https://bugs.freedesktop.org/show_bug.cgi?id=49681
--- Comment #22 from Frederic Germain <[email protected]> --- I have more information about this bug. It happens in the SIMD function in ORC, and it's maybe specific to AMD Geode or MMX only (no SSE/SSE2) CPUs I have it on AMD Geode, when pulseaudio has to apply a volume to the sink. A quickfix to it could be to launch pulseaudio with PULSE_NO_SIMD env variable -> PULSE_NO_SIMD=1 pulseaudio --start But then you loose the optimization you could have with MMX I'm not sure I have time to fix the bug in ORC, but maybe there is a jedi out there who can find the problem quickly :) Step to reproduce on Geode platform (Debian Wheezy x86). Alsa card driver is cs5535audio. > gdb pulseaudio > pacmd set-sink-volume 0 0x100 (volume is really really loud at 100% on my > board) > paplay mymusic.wav In gdb -> Program received signal SIGILL, Illegal instruction. [Switching to Thread 0xb3235b70 (LWP 3239)] 0xae128031 in ?? () (gdb) bt #0 0xae128031 in ?? () #1 0x00003d3c in ?? () > cat /proc/$(pidof pulseaudio)/maps |grep orc ae118000-ae128000 rw-s 00000000 08:01 56769 /tmp/orcexec.Wk0lzw (deleted) ae128000-ae138000 r-xs 00000000 08:01 56769 /tmp/orcexec.Wk0lzw (deleted) It seems gdb cannot rewind stack of orc functions on this architecture, that's pretty sad. (gdb) disas /r 0xae128000,0xae138000 Dump of assembler code from 0xae128000 to 0xae138000: 0xae128000: 55 push %ebp 0xae128001: 8b 6c 24 08 mov 0x8(%esp),%ebp 0xae128005: 0f 6e 85 74 01 00 00 movd 0x174(%ebp),%mm0 0xae12800c: 0f c4 85 94 01 00 00 02 pinsrw $0x2,0x194(%ebp),%mm0 0xae128014: 0f c4 85 96 01 00 00 03 pinsrw $0x3,0x196(%ebp),%mm0 0xae12801c: 0f ef c9 pxor %mm1,%mm1 0xae12801f: 8b 4d 04 mov 0x4(%ebp),%ecx 0xae128022: 89 4d 0c mov %ecx,0xc(%ebp) 0xae128025: 8b 45 14 mov 0x14(%ebp),%eax 0xae128028: 83 7d 0c 00 cmpl $0x0,0xc(%ebp) 0xae12802c: 74 54 je 0xae128082 0xae12802e: 8b 55 0c mov 0xc(%ebp),%edx => 0xae128031: 66 66 66 66 66 66 2e 0f 1f 84 00 00 00 00 00 data32 data32 data32 data32 data32 nopw %cs:0x0(%eax,%eax,1) 0xae128040: 0f 6e 10 movd (%eax),%mm2 0xae128043: 0f ef db pxor %mm3,%mm3 0xae128046: 0f 61 d3 punpcklwd %mm3,%mm2 0xae128049: 0f 6f d9 movq %mm1,%mm3 0xae12804c: 0f 65 da pcmpgtw %mm2,%mm3 0xae12804f: 0f db d8 pand %mm0,%mm3 0xae128052: 0f e4 d0 pmulhuw %mm0,%mm2 0xae128055: 0f fa d3 psubd %mm3,%mm2 0xae128058: 0f 6f d8 movq %mm0,%mm3 0xae12805b: 0f 72 e3 10 psrad $0x10,%mm3 0xae12805f: 0f 6b db packssdw %mm3,%mm3 0xae128062: 0f 6e 20 movd (%eax),%mm4 0xae128065: 0f 6f ec movq %mm4,%mm5 0xae128068: 0f e5 eb pmulhw %mm3,%mm5 0xae12806b: 0f d5 e3 pmullw %mm3,%mm4 0xae12806e: 0f 61 e5 punpcklwd %mm5,%mm4 0xae128071: 0f fe d4 paddd %mm4,%mm2 0xae128074: 0f 6b d2 packssdw %mm2,%mm2 ---Type <return> to continue, or q <return> to quit---q Quit cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 5 model : 10 model name : Geode(TM) Integrated Processor by AMD PCS stepping : 2 microcode : 0x8b cpu MHz : 498.053 cache size : 128 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu de pse tsc msr cx8 sep pge cmov clflush mmx mmxext 3dnowext 3dnow bogomips : 996.10 clflush size : 32 cache_alignment : 32 address sizes : 32 bits physical, 32 bits virtual power management: -- You are receiving this mail because: You are the QA Contact for the bug. You are the assignee for the bug.
_______________________________________________ pulseaudio-bugs mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/pulseaudio-bugs
