[lkml]   [2001]   [Oct]   [31]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Subject[PATCH][CFT] updated preempt-kernel
Penguins Rejoice:  An updated preemptible kernel patch 

Available now at
for kernels 2.4.14-pre5, 2.4.13-ac5, 2.4.13, and 2.4.12.

First, an optimization: we redesigned the preemption counter code to not
use atomic operators. This non-atomicity should result in a modest
performance gain.

Second, we have a mailing list at
(see where we discuss technical issues and the
occasional odd-bug is reported. Anyone with such interests is welcome.

Third, we have ARM support (in need of much testing) by Nico Pitre!
You will need kernel 2.4.12-ac6, Russell's ARM patch available at
the corresponding preempt-kernel patch available at
and finally the ARM-specific patch available at

Finally, the partial change log since the last public post:

- convert to non-atomic model:
o preempt_count is now an int
o memory validating is down via barrier()
o this removes many of the macros
o not using atomic ops should gain us some speed ...
- rename preempt_is_disable to preempt_is_disabled
- remove some extraneous ifdefs
- compile preempt_count into the task_struct regardless of
whether CONFIG_PREEMPT is set. This way the offsets do
not change depending on whether preemption is set.
- rediff all the patches off one codebase to insure no ambiguities,
however I may introduce a typo since the trees do differ.
- update tty race fix

- add a Documentation/preempt-locking.txt
- merge fix to protect tty race on console close. this isn't our
problem but preempt induces it. will try to merge into mainline.


Robert Love

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:12    [W:0.030 / U:2.692 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site