[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: bug#7489: [coreutils] over aggressive threads in sort
From: |
Chen Guo |
Subject: |
Re: bug#7489: [coreutils] over aggressive threads in sort |
Date: |
Fri, 3 Dec 2010 12:18:46 -0800 |
Thanks Jim, that helped a lot.
I'll try out Professor Eggert's suggestion, of switching to mutexes
only at the top level merge. Of the following approaches, which would
you guys consider better practice?
1) void pointer, cast as either mutex or spinlock in lock function
2) union of mutex and spinlock, chosen in lock function
The idea is basically, in the lock function if we see the node is
MERGE_LEVEL_ROOT we'd lock the mutex either via cast or union member,
and likewise for other merge levels for spinlock.
- Re: bug#7489: [coreutils] over aggressive threads in sort, Chen Guo, 2010/12/01
- Re: bug#7489: [coreutils] over aggressive threads in sort, Chen Guo, 2010/12/02
- Re: bug#7489: [coreutils] over aggressive threads in sort, Paul Eggert, 2010/12/02
- Re: bug#7489: [coreutils] over aggressive threads in sort, Jim Meyering, 2010/12/02
- Re: bug#7489: [coreutils] over aggressive threads in sort,
Chen Guo <=
- Re: bug#7489: [coreutils] over aggressive threads in sort, Paul Eggert, 2010/12/03
- Re: bug#7489: [coreutils] over aggressive threads in sort, Chen Guo, 2010/12/06
- Re: bug#7489: [coreutils] over aggressive threads in sort, Paul Eggert, 2010/12/06
- Re: bug#7489: [coreutils] over aggressive threads in sort, Chen Guo, 2010/12/06
- Re: bug#7489: [coreutils] over aggressive threads in sort, Jim Meyering, 2010/12/07
- Re: bug#7489: [coreutils] over aggressive threads in sort, Jim Meyering, 2010/12/07
- Re: bug#7489: [coreutils] over aggressive threads in sort, Chen Guo, 2010/12/07
- [coreutils] multi-threaded sort can segfault (unrelated to the sort -u segfault), Jim Meyering, 2010/12/09
- Re: [coreutils] multi-threaded sort can segfault (unrelated to the sort -u segfault), Jim Meyering, 2010/12/09
- [coreutils] Re: multi-threaded sort can segfault (unrelated to the sort -u segfault), Paul Eggert, 2010/12/09