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

List:       kde-bugs-dist
Subject:    [Bug 191182] VALGRIND_LEAK_CHECK quadratic when big nr of chunks or
From:       <philippe.waroquiers () eurocontrol ! int>
Date:       2009-05-06 18:45:36
Message-ID: 20090506184538.53FFA166E2 () immanuel ! kde ! org
[Download RAW message or body]

https://bugs.kde.org/show_bug.cgi?id=191182





--- Comment #7 from  <philippe waroquiers eurocontrol int>  2009-05-06 20:45:20 ---
It is me which is thanking you for all the nice work you do on valgrind
(including this fix).

One remark for the hash table:
There was no risk of having two exe contexts
erroneously merged, because the HTGen_Construct has two functions:
   * a hash function
   * a equal key function
So, if ever two exe contexts have the same hash key, the "get" function was
in any case verifying if the searched key is equal to the searched key.

With your fix, the hash table change is in any case not needed anymore.

However, I am asking myself if it such a more general hash table could
not avoid quite some code duplication in valgrind.
Doing a grep on */*.c, I find that there are multiple "local" hash table
implementations (e.g. in m_execontext.c, in various callgrind files).
I guess a more general hash table could reduce the code.
But for sure, the HTGenConstruct I did cannot replace all of that
out of the box.

-- 
Configure bugmail: https://bugs.kde.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
[prev in list] [next in list] [prev in thread] [next in thread]