Bug 112167 - assertion failed in VEX (MFence on mmx-ext only CPU)
Summary: assertion failed in VEX (MFence on mmx-ext only CPU)
Status: RESOLVED DUPLICATE of bug 112152
Alias: None
Product: valgrind
Classification: Developer tools
Component: general (show other bugs)
Version: unspecified
Platform: Compiled Sources Linux
: NOR crash
Target Milestone: ---
Assignee: Julian Seward
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-09-07 15:23 UTC by Ivan Kalvachev
Modified: 2005-09-07 15:44 UTC (History)
0 users

See Also:
Latest Commit:
Version Fixed In:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Ivan Kalvachev 2005-09-07 15:23:02 UTC
Version:           3.0.1 (using KDE KDE 3.4.2)
Installed from:    Compiled From Sources
Compiler:          gcc-3.3.6 
OS:                Linux

This is the second time I am having problems with valgrind because I have an Athon thunderbird.
It does have mmx-ext (instructuin like movntq, mfence) that are refered by intel zealots as integer SSE instructions.

I got an assert failer in
VEX/priv/host-x86/hdefs.c::2310

   case Xin_MFence:
      /* see comment in hdefs.h re this insn */
      if (0) vex_printf("EMIT FENCE\n");
      switch (i->Xin.MFence.subarch) {
         case VexSubArchX86_sse0:
-->         vassert(0); /* awaiting test case */  <--
            /* lock addl $0,0(%esp) */
            *p++ = 0xF0; *p++ = 0x83; *p++ = 0x44; 
            *p++ = 0x24; *p++ = 0x00; *p++ = 0x00;
            goto done;

removing the assert() solved the problem for me.

Well, now you have test case, so please stop bugging me!!!
Comment 1 Tom Hughes 2005-09-07 15:44:51 UTC
Calm down... We're not intel zealots and we don't have anything against your processor, we just wanted to make sure the code was right.

*** This bug has been marked as a duplicate of 112152 ***