| Summary: | Unhandled instructions bytes: 0x48 0xF 0xFE 0x7E 0x70 0x48 | ||
|---|---|---|---|
| Product: | [Developer tools] valgrind | Reporter: | Sebastien Cote <sebas642> |
| Component: | vex | Assignee: | Julian Seward <jseward> |
| Status: | REPORTED --- | ||
| Severity: | crash | CC: | cpigat242 |
| Priority: | NOR | ||
| Version First Reported In: | unspecified | ||
| Target Milestone: | --- | ||
| Platform: | Compiled Sources | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed/Implemented In: | ||
| Sentry Crash Report: | |||
| Bug Depends on: | |||
| Bug Blocks: | 253451 | ||
| Attachments: | workaround for rex.W prefix | ||
|
Description
Sebastien Cote
2010-04-07 20:14:32 UTC
I also get another unhandled instruction with IPP (in a different part of the code): vex amd64->IR: unhandled instruction bytes: 0x48 0xF 0xFC 0xD 0x77 0x8A ==21616== valgrind: Unrecognised instruction at address 0x182e0171. I get the following from objdump: 65171: 48 0f fc 0d 77 8a 24 rex.W paddb 0x248a77(%rip),%mm1 # 2adbf0 <const_array_of_127b> I can open another bug for the second instruction if it matters. Created attachment 42847 [details]
workaround for rex.W prefix
Attached workaround for the REX.W prefix. The PMOVMSKB instruction is just ignored for now but with these changes my application can run inside valgrind with memcheck.
|