tags 164588 + moreinfo thanks - please recheck with gcc-3.2_3.2.1-0pre4 (from unstable)
- please attach the preprocessed output Felix Kühling writes: > Package: gcc-3.2 > Version: 1:3.2.1-0pre2 > Severity: normal > > I tracked down this problem in the DRI CVS source tree. When compiling > with -O2 -march=athlon or -O2 -march=pentium2 the generated code > incorrectly mixes MMX and FPU instructions. I generated a stripped > down preprocessed file and added a small main function and 2 printfs > to demonstrate the problem. I will submit it as attachment with a > follow-up mail. This is the exact compiler command line: > > gcc-3.2 -Wall -O2 -march=athlon -o radeon_state2 radeon_state2.i > > The programme is compiled correctly with: > > gcc-3.2 -Wall -O2 -mcpu=athlon -o radeon_state2 radeon_state2.i > > or > > gcc-3.2 -Wall -O2 -march=athlon -mno-mmx -mno-3dnow -o radeon_state2 > radeon_state2.i > > The incorrect version produces this output > Inside if: 133 != nan || 589 != nan > > and a sigsegv since I didn't fake things up enough in the main function ;-) > > The correct version produces this output: > Outside if: 133 != 133.000000 || 589 != 589.000000 > > -- System Information > Debian Release: testing/unstable > Architecture: i386 > Kernel: Linux viking 2.4.19 #6 Fre Sep 27 23:46:47 CEST 2002 i686 > Locale: LANG=de_DE.ISO-8859-1, LC_CTYPE=de_DE.ISO-8859-1 > > Versions of packages gcc-3.2 depends on: > ii binutils 2.13.90.0.4-1 The GNU assembler, linker and > bina > ii cpp-3.2 1:3.2.1-0pre2 The GNU C preprocessor. > ii gcc-3.2-base 1:3.2.1-0pre2 The GNU Compiler Collection > (base > ii libc6 2.2.5-14.3 GNU C Library: Shared libraries > an > ii libgcc1 1:3.2.1-0pre2 GCC support library. > > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]