[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: SEGV (was Re: Internal compiler error on m68k for gcc-4.6)



On 2012-12-02 04:55, Michael Schmitz wrote:

I surely hope not╜ this looks more like a hardware issue,
possibly faulty RAM, to me. (Never saw the ╲The bug is not
reproducible╡ blurb beforeâ•œ)
I've seen that one before - it's been there pretty much since the
stone  ages. gcc does exercise memory more thoroughly than anything
else, and  hardware faults that went undetected otherwise got
triggered.
That's all ancient history though (RAM starved machines were qute
commonn before memory got cheap). Might not matter much now. Then
again, we're dealing with historic HW here.
I can try and run the binary  or binaries faulting on my 060 to make
sure. Same on ARAnyM.

My guess is: schroot and others are segfaulting with "*** glibc detected *** schroot: double free or corruption (fasttop): 0x80119130 ***" and 3 different machines. I think that it is very unlikely that all 3 machines suffers from the same problem, especially because two of them are performing memory tests everytime during boot process. Therefore I suspect a toolchain issue or even an issue in the emulator itself.

I can test with an 040 as well, I guess. If that test passes, it's an 040 vs 060 issue within the compiler, if that fails, it's a generic compiler/toolchain problem. That's my understanding.

--
Ciao...          //    Fon: 0381-2744150
.     Ingo     \X/     http://blog.windfluechter.net

gpg pubkey: http://www.juergensmann.de/ij_public_key.


Reply to: