Re: Kernel Compiling
Matei Conovici ~cls XII~ (cmatei@lbi.sfos.ro)
Wed, 20 Mar 1996 11:28:01 +0200 (EET)
> To add to the din, Martin is correct. Corrupted/defective memory can
> easily lead to attempts to access stupid places in memory. If you have a
> valid pointer and it changes its value (for whatever reason), it no longer
> points to the same place. The point is not that sig 11 means bad memory,
> but that getting SEGV's while compiling the kernel is a good sign you have
> a memory problem. The same could be said of bus errors. (Last time I
> checked, neither gcc nor the kernel poked about in memory randomly:)
I didn't notice memory faults on my machine, however I get a sigsegv once
or twice when I compile a new kernel. This has happened with kernel post
3.64, before that I used 1.2.13 that worked fine and compiled ok.
Load average, before gcc dies with SIGSEGV is going sky-high, and problems
only arise when there is a memory shortage.
> K.
Matei