Thanks Christoph, I'll try to test it in my environment. On Thu, 2010-07-08 at 00:52 +0200, Christoph Hellwig wrote: > Currently the per-bdi writeback thread is only created when there is > dirty any dirty data on the BDI, and it lazy exists when it's been > unused for some time. > > This leads to some very complex code, and the need to keep a forker > thread around. I'd add here a comment that due to the complexity it is not bug-free - e.g., I pointed to the race between thread suicide and wake-up. Also I'd add a comment that it leads to more thread types waking up periodically and consuming power. With just one thread type it is easier to fix this issue. > This patch removes all this code and simply creates the thread as part > of the bdi registration. The downside is that we use up ressoures > for possible unused devices, although that overhead is rather low, > with 8k kernel stack size on x86 and few other, even smaller ressources. s/ressources/resources/ :-) On the other hand, there was a message that process IDs are precious resource on large NUMA systems. So, less suicides, more free IDs :-) But this is just a minor note. > If the overhead is still considered too much I can look into starting > the thread explicitly instead of as part of the bdi registration, but > that will require a bit of code complexity, too. Well, the threads can exit, then we need: 1. when adding bdi works and the thread exited, forr it. 2. the same should be done in __mark_inode_dirty - if the bdi thread is dead, create it. -- Best Regards, Artem Bityutskiy (Артём Битюцкий) -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
