signal 11/memory? problem?

signal 11/memory? problem?

Post by The Ancient Programm » Mon, 08 Jun 1998 04:00:00



I'm having a strange problem which appears to be(?) memory related.
                                   ^^^^^^^^^^^^^
I'm running the "hamm" release of Debian on a Tyan dual CPU
motherboard with 1-166mH Pentium CPU. The problem predated the
installation of the "hamm" release. I was using Fast Page Mode Drams
memory, 2-16Mb + 2-4Mb SIMMS.

Most things ran, however, when I tried to build a new kernel I got
repeated failures with the error message:
"gcc: Internal compiler error: program cc1 got fatal signal 11"

These errors occured randomly in the make process with variable
frequency.

Also dselect gets random Segmentaion violations.

I purchased 2-32Mb SIMMS and replaced the memory. The errors
persisted. I then began trying all combinations of memory. The error
went away only when I ran with the 2-4Mb SIMMS. The computer runs
error free but painfully slow. (With only 8Mb of memory, it's a wonder
that it runs at all)

Has anyone else encountered a similar problem? Does anyone have any
ideas as to what could be causing the problem?

Any suggestions would be greatfully received.

Karl Abrams

 
 
 

signal 11/memory? problem?

Post by David Re » Mon, 08 Jun 1998 04:00:00


I encountered many of the same problems you describe, which were fixed by
buying some high quality RAM.  I've found that some motherboards are more
picky than others about the ram quality.  One way to work around this is to
slow down the ram timings in BIOS.  Set them to the slowest (largest)
settings, and compile the kernel.  If you still get sig11s (even with the
newer EDO ram you bought, I'd start suspecting something else.  If things go
alright, start speeding up the settings slowly, and try recompiling the
kernel again.  Repeat until the start getting sig11s, then back off one
setting.  Of course, you probably won't notice that much of a performance
hit from leaving the ram at their slowest timings, so you can leave them
there if you're lazy.

-Dave


>I'm having a strange problem which appears to be(?) memory related.
>                                   ^^^^^^^^^^^^^
>I'm running the "hamm" release of Debian on a Tyan dual CPU
>motherboard with 1-166mH Pentium CPU. The problem predated the
>installation of the "hamm" release. I was using Fast Page Mode Drams
>memory, 2-16Mb + 2-4Mb SIMMS.

>Most things ran, however, when I tried to build a new kernel I got
>repeated failures with the error message:
>"gcc: Internal compiler error: program cc1 got fatal signal 11"

>These errors occured randomly in the make process with variable
>frequency.

>Also dselect gets random Segmentaion violations.

>I purchased 2-32Mb SIMMS and replaced the memory. The errors
>persisted. I then began trying all combinations of memory. The error
>went away only when I ran with the 2-4Mb SIMMS. The computer runs
>error free but painfully slow. (With only 8Mb of memory, it's a wonder
>that it runs at all)

>Has anyone else encountered a similar problem? Does anyone have any
>ideas as to what could be causing the problem?

>Any suggestions would be greatfully received.

>Karl Abrams



 
 
 

signal 11/memory? problem?

Post by Phil DeBecke » Tue, 09 Jun 1998 04:00:00



> I'm having a strange problem which appears to be(?) memory related.
>                                    ^^^^^^^^^^^^^
> I'm running the "hamm" release of Debian on a Tyan dual CPU
> motherboard with 1-166mH Pentium CPU. The problem predated the
> installation of the "hamm" release. I was using Fast Page Mode Drams
> memory, 2-16Mb + 2-4Mb SIMMS.

> Most things ran, however, when I tried to build a new kernel I got
> repeated failures with the error message:
> "gcc: Internal compiler error: program cc1 got fatal signal 11"

> These errors occured randomly in the make process with variable
> frequency.

> Also dselect gets random Segmentaion violations.

> I purchased 2-32Mb SIMMS and replaced the memory. The errors
> persisted. I then began trying all combinations of memory. The error
> went away only when I ran with the 2-4Mb SIMMS. The computer runs
> error free but painfully slow. (With only 8Mb of memory, it's a wonder
> that it runs at all)

> Has anyone else encountered a similar problem? Does anyone have any
> ideas as to what could be causing the problem?

> Any suggestions would be greatfully received.

> Karl Abrams


Offhand, it sounds like cache might be your problem.  Try disabling the
external (L2) cache from the BIOS and run with your 2 32MB SIMMs.  If
that works, it's most likely a bad cache SRAM chip causing your problem.
 
 
 

signal 11/memory? problem?

Post by friedhelm.mehner » Tue, 09 Jun 1998 04:00:00


: I encountered many of the same problems you describe, which were fixed by
: buying some high quality RAM.  I've found that some motherboards are more
: picky than others about the ram quality.  One way to work around this is to
: slow down the ram timings in BIOS.  Set them to the slowest (largest)
: settings, and compile the kernel.  If you still get sig11s (even with the
: newer EDO ram you bought, I'd start suspecting something else.  If things go
: alright, start speeding up the settings slowly, and try recompiling the
: kernel again.  Repeat until the start getting sig11s, then back off one
: setting.  Of course, you probably won't notice that much of a performance
: hit from leaving the ram at their slowest timings, so you can leave them
: there if you're lazy.

I second that!

You may also slow down the settings for the timings with cache access
within the BIOS.
It makes a difference with some mobo's.

Regards,
Friedhelm

--
Microsoft is NOT the answer. Microsoft is the Question.
The answer is: "NO!"
-------------------------------------------------------------------
Friedhelm Mehnert,  Berliner Allee 42,  22850 Norderstedt,  Germany

-------------------------------------------------------------------

 
 
 

signal 11/memory? problem?

Post by Steve Marti » Wed, 10 Jun 1998 04:00:00




> : I encountered many of the same problems you describe, which were fixed by
> : buying some high quality RAM.  I've found that some motherboards are more
> : picky than others about the ram quality.  One way to work around this is to

How does one identify high quality RAM??  Can you suggest some brand
names?

TIA

--

**********************  Only my opinion!!  *************************
Steve Martin, Software Engineer, Philips-Albuquerque Test Operations
NR7P                                               Ph:  505-822-7274


Linux ---- because reboots are for kernel upgrades...
********************************************************************

 
 
 

1. Virtual Memory and Signal 11...

I've been  running a considerably big simulation ('top' says its 33 megs
big) under 2.2.1 in my Pentium 100 of 48 Megs RAM and the EIDE disk
below:

wdc0: unit 0 (wd0): <Conner Peripherals 1620MB - CFS1621A>
wd0: 1548MB (3171168 sectors), 3146 cyls, 16 heads, 63 S/T, 512 B/S

Virtual memory collapses specially when running X. Sample messages are
appended below. The FAQ says signal 11 is possibly a memory problem. But
I also suspect the hard disk may not keep up with the paging traffic.

Fact: My first pair (2x8megs) of SIMMs is 70ns and the other (2x16megs)
is 60 ns. Could this be the reason ?

Any suggestion on how to check RAM and/or Disk? Would Norton do it?

Thanks a lot,

Ewerton

UC Santa Cruz, CA

----------------------------------------------------------------------
wd0s2b: wdstart: timeout waiting for DRQ writing fsbn 37504 of
37504-37631 (wd0s2 bn 1534592; cn 380 tn 38 sn 38)wd0: status
d1<busy,rdy,seekdone,err> error 40<uncorr>

swap_pager: I/O error - pagein failed; blkno 5256, size 8192, error 5

vm_fault: pager input (probably hardware) error, PID 153 failure

pid 153 (fvwm), uid 1001: exited on signal 11

wd0s2b: hard error reading fsbn 10641 of 10632-10671 (wd0s2 bn 1507729;
cn 373 tn 60 sn 13)wd0: status 59<rdy,seekdone,drq,err> error 40<uncorr>

swap_pager: I/O error - pagein failed; blkno 10632, size 20480, error 5

vm_fault: pager input (probably hardware) error, PID 166 failure

pid 166 (FvwmPager), uid 1001: exited on signal 11 (core dumped)

2. SCSI Tape Drive reading "other" UNIX

3. signal 11, signal 8 Error?

4. Need wu-ftpd

5. signal 4 & signal 11

6. Redhat Linux 6.0, stuck at mouse setup.

7. Memory Fault (11) (sig 11)

8. Is it worth upgrading Slackware3.0 -> Slackware3.1 ?

9. RH5.2 install problems re:signal 11

10. SIGNAL 11 Problem on SuperMIcro Server

11. Problem with linking gpm-1.06 (fatal signal 11)

12. GCC and the 'signal 11' problem, HELP!

13. RH5.0 Install problems ("Signal 11")