Re: 2.6.9-rc2-mm2

[William Lee Irwin III]

Jaakko Hyv?tti <jaakko@xxxxxxxxxxxxxx> wrote:
>> Forgive me for asking a question that probably enough research would
>> answer, but which exact patch of those listed does fix this problem?  I
>> cannot find the right one myself, and I would like to just address this
>> problem that has haunted me at least since 2.6.6, I guess.  Or is the fix
>> too interdependent with other changes?

On Sat, Oct 02, 2004 at 12:49:38AM -0700, Andrew Morton wrote:
> It was wait_on_bit-must-loop.patch.
> But that simply fixes a bug which was introduced into an earlier
> 2.6.9-rcX-mmY kernel.  The bug is certainly not present in any Linus
> kernel, nor in any 2.6.6/7/8 kernel.
> So you're seeing something different.  Please send a full report.

Well, for the record, it's that otherwise one would have to honor the
invariant of no spurious wakeups, which e.g. jbd does not. There
actually was API forethought put into timeouts, but jbd uses a timer
and wake_up_process() instead of schedule_timeout(), and is nontrivial
to convert. It was also anticipated that e.g. other users may trip up.


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