Re: [patch] lockless, scalable get_pid(), for_each_process() elimination, 2.5.35-BK

William Lee Irwin III (wli@holomorphy.com)
Wed, 18 Sep 2002 10:47:29 -0700


On Wed, 18 Sep 2002, William Lee Irwin III wrote:
>> There were only 10K tasks, with likely consecutively-allocated PID's,
>> and some minor background fork()/exit() activity, but there are more
>> offenders on the read side than get_pid() itself.
>> There is no question of PID space: the full 2^30 was configured in
>> the tests done after the PID space expansion.

On Wed, Sep 18, 2002 at 10:46:35AM -0700, Linus Torvalds wrote:
> I bet the lockup was something else. There have been other bugs recently
> with the task state changes, and the lockup may just have been a regular
> plain lockup. Thread exit has been kind of fragile lately, although it
> looks better now.
> Linus

Pretty much the same tasklist iteration issues have been visible
running benchmarks on kernels from 2.4.early up. 2.5.x activity on this
front was somewhat stymied by architecture support issues whose causes
were not discovered until 2.5.20 or thereabouts, and perhaps also the
fact little time was specifically allocated to addressing it.

Bill
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/