| View previous topic :: View next topic |
| Author |
Message |
PT_LAmb
Guru
Joined: 22 Apr 2002
Posts: 332
Location: Lisboa (Lisbon), Portugal
|
 Posted: Fri Jan 17, 2003 1:26 pm Post subject: SegFault after a prematurely aborted boot |
 |
|
Hi,
Yesterday I updated my system which was a month old, or even less.
Everything went ok, until emerge returned Segmentation Fault. I think that happened after emerging gcc-2.3.1-r6, but I'm not sure.
After this, I tried to arange a quick solution, by re-installing my system from scratch. I was hoping that the problem would go away. But after following all the Installation Guide's isntructions, and rebooting the system for the first time, the problem appeared again. INIT got off at runlevel 3, the file systems were read-only, a total mess.
I had the chance to at least saving an ls strace:
| Code: | execve("/bin/ls", ["ls"], [/* 22 vars */]) = 0
uname({sys="Linux", node="(none)", ...}) = 0
brk(0) = 0x80597a8
open("/etc/ld.so.preload", O_RDONLY) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0
close(3) = 0
open("/etc/ld.so.cache", O_RDONLY) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=9263, ...}) = 0
mmap2(NULL, 9263, PROT_READ, MAP_PRIVATE, 3, 0) = 0x25e18000
close(3) = 0
open("/lib/librt.so.1", O_RDONLY) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\360\32"..., 1024) = 1024
fstat64(3, {st_mode=S_IFREG|0755, st_size=35371, ...}) = 0
mmap2(NULL, 69076, PROT_READ|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0x25e1b000
mprotect(0x25e21000, 44500, PROT_NONE) = 0
mmap2(0x25e21000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 3, 0x6) = 0x25e21000
mmap2(0x25e22000, 40404, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x25e22000
close(3) = 0
open("/lib/libc.so.6", O_RDONLY) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0!Z\1\000"..., 1024) = 1024
fstat64(3, {st_mode=S_IFREG|0755, st_size=1425362, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x25e2c000
mmap2(NULL, 1211108, PROT_READ|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0x25e2d000
mprotect(0x25f4d000, 31460, PROT_NONE) = 0
mmap2(0x25f4d000, 16384, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 3, 0x11f) = 0x25f4d000
mmap2(0x25f51000, 15076, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x25f51000
close(3) = 0
open("/lib/libpthread.so.0", O_RDONLY) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0`E\0\000"..., 1024) = 1024
fstat64(3, {st_mode=S_IFREG|0755, st_size=83562, ...}) = 0
mmap2(NULL, 323040, PROT_READ|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0x25f55000
mprotect(0x25f61000, 273888, PROT_NONE) = 0
mmap2(0x25f61000, 12288, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 3, 0xc) = 0x25f61000
mmap2(0x25f64000, 261600, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x25f64000
close(3) = 0
munmap(0x25e18000, 9263) = 0
modify_ldt(0x1, 0x5947d49c, 0x10) = -1 EINVAL (Invalid argument)
--- SIGSEGV (Segmentation fault) ---
+++ killed by SIGSEGV +++ |
Everything looks normal on dmesg.
My CFLAGS are: | Code: | | CFLAGS="-march=pentium4 -O3 -pipe -mfpmath=sse -msse2 -mmmx -fomit-frame-pointer" |
Some packages versions: | Code: | sys-libs/glibc-2.3.1-r2
sys-devel/gcc-3.2.1-r6
sys-apps/portage-2.0.46-r9
sys-kernel/gentoo-sources-2.4.19-r10
|
I really hope some of you can help me. 
Thanks in Advance,
Ricardo Cordeiro |
|
| Back to top |
|
 |
garo
Bodhisattva
Joined: 15 Jul 2002
Posts: 860
Location: Edegem,BELGIUM
|
| Posted: Fri Jan 17, 2003 1:36 pm Post subject: |
|
|
I would try some more "normal" and stable CFLAGS:
| Code: | | CFLAGS="-march=pentium4 -O3 -pipe -fomit-frame-pointer" |
_________________
My favorite links this month:
- Surf Random
- Web-based SSH
- Stop Spam |
|
| Back to top |
|
|
PT_LAmb
Guru
Joined: 22 Apr 2002
Posts: 332
Location: Lisboa (Lisbon), Portugal
|
| Posted: Fri Jan 17, 2003 1:57 pm Post subject: loosing -fomit-frame-pointer |
|
|
Hi, and thanks for the reply.
Ok, I'l loose the -fomit-frame-pointer only, because the other are pretty standard. Don't you think?
After that emerge world -e, from the livecd.
Hope it'll do.
And thanks again,
Ricardo Cordeiro |
|
| Back to top |
|
|
PT_LAmb
Guru
Joined: 22 Apr 2002
Posts: 332
Location: Lisboa (Lisbon), Portugal
|
| Posted: Fri Jan 17, 2003 4:25 pm Post subject: Not yet... |
|
|
| It did work again... I'll just leave it blank, just to check. |
|
| Back to top |
|
|
PT_LAmb
Guru
Joined: 22 Apr 2002
Posts: 332
Location: Lisboa (Lisbon), Portugal
|
| Posted: Fri Jan 17, 2003 9:15 pm Post subject: It's not a problem related to the CFLAGS |
|
|
Well, I tested recompiling the system with no CFLAGS at all, and the error ocurred again...
Anymore suggestion?
Thx in advance,
Ricardo Cordeiro |
|
| Back to top |
|
|
rtn
Guru
Joined: 15 Nov 2002
Posts: 427
|
| Posted: Fri Jan 17, 2003 9:35 pm Post subject: |
|
|
A significant portion of the segfaults I've heard about concerned either:
overclocked processors
processors that ran better slightly underclocked
memory that needs to be run at a more conservative speed
just plain bad broken ram.
Might any of those apply to you? 
--rtn
ps. Your original CFLAGS are redundant. mmx and sse are assumed under
-march=pentium4. |
|
| Back to top |
|
|
PT_LAmb
Guru
Joined: 22 Apr 2002
Posts: 332
Location: Lisboa (Lisbon), Portugal
|
| Posted: Fri Jan 17, 2003 10:40 pm Post subject: It's not overcloking, or bad ram. |
|
|
Hi,
Thanks for the reply rtn.
| Quote: | overclocked processors
processors that ran better slightly underclocked
memory that needs to be run at a more conservative speed
just plain bad broken ram. |
Well, I think none of what you said is really the cause of the problem. Because, I don't get a single Segmentation Fault, when I run from the livecd. I can chroot, and do as I please on that environment, without the minimum threat of getting a Segmentation Fault.
I think it might be something related with the kernel. I'm going to try one from the vanilla branch.
What do you think?
Thanks again,
Ricardo Cordeiro |
|
| Back to top |
|
|
PT_LAmb
Guru
Joined: 22 Apr 2002
Posts: 332
Location: Lisboa (Lisbon), Portugal
|
| Posted: Fri Jan 17, 2003 11:06 pm Post subject: It's the kernel... |
|
|
Well, finally, after a dozen of experiences... The problem is the kernel, that somewhat got in conflict with another package, such has gcc-3.2.1-r6 (don't really know for sure). The kernel that came with the livecd, works perfectly.
Of course, I'm going to compile a costum one.
Thank you garo and rtn. You kept me alive, searching for a solution. 
Goodnight all,
Ricardo Cordeiro |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Other Things Gentoo
