +0000 (GMT)
Date: Sun, 26 Nov 2000 12:12:21 +0000 (GMT)
From: Alan Cox alan@lxorguk.ukuu.org.uk
Subject: [parisc-linux] RPM and hppa
> How old are your glibc and binutils? I made some changes late October
From the 0.5 cdrom
> http://puffin.external.hp.com/mailing-lists/parisc-linux/2000/10-Oct/0146.html
> Does "readelf -d" on your hppa-linux ld.so show you have a DT_TEXTREL tag?
Nope
> Could you put all the objects involved in the link up for ftp
> somewhere? I've just built rpm-2.5.6 without seeing this problem, but I'm
> sure my setup is different to yours...
rpm 4.0. rpm2 doesnt do that 3 way link with db and db3. I'll put them up when
I get a bit of time
From dave@hiauly1.hia.nrc.ca Sun, 26 Nov 2000 11:45:08 -0500 (EST)
Date: Sun, 26 Nov 2000 11:45:08 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: [parisc-linux] RPM and hppa
> On Sat, 25 Nov 2000, Alan Cox wrote:
>
> > - g++ explodes trying to build groff after allocating about 400Mb of RAM.
> > Building -O0 works
>
> Yeah, this is a known issue. I see on the gcc list that rth and others
> have been working on gcc fixes recently that should address the problem.
> I don't have the time/ability to fix it myself.
I think the above is a different problem. The explosion is most likely
a problem with exception edges in the gcse pass. Try `-fno-gcse'. This
also appears in the tFile.cc libio test.
The pa port uses sjlj exceptions via __builtin_setjmp/longjmp. A
nonlocal goto label is used for the target of the longjmp. The flow
analysis assumes that an "exception" could jump to any label in the
procedure rather than just the label associated with the exception
region.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From cwalthall@cwalthall.com Sun, 26 Nov 2000 11:45:38 -0600
Date: Sun, 26 Nov 2000 11:45:38 -0600
From: Carl & Delores Walthall cwalthall@cwalthall.com
Subject: [parisc-linux] RPM and hppa
Does anyone know where I can find the operating system
for the HP9000 J200 Server?
Thanks Carl Walthall
----- Original Message -----
From: "Alan Modra"
To: "Alan Cox"
Cc:
Sent: Sunday, November 26, 2000 6:01 AM
Subject: Re: [parisc-linux] RPM and hppa
> On Sat, 25 Nov 2000, Alan Cox wrote:
>
> > - g++ explodes trying to build groff after allocating about 400Mb of
RAM.
> > Building -O0 works
>
> Yeah, this is a known issue. I see on the gcc list that rth and others
> have been working on gcc fixes recently that should address the problem.
> I don't have the time/ability to fix it myself.
>
> > - the configure script for procmail tries to find the largest argument
set
> > that works (by searching). It crashes the kernel in doing so 8)
> >
> > - ldd is causing page faults in ld.so (kernel logged ones) and dying
with
> > segv. Fortunately it outputs the library list first
>
> How old are your glibc and binutils? I made some changes late October
> that should have fixed this problem. See
>
http://puffin.external.hp.com/mailing-lists/parisc-linux/2000/10-Oct/0146.ht
ml
> Does "readelf -d" on your hppa-linux ld.so show you have a DT_TEXTREL tag?
>
> > - The linker appears to have a problem when resolving symbols between
three
> > shared objects while doing a shared object link.
> >
> > [Example is rpm:
> >
> > rpmlib is linked dynamically with -ldb3 -ldb
> >
> > The linker emits messages about symbols being static and should be
> > built -fPIC. If you dump the libraries they are -fPIC.
> >
> > It looks as if its resolving a symbol between two shared libraries
> > and making a static resolution that then blows up when the third
> > library gets involved
> >
> > ]
>
> Could you put all the objects involved in the link up for ftp
> somewhere? I've just built rpm-2.5.6 without seeing this problem, but I'm
> sure my setup is different to yours...
>
> Regards, Alan Modra
> --
> Linuxcare. Support for the Revolution.
>
>
> --------------------------------------------------------------------------
-
> To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com
with
> `unsubscribe' as the subject.
>
From cwalthall@cwalthall.com Sun, 26 Nov 2000 11:47:24 -0600
Date: Sun, 26 Nov 2000 11:47:24 -0600
From: Carl & Delores Walthall cwalthall@cwalthall.com
Subject: [parisc-linux] RPM and hppa
Hi All
I have a HP9000 J200 server and would like to find the operating system
software for it. Can you tell me where to look on the internet or who I may
call?
The server has a OS installed but no one can remember the user or password
information.
So they gave it to me to take home, is there a way to boot it on a floppy
and change this
information?
Thanks for any information you can give me!
Carl walthall
----- Original Message -----
From: "John David Anglin"
To: "Alan Modra"
Cc: ;
Sent: Sunday, November 26, 2000 10:45 AM
Subject: Re: [parisc-linux] RPM and hppa
> > On Sat, 25 Nov 2000, Alan Cox wrote:
> >
> > > - g++ explodes trying to build groff after allocating about 400Mb of
RAM.
> > > Building -O0 works
> >
> > Yeah, this is a known issue. I see on the gcc list that rth and others
> > have been working on gcc fixes recently that should address the problem.
> > I don't have the time/ability to fix it myself.
>
> I think the above is a different problem. The explosion is most likely
> a problem with exception edges in the gcse pass. Try `-fno-gcse'. This
> also appears in the tFile.cc libio test.
>
> The pa port uses sjlj exceptions via __builtin_setjmp/longjmp. A
> nonlocal goto label is used for the target of the longjmp. The flow
> analysis assumes that an "exception" could jump to any label in the
> procedure rather than just the label associated with the exception
> region.
>
> Dave
> --
> J. David Anglin dave.anglin@nrc.ca
> National Research Council of Canada (613) 990-0752 (FAX:
952-6605)
>
> --------------------------------------------------------------------------
-
> To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com
with
> `unsubscribe' as the subject.
>
From alan@lxorguk.ukuu.org.uk Sun, 26 Nov 2000 22:47:43 +0000 (GMT)
Date: Sun, 26 Nov 2000 22:47:43 +0000 (GMT)
From: Alan Cox alan@lxorguk.ukuu.org.uk
Subject: [parisc-linux] Linker failure with db1
I replaced the libdb1 on the ISO with one built using the toolchain on the ISO
and all is now happy.
From grundler@cup.hp.com Sun, 26 Nov 2000 19:11:41 -0800
Date: Sun, 26 Nov 2000 19:11:41 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] RPM and hppa
"Carl & Delores Walthall" wrote:
> Hi All
>
> I have a HP9000 J200 server and would like to find the operating system
> software for it. Can you tell me where to look on the internet or who I may
> call?
Carl,
The short answer is a port of linux to the J200 is in progress.
See http://www.parisc-linux.org/ for status.
If you have more questions read the FAQ and check the mailing list
archives using http://puffin.external.hp.com/cgi-bin/mailgrep first please.
> The server has a OS installed but no one can remember the user or password
> information. So they gave it to me to take home, is there a way to boot
> it on a floppy and change this information?
Here's how to "fix" this:
During power up/boot press key to get "BOOT ADMIN>" prompt.
type "bo pri isl". At "ISL>" prompt, type "hpux -is".
HPUX should boot to single user. use "mountall" or "mount -a" to
mount all file systems. Then you can "vi /etc/passwd" or "passwd root".
If that doesn't work and you can afford to loan the box out for a
3-4 monthes, you might ask if someone could install parisc-linux
on it for you in exchange for a few monthes use.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From grundler@cup.hp.com Sun, 26 Nov 2000 22:12:10 -0800
Date: Sun, 26 Nov 2000 22:12:10 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] case LEDs
Alex deVries wrote:
> Grant Grundler wrote:
> > I was already asked this weekend to dig up technical info on LED
> > and soft power control. I guess this is my reminder to do that. :^)
I dug up the info but it's not in a form I'm willing to publishing.
However, someone did volunteer to look at this and I've provided them
with this info. So it will make into our CVS source tree and get
published that way.
> Isn't there a PDC call (pdc_chassis?) to do this?
Not AFAICT. PDC_CHASSIS is documented in the pdc32.pdf found on
http://www.parisc-linux.org/documentation.html
But my gut feeling is parisc specific code could make some PDC_CHASSIS
calls to set up "sysstat" field (eg initialize, shutdown, run states).
Does anyone know if the chassis codes used by HPUX are published?
It would be cool if parisc-linux used the same codes where possible...
> Or is the heartbeat LED done by hardware?
Code I've looked at before seems to all do it in software.
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From sanjak@tipas.lt Mon, 27 Nov 2000 09:09:22 +0200
Date: Mon, 27 Nov 2000 09:09:22 +0200
From: Aleksandr Konstantinov sanjak@tipas.lt
Subject: [parisc-linux] how to boot ?
Hello, All.
We have two HP Workstations here (Model 735).
I would like to try linux on one of them. But I have no disk space
to install all the stuff (binutils, compiler, etc) .
I found precompiled kernel on puffin.external.hp.com but it looks like
I also need palo . Does anybody know, where to get precompiled palo
for hpux9, hpux10 or linux-x86 ?
Thanks in advance.
A.K.
From rhirst@linuxcare.com Mon, 27 Nov 2000 09:18:21 +0000
Date: Mon, 27 Nov 2000 09:18:21 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] strace?
Hi Dave,
The source is in cvs on pehc; I can supply a binary if you want
one. I havn't made serious use of it, but it does basically work.
Richard
On Fri, Nov 24, 2000 at 05:01:00PM -0500, Dave O'Neill wrote:
>
> So, I've heard rumours that somewhere there's a working strace for
> parisc.... Can anyone point me in the right direction?
>
> Dave
> --
> Dave O'Neill, Senior Linux Consultant, Linuxcare, Inc.
> desk: (613) 562-9949 fax: (613) 562-9700 cell: (613) 223-0219
> doneill@linuxcare.com http://www.linuxcare.com/
>
> ---------------------------------------------------------------------------
> To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com with
> `unsubscribe' as the subject.
>
>
From alan@lxorguk.ukuu.org.uk Mon, 27 Nov 2000 16:55:36 +0000 (GMT)
Date: Mon, 27 Nov 2000 16:55:36 +0000 (GMT)
From: Alan Cox alan@lxorguk.ukuu.org.uk
Subject: [parisc-linux] RPM and hppa
> I think the above is a different problem. The explosion is most likely
> a problem with exception edges in the gcse pass. Try `-fno-gcse'. This
> also appears in the tFile.cc libio test.
I've not yet had a chance to try this. I see the same behaviour building Qt
so I may try it on that
From marteaut@esiee.fr Mon, 27 Nov 2000 18:11:36 +0000
Date: Mon, 27 Nov 2000 18:11:36 +0000
From: marteau marteaut@esiee.fr
Subject: [parisc-linux]
Hi all,
We have tried to compile the new linux source today and vmlinux
always needs pci.o to link. It works if you delete pci.o from the
objects list in the kernel Makefile. But, we don't know why it was
needed because we did not ask for in our "make config"...
We appreciate any help!
THX,
ESIEE Team
From dave@hiauly1.hia.nrc.ca Mon, 27 Nov 2000 12:13:02 -0500 (EST)
Date: Mon, 27 Nov 2000 12:13:02 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: [parisc-linux] RPM and hppa
>
> > I think the above is a different problem. The explosion is most likely
> > a problem with exception edges in the gcse pass. Try `-fno-gcse'. This
> > also appears in the tFile.cc libio test.
>
> I've not yet had a chance to try this. I see the same behaviour building Qt
> so I may try it on that
Another option that might help if exceptions aren't needed is `-fno-exceptions'.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From grundler@cup.hp.com Mon, 27 Nov 2000 09:57:39 -0800
Date: Mon, 27 Nov 2000 09:57:39 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: No subject
marteau wrote:
> We have tried to compile the new linux source today and vmlinux
> always needs pci.o to link. It works if you delete pci.o from the
> objects list in the kernel Makefile. But, we don't know why it was
> needed because we did not ask for in our "make config"...
Hi Thomas,
The problem is in arch/parisc/kernel/Makefile.
It doesn't use CONFIG_PCI to decide if pci.o is needed.
I'll fix that now - should be simple.
thanks,
grant
From dave@hiauly1.hia.nrc.ca Mon, 27 Nov 2000 13:19:52 -0500 (EST)
Date: Mon, 27 Nov 2000 13:19:52 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: [parisc-linux] RPM and hppa
> > I think the above is a different problem. The explosion is most likely
> > a problem with exception edges in the gcse pass. Try `-fno-gcse'. This
> > also appears in the tFile.cc libio test.
>
> I've not yet had a chance to try this. I see the same behaviour building Qt
> so I may try it on that
If the problem is in fact the exception handling method, I think we should
look at implementing the DWARF2 unwind mechanism and exception handling using
this unwind mechanism. The default method of handling exceptions is sjlj.
See the description of DWARF2_UNWIND_INFO and INCOMING_RETURN_ADDR_RTX in
gcc/tm.texi for more info. Alan Modra may want this for gdb as well.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From alan@lxorguk.ukuu.org.uk Mon, 27 Nov 2000 18:34:23 +0000 (GMT)
Date: Mon, 27 Nov 2000 18:34:23 +0000 (GMT)
From: Alan Cox alan@lxorguk.ukuu.org.uk
Subject: [parisc-linux] gcc crashes/out of memory and groff
With groff -fno-gcse does not help but -O0 does. Without -O0 you get an
internal error.
g++ -I. -I/usr/src/redhat/BUILD/groff-1.16/src/roff/troff -I/usr/src/redhat/BUILD/groff-1.16/src/include -I/usr/src/redhat/BUILD/groff-1.16/src/include -DHAVE_UNISTD_H=1 -DHAVE_DIRENT_H=1 -DHAVE_LIMITS_H=1 -DHAVE_SYS_DIR_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRINGS_H=1 -DSTDLIB_H_DECLARES_PUTENV=1 -DSTDIO_H_DECLARES_POPEN=1 -DSTDIO_H_DECLARES_PCLOSE=1 -DHAVE_SYS_NERR=1 -DHAVE_SYS_ERRLIST=1 -DHAVE_CC_LIMITS_H=1 -DRETSIGTYPE=void -DHAVE_STRUCT_EXCEPTION=1 -DHAVE_UNISTD_H=1 -DHAVE_GETPAGESIZE=1 -DHAVE_MMAP=1 -DHAVE_FMOD=1 -DHAVE_STRTOL=1 -DHAVE_GETCWD=1 -DHAVE_STRERROR=1 -DHAVE_PUTENV=1 -DHAVE_RENAME=1 -DHAVE_MKSTEMP=1 -DHAVE_STRCASECMP=1 -DHAVE_STRSEP=1 -DHAVE_STRDUP=1 -DSYS_SIGLIST_DECLARED=1 -O2 -mpa-risc-1-0 -fno-gcse -c env.cc
env.cc: In member function `void environment::output_line (node *, hunits)':
env.cc:1582: Internal error: Segmentation fault.
Please submit a full bug report.
See for instructions.
make[2]: *** [env.o] Error 1
make[2]: Leaving directory `/usr/src/redhat/BUILD/groff-1.16/src/roff/troff'
Alan
From marteaut@esiee.fr Mon, 27 Nov 2000 21:31:09 +0100
Date: Mon, 27 Nov 2000 21:31:09 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] Include trouble
Hi folks,
Just to know if it is a local problem. We have this warning when we
cross compile the kernel:
/linux/cvs/linux/include/linux/string.h:61: warning: conflicting types
for built-in function `memset'
/linux/cvs/linux/include/linux/string.h:64: warning: conflicting types
for built-in function `memcpy'
/linux/cvs/linux/include/linux/string.h:73: warning: conflicting types
for built-in function `memcmp'
Can we have your point of view?
Bye, ESIEE Team
From marteaut@esiee.fr Tue, 28 Nov 2000 00:17:24 +0100
Date: Tue, 28 Nov 2000 00:17:24 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] Leds
Hi all,
We tried to implement the led power of 712 hp boxes. We call leds_init
in main.c in init directory. Even though it is working, we are not sure
where to put the source that must be completed for others models.
We could make a LEDS directory in drivers
put the file in the kernel directory
We also put the source for the initialisation of keyboard leds in
lasi_ps2_reset but we admit that no led is on. Is it true?
How can we know?
THX, ESIEE Team
From alan@linuxcare.com.au Tue, 28 Nov 2000 12:12:03 +1100 (EST)
Date: Tue, 28 Nov 2000 12:12:03 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] gcc crashes/out of memory and groff
On Mon, 27 Nov 2000, Alan Cox wrote:
> With groff -fno-gcse does not help but -O0 does. Without -O0 you get an
> internal error.
Cross compiling from an x86-linux box with enough memory+swap (256M + 2G),
env.cc eventually compiled for me using the default -O2 optimisation
level. I had a crash later when compiling preproc/tbl/table.cc, so that's
nothing to boast about.
table.cc: In member function `void table::build_span_list ()':
table.cc:2009: Internal error: Segmentation fault.
Worse, when I added "-Q -da" to see what was happening, the compile
succeeded - and also succeeded with either of -Q or -da alone. So I
ran cc1plus under gdb, and that too failed to crash. :-(
Maybe a garbage collection or uninitialised var bug?
Alan Modra
--
Linuxcare. Support for the Revolution.
From alan@linuxcare.com.au Tue, 28 Nov 2000 15:59:54 +1100 (EST)
Date: Tue, 28 Nov 2000 15:59:54 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] Include trouble
On Mon, 27 Nov 2000, Thomas Marteau wrote:
> Can we have your point of view?
Your version of gcc expects the size_t parameter to all these functions to
be "unsigned int", whereas the 2000/11/18 changes to
linux/include/asm-parisc/posix_types.h made __kernel_size_t
"unsigned long".
As far as I understand, gcc's cpp has a built-in definition of size_t,
__SIZE_TYPE__, and it ultimately gets it's idea of the definition from the
kernel includes on the machine where gcc was compiled. ie. recompile gcc
with the new kernel headers installed, and the problem should go away.
For 32-bit hppa-linux, sizeof(int) == sizeof(long) so there shouldn't be
any practical consequence other than these warning messages. There might
be some "interesting" problems on hppa64-linux - I'm not sure.
Alan Modra
--
Linuxcare. Support for the Revolution.
From alan@linuxcare.com.au Tue, 28 Nov 2000 20:19:20 +1100 (EST)
Date: Tue, 28 Nov 2000 20:19:20 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] Include trouble
On Tue, 28 Nov 2000, Alan Modra wrote:
> As far as I understand, gcc's cpp has a built-in definition of size_t,
> __SIZE_TYPE__, and it ultimately gets it's idea of the definition from the
> kernel includes on the machine where gcc was compiled. ie. recompile gcc
> with the new kernel headers installed, and the problem should go away.
No, that's wrong. __SIZE_TYPE__ comes from `#define SIZE_TYPE' in
gcc/config/pa/pa-linux{,64}.h I'll check in a patch to gcc in a few
moments.
Alan
--
Linuxcare. Support for the Revolution.
From marteaut@esiee.fr Tue, 28 Nov 2000 16:07:04 +0100
Date: Tue, 28 Nov 2000 16:07:04 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] Mouse driver for PS/2
Hi all,
We appreciate if someone can explain where we can find request_irq and
request_region in hp_psaux.c and also,
what are they doing?
Thanks,
ESIEE Team
From deller@gmx.de Tue, 28 Nov 2000 18:21:34 +0100
Date: Tue, 28 Nov 2000 18:21:34 +0100
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] Mouse driver for PS/2
On Tuesday 28 November 2000 16:07, Thomas Marteau wrote:
> Hi all,
Hi Thomas,
>
> We appreciate if someone can explain where we can find request_irq
#include and /arch/parisc/kernel/irq.c
request_irq() binds the given interrupt-number to a function (if possible).
> and
> request_region
#include
request_region only marks (and tests) an I/O-region as used by a driver and
makes this information visible via /proc/iomem and /proc/ioports.
> in hp_psaux.c and also,
> what are they doing?
>
> Thanks,
> ESIEE Team
From wferguson@server01.chatspace.com Tue, 28 Nov 2000 09:35:36 -0800
Date: Tue, 28 Nov 2000 09:35:36 -0800
From: William Ferguson wferguson@server01.chatspace.com
Subject: [parisc-linux] PDC firmware revision FAQ update
This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.
------_=_NextPart_001_01C05961.9E1AB8A8
Content-Type: text/plain;
charset="iso-8859-1"
Is the web server at www.parisc-linux.org down? Can't seem to connect from
San Diego.
-----Original Message-----
From: Grant Grundler [mailto:grundler@cup.hp.com]
Sent: Friday, November 17, 2000 1:51 PM
To: parisc-linux@puffin.external.hp.com
Subject: [parisc-linux] PDC firmware revision FAQ update
Hi folks,
The issue of PDC/firmware revs came up locally and sounded like a FAQ.
I added
"10. How can I check if the PDC (firmware) revision is the latest?"
and what I knew/could find to our FAQ at:
http://www.parisc-linux.org/faq.html
The new FAQ entry should be visible to the world in the next hour or so.
**** WARNING ****
Firmware upgrades can *kill* your machine!
**** WARNING ****
Don't do it just because. Read the FAQ carefully. Take the
time to figure out why you might need the upgrade and expose
yourself to this risk.
grant
ps. please don't ask me why your favorite machine's PDC isn't listed.
I don't run the referenced sites.
---------------------------------------------------------------------------
To unsubscribe: send e-mail to parisc-linux-request@thepuffingroup.com with
`unsubscribe' as the subject.
------_=_NextPart_001_01C05961.9E1AB8A8
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
RE: [parisc-linux] PDC firmware revision FAQ update
Is the web server at www.parisc-linux.org down? =
Can't seem to connect from San Diego.
-----Original Message-----
From: Grant Grundler [mailto:grundler@cup.hp.com]
Sent: Friday, November 17, 2000 1:51 PM
To: parisc-linux@puffin.external.hp.com
Subject: [parisc-linux] PDC firmware revision FAQ =
update
Hi folks,
The issue of PDC/firmware revs came up locally and =
sounded like a FAQ.
I added
"10. How can I check if the =
PDC (firmware) revision is the latest?"
and what I knew/could find to our FAQ at:
http://www.parisc-linux.org/faq.html
The new FAQ entry should be visible to the world in =
the next hour or so.
**** WARNING ****
Firmware upgrades can *kill* your =
machine!
**** WARNING ****
Don't do it just because. Read the FAQ carefully. =
Take the
time to figure out why you might need the upgrade =
and expose
yourself to this risk.
grant
ps. please don't ask me why your favorite machine's =
PDC isn't listed.
I don't run the referenced =
sites.
---------------------------------------------------------------=
------------
To unsubscribe: send e-mail to =
parisc-linux-request@thepuffingroup.com with
`unsubscribe' as the subject.
------_=_NextPart_001_01C05961.9E1AB8A8--
From dave@hiauly1.hia.nrc.ca Tue, 28 Nov 2000 12:37:28 -0500 (EST)
Date: Tue, 28 Nov 2000 12:37:28 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: [parisc-linux] Include trouble
> On Tue, 28 Nov 2000, Alan Modra wrote:
>
> > As far as I understand, gcc's cpp has a built-in definition of size_t,
> > __SIZE_TYPE__, and it ultimately gets it's idea of the definition from the
> > kernel includes on the machine where gcc was compiled. ie. recompile gcc
> > with the new kernel headers installed, and the problem should go away.
>
> No, that's wrong. __SIZE_TYPE__ comes from `#define SIZE_TYPE' in
> gcc/config/pa/pa-linux{,64}.h I'll check in a patch to gcc in a few
> moments.
Hmm. Most 32 bit systems use "unsigned int" rather than "unsigned long".
Using "unsigned long" might cause problems with packages like libio.
I know there is a problem if off_t is not the same as size_t.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From grundler@cup.hp.com Tue, 28 Nov 2000 09:49:26 -0800
Date: Tue, 28 Nov 2000 09:49:26 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] Mouse driver for PS/2
Thomas Marteau wrote:
> Hi all,
>
> We appreciate if someone can explain where we can find request_irq and
> request_region in hp_psaux.c and also, what are they doing?
include/linux/sched.h:extern int request_irq(unsigned int, ...)
(Implementation is in arch/parisc/kernel/irq.c)
The request_irq() "allocates" an IRQ line for use by the device - this
program the PIC (or APIC) on x86 platforms. request_irq() is also how
an interrupt handler is associated with a specific IRQ line. Since
PA Risc CPU's don't have IRQ lines going into them, IRQ's are virtualized
and don't always represent a physical IRQ line.
Under LASI, see gsc_alloc_irq(&gsc_irq) usage in drivers/gsc/lasi.c.
lasi_find_irq() helps associate the PS/2 interrupt handler with
the correct IRQ line which is internal to lasi.
include/linux/ioport.h:#define request_region(start,n,name) ...
request_region() will reserve a range of I/O port address from the
global I/O space. request_mem_region() does the same for MMIO.
Drivers that use inb/outb (as defined in include/asm-parisc/io.h)
must use request_region(). Drivers that use gsc_readb/gsc_writeb
(as defined in include/asm-parisc/gsc.h) must use request_mem_region().
Collisions are probably occuring where a driver originally used inb/outb
and those were redefined to use gsc_readb/writeb functions. But the
driver is still using request_region().
And it doesn't help that I may have broken some of the resource mgt
with some code I committed last night. It worked on my boxes (A180/C3K)
but broke on other folks. Paul Bame helped find/fix one bug but
I shouldn't be surprised if more bugs are still out there.
I will be fixing some known resource failure problems on A500.
Please post problems on other platforms to parisc-linux list as well.
hope this helps,
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From grundler@cup.hp.com Tue, 28 Nov 2000 09:53:37 -0800
Date: Tue, 28 Nov 2000 09:53:37 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] PDC firmware revision FAQ update
William Ferguson wrote:
> Is the web server at www.parisc-linux.org down? Can't seem to connect from
> San Diego.
Yes. Linuxcare is having some DNS problems right now and
they host that site. Any forecasts on when that will be back up?
grant
From bame@noam.fc.hp.com Tue, 28 Nov 2000 12:27:09 -0700
Date: Tue, 28 Nov 2000 12:27:09 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] Include trouble
= Hmm. Most 32 bit systems use "unsigned int" rather than "unsigned long".
I will be happy to change it back to unsigned int. The only reason
I used unsigned long is because it seems off_t wants to be, to a
first approximation, the word length of the machine, and 'long' does
that.
= Using "unsigned long" might cause problems with packages like libio.
= I know there is a problem if off_t is not the same as size_t.
What problem is that? I'm working on a proposal for the palinux
type sizes at the moment.
One dilema is that off_t is supposed to
be good for file offsets, which these days means it should be 64 bits.
size_t refers to the sizes of objects which are expected to fit in
RAM, so should be the word size of the machine. So there's a conflict
because one logic suggests 64-bit off_t and 32-bit size_t on 32-bit palinux,
but your statement suggests they should be the same size. Any ideas?
However I'm leaning towards leaning with the older 32-bit off_t because
we might not want to be the first ones to fix all the problems with
making it 64 bits on a 32-bit machine.
-P
From marteaut@esiee.fr Tue, 28 Nov 2000 20:30:30 +0100
Date: Tue, 28 Nov 2000 20:30:30 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] Mouse driver for PS/2
Hi all,
We have a sample of code for the mouse driver. But, we would like to
know how the Puffin would like the implementation of the driver. Because
of drag & drop..., do you want two different interrupt functions or a
single that does a redirection. Also , we have seen that busmouse.c is
quite interesting for our driver. Do we make a copy and call it
hp_mouse.c?
Thanks for your answer,
ESIEE Team
For a better future with a mouse!
From alan@lxorguk.ukuu.org.uk Tue, 28 Nov 2000 20:02:26 +0000 (GMT)
Date: Tue, 28 Nov 2000 20:02:26 +0000 (GMT)
From: Alan Cox alan@lxorguk.ukuu.org.uk
Subject: [parisc-linux] Include trouble
> One dilema is that off_t is supposed to
> be good for file offsets, which these days means it should be 64 bits.
off_t should be a natural type. So it should be 32bits. glibc 2.2 deals with
the 64bit I/O stuff nicely
> However I'm leaning towards leaning with the older 32-bit off_t because
> we might not want to be the first ones to fix all the problems with
> making it 64 bits on a 32-bit machine.
Go 32bit. Linus will probably refuse to touch a 32bit port using longlong
internally for off-t
From dave@hiauly1.hia.nrc.ca Tue, 28 Nov 2000 15:24:26 -0500 (EST)
Date: Tue, 28 Nov 2000 15:24:26 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: [parisc-linux] Include trouble
> = Hmm. Most 32 bit systems use "unsigned int" rather than "unsigned long".
>
> I will be happy to change it back to unsigned int. The only reason
> I used unsigned long is because it seems off_t wants to be, to a
> first approximation, the word length of the machine, and 'long' does
> that.
Agreed.
> = Using "unsigned long" might cause problems with packages like libio.
>
>
> = I know there is a problem if off_t is not the same as size_t.
>
> What problem is that? I'm working on a proposal for the palinux
> type sizes at the moment.
It is simply dumb coding that hasn't been fixed. There are inconsistencies
between the interface declarations and implementations. The old libio
is on the way out.
> One dilema is that off_t is supposed to
> be good for file offsets, which these days means it should be 64 bits.
> size_t refers to the sizes of objects which are expected to fit in
> RAM, so should be the word size of the machine. So there's a conflict
> because one logic suggests 64-bit off_t and 32-bit size_t on 32-bit palinux,
> but your statement suggests they should be the same size. Any ideas?
Only, because of poorly written legacy code. I agree that off_t should
be 64 bits on 32 bit platforms today.
> However I'm leaning towards leaning with the older 32-bit off_t because
> we might not want to be the first ones to fix all the problems with
> making it 64 bits on a 32-bit machine.
Maybe it isn't that bad because I noticed at least one port used unsigned
long long for off_t.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From dave@hiauly1.hia.nrc.ca Tue, 28 Nov 2000 15:33:52 -0500 (EST)
Date: Tue, 28 Nov 2000 15:33:52 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: [parisc-linux] Include trouble
> Maybe it isn't that bad because I noticed at least one port used unsigned
> long long for off_t.
Take that back. It was __kernel_loff_t. Just go with
typedef unsigned int __kernel_size_t;
typedef long __kernel_off_t;
#ifdef __GNUC__
typedef long long __kernel_loff_t;
#endif
for the 32 bit port.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From deller@gmx.de Tue, 28 Nov 2000 21:41:53 +0100
Date: Tue, 28 Nov 2000 21:41:53 +0100
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] Mouse driver for PS/2
On Tuesday 28 November 2000 20:30, Thomas Marteau wrote:
> Hi all,
Hi Thomas,
>
> We have a sample of code for the mouse driver. But, we would like to
> know how the Puffin would like the implementation of the driver. Because
> of drag & drop..., do you want two different interrupt functions or a
> single that does a redirection. Also , we have seen that busmouse.c is
> quite interesting for our driver. Do we make a copy and call it
> hp_mouse.c?
I would propose, that you check in hp_mouse.c and use different interrupt
functions for now. Changing it later shouldn't be too difficult.
Greetings,
Helge.
>
> Thanks for your answer,
> ESIEE Team
> For a better future with a mouse!
From doneill@linuxcare.com Tue, 28 Nov 2000 16:08:49 -0500 (EST)
Date: Tue, 28 Nov 2000 16:08:49 -0500 (EST)
From: Dave O'Neill doneill@linuxcare.com
Subject: [parisc-linux] PDC firmware revision FAQ update
On Tue, 28 Nov 2000, Grant Grundler wrote:
> Yes. Linuxcare is having some DNS problems right now and
> they host that site. Any forecasts on when that will be back up?
Well, it looks like our DNS issues are fixed... the site should be
accessible now.
Dave
--
Dave O'Neill, Senior Linux Consultant, Linuxcare, Inc.
desk: (613) 562-9949 fax: (613) 562-9700 cell: (613) 223-0219
doneill@linuxcare.com http://www.linuxcare.com/
From bame@noam.fc.hp.com Tue, 28 Nov 2000 14:21:05 -0700
Date: Tue, 28 Nov 2000 14:21:05 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] Include trouble
= Go 32bit. Linus will probably refuse to touch a 32bit port using longlong
= internally for off-t
Hmmm, too bad, since we have few palinux backwards compatibility issues and
could just have the __USE_FILE_OFFSET64 glibc magic be a no-op rather
than supporting all those extra *64 syscalls. Plus we'd need
considerably fewer syscall translators to run 32-bit apps on 64-bit
kernel (but might need more for 32-bit hpux apps). It seems illogical
to make a file-system-related data type different based on cpu
word size but I understand this isn't simple -- oh well.
Seems like the consensus is 32-bit off_t on 32-bit kernel and just
live with all those *64 syscalls -- many not supported on palinux
yet I notice.
-P
From alan@lxorguk.ukuu.org.uk Tue, 28 Nov 2000 21:58:56 +0000 (GMT)
Date: Tue, 28 Nov 2000 21:58:56 +0000 (GMT)
From: Alan Cox alan@lxorguk.ukuu.org.uk
Subject: [parisc-linux] Include trouble
> Seems like the consensus is 32-bit off_t on 32-bit kernel and just
> live with all those *64 syscalls -- many not supported on palinux
> yet I notice.
Remember providing you use off_t you can just tell glibc to do all the work
for you and write with a 64bit off_t to userspace
From alan@linuxcare.com.au Wed, 29 Nov 2000 10:27:20 +1100 (EST)
Date: Wed, 29 Nov 2000 10:27:20 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] Include trouble
On Tue, 28 Nov 2000, John David Anglin wrote:
> Hmm. Most 32 bit systems use "unsigned int" rather than "unsigned long".
There's some precedent for using "unsigned long", as that is what is used
on 32-bit hpux11 and osf gcc targets.
current sourceware CVS gcc/config/pa/:
$ grep SIZE_TYPE *
grep: CVS: Is a directory
pa-64.h:#undef SIZE_TYPE
pa-64.h:#define SIZE_TYPE "long unsigned int"
pa-hpux.h:#undef SIZE_TYPE
pa-hpux.h:#define SIZE_TYPE "unsigned int"
pa-hpux11.h:#undef SIZE_TYPE
pa-hpux11.h:#define SIZE_TYPE "long unsigned int"
pa-hpux7.h:#undef SIZE_TYPE
pa-hpux7.h:#define SIZE_TYPE "unsigned int"
pa-linux.h:#undef SIZE_TYPE
pa-linux.h:#define SIZE_TYPE "unsigned int"
pa-osf.h:#undef SIZE_TYPE
pa-osf.h:#define SIZE_TYPE "long unsigned int"
pa-pro-end.h:#undef SIZE_TYPE
pa-pro-end.h:#define SIZE_TYPE "unsigned int"
pa.h:#define SIZE_TYPE "unsigned int"
Some further grepping shows quite a number of other 32-bit gcc targets
using "unsigned long", but I didn't see any 32-bit linux targets in the
list. Paul, If you change back to "unsigned int", please change
gcc/config/pa/pa-linux.h too.
Alan
--
Linuxcare. Support for the Revolution.
From Frank.Butter@otto.de Wed, 29 Nov 2000 11:50:39 +0100
Date: Wed, 29 Nov 2000 11:50:39 +0100
From: Butter, Frank Frank.Butter@otto.de
Subject: [parisc-linux] hardware
to all here I was annaouncing an offer for hp hardware recently:
it'll take longer than initially expected
to convince our bosses ;-/
please stay patient...
frank
From matthias@archimed.math.uni-mannheim.de Wed, 29 Nov 2000 20:52:38 +0100 (MET)
Date: Wed, 29 Nov 2000 20:52:38 +0100 (MET)
From: Matthias Juchem matthias@archimed.math.uni-mannheim.de
Subject: [parisc-linux] parisc-0.5.iso and stack trace
Hi there.
I've just downloaded the ISO images and tried to boot an 9000/705. I keep
getting a stack trace. Can you tell me which part of the trace I can send
you so that you could eventually tell me what is wrong?
TIA,
Matthias
From dave@hiauly1.hia.nrc.ca Wed, 29 Nov 2000 17:05:31 -0500 (EST)
Date: Wed, 29 Nov 2000 17:05:31 -0500 (EST)
From: John David Anglin dave@hiauly1.hia.nrc.ca
Subject: [parisc-linux] gcc crashes/out of memory and groff
> With groff -fno-gcse does not help but -O0 does. Without -O0 you get an
> internal error.
>
> g++ -I. -I/usr/src/redhat/BUILD/groff-1.16/src/roff/troff -I/usr/src/redhat/BUILD/groff-1.16/src/include -I/usr/src/redhat/BUILD/groff-1.16/src/include -DHAVE_UNISTD_H=1 -DHAVE_DIRENT_H=1 -DHAVE_LIMITS_H=1 -DHAVE_SYS_DIR_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRINGS_H=1 -DSTDLIB_H_DECLARES_PUTENV=1 -DSTDIO_H_DECLARES_POPEN=1 -DSTDIO_H_DECLARES_PCLOSE=1 -DHAVE_SYS_NERR=1 -DHAVE_SYS_ERRLIST=1 -DHAVE_CC_LIMITS_H=1 -DRETSIGTYPE=void -DHAVE_STRUCT_EXCEPTION=1 -DHAVE_UNISTD_H=1 -DHAVE_GETPAGESIZE=1 -DHAVE_MMAP=1 -DHAVE_FMOD=1 -DHAVE_STRTOL=1 -DHAVE_GETCWD=1 -DHAVE_STRERROR=1 -DHAVE_PUTENV=1 -DHAVE_RENAME=1 -DHAVE_MKSTEMP=1 -DHAVE_STRCASECMP=1 -DHAVE_STRSEP=1 -DHAVE_STRDUP=1 -DSYS_SIGLIST_DECLARED=1 -O2 -mpa-risc-1-0 -fno-gcse -c env.cc
> env.cc: In member function `void environment::output_line (node *, hunits)':
> env.cc:1582: Internal error: Segmentation fault.
> Please submit a full bug report.
> See for instructions.
> make[2]: *** [env.o] Error 1
> make[2]: Leaving directory `/usr/src/redhat/BUILD/groff-1.16/src/roff/troff'
This error is also exception related but appears different from the
memory explosion. It occurs in scan_region in except.c. The memory
explosion that occurs without -fno-gcse does seem to occur in the
gcse pass as I suspected. I need to rebuild gcc with debugging to
get further.
I was able to build the groff package with `CXXFLAGS="-O3 -fno-exceptions"'.
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)
From delyra@latt.if.usp.br Wed, 29 Nov 2000 20:22:47 -0200 (BRST)
Date: Wed, 29 Nov 2000 20:22:47 -0200 (BRST)
From: Jorge L. deLyra delyra@latt.if.usp.br
Subject: [parisc-linux] HP 9000/735-125
I just booted the palinux-0.5.iso CD on a HP 9000 model 735 workstation.
It went on quite a bit before hanging immediatelly after reporting the
serial ports. Gave large amount of what looks like traceback or state
information. Question: would it do anyone any good if I tried to get
everything the kernel said before the hang into a message in this list?
Cheers,
----------------------------------------------------------------
Jorge L. deLyra, Associate Professor of Physics
The University of Sao Paulo, IFUSP-DFMA
For more information: finger delyra@latt.if.usp.br
----------------------------------------------------------------
From grundler@cup.hp.com Wed, 29 Nov 2000 15:19:10 -0800
Date: Wed, 29 Nov 2000 15:19:10 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] HP 9000/735-125
"Jorge L. deLyra" wrote:
> I just booted the palinux-0.5.iso CD on a HP 9000 model 735 workstation.
> It went on quite a bit before hanging immediatelly after reporting the
> serial ports. Gave large amount of what looks like traceback or state
> information. Question: would it do anyone any good if I tried to get
> everything the kernel said before the hang into a message in this list?
Certainly. I won't be able to debug the problems since I (a) don't have
the time too (unless it's obvious) and (b) don't have a 735 setup
for testing.
This is becoming a FAQ:
"My system crashed after ... What should I do next?"
I'm thinking people in the support space would know how to write
this one really well :^)
Is I get one in the mail, I'll add it to our FAQ.
thanks,
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From smoret@uci.edu Wed, 29 Nov 2000 15:28:20 -0800
Date: Wed, 29 Nov 2000 15:28:20 -0800
From: Steve Moret smoret@uci.edu
Subject: [parisc-linux] HP 9000/735-125
Jorge,
I have a 735-125 and am able to net-boot it on a custom configured kernel as
long as I disable the ASP parallel ports. It works quite well using an NFS
root as I cannot get the SCSI hard drives to mkfs. I was going to try and
debug it but have yet to find the free time. I can e-mail you my working
.config for the kernel, or build kernels (tested on my own 735) for people
if they need them.
--
Steve Moret
smoret@uci.edu
> -----Original Message-----
> From: Grant Grundler [mailto:grundler@cup.hp.com]
> Sent: Wednesday, November 29, 2000 3:19 PM
> To: Jorge L. deLyra
> Cc: parisc-linux@thepuffingroup.com
> Subject: Re: [parisc-linux] HP 9000/735-125
>
>
> "Jorge L. deLyra" wrote:
> > I just booted the palinux-0.5.iso CD on a HP 9000 model 735 workstation.
> > It went on quite a bit before hanging immediatelly after reporting the
> > serial ports. Gave large amount of what looks like traceback or state
> > information. Question: would it do anyone any good if I tried to get
> > everything the kernel said before the hang into a message in this list?
>
> Certainly. I won't be able to debug the problems since I (a) don't have
> the time too (unless it's obvious) and (b) don't have a 735 setup
> for testing.
>
> This is becoming a FAQ:
> "My system crashed after ... What should I do next?"
>
> I'm thinking people in the support space would know how to write
> this one really well :^)
> Is I get one in the mail, I'll add it to our FAQ.
>
> thanks,
> grant
>
>
> Grant Grundler
> Unix Systems Enablement Lab
> +1.408.447.7253
>
> ------------------------------------------------------------------
> ---------
> To unsubscribe: send e-mail to
> parisc-linux-request@thepuffingroup.com with
> `unsubscribe' as the subject.
>
>
From deller@gmx.de Thu, 30 Nov 2000 01:10:51 +0100
Date: Thu, 30 Nov 2000 01:10:51 +0100
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] HP 9000/735-125
X
On Thursday 30 November 2000 00:28, Steve Moret wrote:
> I have a 735-125 and am able to net-boot it on a custom configured kernel
> as long as I disable the ASP parallel ports. ....
Steve,
I really would like to get the parallel-port problems on ASP get fixed as
soon as possible.
Could you please mail me your bootlog (with parport enabled), so I can try to
track down the problem.
Maybe you can also check out CVS again, and test if this version works for
you ?
Thanks,
Helge Deller
From alan@lxorguk.ukuu.org.uk Thu, 30 Nov 2000 02:38:56 +0000 (GMT)
Date: Thu, 30 Nov 2000 02:38:56 +0000 (GMT)
From: Alan Cox alan@lxorguk.ukuu.org.uk
Subject: [parisc-linux] XFree status
I have a server linked. inb/inw/outb/outw and friends are right now null
functions until I fill them in. Thats not a big deal. Initially I'll probably
use /dev/port but for speed I hope everyone uses mmio based hardware.
Also is this bad ?
/usr/bin/ld: lbxdix.o(.text+0x6fc): fixing R_PARISC_DPREL21L
/usr/bin/ld: lbxprop.o(.text+0xde4): fixing R_PARISC_DPREL21L
/usr/bin/ld: lbxprop.o(.text+0xdf4): fixing R_PARISC_DPREL21L
Alan
phux:/usr/src/redhat/BUILD/XFree86-4.0.1/xc/programs/Xserver# ./XFree86
XFree86 Version 4.0.1a / X Window System
(protocol Version 11, revision 0, vendor release 6400)
Release Date: 2 August 2000
If the server is older than 6-12 months, or if your card is newer
than the above date, look for a newer version before reporting
problems. (see http://www.XFree86.Org/FAQ)
Operating System: Linux 2.4.0-test6 parisc [ELF]
(==) Log file: "/var/log/XFree86.0.log", Time: Wed Nov 29 19:40:16 2000
(==) Using config file: "/etc/X11/XF86Config"
Parse warning on line 77 of section Keyboard in file /etc/X11/XF86Config
Ignoring obsolete keyword "LeftAlt".
Parse error on line 77 of section Keyboard in file /etc/X11/XF86Config
"Meta" is not a valid keyword in this section.
(EE) Problem parsing the config file
(EE) Error from xf86HandleConfigFile()
Fatal server error:
no screens found
When reporting a problem related to a server crash, please send
the full server output, not just the last messages.
This can be found in the log file "/var/log/XFree86.0.log".
Please reports problems to xfree86@xfree86.org.
From alan@linuxcare.com.au Thu, 30 Nov 2000 14:41:48 +1100 (EST)
Date: Thu, 30 Nov 2000 14:41:48 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: [parisc-linux] XFree status
On Thu, 30 Nov 2000, Alan Cox wrote:
> Also is this bad ?
>
> /usr/bin/ld: lbxdix.o(.text+0x6fc): fixing R_PARISC_DPREL21L
No.
It's a symptom of a variable's "constness" being declared differently from
the way the variable is defined. For instance, referring to "extern int foo"
in one object with foo defined as "const int foo" in another. I'll be
removing the linker warning at some stage.
Alan Modra
--
Linuxcare. Support for the Revolution.
From grundler@cup.hp.com Wed, 29 Nov 2000 21:18:14 -0800
Date: Wed, 29 Nov 2000 21:18:14 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] XFree status
Alan Cox wrote:
> I have a server linked.
Alan - that's Cool! Wow!
> inb/inw/outb/outw and friends are right now null
> functions until I fill them in. Thats not a big deal. Initially I'll probably
> use /dev/port but for speed I hope everyone uses mmio based hardware.
All HP graphics for PARISC are memory mapped. Both "GSC" and PCI flavors.
But I'm pretty clueless how linux X server finds/talks to a frame buffer.
For HPUX, the graphics driver supports some ioctl()'s.
Any references to describe how it works for linux?
parisc-linux doesn't create kernel virtual addresses for MMIO.
Which interface is used to create user virtual addresses for MMIO?
(In general - not just for frame buffers)
Is the PCI address passed to user space?
I'm wondering if/how "bus_to_virt()" type translations take place.
sorry for the many questions...
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From gibreel@pobox.com 30 Nov 2000 00:09:03 -0800
Date: 30 Nov 2000 00:09:03 -0800
From: Stephen Zander gibreel@pobox.com
Subject: [parisc-linux] CVS linux Vs. -test10
>>>>> "Grant" == Grant Grundler writes:
Grant> For the record, the second issue bdale made clear was we
Grant> need "boot floppies" debian package working. We don't need
Grant> more ISO images (no offense to pjlahaie for his good
Grant> work). "Boot floppies" is a pre-requisite to becoming part
Grant> of the next debian release. Given I still don't have a clue
Grant> how to build a debian package and I can still contribute
Grant> alot in other areas, it doesn't make sense for me to do it
Grant> myself.
Oooh, there's a reason for me to finally get the 712/80 under my desk
to be more than a foot-rest. I'll see what I can do about this.
Note that the likelihood of Debian releasing woody anytime soon is
vanishingly small, so this dosen't have to happen Right Now.
--
Stephen (debian developer)
"Strange women lying in ponds distributing swords is no basis for a
system of government"
From smoret@uci.edu Thu, 30 Nov 2000 01:20:28 -0800
Date: Thu, 30 Nov 2000 01:20:28 -0800
From: Steve Moret smoret@uci.edu
Subject: [parisc-linux] HP 9000/735-125
Helge,
My mistake! I did a full build with todays CVS and parport didn't die. So
somewhere between the 17th and now parport must have been fixed.
Now, maybe you can help me identify my SCSI problem. I don't know if it is
because of driver issues or a bad disk (completly likely). Do other people
have the Fast SCSI2 working on their 735s?
Whenever I do a mke2fs (after partitioning the drive with fdisk ok) it dies.
At bootup I get:
SCSI subsystem driver Revision: 1.00
sim700: Couldn't get consistent shared memory
sim700: Configuring 53c700 (SCSI-ID 7) at f0825100, IRQ 86
scsi0: Revision 0x0
Post test1, istat 05, sstat0 00, dstat 84
sim700: WARNING IRQ probe failed, (returned 0)
scsi0: WARNING: target data areas are not dma coherent!
scsi0: test 1 completed ok.
scsi0: sim700_intr_handle() called with no interrupt
scsi0 : LASI/Simple 53c7xx
Vendor: HP Model: C2235 Rev: 0B11
Type: Direct-Access ANSI SCSI revision: 02
Vendor: HP Model: C2235 Rev: 0B11
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi disk sda at scsi0, channel 0, id 5, lun 0
Detected scsi disk sdb at scsi0, channel 0, id 6, lun 0
SCSI device sda: 825012 512-byte hdwr sectors (422 MB)
Partition check:
sda: sda1 sda2
SCSI device sdb: 825012 512-byte hdwr sectors (422 MB)
sdb: sdb1 sdb2
And then if I try to mke2fs the disk I get:
hp735:~# fdisk -l /dev/sda
Disk /dev/sda: 13 heads, 62 sectors, 1023 cylinders
Units = cylinders of 806 * 512 bytes
Device Boot Start End Blocks Id System
/dev/sda1 1 910 366699 83 Linux
/dev/sda2 911 1023 45539 82 Linux swap
hp735:~# mke2fs /dev/sda1
mke2fs 1.19, 13-Jul-2000 for EXT2 FS 0.5b, 95/08/09
Filesystem label=
OS type: Linux
Block size=1024 (log=0)
Fragment size=1024 (log=0)
91800 inodes, 366699 blocks
18334 blocks (5.00%) reserved for the super user
First data block=1
45 block groups
8192 blocks per group, 8192 fragments per group
2040 inodes per group
Superblock backups stored on blocks:
8193, 24577, 40961, 57345, 73729, 204801, 221185
Writing inode tables: done
Writing superblocks and filesystem accounting information: scsi0: Unable to
abort command for target 5
scsi0: Unable to send Bus Device Reset for target 5
scsi0: Unable to do SCSI bus reset
scsi0: >>>>>>>>>>>> Host reset <<<<<<<<<<<<
scsi0: istat = 0c, sstat0 = 00, sstat1 = 00, dstat = 00
scsi0: dsp = 0cf15038 (script[0x140e]), dsps = 0cf15cde, target = 0
scsi0: Failing command for ID5
scsi0: sim700_intr_handle() called with no interrupt
pa11_dma_map_single(PCI_DMA_NONE) called by c01cb6d4
kernel BUG at pci-dma.c:392!
pa11_dma_unmap_single(PCI_DMA_NONE) called by c01ca3bc
kernel BUG at pci-dma.c:403!
SCSI disk error : host 0 channel 0 id 5 lun 0 return code = 2
I/O error: dev 08:01, sector 268
I/O error: dev 08:01, sector 270
I/O error: dev 08:01, sector 396
I/O error: dev 08:01, sector 16396
I/O error: dev 08:01, sector 16524
I/O error: dev 08:01, sector 16652
Of course the I/O errors continue on for a long time. Are these bad drives?
Or is there a problem with the driver that still needs to be worked out?
Thanks for all your help, I hope my spews of debug output are helpful,
--
Steve Moret
smoret@uci.edu
> -----Original Message-----
> From: Helge Deller [mailto:deller@gmx.de]
> Sent: Wednesday, November 29, 2000 4:11 PM
> To: Steve Moret
> Cc: parisc-linux@thepuffingroup.com
> Subject: Re: [parisc-linux] HP 9000/735-125
>
> I really would like to get the parallel-port problems on ASP get fixed as
> soon as possible.
> Could you please mail me your bootlog (with parport enabled), so
> I can try to
> track down the problem.
> Maybe you can also check out CVS again, and test if this version
> works for
> you ?
From delyra@latt.if.usp.br Thu, 30 Nov 2000 08:58:22 -0200 (BRST)
Date: Thu, 30 Nov 2000 08:58:22 -0200 (BRST)
From: Jorge L. deLyra delyra@latt.if.usp.br
Subject: [parisc-linux] HP 9000/735-125
> > information. Question: would it do anyone any good if I tried to get
> > everything the kernel said before the hang into a message in this list?
>
> Certainly. I won't be able to debug the problems since I (a) don't have
> the time too (unless it's obvious) and (b) don't have a 735 setup
> for testing.
OK, here goes. I want to congratulate you all on this effort. We have five
of these HP-9000 stations here, quite old by now. They used to be our main
number crunching force. They are wonderfully built hardware in bad need of
a wonderful system on them! |:-)
----------------------------------------------------------------
Jorge L. deLyra, Associate Professor of Physics
The University of Sao Paulo, IFUSP-DFMA
For more information: finger delyra@latt.if.usp.br
----------------------------------------------------------------
Seyon Copyright (c) 1992-1993 Muhammad M. Saggaf. All rights reserved.
Version 2 rev. 20c i586-Linux steve@hammer 05/23/99 20:19:22.
Locating Modems...
Modem `/dev/ttyS0' is Available.
(c) Copyright. Hewlett-Packard Company. 1992.
All rights reserved.
PDC ROM rev. 2.7
IODC ROM rev. 1.1
80 MB of memory configured and tested.
Searching for Potential Boot Devices.
To terminate search, press and hold the ESCAPE key.
Device Selection Device Path Device Type
----------------------------------------------------------------------------
P0 scsi.6.0 HP C3725S
P1 scsi.4.0 TOSHIBA CD-ROM XM-3301TA
b) Boot from specified device
s) Search for bootable devices
a) Enter Boot Administration mode
x) Exit and continue boot sequence
?) Help
Select from menu: b p1
Trying scsi.4.0
Boot path initialized.
Attempting to load IPL.
Soft booted.
palo ipl bame@noam Tue Oct 31 14:18:02 MST 2000
0/vmlinux 2140145 bytes @ 0x6f9800
0/palo-cmdline '0/vmlinux ROOT=/ TERM=LINUX root=/dev/scd0'
Kernel: partition 0 file /vmlinux
ELF32 executable
Entry 00100150 first 00100000 n 4
Segment 0 load 00100000 size 1460344 mediaptr 0x1000
Segment 1 load 00266000 size 179048 mediaptr 0x166000
Segment 2 load 00294000 size 109876 mediaptr 0x192000
Segment 3 load 002b0000 size 8192 mediaptr 0x1ad000
branching to kernel entry point 0x00100150
PDC Console Initialized
The 32-bit Kernel has started...
Enabled FP coprocessor
Free memory starts at: 0xc02da000
(0x504d6c,0x504d6c,0x0,0x0)
PALO command line: 'ROOT=/ TERM=LINUX root=/dev/scd0'
PALO initrd 0-0
model 00002060 00000481 00000000 00000000 77f451b0 ffffffff 00000004 0000000a 0000000a
vers 00000015
CPUID vers 0 rev 0
Searching for devices in PDC firmware... processor hpa 0xfffbe000
an older box...
Found devices:
1. Outfield Core BA (11) at 0xf082f000, versions 0x9, 0x0, 0x70, 0x0, 0x0
2. Outfield Core SCSI (10) at 0xf0825000, versions 0x9, 0x0, 0x71, 0x0, 0x0
3. Outfield Core LAN (802.3) (10) at 0xf0826000, versions 0x9, 0x0, 0x72, 0x0, 0x0
4. Outfield Core HIL (10) at 0xf0821000, versions 0x9, 0x0, 0x73, 0x0, 0x0
5. Outfield Core RS-232 (10) at 0xf0823000, versions 0x9, 0x0, 0x75, 0x0, 0x0
6. Outfield Core RS-232 (10) at 0xf0822000, versions 0x9, 0x0, 0x75, 0x0, 0x0
7. Outfield Core Centronics (10) at 0xf0824000, versions 0x9, 0x0, 0x74, 0x0, 0x0
8. Outfield FW SCSI (10) at 0xf0830000, versions 0x9, 0x0, 0x7c, 0x0, 0x0
9. Outfield Audio (10) at 0xf1000000, versions 0x9, 0x0, 0x7f, 0x0, 0x0
10. Cobra EISA BA (11) at 0xfc000000, versions 0x4, 0x0, 0x76, 0x0, 0x0
11. Snake Cheetah (735/130) (0) at 0xfffbe000, versions 0x206, 0x0, 0x4, 0x0, 0x81
12. Snake Cheetah (1) at 0xfffbf000, versions 0x37, 0x0, 0x9, 0x0, 0x0
That's a total of 12 devices.
CPU(s): 1 x PA7100 (PCX-T) at 125.000000 MHz
Linux version 2.4.0-test6 (pjlahaie@elenuial.thepuffingroup.com) (gcc version 2.96 20000925 (experimental)) #32 Mon Nov 6 10:20:58 EST 2000
free_bootmem(0x2daa00, 0x4d25600)
initrd: 00000000-00000000
pagetable_init
On node 0 totalpages: 20480
zone(0): 10240 pages.
zone(1): 10240 pages.
zone(2): 0 pages.
Kernel command line: ROOT=/ TERM=LINUX root=/dev/scd0
trap_init
Calibrating delay loop... 124.52 BogoMIPS
Memory: 77468k available
Dentry-cache hash table entries: 16384 (order: 5, 131072 bytes)
Buffer-cache hash table entries: 4096 (order: 2, 16384 bytes)
Page-cache hash table entries: 32768 (order: 5, 131072 bytes)
Inode-cache hash table entries: 8192 (order: 4, 65536 bytes)
POSIX conformance testing by UNIFIX
ASP version 20 at 0xf0800000 found.
Found i82596 at 0xf0826000, IRQ 87
early initialization of device eth0 is deferred
Found HIL at 0xf0821000, IRQ 94
HIL: no keyboard present.
Warning : device (10, 0x9, 0x0, 0x73, 0x0) NOT claimed by HIL 712, 715 or similiar
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP
IP: routing cache hash table of 512 buckets, 4Kbytes
TCP: Hash tables configured (established 8192 bind 8192)
Starting kswapd v1.7
pty: 256 Unix98 ptys configured
lp: driver loaded but no devices found
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
sim700: Couldn't get consistent shared memory
sim700: Configuring 53c700 (SCSI-ID 7) at f0825100, IRQ 86
scsi0: Revision 0x0
Post test1, istat 05, sstat0 00, dstat 84
sim700: WARNING IRQ probe failed, (returned 0)
scsi0: WARNING: target data areas are not dma coherent!
scsi0: test 1 completed ok.
scsi0: sim700_intr_handle() called with no interrupt
scsi0 : LASI/Simple 53c7xx
scsi : 1 host.
Vendor: TOSHIBA Model: CD-ROM XM-3301TA Rev: 1651
Type: CD-ROM ANSI SCSI revision: 02
Detected scsi CD-ROM sr0 at scsi0, channel 0, id 4, lun 0
Vendor: HP Model: C3725S Rev: 6019
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi disk sda at scsi0, channel 0, id 6, lun 0
scsi : detected 1 SCSI cdrom 1 SCSI disk total.
Uniform CD-ROM driver Revision: 3.11
SCSI device sda: hdwr sector= 512 bytes. Sectors= 4194058 [2047 MB] [2.0 GB]
Partition check:
sda: unknown partition table
82596.c: MAC of HP700 LAN blindely read from the prom!
eth0: Couldn't get consistent shared memory
eth0: 82596 at 0xf0826000, 08 00 09 0B DB EB IRQ 87.
82596.c $Revision: 1.14 $
Serial driver version 5.01 (2000-05-29) with MANY_PORTS SHARE_IRQ SERIAL_PCI enabled
Dumping Stack from c4f9c000 to c4f9cbc0:
c000 00000000 00000140 00000000 00000000 c027a46c 00000001 00000000 ffffffff
c020 00000000 00000000 00000000 00000000 00000000 00000000 ffffffff c027a384
c040 c027a384 c4f90000 c02b0000 c028060c 00000000 00000000 00000000 00000000
c060 00000000 00000000 00000001 00000000 00000000 00000000 00000000 c02b0000
c080 c02b0000 c4f50000 00000000 00000000 00000000 c02c9ab8 00000000 c4f9c09c
c0a0 c4f9c09c c4f9c0a4 c011a6e4 c4f9cac8 00000000 00000000 00000000 00000000
c0c0 00000000 00000000 00000000 00000000 00000000 00000000 c4f9c000 c011d4a8
c0e0 00000000 00000037 00000000 00000000 00000024 00000000 0000005b 00000000
c100 00000000 00000000 00000000 00000000 00000000 80000000 00000000 00000000
c120 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c140 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c160 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c180 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c1a0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 fffffeff
c1c0 00000000 ffffffff 00000000 c027afb4 ffffffff ffffffff ffffffff ffffffff
c1e0 ffffffff ffffffff 00800000 05000000 00000000 ffffffff ffffffff ffffffff
c200 00000500 00000500 00000400 00000400 ffffffff ffffffff ffffffff ffffffff
c220 00007377 61707065 72000000 00000000 00000000 00000000 00000000 00000000
c240 00000000 00000000 00005000 c0267054 c0267054 c013d1e8 00010000 c4ffeba0
c260 c02238c4 c0236708 c02d9800 00504d6c 00000000 c011bb88 c0267000 00005000
c280 c0267054 c0267000 0000003e c027a000 00000001 c02b61eb 00000004 c02b61c7
c2a0 00000023 c02b61eb 00000000 00000000 c0100290 0000003e 00000000 00000024
c2c0 0000000b c027a4ac c027a000 08000059 00000000 000000ff 00000060 00000000
c2e0 00000060 00000002 002b2080 00000008 002b50c0 c0266000 00000000 023c3460
c300 c02b08c0 00000001 08000000 00000000 00000000 00000000 00856606 00000000
c320 00000000 00000000 42780000 00000000 431c0000 00000000 4471cccc 00000000
c340 000003c7 00000000 fffffde0 fffffde0 7f7fffff ffffffff 7f7fffff ffffffff
c360 00000000 00000000 00000000 00000000 41800000 00000000 00000010 00000010
c380 00000000 00000000 fffffde0 fffffde0 41000000 00000000 40800000 00000000
c3a0 fffffde0 fffffde0 41000000 00000000 fffffde0 fffffde0 40800000 00000000
c3c0 41000000 00000000 40300000 00000000 40200000 00000000 40200000 00000000
c3e0 41800000 fffffde0 40000000 00000000 40000000 00000000 40800000 00000000
c400 41000000 00000000 00000000 00000000 c4f9cb80 c0103cf4 00000000 00000000
c420 00000000 00000000 00000000 00000000 c011bb78 c011bb7c 40800000 00000000
c440 00281000 00000000 c0280040 c0280064 00000000 c0280204 00000000 00000000
c460 00000000 00000000 00000000 c4f9c468 00000000 00000000 00000000 00000000
c480 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c4a0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c4c0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c4e0 00000000 00000000 00000000 c0103c48 00000000 00000000 00000000 00000000
c500 c4f9c000 c028060c 00000000 00000000 00000000 00000000 00000000 00000000
c520 00000000 00000000 00000000 c01002a4 00000000 00000000 00000000 00000000
c540 00000000 00000000 00000000 00000000 c02b06c0 c4f9c000 c028060c 00000000
c560 c02b0000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c580 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c5a0 00000000 00000000 00000000 c02950a4 00000000 00000000 00000000 00000000
c5c0 00000001 c0284000 00000000 00000000 00000002 c027a4a0 c027a4a0 c0258bbc
c5e0 00000000 00000000 00000000 c0294fe8 00000000 00000000 00000000 00000000
c600 00000000 08000059 00000000 00000000 f00012a0 000ff000 000a5a59 000a5a59
c620 c0295794 c02d9800 00504d6c c02cc000 c0266000 c02c8000 c02aed34 c02aed24
c640 c02aed34 c02aed20 00000000 00000000 000ff000 000a5a59 000a5a59 c0295794
c660 c02d9800 00504d6c c02cc000 c029bc3c c02c8000 c02aed34 c02aed04 c02c8000
c680 c02c5fe8 c02c60a4 c028758c 00000040 c0244ed8 c0244a0c c0244b80 c0244edc
c6a0 01234567 c4f9c000 00000000 c02a6e64 c4f9c698 00000000 c02cc000 c0266000
c6c0 10000080 c02c5fe8 c02c60a4 c028758c 00000040 c4ffeea0 f00012a0 000ff000
c6e0 000a5a59 000a5a59 c0295794 c0155890 00504d6c c02cc000 c0266000 c02c8000
c700 c02c6164 00000100 c0244c38 00000000 c0245000 00000000 c02c5fe8 c02c5fe8
c720 c01e55ec c028be04 c02c9a20 c0109e74 c4ffc200 c028b474 c4f4a000 c028bf60
c740 00000004 c02aeab4 c02c8d20 c02b621f 00000004 c02b61ca 00000054 c02b621f
c760 00000000 00000001 c027a000 c02a6de4 0000003c 00000058 0000000b c027a4ac
c780 0000005a c4ff74a0 00000000 000000ff 00000060 00000000 00000060 00000002
c7a0 002b2080 00000008 002b50c0 c019324c 00000000 023c3460 c4f9c980 00000001
c7c0 08000000 00000000 00005301 f0823800 00856606 00000000 00000000 c028478c
c7e0 42780000 00000000 431c0000 00000000 4471cccc 00000000 000003c7 00000000
c800 fffffde0 fffffde0 7f7fffff ffffffff 7f7fffff ffffffff 00000000 00000000
c820 00000000 00000000 41800000 00000000 00000010 00000010 f0823800 00000000
c840 00000003 fffffde0 41000000 00000000 40800000 00000000 fffffde0 fffffde0
c860 41000000 00000000 fffffde0 fffffde0 40800000 00000000 41000000 00000000
c880 40300000 00000000 40200000 00000000 40200000 00000000 c018ffb8 c02846d8
c8a0 c02c8800 c026c000 c02c8d20 0000000b c028478c 00000000 41000000 00000000
c8c0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c8e0 00000000 00000000 c011bb78 c0192bd4 4471cccc 00000000 000003c7 00000000
c900 0000000a c028478c c4f9c7c8 00000090 00000006 2b6a0000 00000000 00000000
c920 00000000 00000000 41800000 00000000 00000010 00000010 00000000 00000000
c940 fffffde0 fffffde0 41000000 00000000 40800000 00000000 fffffde0 fffffde0
c960 41000000 00000000 fffffde0 c011e9bc 40800000 00000000 41000000 00000000
c980 0004000a c0221800 c011e9f8 c4ff6520 c4ff6200 c0244f10 00000008 f0823800
c9a0 00000003 00000007 c0191568 c02c60a4 fffffffc c0245000 c0245000 c02d72b8
c9c0 c0245000 c0245000 c02c6028 00000000 c4ff6234 f0823807 f0823800 0000000a
c9e0 ffffffff c4ff6520 c4ff6200 c0266000 ffffffff 00000340 c4f9cbc0 c012dfc0
ca00 08000000 00000000 00000000 00000000 00856606 00000000 00000000 00000000
ca20 42780000 00000000 431c0000 00000000 4471cccc 00000000 000003c7 00000000
ca40 fffffde0 fffffde0 7f7fffff ffffffff 7f7fffff ffffffff 00000000 00000000
ca60 00000000 00000000 41800000 00000000 00000010 00000010 00000000 00000000
ca80 fffffde0 fffffde0 41000000 00000000 40800000 00000000 fffffde0 fffffde0
caa0 41000000 00000000 fffffde0 fffffde0 40800000 00000000 41000000 00000000
cac0 40300000 00000000 40200000 00000000 40200000 00000000 41800000 fffffde0
cae0 40000000 00000000 40000000 00000000 40800000 00000000 41000000 00000000
cb00 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
cb20 00000000 00000000 c011e710 c011e714 00000000 00000000 00000000 00000000
cb40 00000000 00000058 c4f9c740 c02caac8 00000007 0f881093 00000000 00000003
cb60 c02c9800 c02c9a60 c02c9a20 00000000 00000000 00000400 c4f9cd40 c01d1c98
cb80 0000003c 0000003e c027a000 00000001 c02b61e0 00000004 c02b61c7 00000018
cba0 c02b61e0 00000000 431c0000 c01046e0 4471cccc 00000000 000003c7 00000000
Data access rights fault in kernel: Code=26 regs=c4f9c980 (Addr=00000003)
YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000001000000000000001010
r0-3 00000000 c0221800 c011e9f8 c4ff6520
r4-7 c4ff6200 c0244f10 00000008 f0823800
r8-11 00000003 00000007 c0191568 c02c60a4
r12-15 fffffffc c0245000 c0245000 c02d72b8
r16-19 c0245000 c0245000 c02c6028 00000000
r20-23 c4ff6234 f0823807 f0823800 0000000a
r24-27 ffffffff c4ff6520 c4ff6200 c0266000
r28-31 ffffffff 00000340 c4f9cbc0 c012dfc0
sr0-4 00000000 00000000 00000000 00000000
sr4-8 00000000 00000000 00000000 00000000
IASQ: 00000000 00000000 IAOQ: c011e710 c011e714
IIR: 0f881093 ISR: 00000000 IOR: 00000003
ORIG_R28: 00000058
From delyra@latt.if.usp.br Thu, 30 Nov 2000 09:15:11 -0200 (BRST)
Date: Thu, 30 Nov 2000 09:15:11 -0200 (BRST)
From: Jorge L. deLyra delyra@latt.if.usp.br
Subject: [parisc-linux] HP 9000/735-125
> I have a 735-125 and am able to net-boot it on a custom configured kernel as
> long as I disable the ASP parallel ports. It works quite well using an NFS
> root as I cannot get the SCSI hard drives to mkfs. I was going to try and
> debug it but have yet to find the free time. I can e-mail you my working
> .config for the kernel, or build kernels (tested on my own 735) for people
> if they need them.
Well, looks like the problem is known, good! But we have a queer little
problem with our machines here, we are unable to boot from the network. I
think we did everything right, in fact, we are trying to use a server here
other machines boot from, it has dhcp, boopt, tftp, rarp, the works. We
set it all up, put the kernel on the tftpboot area, tested all that we
could by other means but, when we try to net boot the HPs, we are faced
with complete silence on the network. No logs on the server, nothing.
A little explanation might be needed: these are not really native 735-125
models, they were upgraded from older 720-50 models by card swapping. I am
not sure whether all cards were changed, maybe some aspects of the machine
are still old. The syntax of the net boot procedure in them is strange,
you have to put in the hardware address of the server, which is unusual.
I _suspect_ that this bios does not use tcp/ip for the net boot, but some
HP proprietary protocol relating to that clustering software they have or
used to have for these machines, in which you ran several stations out of
the disks of a single one. In any case, a listing of what happens on the
console on a net-boot trial goes below. We had a tail -f on the system log
of the server, we had tcpdump listening, but nothing at all shows up...
----------------------------------------------------------------
Jorge L. deLyra, Associate Professor of Physics
The University of Sao Paulo, IFUSP-DFMA
For more information: finger delyra@latt.if.usp.br
----------------------------------------------------------------
Seyon Copyright (c) 1992-1993 Muhammad M. Saggaf. All rights reserved.
Version 2 rev. 20c i586-Linux steve@hammer 05/23/99 20:19:22.
Locating Modems...
Modem `/dev/ttyS0' is Available.
(c) Copyright. Hewlett-Packard Company. 1992.
All rights reserved.
PDC ROM rev. 2.7
IODC ROM rev. 1.1
80 MB of memory configured and tested.
Searching for Potential Boot Devices.
To terminate search, press and hold the ESCAPE key.
Device Selection Device Path Device Type
----------------------------------------------------------------------------
P0 scsi.6.0 HP C3725S
P1 scsi.4.0 TOSHIBA CD-ROM XM-3301TA
b) Boot from specified device
s) Search for bootable devices
a) Enter Boot Administration mode
x) Exit and continue boot sequence
?) Help
Select from menu: a
BOOT_ADMIN> help boot
Boot from specified device or path:
BOOT IPL boots Initial Program Loader (interactive mode)
BOOT boots default boot utility
may be one of the following:
pri primary path in Stable Storage
alt alternate path in Stable Storage
Pn Device selection from SEARCH
eisa. EISA adapter
fwscsi. On-board FASTWIDE SCSI interface
lan. Slider-card LAN interface
scsi. On-board SCSI interface
For more information on boot selection options, type HELP
where = eisa, lan, scsi, etc ...
BOOT_ADMIN> help lan
LAN (IEEE 802.3/Ethernet LAN) Path Specification
lan...
12 digit (hex) LAN server address
max number of times to try a boot request
(0 = default, 255 = infinite)
max number of times to try a read request
(0 = default, 255 = infinite)
Example: to specify LAN address 123456-78ABCD with infinite
initialization retries and default I/O retries,
lan.123456-78ABCD.255.0
If one or more parameters are not specified, the following
defaults will be used: = 000000-000000
= 3 tries
= 6 tries
BOOT_ADMIN> boot lan.0000f8-01abb1.0.0
Trying lan.0000f8-01abb1.0.0
Failed to initialize lan.0000f8-01abb1.3.0
ENTRY_INIT status = -7
00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
00000008 00000000 00000000 00000000 00000000 00000000 00000000 00000000
00000000 00000000 A000800C F0002898 00000000 0800090B DBEB0000 00000000
Searching for Potential Boot Devices.
To terminate search, press and hold the ESCAPE key.
Device Selection Device Path Device Type
----------------------------------------------------------------------------
P0 scsi.6.0 HP C3725S
P1 scsi.4.0 TOSHIBA CD-ROM XM-3301TA
b) Boot from specified device
s) Search for bootable devices
a) Enter Boot Administration mode
x) Exit and continue boot sequence
?) Help
Select from menu:
From rhirst@linuxcare.com Thu, 30 Nov 2000 11:19:30 +0000
Date: Thu, 30 Nov 2000 11:19:30 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] HP 9000/735-125
On Thu, Nov 30, 2000 at 08:58:22AM -0200, Jorge L. deLyra wrote:
> OK, here goes. I want to congratulate you all on this effort. We have five
> of these HP-9000 stations here, quite old by now. They used to be our main
> number crunching force. They are wonderfully built hardware in bad need of
> a wonderful system on them! |:-)
> Serial driver version 5.01 (2000-05-29) with MANY_PORTS SHARE_IRQ SERIAL_PCI enabled
>
> Dumping Stack from c4f9c000 to c4f9cbc0:
This has been reported on 715/50 (Robert Duncan) and 715/75 (me) round
about November 15th. At the time I tried a current cvs kernel on my
715/75 and it was even worse (IIRC). I'll have another look at it.
Richard
From rhirst@linuxcare.com Thu, 30 Nov 2000 11:27:39 +0000
Date: Thu, 30 Nov 2000 11:27:39 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] HP 9000/735-125
On Thu, Nov 30, 2000 at 01:20:28AM -0800, Steve Moret wrote:
> Now, maybe you can help me identify my SCSI problem. I don't know if it is
> because of driver issues or a bad disk (completly likely). Do other people
> have the Fast SCSI2 working on their 735s?
>
> Whenever I do a mke2fs (after partitioning the drive with fdisk ok) it dies.
I have a 53c700 on my 715/75, and the driver was having trouble
with a CRDOM I attached, so it has some problems. I wrote the
driver, but havn't used the 715/75 for a while since latest kernels
wouldn't boot for me. I'll get it going again and see if the scsi
driver still works.
Richard
From deller@gmx.de Thu, 30 Nov 2000 13:04:49 +0100 (MET)
Date: Thu, 30 Nov 2000 13:04:49 +0100 (MET)
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] HP 9000/735-125
Hi Steve,
> My mistake! I did a full build with todays CVS and parport didn't die.
> So
> somewhere between the 17th and now parport must have been fixed.
I checked in yesterday a modified version of
/drivers/parport/parport_gsc.c with an "#if 0 ... #endif" in the code.
Could you try to boot again with "#if 1" (search for the text which says
something like "enable bidirectional PS/2 mode") and try again ?
If this hangs your machine I will implement a better work-around.
>
> Now, maybe you can help me identify my SCSI problem. I don't know if it
> is
> because of driver issues or a bad disk (completly likely).
I think Richard Hirst can help you much more with your SCSI-problems than
me.
NB: Does your chassis LEDs work with the new kernel, and if not, could you
send me your bootlog (or "dmesg | grep led") ?
Greetings,
Helge
--
Sent through GMX FreeMail - http://www.gmx.net
From rbradetich@uswest.net Thu, 30 Nov 2000 07:01:53 -0800
Date: Thu, 30 Nov 2000 07:01:53 -0800
From: Ryan Bradetich rbradetich@uswest.net
Subject: [parisc-linux] HP 9000/735-125
"Jorge L. deLyra" wrote:
> > I have a 735-125 and am able to net-boot it on a custom configured kernel as
> > long as I disable the ASP parallel ports. It works quite well using an NFS
> > root as I cannot get the SCSI hard drives to mkfs. I was going to try and
> > debug it but have yet to find the free time. I can e-mail you my working
> > .config for the kernel, or build kernels (tested on my own 735) for people
> > if they need them.
>
> Well, looks like the problem is known, good! But we have a queer little
> problem with our machines here, we are unable to boot from the network. I
> think we did everything right, in fact, we are trying to use a server here
> other machines boot from, it has dhcp, boopt, tftp, rarp, the works. We
> set it all up, put the kernel on the tftpboot area, tested all that we
> could by other means but, when we try to net boot the HPs, we are faced
> with complete silence on the network. No logs on the server, nothing.
>
> A little explanation might be needed: these are not really native 735-125
> models, they were upgraded from older 720-50 models by card swapping. I am
> not sure whether all cards were changed, maybe some aspects of the machine
> are still old. The syntax of the net boot procedure in them is strange,
> you have to put in the hardware address of the server, which is unusual.
>
> I _suspect_ that this bios does not use tcp/ip for the net boot, but some
> HP proprietary protocol relating to that clustering software they have or
> used to have for these machines, in which you ran several stations out of
> the disks of a single one. In any case, a listing of what happens on the
> console on a net-boot trial goes below. We had a tail -f on the system log
> of the server, we had tcpdump listening, but nothing at all shows up...
I believe the 735/755 type machines require rbootd to boot from the network.
rbootd is available at:
ftp://puffin.external.hp.com/pub/parisc/binaries/tgz/rbootd-2.0-2.tar.gz
- Ryan
From rhirst@linuxcare.com Thu, 30 Nov 2000 14:03:10 +0000
Date: Thu, 30 Nov 2000 14:03:10 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] 715/75 dies in pdc_iodc_read()
Hi,
My 715/75 no longer boots with cvs kernel. It hangs while
searching for devices in PDC firmware, after finding the first
device. The beta 0.5 cd kernel gets past this point. The device
list is:
Searching for devices in PDC firmware... processor hpa 0xfffbe000
an older box...
Found devices:
1. Stinger Optional Graphics (10) at 0xf4000000, versions 0x6, 0x0, 0x77, 0x0, 0x0
2. Scorpio Sr. Core BA (11) at 0xf082f000, versions 0x19, 0x0, 0x70, 0x0, 0x0
3. Scorpio Sr. Core SCSI (10) at 0xf0825000, versions 0x19, 0x0, 0x71, 0x0, 0x0
4. Scorpio Sr. Core LAN (802.3) (10) at 0xf0826000, versions 0x19, 0x0, 0x72, 0x0, 0x0
5. Scorpio Sr. Core HIL (10) at 0xf0821000, versions 0x19, 0x0, 0x73, 0x0, 0x0
6. Scorpio Sr. Core RS-232 (10) at 0xf0823000, versions 0x19, 0x0, 0x75, 0x0, 0x0
7. Scorpio Sr. Core RS-232 (10) at 0xf0822000, versions 0x19, 0x0, 0x75, 0x0, 0x0
8. Scorpio Sr. Core Centronics (10) at 0xf0824000, versions 0x19, 0x0, 0x74, 0x0, 0x0
9. Scorpio Sr. Audio (10) at 0xf1000000, versions 0x19, 0x0, 0x7b, 0x0, 0x0
10. Scorpio Sr. EISA BA (11) at 0xfc000000, versions 0x19, 0x0, 0x76, 0x0, 0x0
11. Unknown device (10) at 0xfc001000, versions 0x0, 0x0, 0xfff, 0x0, 0x0
12. Scorpio Sr.(715/75) (0) at 0xfffbe000, versions 0x316, 0x0, 0x4, 0x0, 0x81
13. Scorpio Sr. (1) at 0xfffbf000, versions 0x27, 0x0, 0x9, 0x0, 0x0
That's a total of 13 devices.
CPU(s): 1 x PA7100 (PCX-T) at 75.000000 MHz
So, with lastest cvs source, it gets in to the loop in
really_do_oldhw_inventory(), and first time through pdc_mem_map_hpa()
returns r_addr.hpa=0xf4000000 (the Stinger Optional Graphics). Next
time round, mod=1, pdc_mem_map_hpa() returns r_addr.hpa=0xf8000000,
and the subsequent call to pdc_iodc_read() hangs.
I made the code skip the loop where mod=1, and it then goes on to
discover all the devices without problems.
At power on, the machine reports:
Warning: one or more EISA cards could not be configured.
Autoselect and search will ignore unconfigured cards.
Which I assume relates to an EISA SCSI card in the machine, which
I assume is item 11 in the above list.
Richard
From delyra@latt.if.usp.br Thu, 30 Nov 2000 13:04:46 -0200 (BRST)
Date: Thu, 30 Nov 2000 13:04:46 -0200 (BRST)
From: Jorge L. deLyra delyra@latt.if.usp.br
Subject: [parisc-linux] HP 9000/735-125
> I believe the 735/755 type machines require rbootd to boot from the network.
> rbootd is available at:
>
> ftp://puffin.external.hp.com/pub/parisc/binaries/tgz/rbootd-2.0-2.tar.gz
Ah! This is it, then! Had never heard of this beast before. I see it is
available for i386, nice, our server is a Pentium. Well, thanks a whole
lot for the tip, we are certainly trying this. Well, back to work...
----------------------------------------------------------------
Jorge L. deLyra, Associate Professor of Physics
The University of Sao Paulo, IFUSP-DFMA
For more information: finger delyra@latt.if.usp.br
----------------------------------------------------------------
From delyra@latt.if.usp.br Thu, 30 Nov 2000 13:49:56 -0200 (BRST)
Date: Thu, 30 Nov 2000 13:49:56 -0200 (BRST)
From: Jorge L. deLyra delyra@latt.if.usp.br
Subject: [parisc-linux] HP 9000/735-125
OK, rbootd is up and running, the station now establishes instantaneous
communication with the server. But it says "bad LIF magic" and does not
boot. I presume I can't just put the precompiled kernel in there. Since
I cannot cross-compile for the time being (bad HP server disk crash) is
there a net-boot kernel somewhere that I can download and try?
Cheers,
----------------------------------------------------------------
Jorge L. deLyra, Associate Professor of Physics
The University of Sao Paulo, IFUSP-DFMA
For more information: finger delyra@latt.if.usp.br
----------------------------------------------------------------
From randolph@tausq.org Thu, 30 Nov 2000 08:44:18 -0700
Date: Thu, 30 Nov 2000 08:44:18 -0700
From: Randolph Chung randolph@tausq.org
Subject: [parisc-linux] CVS linux Vs. -test10
> Oooh, there's a reason for me to finally get the 712/80 under my desk
> to be more than a foot-rest. I'll see what I can do about this.
>
> Note that the likelihood of Debian releasing woody anytime soon is
> vanishingly small, so this dosen't have to happen Right Now.
So little faith... :P
Let me know what I can do to help.... it takes my dual-400MHz i386 box
about an hour to build boot-floopies; i don't even wait to think how
long it will take on the 712/80 :-)
randolph
--
@..@ http://www.TauSq.org/
(----)
( >__< )
^^ ~~ ^^
From rhirst@linuxcare.com Thu, 30 Nov 2000 16:11:15 +0000
Date: Thu, 30 Nov 2000 16:11:15 +0000
From: Richard Hirst rhirst@linuxcare.com
Subject: [parisc-linux] HP 9000/735-125
On Thu, Nov 30, 2000 at 11:27:39AM +0000, Richard Hirst wrote:
> On Thu, Nov 30, 2000 at 01:20:28AM -0800, Steve Moret wrote:
> > Now, maybe you can help me identify my SCSI problem. I don't know if it is
> > because of driver issues or a bad disk (completly likely). Do other people
> > have the Fast SCSI2 working on their 735s?
> >
> > Whenever I do a mke2fs (after partitioning the drive with fdisk ok) it dies.
>
>
> I have a 53c700 on my 715/75, and the driver was having trouble
> with a CRDOM I attached, so it has some problems. I wrote the
> driver, but havn't used the 715/75 for a while since latest kernels
> wouldn't boot for me. I'll get it going again and see if the scsi
> driver still works.
I got my 715/75 to boot. The scsi driver still has problems with my
CDROM drive, but the disk seems ok. I ran mke2fs of a 1Gig partition
a few times and then copied 200MB of files from the network to it.
Richard
From bame@noam.fc.hp.com Thu, 30 Nov 2000 09:18:51 -0700
Date: Thu, 30 Nov 2000 09:18:51 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] CVS linux Vs. -test10
= Grant> need "boot floppies" debian package working. We don't need
=
= Oooh, there's a reason for me to finally get the 712/80 under my desk
= to be more than a foot-rest. I'll see what I can do about this.
=
= Note that the likelihood of Debian releasing woody anytime soon is
= vanishingly small, so this dosen't have to happen Right Now.
Well yes and no. We want to release parisc-linux sooner than woody
will be ready for public stable release, so we'll want to do the
boot floppies work sooner than other architectures need it and
can probably therefore provide early testing too. Since
we aren't going to time travel back to Debian potato, woody
boot floppies are increasingly interesting to replace our manual
install process (hey at least it's documented!).
-P
From alan@lxorguk.ukuu.org.uk Thu, 30 Nov 2000 16:10:46 +0000 (GMT)
Date: Thu, 30 Nov 2000 16:10:46 +0000 (GMT)
From: Alan Cox alan@lxorguk.ukuu.org.uk
Subject: [parisc-linux] XFree status
> Alan Cox wrote:
> > I have a server linked.
> Alan - that's Cool! Wow!
Its still got its atoms in a twist, so its bombing out in Xnest loading the
first font.
> > inb/inw/outb/outw and friends are right now null
> > functions until I fill them in. Thats not a big deal. Initially I'll probably
> > use /dev/port but for speed I hope everyone uses mmio based hardware.
>
> All HP graphics for PARISC are memory mapped. Both "GSC" and PCI flavors.
Yes, but if I want to say put an S3 Trio64 in my A180 and a USB card for
keyboard mouse..,, (and yes these are sitting on my desk)
> But I'm pretty clueless how linux X server finds/talks to a frame buffer.
> For HPUX, the graphics driver supports some ioctl()'s.
> Any references to describe how it works for linux?
The OS specific X code in XFree86 knows several ways to talk to Linux
Memory:
1. Directly mmap()ing /dev/mem or /dev/kmem to get access to the
mmio space of the card and frame buffer memory.
2. Mapping the pci space via a kernel frame buffer device (/dev/fb*)
3. Arbitary other mmap based code that you plug into it (harder to do)
I/O
1. Use of iopl/ioperm on x86
2. Use of mmap to map the PCI I/O space regions on different
platforms (which doesnt work on some PA kit)
3. Arbitary other code we plug into it
For I/O it seems that on things like the A180 the only way to do it is to use
/dev/port and pread/pwrite the file handle for each port I/O. This is slow but
hopefully primarily used for booting the card (XFree 4.0 has a X86 emulator
for booting the BIOS firmware on PCI cards)
For most machines I imagine we would be using mmio and the /dev/fb interface.
Alan
From alan@lxorguk.ukuu.org.uk Thu, 30 Nov 2000 16:16:18 +0000 (GMT)
Date: Thu, 30 Nov 2000 16:16:18 +0000 (GMT)
From: Alan Cox alan@lxorguk.ukuu.org.uk
Subject: [parisc-linux] CVS linux Vs. -test10
> Oooh, there's a reason for me to finally get the 712/80 under my desk
> to be more than a foot-rest. I'll see what I can do about this.
>
> Note that the likelihood of Debian releasing woody anytime soon is
> vanishingly small, so this dosen't have to happen Right Now.
My experiences with building Red Hat 7 so far are mostly good. I don't think
there will be many actual changes needed to build Debian packages either. I've
made very little that isnt 'use -fPIC' 'dont optimise C++'
From marteaut@esiee.fr Thu, 16 Nov 2000 21:16:22 +0100
Date: Thu, 16 Nov 2000 21:16:22 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] Troubles with read only file system
Hi all,
We have done everything well but the NFSroot and the bootable disk say that
we have a read only file system ???
What is the recipe for making a bootable disk because ours can not be the
good one...
Thanks,
Thomas
From marteaut@esiee.fr Fri, 17 Nov 2000 17:59:14 +0100
Date: Fri, 17 Nov 2000 17:59:14 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] Trouble with new STI
This is a multi-part message in MIME format.
------=_NextPart_000_000F_01C050C0.18E3D060
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi all,
We've tried the new STI driver and our 712 dies with this message:
Kernel Panic: VFS: Unable to mount root fs on 01:00
We tried with Ramdisk, nfs and hard disk support and always the same =
end.
Also, the death happens when you 've passed bootp request...
Help!
Thomas
------=_NextPart_000_000F_01C050C0.18E3D060
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi all,
We've tried the new STI driver and our =
712 dies=20
with this message:
Kernel Panic: VFS: Unable to mount root =
fs on=20
01:00
We tried with Ramdisk, nfs and hard =
disk support=20
and always the same end.
Also, the death happens when you 've =
passed bootp=20
request...
Help!
Thomas
------=_NextPart_000_000F_01C050C0.18E3D060--
From marteaut@esiee.fr Sun, 19 Nov 2000 13:08:45 +0100
Date: Sun, 19 Nov 2000 13:08:45 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] Booting 712 STI and nfs
This is a multi-part message in MIME format.
------=_NextPart_000_0047_01C05229.D8DA5D20
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi all,
We managed to boot on the STI-console with a 712/80 in changing the =
parameters in inittab and in securetty (if you want to login as root).=20
> cat inittab (! just the interesting thing!)
# /sbin/getty invocations for the runlevels.
#
# The "id" field MUST be the same as the last
# characters of the device (after "tty").
#
# Format:
# :::
1:2345:respawn:/sbin/getty 38400 tty1
#2:23:respawn:/sbin/getty 38400 tty2
#3:23:respawn:/sbin/getty 38400 tty3
#4:23:respawn:/sbin/getty 38400 tty4
#5:23:respawn:/sbin/getty 38400 tty5
#6:23:respawn:/sbin/getty 38400 tty6
=20
> cat securetty
# /etc/securetty: list of terminals on which root is allowed to login.
# See securetty(5) and login(1).
ttyS0
tty1
Also, in order to boot with the STI console, you need to have the module =
for STI-console and not for serial console support...
But, we have troubles with the rpc that print somestimes
nfs: server X.X.X.165 not responding, still trying
nfs: server X.X.X.165 OK
nfs: server X.X.X.165 not responding, still trying
nfs: server X.X.X.165 OK
nfs: server X.X.X.165 not responding, still trying
nfs: server X.X.X.165 OK
We don't understand where comes from the problem!!
Bye,
ESIEE Port Team
------=_NextPart_000_0047_01C05229.D8DA5D20
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi all,
We managed to boot on the STI-console =
with a 712/80=20
in changing the parameters in inittab and in securetty (if you want to =
login as=20
root).
> cat inittab (! just the =
interesting=20
thing!)
# /sbin/getty invocations for the=20
runlevels.
#
# The "id" field MUST be the same as the last
# =
characters=20
of the device (after "tty").
#
# Format:
# =20
<id>:<runlevels>:<action>:<process>
1:2345:res=
pawn:/sbin/getty=20
38400 tty1
#2:23:respawn:/sbin/getty 38400 =
tty2
#3:23:respawn:/sbin/getty=20
38400 tty3
#4:23:respawn:/sbin/getty 38400 =
tty4
#5:23:respawn:/sbin/getty=20
38400 tty5
#6:23:respawn:/sbin/getty 38400 tty6
> cat securetty
# /etc/securetty: list of terminals on =
which root=20
is allowed to login.
# See securetty(5) and=20
login(1).
ttyS0
tty1
Also, in order to boot with the STI =
console, you=20
need to have the module for STI-console and not for serial console=20
support...
But, we have troubles with the rpc that =
print=20
somestimes
nfs: server X.X.X.165 not responding, =
still=20
trying
nfs: server X.X.X.165 OK
nfs: server X.X.X.165 not responding, =
still=20
trying
nfs: server X.X.X.165 OK
nfs: server X.X.X.165 not responding, =
still=20
trying
nfs: server X.X.X.165 OK
We don't understand where comes from the problem!!
Bye,
ESIEE Port Team
------=_NextPart_000_0047_01C05229.D8DA5D20--
From marteaut@esiee.fr Mon, 20 Nov 2000 15:32:55 +0100
Date: Mon, 20 Nov 2000 15:32:55 +0100
From: Thomas Marteau marteaut@esiee.fr
Subject: [parisc-linux] A new file system
This is a multi-part message in MIME format.
------=_NextPart_000_0009_01C05307.27222700
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi all,
Since we have the STI, we have produced a file system quite =
interesting for the 712. You can
have a network support till you configure the file.conf like resolv... =
Also, we have noticed that
the sti on B180 does not work well, it look like it can not find the =
rom. But we have to work on...
To download the archives go there:
http://www.esiee.fr/~djoudim/code/suitable.html
It is incredible what can do a 712 with it!
Bye,
ESIEE Port Team
------=_NextPart_000_0009_01C05307.27222700
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi all,
Since we have the =
STI, we have=20
produced a file system quite interesting for the 712. You =
can
have a network support till you =
configure the=20
file.conf like resolv... Also, we have noticed that
the sti on B180 does not work well, it =
look like it=20
can not find the rom. But we have to work on...
To download the archives go =
there:
It is incredible what can do a 712 with =
it!
Bye,
------=_NextPart_000_0009_01C05307.27222700--
From kailasr@webcash.com Tue, 31 Oct 2000 12:58:41 -0800
Date: Tue, 31 Oct 2000 12:58:41 -0800
From: Kailashnath V Rampure kailasr@webcash.com
Subject: [parisc-linux] unable to run palo on nfs root
Hi Paul,
After booting the Server from NFSroot I need to Initialize the harddisk on
the server. so I copied the palo and linux in to the nfsroot. I am trying
to Run the following command:
$ palo -I -k /boot/vmlinux -b /boot/iplboot -c '2/boot/vmlinux TERM=linux
HOME=/ root=/dev/sda2' /dev/sda
fisrt partition is 10M of type f0
second partition is /dev/sda2 of type linux native.
Third partition is /dev/sda3 of type linux Swap
Then I get the cannot execute binary file.
second question is that what is the Terminal type I should set as I am
unable to open vi.
Please suggest the correct steps.
Thanks.
Kailas
At 09:54 AM 10/31/00 -0700, Paul Bame wrote:
>= Hi,
>=
>= I have booted HP A class server from nfsroot. but when I try to run palo on
>= the server to initialize boot loader to the hdd I get
>= "cannot execute the binary file"
>=
>= can any one help me to make the hdd bootable
>
>It would help me to see the actual command you used and the actual
>error message(s) printed. With the info you give so far it could be
>"a million" things.
>
> -P
From bame@noam.fc.hp.com Tue, 31 Oct 2000 14:26:00 -0700
Date: Tue, 31 Oct 2000 14:26:00 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] NEW NATIVE PALO TARBALL - REQUIRED
= Hi Paul,
=
= After booting the Server from NFSroot I need to Initialize the harddisk on
= the server. so I copied the palo and linux in to the nfsroot. I am trying
= to Run the following command:
= $ palo -I -k /boot/vmlinux -b /boot/iplboot -c '2/boot/vmlinux TERM=linux
= HOME=/ root=/dev/sda2' /dev/sda
= fisrt partition is 10M of type f0
= second partition is /dev/sda2 of type linux native.
= Third partition is /dev/sda3 of type linux Swap
That looks great.
= Then I get the cannot execute binary file.
Ok I think we changed the kernal such that the old palo bits won't run
any more (we removed the old gateway page). I uploaded a new version:
ftp://puffin.external.hp.com/pub/parisc/binaries/userspace/palo-200001031.tgz
-P
From grundler@cup.hp.com Tue, 31 Oct 2000 13:53:04 -0800
Date: Tue, 31 Oct 2000 13:53:04 -0800
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] unable to run palo on nfs root
Kailashnath V Rampure wrote:
> fisrt partition is 10M of type f0
> second partition is /dev/sda2 of type linux native.
> Third partition is /dev/sda3 of type linux Swap
Kailashnath,
Even though your partitioning should work fine, you really want the
swap on the lowest numbered SCSI block you can get away with.
Several reasons for this:
o lowest block number is on the outside of the SCSI disk were data
xfer rate is typically 80% faster than the inside.
(someday try "time dd if=/dev/sda of=/dev/null bs=8k count=50"
with and with out skip parameter).
o Eventualy we will have kernel dumps to swap space - and IODC
will likely have the same limitations where on the disk
dump can be as we have with booting from a disk (as discussed in
the PALO documentation).
grant
Grant Grundler
Unix Systems Enablement Lab
+1.408.447.7253
From taggart@carmen.fc.hp.com Tue, 31 Oct 2000 16:16:09 -0700
Date: Tue, 31 Oct 2000 16:16:09 -0700
From: Matt Taggart taggart@carmen.fc.hp.com
Subject: [parisc-linux] New cross-compiler
I have placed a new i386-linux -> hppa32/64-linux cross compiler (with 32bit
glibc) in,
ftp://puffin.external.hp.com/pub/parisc/binaries/LinuxX86/xc-20001031.tar.gz
It includes Alan Modra's recent fix for the 64 bit compiler,
http://puffin.external.hp.com/mailing-lists/parisc-linux-cvs/2000/10-Oct/0233.h
tml
--
Matt Taggart
taggart@fc.hp.com
From pdbeal@bealz.net Tue, 31 Oct 2000 18:43:06 -0500
Date: Tue, 31 Oct 2000 18:43:06 -0500
From: Phillip Beal pdbeal@bealz.net
Subject: [parisc-linux] 735 and Thin Lan
Hey,
I've obtained two HP 735's and I'd like to try the linus port on them.
I have compiled a kernel, but it never boots the nfsroot disk. It has
the same problem that the HP 755 that I've worked with. However, it has
a different ethernet connection. The 735 has a thin lan connection.
What network device should be used in the kernel to be able to use the
thin lan adapter? Here's the info on the Thin lan:
Outfield Core LAN (802.3) (10) at 0xf0826000 versions 0x9, 0x0, 0x72,
0x0, 0x0
Thanks,
--
Phillip Beal
S+LUG Vice-President
From deller@gmx.de Wed, 1 Nov 2000 00:57:22 +0100
Date: Wed, 1 Nov 2000 00:57:22 +0100
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] 735 and Thin Lan
On Wednesday 01 November 2000 00:43, Phillip Beal wrote:
> Hey,
>
> I've obtained two HP 735's and I'd like to try the linus port on them.
> I have compiled a kernel, but it never boots the nfsroot disk. It has
> the same problem that the HP 755 that I've worked with. However, it has
> a different ethernet connection. The 735 has a thin lan connection.
> What network device should be used in the kernel to be able to use the
> thin lan adapter? Here's the info on the Thin lan:
>
> Outfield Core LAN (802.3) (10) at 0xf0826000 versions 0x9, 0x0, 0x72,
> 0x0, 0x0
>
> Thanks,
> Phillip Beal
> S+LUG Vice-President
Hi Phillip,
I think the "Lasi ethernet"-driver (enabled by default in our configuration)
should work on both of your boxes.
Helge.
From deller@gmx.de Wed, 1 Nov 2000 01:45:52 +0100
Date: Wed, 1 Nov 2000 01:45:52 +0100
From: Helge Deller deller@gmx.de
Subject: [parisc-linux] The new PS/2 Keyboard Driver
On Friday 27 October 2000 01:50, Helge Deller wrote:
> On Thursday 26 October 2000 22:05, Thomas Marteau wrote:
>
> > >
> > Hello everyone,
> >
> > We've just updated the PS/2 keyboard driver. The leds and interrupt
> > functions work really well on a 712 workstation and also B132 now. The
> > updated driver files are available on our website. It works better than
> > under HP UX for the B 132 ;->
> >
> > http://www.esiee.fr/~djoudim
> >
> > The ESIEE Port Team in Paris.
> >
> > Here is the patch:
> >
> > diff -urN linux/drivers/char/gsc_ps2.c linux-parisc/drivers/char/gsc_ps2.c
> > --- linux/drivers/char/gsc_ps2.c Thu Oct 26 21:06:54 2000
> > +++ linux-parisc/drivers/char/gsc_ps2.c Thu Oct 26 21:34:00 2000
> > @@ -7,6 +7,11 @@
> > [.............]
>
> > diff -urN linux/drivers/char/keyb_at.c linux-parisc/drivers/char/keyb_at.c
> > --- linux/drivers/char/keyb_at.c Thu Oct 26 21:07:00 2000
> > +++ linux-parisc/drivers/char/keyb_at.c Thu Oct 26 21:23:16 2000
> > [........]
>
> Hi Thomas,
>
> Thanks for your patch.
> But I don't think it's a good idea to change a common file like keyb_at.c,
> which is used in most other arches too. This patch surely breaks their
> keyboard support and more than that I'm sure, that Linus will not accept
this
> patch, when the time is come to integrate parisc into the official kernel.
>
> Isn't there any other solution as for example to #ifdef the code or create
a
> new keyb_at.c for parisc (Yes I know, both of those aren't clean too.) ?
>
> Helge Deller
Hi folks,
I need to correct myself on this topic. The ESIEE-team made a great patch and
didn't changed any globally used file. keyb_at.c is just used in the current
parisc-port, and so it's ok to change that file.
I just committed their changes to the CVS, and in the same cycle tried to
clean up the code. In the same step I renamed the original filenames to some
hopefully better ones.
Since I don't own myself a real HP PS/2 keyboard (it's just an PC-AT one with
a small DIN to PS/2-connector), it would be great to get some feedback if I
did the Right Thing.
Helge.
From bri@mojo.calyx.net Wed, 1 Nov 2000 02:48:02 -0500 (EST)
Date: Wed, 1 Nov 2000 02:48:02 -0500 (EST)
From: Brian S. Julin bri@mojo.calyx.net
Subject: [parisc-linux] The new PS/2 Keyboard Driver
Probably best not to worry about cleaning keyboard drivers up too much.
The current USB code will be followed by linux-input style drivers at
some point. In fact I started a HIL linux-input style driver, which
would abstract the PS2 port/HIL ports and allow a standard keyboard module
to be hooked into the abstracted serio port. I will work on it more when
time permits.
--
Brian S. Julin
From jsm@udlkern.fc.hp.com Wed, 1 Nov 2000 02:37:58 -0700 (MST)
Date: Wed, 1 Nov 2000 02:37:58 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: Elf Header change proposal
Paul and Alan pointed out that I goofed. I analyzed the vmlinux in my 64
bit build area to get the current flags, but it turns out that the vmlinux
I looked at was a 32 bit version. I had recently made some 64 bit checkins,
and I like to rebuild everything for 32 bit before doing that, to make
sure that I haven't broken the 32 bit path by making 64 bit changes.
So, the bad news is that I wasted time writing an unecessarily long
design proposal. The good news is that the only thing that needs to
change is value in the e_ident[EI_OSABI] field, so that we can
differentiate 64 bit HP-UX binaries from 64 bit Linux binaries.
We should also change the field for 32 bit Linux also.
John
From jsm@udlkern.fc.hp.com Wed, 1 Nov 2000 03:35:49 -0700 (MST)
Date: Wed, 1 Nov 2000 03:35:49 -0700 (MST)
From: John Marvin jsm@udlkern.fc.hp.com
Subject: space registers
Matthew Wilcox wrote:
> a small optimisation just occurred to me.
>
> right now, when we switch to kernel space, we set all of sr4-sr7 to 0
> (for the kernel mapping). we don't need to do that since the kernel is
> entirely in sr7's domain. this has the added bonus that badly written
> drivers which blindly dereference userspace pointers will work on
> parisc as well as x86. we can also lazily restore sr4-6 on exit from
> kernel space if we're switching back to the same task which called us.
> this optimisation may not be worthwhile, but i think setting sr4-6
> on entry to the kernel is unnecessary.
True for now. But it won't always be true. It is a desired goal to be
able to support large (~3.5 Gb) physical memory for the 32 bit port. To
do this we will move the kernel down to around virtual address 0, so the
kernel address space will then be controlled by sr4, and depending on the
amount of physical memory in the machine, possibly sr5, sr6, and sr7. We
could do this based on a test of the amount of physical memory in the
machine, but once you load something from memory to do the test, and then
actually do the test, you wouldn't have any advantage over just writing
zero's to the space registers.
This might be worth considering for the 64 bit port, since both the kernel
and user space will reside entirely within the linear address space
covered by sr4 (We would have to go to a 1 Mb page size to be able to
support greater than a 62 bit address space with three level page tables).
sr5,sr6, and sr7 will only be used when we are running 64 bit HP-UX
processes (with its address space broken up into 4 segments). Note that
since we just store the users space in sr3 while we are in the kernel, its
not clear that any kind of test with a branch would be a performance gain,
especially if it required that we load something from memory to do that
test. We may be able cover this by managing sr5, sr6 and sr7 only in the
HP-UX specific parts of the syscall path.
John
From alan@linuxcare.com.au Thu, 2 Nov 2000 00:11:11 +1100 (EST)
Date: Thu, 2 Nov 2000 00:11:11 +1100 (EST)
From: Alan Modra alan@linuxcare.com.au
Subject: Elf Header change proposal
On Wed, 1 Nov 2000, John Marvin wrote:
> So, the bad news is that I wasted time writing an unecessarily long
> design proposal. The good news is that the only thing that needs to
> change is value in the e_ident[EI_OSABI] field, so that we can
> differentiate 64 bit HP-UX binaries from 64 bit Linux binaries.
> We should also change the field for 32 bit Linux also.
Some other bad news is that the change isn't quite trivial, due to not
wishing to break other bfd targets. The good new is I've made the
change. Compiling at the moment.
Alan Modra
--
Linuxcare. Support for the Revolution.
From bame@noam.fc.hp.com Wed, 01 Nov 2000 10:59:34 -0700
Date: Wed, 01 Nov 2000 10:59:34 -0700
From: Paul Bame bame@noam.fc.hp.com
Subject: [parisc-linux] new method for 64-bit parisc tree
I want to propose/discuss a new method for maintaining our 64-bit parisc
tree in relation to the 32-bit tree. I have prototyped this and so
far it seems pretty useful.
Most of the files in the current parisc64 tree only contain one
line, a #include of the same file from the parisc tree. This confuses
'make dep', causes some compile errors to have nonsense line numbers,
and doesn't allow direct editing of the source files in the parisc64 tree.
The method I'm proposing works like this:
The future parisc64 tree ONLY contains files which are different from,
or in addition to, those in the parisc tree. When you 'make config'
or 'make oldconfig', each file in the parsic tree is symbolically
linked as the same file in the parisc64 tree. This enables all
the rest of the tools/build to work normally. 'make distclean' includes
a step to remove all the symlinks.
The ugliest "feature" is that even though you can edit source files
in the parisc64 tree, 'cvs commit' will fail on those which are
symbolic links. To reduce this problem, I'm dropping a symbolic link
called '...' in each parisc64 directory which is a pointer to the
corresponding parisc directory, so 'cd ...; cvs commit foo.c' will
work and not be too onerous.
We should additionally consider a naming convention or something so
that maintainers in the parisc tree know whether files are shared with
parisc64 or not.
I prototyped this as a fictional new "architecture" called "p64". To
try it out, grab the tarball (only about 30 files -- can be fewer)
ftp://puffin.external.hp.com/pub/parisc/
and unpack in your top-level linux source tree directory. Then in
your top-level Makefile, change ARCH := parisc64 to ARCH := p64, then
make oldconfig or whatever you usually do. Let me know of any problems.
Is this something we should adopt for the real parisc64 tree?
-Paul Bame
From kailasr@webcash.com Wed, 01 Nov 2000 12:30:40 -0800
Date: Wed, 01 Nov 2000 12:30:40 -0800
From: Kailashnath V Rampure kailasr@webcash.com
Subject: [parisc-linux] NEW NATIVE PALO TARBALL - REQUIRED
Thanks paul I was able to run the command.
Can you let me know what I should type at ISL prompt as its trying to boot
from /stand/vmunix instead of /boot/vmlinux. Also I am unable to open vi as
it says vi:LINUX:unknown terminal type.
Please suggest.
Regards
At 02:26 PM 10/31/00 -0700, Paul Bame wrote:
>= Hi Paul,
>=
>= After booting the Server from NFSroot I need to Initialize the harddisk on
>= the server. so I copied the palo and linux in to the nfsroot. I am trying
>= to Run the following command:
>= $ palo -I -k /boot/vmlinux -b /boot/iplboot -c '2/boot/vmlinux TERM=linux
>= HOME=/ root=/dev/sda2' /dev/sda
>= fisrt partition is 10M of type f0
>= second partition is /dev/sda2 of type linux native.
>= Third partition is /dev/sda3 of type linux Swap
>
>That looks great.
>
>= Then I get the cannot execute binary file.
>
>Ok I think we changed the kernal such that the old palo bits won't run
>any more (we removed the old gateway page). I uploaded a new version:
>ftp://puffin.external.hp.com/pub/parisc/binaries/userspace/palo-200001031.tgz
>
> -P
From matthew@wil.cx Thu, 2 Nov 2000 00:13:06 +0000
Date: Thu, 2 Nov 2000 00:13:06 +0000
From: Matthew Wilcox matthew@wil.cx
Subject: linux bame
On Wed, Nov 01, 2000 at 01:53:24PM -0700, Paul Bame wrote:
> CVSROOT: /home/cvs/parisc
> Module name: linux
> Changes by: bame 00/11/01 13:53:24
>
> Modified files:
> include/asm-parisc64: posix_types.h
>
> Log message:
> Don't need a separate copy of this one either
err.. are you sure? we used to get a lot of prototype problems when they
were the same file. what's changed that they're now able to be the same?
--
Revolutions do not require corporate support.
From bame@riverrock.org Wed, 01 Nov 2000 23:18:43 -0700
Date: Wed, 01 Nov 2000 23:18:43 -0700
From: bame@riverrock.org bame@riverrock.org
Subject: linux bame
= On Wed, Nov 01, 2000 at 01:53:24PM -0700, Paul Bame wrote:
= > CVSROOT: /home/cvs/parisc
= > Module name: linux
= > Changes by: bame 00/11/01 13:53:24
= >
= > Modified files:
= > include/asm-parisc64: posix_types.h
= >
= > Log message:
= > Don't need a separate copy of this one either
=
= err.. are you sure? we used to get a lot of prototype problems when they
= were the same file. what's changed that they're now able to be the same?
I changed the parisc version so that the data types would compile to
the same size in both wide and narrow mode. Unfortunately there is
at least one issue which will probably require this scheme to change :-(
-P
From grundler@cup.hp.com Thu, 2 Nov 2000 00:21:36 -0800 (PST)
Date: Thu, 2 Nov 2000 00:21:36 -0800 (PST)
From: Grant Grundler grundler@cup.hp.com
Subject: [parisc-linux] a500.out16
Hi Richard (et al),
I finally think I understand how pcibios_align_resource() is used...
that definitely was the problem. Everything on A500 but PCI-PCI bridge
seems to be assigned I/O port and MMIO addresses correctly.
I'll look at tulip code tomorrow to see why it's not happy.
6 Tulips are behind PCI-PCI Bridges and that's part of
the problem. But the complaints about "MMIO resource"
list I/O Port addresses instead...and those look fine to me
if they are treated like I/O port addresses...
I'd also like to connect some more SCSI disks....but any clue
what the "CACHE TEST FAILED" is about?
But instead of putting out another tar ball, I'm committing my code.
I haven't tested on c3k/j5k yet and it might be broken.
32-bit still builds and any fix should be simple - either
cvs update back to an older date or put "if (pdc_pat) { }"
around anything that I added that shouldn't be.
Tell me to test/fix any brokeness you might find and I'll
make time to do it.
aplogies for the long delay,
grant
Firmware Version 40.32
Duplex Console IO Dependent Code (IODC) revision 1
------------------------------------------------------------------------------
(c) Copyright 1995-1998, Hewlett-Packard Company, All rights reserved
------------------------------------------------------------------------------
Processor Speed State CoProcessor State Cache Size
Number State Inst Data
--------- -------- --------------------- ----------------- ------------
0 440 MHz Active Functional 512 KB 1 MB
1 440 MHz Idle Functional 512 KB 1 MB
Central Bus Speed (in MHz) : 111
Available Memory : 262144 KB
Good Memory Required : 11468 KB
Primary boot path: 0/0/1/1.15
Alternate boot path: 0/0/2/1.15
Console path: 0/0/4/0.0
Keyboard path: 0/0/4/0.0
WARNING: The non-destructive test bit was set, so memory was not tested
destructively. Information only, no action required.
---- Main Menu ---------------------------------------------------------------
Command Description
------- -----------
BOot [PRI|ALT|] Boot from specified path
PAth [PRI|ALT] [] Display or modify a path
SEArch [DIsplay|IPL] [] Search for boot devices
COnfiguration menu Displays or sets boot values
INformation menu Displays hardware information
SERvice menu Displays service commands
DIsplay Redisplay the current menu
HElp [