Re: df -l broken
In my dual Pentium3/1GHz box: % uname -a FreeBSD mako.kobe1995.net 5.0-CURRENT-20010830-JPSNAP FreeBSD 5.0-CURRENT-20010830-JPSNAP #9: Sat Nov 3 17:05:25 JST 2001 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/KOBE5SMP i386 % df -l Filesystem 1K-blocks UsedAvail Capacity Mounted on /dev/da0s1a127023572155964749%/ devfs 110 100%/dev /dev/da0s1e 3935347 112806 3507714 3%/usr /dev/da0s2a127023346898217330%/altroot procfs 440 100%/proc % In article [EMAIL PROTECTED] [EMAIL PROTECTED] writes: If my patch is exact, then the bug should manifest itself only if there are no network filesystems mounted. Do you have any network fs mounted on your box ? % df Filesystem1K-blocks UsedAvail Capacity Mounted on /dev/da0s1a 127023572155964749%/ devfs 110 100%/dev /dev/da0s1e 3935347 112806 3507714 3%/usr /dev/da0s2a 127023346898217330%/altroot procfs440 100%/proc ace:/home 2318686 1079119 105407351%/home ace:/var/mail 63567 6558417 0%/var/mail safa:/mnt 113956615 1048400860 100%/mnt safa:/usr/local 1016047 9605325551595%/usr/local safa:/usr/X11R6 1016047 9605325551595%/usr/X11R6 safa:/mnt/ftp/pub/obj 113956615 1048400860 100%/usr/obj -- mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE http://kobe1995.net/~kaz/index-e.html - Break the hate chain. No more kill! To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: adding athlon xp to bsd.cpu.mk
In article [EMAIL PROTECTED]
[EMAIL PROTECTED] writes:
how about the following patch (untested) regarding the newer athlon xp
processor type ? if needed, I could submit a PR.
AthlonXP is same as AthlonMP. It's better of the name palomino,
isn't it? (But morgan Duron will have same feature such as SSE.)
What is the best name of the AMD's CPU core that can use SSE?
Index: bsd.cpu.mk
===
RCS file: /home/ncvs/src/share/mk/bsd.cpu.mk,v
retrieving revision 1.2.2.4
diff -u -r1.2.2.4 bsd.cpu.mk
@@ -28,8 +30,10 @@
.if !defined(NO_CPU_CFLAGS) || !defined(NO_CPU_COPTFLAGS)
. if ${MACHINE_ARCH} == i386
-. if ${CPUTYPE} == k7
+. if ${CPUTYPE} == xp
_CPUCFLAGS = -march=k6# gcc doesn't support athlon yet, but it will
+. elif ${CPUTYPE} == k7
+_CPUCFLAGS = -march=k6# gcc doesn't support athlon yet, but it will
. elif ${CPUTYPE} == k6-2
_CPUCFLAGS = -march=k6
. elif ${CPUTYPE} == k6
My dual AthlonXP is offen better performance at -mcpu=i686 than -mk6.
(gcc 2.95.3)
My dual AthlonXP's WWW page(in Japanese,sorry):
http://kobe1995.net/~kaz/FreeBSD/AthlonDual.html
Thank you sorry for my poor English.
--
mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE
http://kobe1995.net/~kaz/index-e.html
- Break the hate chain. No more kill!
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
panic: when received rup
used /etc/nsswitch.conf will be used instead hostname domain. ed0: flags=8843UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST mtu 1500 inet 192.168.0.9 netmask 0xff00 broadcast 192.168.0.255 inet6 fe80::260:67ff:fe48:61f2%ed0 prefixlen 64 tentative scopeid 0x1 ether 00:60:67:48:61:f2 lo0: flags=8049UP,LOOPBACK,RUNNING,MULTICAST mtu 16384 inet6 ::1 prefixlen 128 inet6 fe80::1%ed0 prefixlen 64 scopeid 0x1 inet 127.0.0.1 netmask 0xff00 add net default: gateway 192.168.0.1 Additional routing options: TCP keepalive=YES. Routing daemons:. Mounting NFS file systems:. Additional daemons: syslogd. Doing additional network setup: ntpdate ntpd rpcbind ypbind. Starting final network daemons: mountd nfsd NFS access cache time=2 rwhod. ELF ldconfig path: /usr/lib /usr/lib/compat /usr/X11R6/lib /usr/local/lib a.out ldconfig path: /usr/lib/aout /usr/lib/compat/aout /usr/X11R6/lib/aout Starting standard daemons: inetd cron sshd usbdOct 13 17:48:18 pilo sshd[307]: fatal: rsa_generate_private_key: key generation failed. . Initial rc.i386 initialization:. Configuring syscons: keymap blanktime. Additional ABI support:. Local package initialization:. Additional TCP options:. Starting background filesystem checks Sat Oct 13 17:48:19 JST 2001 (snip) % rup (from both local or remote host causes this problem) Fatal trap 12: page fault while in kernel mode cpuid = 0; lapic.id = 0100 fault virtual address = 0x5c fault code = supervisor read, page not present instruction pointer = 0x8:0xc0259272 stack pointer = 0x10:0xccb4faf0 frame pointer = 0x10:0xccb4fbc4 code segment= base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags= interrupt enabled, resume, IOPL = 0 current process = 385 (rpc.rstatd) kernel: type 12 trap, code=0 Stopped at sysctl_ifdata+0x3a: movl0x5c(%eax),%ebx db db Context switches not allowed in the debugger. db trace sysctl_ifdata(c03ab600,ccb4fcc8,2,ccb4fc18,0) at sysctl_ifdata+0x3a sysctl_root(0,ccb4fcb8,6,ccb4fc18,0) at sysctl_root+0x113 userland_sysctl(ccb1c004,ccb4fcb8,6,bfbff46c,bfbff428) at userland_sysctl+0x12d __sysctl(ccb1c004,ccb4fd20,28127c28,0,bfbff438) at __sysctl+0x7e syscall(2f,2f,2f,bfbff438,0) at syscall+0x209 syscall_with_err_pushed() at syscall_with_err_pushed+0x1b --- syscall (202, FreeBSD ELF, __sysctl), eip = 0x280c3f2c, esp = 0xbfbff37c, ebp = 0xbfbff3b8 --- db panic panic: from debugger cpuid = 0; lapic.id = 0100 Debugger(panic) Stopped at sysctl_ifdata+0x3a: movl0x5c(%eax),%ebx db panic panic: from debugger cpuid = 0; lapic.id = 0100 boot() called on cpu#0 Uptime: 3m50s Automatic reboot in 15 seconds - press a key on the console to abort -- Press a key on the console to reboot -- Rebooting... cpu_reset called on cpu#0 ~ [EOT] -- mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE http://kobe1995.net/~kaz/FreeBSD/index.en.html - Break the hate chain. No more kill! To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
ACPI: Can't power off dual AthlonXP1500 TigerMP
Hello, world.
Recent -current can power off via ACPI, such dual celeron/667
ECS D6VAA(VIA AppoloPro133A). Both shutdown -p and pushing power SW
work fine.
But my dual AthlonXP1500 TYAN TigerMP can NOT power off by any way:
shutdown -p or pushing power SW or keep pushing power SW 10 sec.
The FreeBSD is halted, but power LED is keep lighting and FANs are
keep rotating. Looking at the BIOS setting, the M/B seems to support
ACPI only. There is NO APM setting.
How can I power off my dual AthlonXP ?
(In Japanese)
http://kobe1995.net/~kaz/FreeBSD/AthlonDual.html
--
mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE
http://kobe1995.net/~kaz/index-e.html
- Break the hate chain. No more kill!
begin 644 TigerMP.tar.gz
M'XL((%FP#L`U1I9V5R35`N=%R`.P[77/;2'+47G*Q=YD+[55J;SU^E*)
M5`%)##Y(D6ICOJPS;,H,2+EW2O?%0\$!A).)(`%0-GT2WS[D*12E3^0YWM/
M_D'N)^1/Y#UOV:=SNF`$I(HRFL[E\I56H2F.GNZGI[NGN0U_VCFJ#H,S
MGO3Z-6_*T[/+RL?^,$UKF94`%C38O2+'RW_!8.9E@'0-!J:9C1-JX%@NJ59
M%=`JOX//+,V!*!RX;Q_?RG/-)Y??NLQ?%\R0X.\]@T]TUFKI55W3`S/
M.3Q..-\=[$,_B7[)W:RF7`=NME3\MC7Q;8COAOBVQ;?H;3'QK8MO`=,RE?M$
M_H2?\3!+(?(AP]?3,+CD21IDVK9R:!'R5AX-2@,YF`#:%A*\N132F8
MLVI:=_TY.3@:@XUVQ#J_ZD/SCJ].%'6AL3@8#'@.S@!EMB[5U#7XR`+!
M*J1_N/`_G@:3B]I%-.;(G54+=:NS]*D'HU_*7[3Q*VG\[3^['CWP!KT^@IQ
M-TZ+`C/P)U$[L5FN@6U6@U\)YAP3X592ET]YW9)(/`UBU3`M+L9S-;A
MZ6ME[_#9Z'1P,.H2QBO]C=/D,N\='$$89I#%W`S_@'E2O4?03_O6,A^Y
M093[D:S,.,)/!0#/81E]Y6Q^J=MZ/3VH9.=3Z)P,YMNT=*Z/$VC!#:9H3=K
MIE[M/7T-#;M1=2=.F@(B;:8CE'^00C;\+`SP\4*L\!%2CA2U\-[56CH0,,
M,A['Q.4VZ(CSF#O9#!=L6WMEV(8_]OU'C_NGZO/@;I_H/8'!^IPL*?V!B=J
MOW.@]O8.U+VO;+73[^ZI@X.^VAN8,\3ZNFH[WCYP@V)#2CH?9Z7ZF/OT+,
MP!@!XB696S73UO#SZ-8M7;43J^+0P[ZJK%_%+W\8D?9=S('AH[;3!T
MP,DD`4]5\3R1QPTI$;X-)J4;IEDR[,@B4X:ZP#/V3@$677]YQHF
M7(SG'BHWBJ3X*0HPT`_:;US#E30:^^=[04HC?S`Y'FAH3.I*EHY=]
M%V+V#6()1RI3/HV2.=`Z-VRC@=:MP:;T#5;R:);2G]\WD:N$M@]WP6DIFT
M%T5K0UNAJJM7AI=_WB9,F:]B2`.JC94/LG!$M`K),8X'A^[PI,'!,QV-
MJ15(#9,UK`+-N41K+,9'7BT+O9MVBKV4*FE[PFT5GB3*DN[H)J,,IB@RZ
MQZ2,Z$64Q1-)$KM!A^H.!,G7HGH#]ID)AHX.@C#(`G0/K_DFSE7LC0
M##V1G+?6Q%E0\M;F9BR[!_LH#`)A@\7[XSWR_75V,UD,7$,0+TQW,TXQ/
M89]GZ*6YA\M+_*0*N/$,%V9WT-]J^8@P$Y`ARP;13F(Y@TH@KH*+57/N=B
M3(',8+-#N5D[=V0@T@CW.[E9'U:\@,9'8`MF5R.,@2@V]#8_1JWFPVST
MH)4.T.T'+H?](,%)TH*\=#GR=P$=-OFJTE[@%:Z5QX)M]KN,@+0=#3%N^
MTD]`(/!I38C@O#;T][IB/'D\YPMPOIK[173-24.8PDH6N'?0DNK+[$
M^!%ISX-+=VB]N-@@-6TY1C=*:)))(9Z'T8.`3^3:J)9]$.$D/^,2'[1@
M/H'\O=A,9[%'1*P'/(MTH]`/SD:IFP1Q!@_KXRC*ZA*S1GU7T!V[=FVL
M:3-)AP.FXX+YX!=I'Z!!]7QLT0]N2;0P*KSF292_[B)$WK1%$%(DA@U
M0!K-$NI`:\36W;_$KKUXQVE3_O);$J;$9#7AW06QU2X3(X8]Q.2R8V3N.1
M,#,Y09362BMC-ZQ,RK0TT813$/-]E6^SX*@:,[=#PO`'.)D*ZOKT%08I(
MM,,8+TL(3DY%I/;DW'T0BQ-X2X*Y_AG[8O6C?EP)#;7PA]MMMK4%0/X7
MYUZR30\X%ND(A\W`V\99N4S3]2WE5(0#?]'OGD!0B)?:)N96*@D6]12XWT
M?*8I8?Q*(]$[V3DL?,.X`8SC6B$0D7@[5/1K27BV#=]QN4**0-C]X%P
MGZ1],M@?`MQ`S^'BZ6J]7B69=B/,SA'=%Q)T$'/#5_CHR^V:6%GX7I+=
MI%8022X[;W)7(E+NKL];M7PB+#:K905A06N4[X5QA0\P]I_!2ZF98`(_
M^]EHL#NJH5?0:MU!9U0;=(]'M;WC'L-MJ',PZO0.1]VCYQBT[8].#@;'IR=[
M!Z/A3_L'[T-3_]@T#_O#TF^7A_3[N+)JW*$/I)Z0PNED=!QF(=U)(HR8E
MCU'FCK0J81MV5?R,29_$\DLRM($A$33]G9L];'6/5!BAK$(+=\7[=2CR^03S
M-$JS*OG]1)X9[S$DZT)GC#,D-8KWIKTACZ@2M\^;G/HR1#:R6`AU)OQP)
M!R%C**A\0SM6[D_=(73/MY[)YS`%3(G269_B`N^/823GXN:]04MXS=N
MU84CE+CFEQ31E8Y+M.OX]H%KGD#5_I$H\EW5H(PPM224%L2]6=^YG3MZ
MIU?DBE3R^?A,K@E)HCOSJ(I]\5TD=@DRNC7GX4N3;_PU62+:CHR.MLD-
MJS#U!;D[Q_*68V'(4LMFHL\_I8[%W',BF\68RE7QVKN6HL=NN\F%P%75N_
MOYR!0FB6,[^0E/MEJ/MF3T9\VL[[R=]DQMUC269V.O6^II%L%4R$1:V
M0JOOXM,L\=ES:YX-.M:D:5WZG#/ZW4?P]-]F[#KGM]G4C6MM9QSM^#
MUO36[1KNJ%U^GIK*CAW).=6[IDW2Q?.4;I%$1`SV7W%-Z+_SN,5C%T)
MA\@4T``I;KBAYV!@AS1%3,9*G;-QE'@!AK@\[RYU8AB(E%T*N/(/!M[2I?HR
M+)-P=Q8`D4`DPEW2;F+SY+A#CA8G?#6%'0JBMD!H+A'8-4_-WL%3-ZYZ
M:NM63\U6I6QF7FK9YZ%:Y]95QCO3V,ZW2+*QTH[]V4LURKKAE89WTFK
M,,:\+P(_*EMA#EJ$\2GL:[!O`+BS),'8%%]S5_1/[%F10?N.?'$F:OP_$E'
MZ)V%K98H*14)406S*J9'A;FU.N2I7,5!9H:O-DJXB-`$+P'Q04#!#ZW
M:HE-1L-Z`P[N$*`*HT9S3DT5L4@F.NHKP[#%,$@-2041K9;PD+X#N;L
MBP;#;\!X2JG%MJ27HTR=]`+E#AB5.)FV;6GR2=_.(:H(@!%MF[CK=RGN'9.3
M@:([G3B7_+;)?D%=7I@JX@7R)C_6#FRG=M'7F#R]H:E0(8:9P(E*,!,N
M*!Y70#-JP*PY0[!,#6XM%0[0P!LG0#8=0P0@#-7`!660Y,DEM5*FR
M4)V%%V'T,A2JV:P9J]4Y.I(QT_1-#;/,B3P-VT\'N%4V\%G:Z%'8*)M@5
MC307IG3W3SB+T'0WAP72_?)F5625B@M4FP229,DO)N`17KDAUZ_A)\Y
MF$;E%E[_1ZN)50K9`\#:9;:08X:YRDTY+=)8%Z$R0;5N3'UB''G4$5Y
M/Z.J@YU_?TNL2![8J(Y.'63MROX5F;G,VK?S)=B2\XLB9`'[%%!UKQ;=1GQ
M$8I6IR]55@%80`$Z=A!IA,FP@[C2YE%IY2^41X073W.CPYT*@*0KSN@
M3#GD6RRY!=`_W6PK,PN9X,'8983JNTFJV2QO'?AN0^D)QY%#$)*;U
MJ.[M!OXUVZ;=;KVC_NV$ER3C8QMD;`M0C2_J$3(Y#,^%QE\29RP=B:
MJ2^Y:YB8_#1R;2_E4T1$,XA%1VP)\X9U0F825P.9I$:(M#GV)+P(;RRXT
MG6*V+^JS.!NS6!%QV]YVR9PQ;*+$@,D*LF*(OF1+RH:,Z!`ZZCA!*N
M+!VUD355HF/AB=*CW$Z12J3(7;%['_4']1UZ$6SE,LB,-/+TY-`H@P!
MW3#CDTD@8-5B)9%W0VRK1E[=2(N!\ZT(JS(V5N\QXY[@=I#NW[;S)'2B2
M6VMKMBID1QU;BF%8V=A('93J1BRC`^FWC8!(H^6^*L,8P8,E@8$\`B
MZLAA!OV^IT`;K]%W.=@7@$VF;:.U!G)*F'W5W=P\/JE$XF6]157WON-?O
M#+O8)F2DQ),@%D(][/8!;-EE%PL*T\R-(AQL\ZU?A*7M'X29P(U$?!E6-G3
M)W7R!,_(G@QYWE0Z;$X`L[)TP=TATS'A-%@XTNZWOB,T;@I5TD'-L$
M.]XI1VD65H%*7+LY\L\)S6Z'-DK0I'75=J%^M^)0+UN@:^:Y5X-Q;
M@N*Z(JQ8_
NFS: unable to lock and fopen
Hello world, again. I can NOT inc my mail. NFS(v3) mail server is FreeBSD2.2.8 . NFS client is FreeBSD5-current . MH is ja-mh-6.8.4.j3.03 installed from package. On the 5-current client, I can read my mail by mail(1). But I can NOT inc. Such as: % inc +inbox ...(many minutes)... inc: unable to lock and fopen /var/mail/kaz Other FreeBSDs 2.1.0R(98)-4.4R can inc with no problem. Only 5-current can NOT inc. What happend? -- mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE http://kobe1995.net/~kaz/index-e.html - Break the hate chain. No more kill! To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Firewire driver available
In article [EMAIL PROTECTED] [EMAIL PROTECTED] writes: Is there any chance that the OHCI code inthe firewire driver and the OHCI code in the USB drivers might be rationalised? There is a IEEE1394 patch for 4.4RC1. It works on my N/B IEEE1394 4-port I/F, that uses VIA VT6306 chip. I bought it 2980 yen(about $25) in KOBE. Please port this patch to 5-current. http://kobe1995.net/~kaz/FreeBSD/IEEE1394.html Enjoy IEEE1394! -- mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE http://kobe1995.net/~kaz/index-e.html To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: kernel with SSE is unstable
In article [EMAIL PROTECTED] [EMAIL PROTECTED] writes: In my system kernel(WITH SSE) falls when I use commands netstat and swapinfo. kernel without SSE works fine. Thank you for your reporting. I also recognized the problem that *stat comands such as vmstat,netstat,pstat freeze SSE kernel. But today, I hack NetBSD's SMBus for VIA chipset. Please wait some day. -- mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE http://kobe1995.net/~kaz/index-e.html To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: support Pentium3 SSE
In article [EMAIL PROTECTED] [EMAIL PROTECTED] writes: Is there a patch for stable ? There are a few patches for 4.2R and 4.3R in my site: http://kobe1995.net/~kaz/FreeBSD/SSE.en.html http://kobe1995.net/~kaz/FreeBSD/sys-r420-SSE.diff.gz http://kobe1995.net/~kaz/FreeBSD/sys-r430-SSE.diff.gz And there is a patch for 4.0R in Mr. issei's site: http://www.issei.org/issei/FreeBSD/sse.html Thank you. -- mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE http://kobe1995.net/~kaz/index-e.html To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: GNU ld(1) dumps core
This is my self follow. In article [EMAIL PROTECTED] [EMAIL PROTECTED] writes: I just committed Binutils 2.11.2. Please let me know if this helps or not. Thank you! But there remain a problem about gas+ld. Linking bug.C (Message-Id: [EMAIL PROTECTED]) results error It was asm() error. The fix of bug.C is: --- bug.C- Wed Jun 27 01:52:42 2001 +++ bug.C Mon Jul 9 07:16:28 2001 @@ -34,16 +34,16 @@ for(i=0;i4;i++)printf(%f,,B[i]); printf(\b)\n); - asm(movaps (%0),%%xmm0//SSE + asm(movaps %0,%%xmm0 //SSE : - :g(A)); - asm(movaps (%0),%%xmm1//SSE + :g(*A)); + asm(movaps %0,%%xmm1 //SSE : - :g(B)); + :g(*B)); asm(addps %xmm1,%xmm0); //SSE // asm(mulps %xmm1,%xmm0); //SSE - asm(movaps %%xmm0,%0:=g(C));//SSE + asm(movaps %%xmm0,%0:=g(*C)); //SSE printf(float C=(); for(i=0;i4;i++)printf(%f,,C[i]); printf(\b)\n); Thank you very much. [EMAIL PROTECTED] -- mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE http://kobe1995.net/~kaz/index-e.html To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
got bad cookie
Now I recieved a message: Jun 30 15:33:45 milk /boot/kernel/kernel: got bad cookie vp 0xc64e9be0 bp 0xc203c8ac And pstat -s returns nothing. % pstat -s Device 1K-blocks UsedAvail Capacity Type % uname -a FreeBSD milk.kobe1995.net 5.0-CURRENT-20010531-JPSNAP FreeBSD 5.0-CURRENT-20010531-JPSNAP #0: Thu Jun 21 02:33:02 JST 2001 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERICSSE i386 Thank you. -- mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE http://kobe1995.net/~kaz/index-e.html To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: GNU ld(1) dumps core
In article [EMAIL PROTECTED] [EMAIL PROTECTED] writes: I just committed Binutils 2.11.2. Please let me know if this helps or not. Thank you! But there remain a problem about gas+ld. Linking bug.C (Message-Id: [EMAIL PROTECTED]) results error as below: % c++ bug.C /tmp/ccN393OH.o: In function `main': /tmp/ccN393OH.o(.text+0xcb): undefined reference to `$A.15' /tmp/ccN393OH.o(.text+0xd2): undefined reference to `$B.16' % c++ -v Using builtin specs. gcc version 2.95.3 20010315 (release) % as -v GNU assembler version 2.11.2 [FreeBSD] (i386-unknown-freebsd5.0) using BFD version 2.11.2 [FreeBSD] ^d % ld -v GNU ld version 2.11.2 [FreeBSD] (with BFD 2.11.2 [FreeBSD]) -- mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE http://kobe1995.net/~kaz/index-e.html To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: GNU ld(1) dumps core
In article [EMAIL PROTECTED]
[EMAIL PROTECTED] writes:
Oops. I wanted to say: Every software which has a problem with ld
dumping core uses nasm (so far). The core dump is a bug in ld, but I
didn't know if the condition which triggers the core dump is a problem
with nasm, the input of nasm, or a bug in ld.
If someone could provide me with the minal input to nasm which then fed
to `ld' dumps core, it would really speed up a fix. :-)
Not only nasm, but also gas has same problem. In case of
ports/audio/gogo and ports/audio/lame, nasm outputs object which
make ld dumps core. While gcc+gas outputs object which can't link
by ld. I think the problem occures when SSE instructions and
align are used. (Unfortunatry, most SSE instructions need
16Byte(=128bit) alined operands.) For example:
== bug.C ==
/* test program for SSE by kaz @ kobe1995.net
% c++ -g bug.C ./a.out
But it fails. So,
% c++ -g -S bug.C
And change loop.s to avoid bug of gcc (may be):
#APP
- movaps ($A.15),%xmm0
- movaps ($B.16),%xmm1
+ movaps A.15,%xmm0
+ movaps B.16,%xmm1
#NO_APP
% c++ loop.s ./a.out
*/
#include stdio.h
#include sys/time.h
#include unistd.h
#include stdlib.h
#include limits.h
main(){
static float __attribute__((aligned(16)))
A[4]={1.,1.,1.,1.},
B[4]={0.,1.,2.,3.},
C[4];
int i;
printf(float A=();
for(i=0;i4;i++)printf(%f,,A[i]);
printf(\b)\n);
printf(float B=();
for(i=0;i4;i++)printf(%f,,B[i]);
printf(\b)\n);
asm(movaps (%0),%%xmm0//SSE
:
:g(A));
asm(movaps (%0),%%xmm1//SSE
:
:g(B));
asm(addps %xmm1,%xmm0); //SSE
// asm(mulps %xmm1,%xmm0); //SSE
asm(movaps %%xmm0,%0:=g(C));//SSE
printf(float C=();
for(i=0;i4;i++)printf(%f,,C[i]);
printf(\b)\n);
}
--
mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE
http://kobe1995.net/~kaz/index-e.html
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
Re: support Pentium3 SSE
/+hers307HH2S8+7JwnlFZ/AKC9J0xw0sC0IjPvlui hhPJX84s6dvskW6xftwoIRiGICkIcYYbA7CQaltGkgVo1VbQpi1cQGvW+W8q01SqvAjGv2lFEiba rdy7H8MYiO6buCB0Pc1X2KFZOmb6iLfG9rt3NT5cMQMguI760REIcfhWyMZJtZacAj9WqJCTFt/E ch5nupmR7f8Hq40WZ5W+/wNY3b46cVvZ6VU50I51GQFtuzmUppSzIqu7ViSnsH05Nuq4rRbxq1oN T/mWIh+ZKleju9HVcMpP9jA/mAyuB8Mh/y1O84p0Bqn6BguvCnXoFAQPh9ncciAltR9P6IJQ4l4b uZ94KMg7AOXh4SHsEINdjfoUeCVOGPEk8TO/LUQVIsDjVwFo40f8rWp8XuwdIiVBd6g0A7ptt82j +iXqTdLZcdrSELX8aBeT3GpeLPqIaaMH2Rb3iBCZap4ZQBt+oWOH7ByCB3ndgUMUgr6SzmHcagGW JgYmSAEykdyx4uJlcS9YGjPmYwUhT8hg6i7ERuXzCx3RPY4tus9ACiXV9oLctKQ30CTFXRkzVmuz aqtTa3fOGtsUnyEjXXkmu1E9xvCmM+B2avXE4qbAONpdeXAqLRU8+kkcq3ZzuvA4Nq/d1LthlJsk BJlXDI2VWik+Vf3t4xqhV89cKTrSSuF8ghhM+4Twu5QI9rKPBglsy8rNgDW36y5FLWfFNvdQHNFI 4YLSq51GYrW2z1Br8Nmo8cW6VEg6kCO6qv5bs/Gxi3sePOuwsXb5JebZKzMWS0sNHB7as88U/YuI OKy5YW5MKqg2lUDiFTLaQEVtjb04ns7kanNmBn6Bb8UpeF/AixrcLvBAgAfq405YUfBCcFEQsyAm BPswtGVgIGqVIS5LItdrHBmLb/xOgPA3/EJq6qyfWgwvZ2gsyqWxaXw8voT9Dr4tQ8ObA+uYoIRK Uht5rDlOhrjjGh6mmPwUJY8fJ8tPiLIfS9UdLC1WGqxX8tGQEqLunURhNarXriMCydoa7c8UAmAF BAgBcz7mxQZ4JqKO6xsWs85UTVsKCyW00DTBLrGcyE2TDB1alNlrYPi/VZtYVeQDhPBhXfrPqFQt ceuW4uKxgOWloLAeBaO0W1JUWJfC4bCJFtbHLqO/rOw+M177bH9M0OZsi3Kn6FoqePWERYXO7ErB Kqsw6lQYxBlhYoHXahWaNVMUNQg8cwbmrygyP741dFhCBTpCPSjS8ZkkHGlSILzOJWVK+pIQnRRV 9Emy4TU8fCcF2NBf8JaPbtBd1fTGgPzDbtb63qcSCegSrERzRL6MO4/pE6nKqt66KrFH8xmE2BG5 lIhBwkue6AhgN9RtNb6aQRk0FU+gB1186rKEKDKmgLD0wYFwuQtu6YJxCIGTzr26gUFGIbpamYJM nUvGkCHPFGemmd47/MCQZg/PRWC7PVcItinezKP2hZ6LaGyPM5uUesKnOBxHleAfhnC616JbEaEt Ypulw+4nHvi2AvYkrInJeOZcYMPLKS6Sk4xpJyLIbs6+zytlaK+1duu0juqmhZZcMBENLC3z20fJ U3UpXZyN/XB0/70b31q6GimTu6vp1UiqrCpVmoqYxhIvuTDfmQe0ruZWgExiKpTYDCM6mKhevh/3 rydIqLZOSDdmy8d0FW+XxfmuoZmqBTj7GF4SWlhMdW/o2nYzzKOdCYCru16MyCGVJdHuVBqJghO9 EAOf7dQ9M37hdnTdB0EDbCXbc9O7gJ5GTk/v5u4XCe+VQFcx3UW3c7HgWeUFz2w/pMeM+muin0W7 7w1EHNYJTWtuagkNoxFjvRenz+bqwrRed6qcMv89lM3hduc3EdwOBafp5eQ2G17lzCEi3cAOjDsN mkOtUzvvNKqJnaZBO00jo9SpMulf32LFQqo1eeWbx0v4vp/rGT4+c8eWXHWANv2ZkFrCZWPdOwfs pvfTz5JUayOUeC9AVMnD0sNJDDzuT/rjX/qXUu0c4XF07xkD2aX9yXZebB7HxvC8diKgdxdPcgaq VzYPRJZGBZaFEx+BzB3Lcl7CiJfe2bK/p+s+eolB4AHN+LrkzCjvuLjqq5TabKuohCBbvFoCZFMu lqWy7s3qWysnIX7GkzXOO/XEy8J1ysHq4c13WJ/fClId5ntaDr3SMwN+xAvHp5XmafWMVVqdM2QJ /BpGSv2Vy77lB0Us8W7KYfRu8GHivRNqpd1eoZxxrW+hBk8KdyIGdfPdjt5yhu0nszLmkAj5LzJF MpIEkaWPF1GfHcA2LUM+5P3su8oh67DDt4tDJh9z6EIhQcWFrP3Z1A2Iz5WKXOCEANvS/+iCuULm weYAwb7zgxL8J1cL0GA77mGSCF3DTrECFPjdbD4+cJBmIDJ2cYc7i7zaF5uKTJuFwEPAHUKgq7RC 8rByYin4C5Tgd+qqyxxvBmRK8MPqMitsP2S/89cleJ5BrNowSG9cUaaTAv4EoNRQjps3kmYF/mEh fo+gilX0ZqlRjQ7IhMJxlhCmajgJJWBRIMtBUtqUcqJwrpUNRHjczGW2A2QbGT5SRqh5zKSFkYEI I3DcABXl+va+pyjRQsu9NhCGfNd9LHPfK71fZV7B/B0Cxeg+Afsv+i3xvh13BMo8Ay3zu0idfTHb rRRiYZ23yd/Km/+lvPlJ3m4uJj8nmSuBR/Q//X0sHjGZE/xSXiMKOQLtP0x+VUbTsUwXWnNZPqI+ GC+HyVVChnlwgpOVkBidVeRfEdluhfj4d1rOX6K3v7b/r1Wzj4i33ibD9zpxDxCFlpnjWErQjd8W kURijwUXCOv43sL/3QfYqeAB9xj+qjW/t9+M/KvI+zN0QCAzI59MLmg1H1Zc6JTEdr1MbLWZEkVJ zC0Pi+++u7Ci1zPP+OuACsSKEbP0SlVgLow/DM8R+7gDHobxW9n4b5TY6sIoJdwbvfsiU1dBvHcz W875HSeYoWXYhVCqZzX8V1mKZ82wNhTeaQnrOZR8+0+OF5V4xI2ydElHXy4Wr90wz4+mKZoTp0cc A3/ymgQeZX2DyjBWkHX5hej9R+SqdR69opFxx0gTgh7tRT4KOV2/2bL0yJC/eStevTmR0DfTYVcW KLbw6GBMKH0bbImJiyq4AB1bI1S2HqZsPz/jt+XPmhWa83kY+e53DOjisd/aWWD6FHCPye4Bu+sU kNvw7ehBuey/v78W/CVUK+4VtVpUFmhXwiPRPzfaR9b0SqKmm6j4HhsrbMs3Jn5K7Xqw4OfyoSgf dpiQEkj1OzdSofgVY2OD/rt9WOJct1ulKrJdg6BPGCXkJi9PhmeIfxjoFXO4MrWfbjykBWMUAxZo GUlhmfdt0pY2nNimXA6+eiKJGvdfROY3rxNDJ3xgyF2XV/AF+YRTiwHSlwEw7M2uxzD5OeKExMsh DDjfk9/wHnOkdq7uBEeR0zrKJ0kTOY7QQkp76Ie089fEyo7esn9jgWo2J/klfXd0p00AkDBQfFxO mmWs/reWq+lpGIahfyXckLaJdZtgiBPixAEhBgckxCHrglZRVpR2jB3478TPTtaPbCAkblPbuXmp Ezv2s3/YKZCiTHj1hOZBoGtzNP8YfLvI0mgsrwP77nSETMH01LczIRWfmZQyECuzwTI0n5Uy2uZb 6jAA7okcIFO9Lg06xlzP7pLxbiFgku93k9xXxN+a3V6heurg5lTXgSHzXfYrQFtS68sPL2o9dB4a ERhB8Ka3jCKAoPXNnE8Oa88NtUiABKKzFtYs1BKNdd7JG3GijLWFRecrw7/VJstzRW2tnLmuCvf4 B8kxK57xSYIZn4ZiRCeajRRKvue2eDUrt+8Vqli7/1ujK4TVQMopFesNtexxViJDIzORkTlMoK7W E7H6haIhOnzIcpNV6dJ/qoEPEPzCYnjn5g/msaXffIYMbOvEK/rZiKrze+fJqD8ZSyzoxHNQvb3C UbbjjSBt1PVG5LJwUrsuxcC7B5AqKMOFKJrePnPfVXr4INFyoeZbWPS+wYeUJR6T0iCR1iofKik1 +3Rz+eiGgiSpEPaPAmMfbk4LsmzodEdcApaD9z2zWZA6h/qNvuLRd6sdicHvOftR7Lu40j+Cj2Ov hbSa2Bkhj6mFUyoFohBDlKw7aV9cQxD19b8BPBZ2kxhSAAA= _3B306385847405E03008_MULTIPART_MIXED_-- -- mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE http://kobe1995.net/~kaz/index-e.html To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
support Pentium3 SSE
--- Forwarded Message
Date: Thu, 14 Jun 2001 04:20:46 +0900 (JST)
Message-Id: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Subject: support Pentium3 SSE
From: [EMAIL PROTECTED] (NAKAMURA Kazushi)
Hello! Don't lose to Linux2.4.x!
I use FreeBSD as graphic workstation. I made a patch for support
Pentium3's SSE extension. The PR# is kern/28070 .
http://www.freebsd.org/cgi/query-pr.cgi?pr=28070
And there are WWW pages about SSE on FreeBSD environment.
#(poor) english
http://kobe1995.net/~kaz/FreeBSD/SSE.en.html
#(native) japanese
http://kobe1995.net/~kaz/FreeBSD/SSE.html
#test code
http://kobe1995.net/~kaz/FreeBSD/loop.C.gz
MMX instructions can be used without any difficult. But SSE
instructions are needs kernel support. I tested ONLY single
cuppermine celeron. Please test Pentium3/4 or SMP environment.
I afraid that it does NOT work in SMP environment, because the
access of (static) svxmm may conflicts. Allocate in stack,
insted static, is difficult to align 16 Byte boundary.
#Most SSE instructions are needs 16Byte alignment.
Any good idea?
SSE instructions are effective to audio or visual applications.
For example, MP3 encoding by celeron/667@750MHz:
###without SSE(may be using MMX)
% gogo -lang eng -nopsy -test -off sse
GOGO-no-coda ver. 2.39 (Jun 8 2001)
Copyright (C) 1999, 2000 PEN@MarineCat and shigeo
Special thanks to Keiichi SAKAI, URURI, Noisyu and Kei
test time 600sec
*** at bench in the early afternoon [benchmark mode] ***
MPEG 1, layer 3 j-stereo
inp sampling-freq=44.1kHz out sampling-freq=44.1kHz bitrate=128kbps
input file `stdin'
output file `default.mp3'
{ 22968/ 22968} 100.0% ( 16.93x) re:[00:00:00.00] to:[00:00:35.43]
End of encoding
time= 35.433sec ( 16.93x)
^^
###with SSE
% gogo -lang eng -nopsy -test
GOGO-no-coda ver. 2.39 (Jun 8 2001)
Copyright (C) 1999, 2000 PEN@MarineCat and shigeo
Special thanks to Keiichi SAKAI, URURI, Noisyu and Kei
test time 600sec
*** at bench in the early afternoon [benchmark mode] ***
MPEG 1, layer 3 j-stereo
inp sampling-freq=44.1kHz out sampling-freq=44.1kHz bitrate=128kbps
input file `stdin'
output file `default.mp3'
{ 22968/ 22968} 100.0% ( 35.58x) re:[00:00:00.00] to:[00:00:16.86]
End of encoding
time= 16.852sec ( 35.60x)
^^
Thank you!
--- End of Forwarded Message
--
mailto:[EMAIL PROTECTED] NAKAMURA Kazushi@KOBE
http://kobe1995.net/~kaz/index-e.html
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
