Bug 268314

Summary: unhandled instruction bytes: 0xC5 0xFA 0x10 0x5 0x4F 0xC2
Product: [Developer tools] valgrind Reporter: bugs
Component: vexAssignee: Julian Seward <jseward>
Severity: crash CC: Enrico_m, kdebugs, me
Priority: NOR    
Version: 3.6.0   
Target Milestone: ---   
Platform: Arch Linux   
OS: Linux   
Latest Commit: Version Fixed In:

Description bugs 2011-03-12 21:21:54 UTC
Version:           3.6.0 (using KDE 4.6.1) 
OS:                Linux

vex amd64->IR: unhandled instruction bytes: 0xC5 0xFA 0x10 0x5 0x4F 0xC2

Reproducible: Sometimes

valgrind version is really 3.6.1-1

g++ 4.5.2 20110127 (prerelease)
/proc/cpuinfo (truncated to 1 instance of the processor)
processor	: 3
vendor_id	: GenuineIntel
cpu family	: 6
model		: 42
model name	: Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz
stepping	: 7
cpu MHz		: 3292.141
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 3
cpu cores	: 4
apicid		: 6
initial apicid	: 6
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 popcnt aes xsave avx lahf_lm ida arat epb xsaveopt pln pts dts tpr_shadow vnmi flexpriority ept vpid
bogomips	: 6587.08
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management:
Comment 1 Joern Heissler 2011-05-11 03:45:06 UTC
I'm encountering similar problems with real-world code.

gcc version 4.6.1 20110507 (prerelease) (Debian 4.6.0-7) 
model name	: Intel(R) Core(TM) i5-2410M CPU @ 2.30GHz

How to reproduce:

int main(void)
    int foo = 0;
    double bar = foo;
    return 0;

gcc -O0 -march=native foo.c -S

results in this line:
vcvtsi2sd   -4(%rbp), %xmm0, %xmm0

or assembled + objdump:
   b:	c5 fb 2a 45 fc       	vcvtsi2sdl -0x4(%rbp),%xmm0,%xmm0

-O0 is needed because the test case is too simple. With real code, -O2 etc. cause the same problem.
I did not find a more specific flag for -march.

Joern Heissler
Comment 2 Peter Grasch 2011-08-23 10:22:30 UTC
Yes, I can reproduce this problem with a 2nd gen i5 CPU.

Valgrind from 3.7 SVN still crashes (gcc 4.5.3).
Comment 3 Julian Seward 2011-10-12 09:47:30 UTC

*** This bug has been marked as a duplicate of bug 273475 ***