Le Sun, 30 Nov 2003 11:34:22 -0800 (PST) "Steven M. Schultz" <[EMAIL PROTECTED]> écrivait :
[...] > Nicolas Boos is having the same problem with his glibc 2.1.2 > system - in fact the segfault is happening on the same line > ' 309 mulps_m2r( *(mmx_t*)&piqf[0], xmm2 );'! Hey, mine is 2.2.5 (Debian Woody) ! > There were two problems, the first was the use of > posix_memalign() - that is known to be broken in glibc 2.1.2. For the moment, I'm using "#undef HAVE_POSIX_MEMALIGN" instead of "#define HAVE_POSIX_MEMALIGN 1" in configure as workaround. [...] > The problem seems to be related to SSE being used with glibc > 2.1.2. Signal handling related perhaps? Malloc alignment > issues would be another good possibility. > > Not sure how this can be dealt with. If the exact breakage were > known it might be (but then again it might not be) possible to > work around the problem. Working around libc bugs is difficult > - at some point programs need to trust the system components to > work as expected. Yes, It's probably SSE related (mpeg2enc runs fine on my old Celeron), but don't know exactly for the glibc case. Cheers, Nicolas ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
