On Tue, Apr 29, 2008 at 06:03:40PM +0200, Andrea Arcangeli wrote:In case you didn't notice this already, for a further explanation of why semaphores runs slower for small critical sections and why the conversion from spinlock to rwsem should happen under a config option, see the "AIM7 40% regression with 2.6.26-rc1" thread. --
| Artem Bityutskiy | [PATCH 00/22 take 3] UBI: Unsorted Block Images |
| Greg Kroah-Hartman | [PATCH 022/196] adb: Convert from class_device to device |
| Parag Warudkar | BUG: soft lockup - CPU#1 stuck for 15s! [swapper:0] |
| Ingo Molnar | Re: pthread_create() slow for many threads; also time to revisit 64b context switc... |
git: | |
| Bill Lear | Meaning of "fatal: protocol error: bad line length character"? |
| Ludovic | `git-send-email' doesn't specify `Content-Type' |
| Miles Bader | multiple-commit cherry-pick? |
| JD Guzman | C# Git Implementation |
| Martin Schröder | Re: Real men don't attack straw men |
| Jacob Meuser | Re: esd + mpd |
| Daniel Ouellet | identifying sparse files and get ride of them trick available? |
| Todd Pytel | IDE or SCSI virtual disks for VMWare image? |
| David Miller | Re: kernel oops when system under network stress |
| Denys Fedoryshchenko | thousands of classes, e1000 TX unit hang |
| Eric Dumazet | [PATCH 6/6] fs: Introduce kern_mount_special() to mount special vfs |
| Леонид Юрьев | [r8169] patch for RTL8102 (5 new MAC/PHY) |
