[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-bugs-dist
Subject:    [valgrind] [Bug 396290] [PATCH] Possible tool - allocfail
From:       Philippe Waroquiers <bugzilla_noreply () kde ! org>
Date:       2018-07-29 20:34:36
Message-ID: bug-396290-17878-wVFKs1NhyU () http ! bugs ! kde ! org/
[Download RAW message or body]

https://bugs.kde.org/show_bug.cgi?id=396290

Philippe Waroquiers <philippe.waroquiers@skynet.be> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |philippe.waroquiers@skynet.
                   |                            |be

--- Comment #2 from Philippe Waroquiers <philippe.waroquiers@skynet.be> ---
I took a (very) quick look at the patch.

IIUC, the tool stores a sha3 of the stack traces for which it
already produced an alloc failure, so as to not make the same
stacktrace fail.
Why not store the stack traces themselves in a file?
I guess we do not mind (too much) about disk space ?
(we now that if we run the same application under e.g. memcheck, the
whole set of stacktraces will fit in valgrind memory, so the size should
be reasonable).

I guess you might then use pub_tool_execontext.h functions to store the
stacktraces,and check if a stacktrace is already stored.

Also, probably better to have a few regression tests for a tool to be
merged.
By no way this last comment is a promise that the tool will be merged :).

-- 
You are receiving this mail because:
You are watching all bug changes.=
[prev in list] [next in list] [prev in thread] [next in thread]