| Summary: | Need library load address in log files | ||
|---|---|---|---|
| Product: | [Developer tools] valgrind | Reporter: | Dan Kegel <dank> |
| Component: | general | Assignee: | Julian Seward <jseward> |
| Status: | REPORTED --- | ||
| Severity: | normal | CC: | zzam |
| Priority: | NOR | ||
| Version First Reported In: | unspecified | ||
| Target Milestone: | --- | ||
| Platform: | Compiled Sources | ||
| OS: | Other | ||
| Latest Commit: | Version Fixed/Implemented In: | ||
| Sentry Crash Report: | |||
| Attachments: |
Draft patch to add load address info to xml file
Update the patch to apply to Valgrind r12230 |
||
|
Description
Dan Kegel
2009-08-24 20:35:11 UTC
Doesn't -v do what you want? --21467-- Reading syms from /home/sewardj/VgTRUNK/trunk/Inst/lib/valgrind /vgpreload_core-x86-linux.so (0x4021000) --21467-- Reading syms from /home/sewardj/VgTRUNK/trunk/Inst/lib/valgrind /vgpreload_memcheck-x86-linux.so (0x4024000) Those are the addresses of where the text segment got mapped. Yes, the bit that's missing is getting it in the xml file. For now I can either grub it out of the text log or bite the bullet and run dsymutil (which might only take 20 minutes rather than the hour I remember). Created attachment 36453 [details]
Draft patch to add load address info to xml file
Here's a quick try at the XML change.
Does it need to be conditional?
Right now, the <load_obj></load_obj> records it
emits are at top level, not enclosed in a scope like the
<suppcounts> section. That's kind of messy, the final
patch might want to do better.
Created attachment 80275 [details]
Update the patch to apply to Valgrind r12230
|