lkml.org 
[lkml]   [1996]   [Apr]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRE: Compile problem
Date
If anybody wants a semi reliable way to recreate this problem I have a setup that should do it for you.  I had this same message coming up when I pushed a AMD 5x86-133 to 160 (Just to see what the processor could take).  The system had no problems while running, but would give random sig11 errors while compiling.  After I swtiched back to 133mhz the sig11 errors went away.  I have not had any ram problems on that system, so I think that can be elimintated as the cause.

----------
From: Kevin M Bealer[SMTP:kmb203@psu.edu]
Sent: Sunday, March 31, 1996 5:18 PM
To: root
Cc: linux-kernel@vger.rutgers.edu
Subject: Re: Compile problem

On Sun, 31 Mar 1996, root wrote:

>
> Hiya
>
> I do not know if I am encountering a Linux specific problem or a
> bug with GCC. This is my setup:
>
> Linux 1.3.58 and 1.3.79 (same problem on both) on an i586. Using
> gcc 2.7.2 and 2.7.2p.
>
> For a long time I have gotten what seems like random errors when compiling,
> for example when compiling the Linux kernel, I get:
>
> gcc [snip...]
> cpp: output pipe has been closed
> gcc: Internal compiler error: Program cc got fatal signal 11
>
> (note, 11 is SEGV somtimes its signal 6)
>
> sometimes the "cpp: output pipe has been closed" does not appear.
>
> (an invalid .o file is left as a result causing ld to fail its linking).
>
> The error seems to happen in particular on certain .c files. For example
> the drivers/block/floppy.c does this over and over, but perhaps one out
> of 10 compilations succeed. Sometimes logging out & back in solves
> the problem.
>
> What should I do?
>
> --Michael

For several days, the great gurus of linux have been arguing... this may be
a memory problem (bad ram) or this may be a misconfigured memory system (try
disabling your cache) or this may be a bug in gcc or it's friends.

Sig 11 is known to be ambiguous in this way, signal 6 is variously defined
as SIGIOT and SIGABRT, but when I do this:

cat
^Z
ps (look for the pid of cat)
kill -6 (that pid)
fg

it gives me IOT trap/Abort, so I guess this signal is being redefined or
includes more stuff?

So fiddle with the hardware, and see what happens.


__kmb203@psu.edu_____________________________Debian/GNU__Linux__1.3.77___
"The C Programming Language -- A language which combines the
flexibility of assembly language with the power of assembly language."






\
 
 \ /
  Last update: 2005-03-22 13:36    [W:0.034 / U:1.152 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site