From mboxrd@z Thu Jan 1 00:00:00 1970 Subject: Re: Why kmem_cache_free occupy CPU for more than 10 seconds? From: Peter Zijlstra In-Reply-To: <20070412005535.951ccb4b.akpm@linux-foundation.org> References: <20070411153040.a7e6c3b8.akpm@linux-foundation.org> <1176363565.6893.73.camel@twins> <20070412005535.951ccb4b.akpm@linux-foundation.org> Content-Type: text/plain Date: Thu, 12 Apr 2007 11:19:56 +0200 Message-Id: <1176369596.6893.105.camel@twins> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: Andrew Morton Cc: Zhao Forrest , linux-mm@kvack.org, linux-kernel@vger.kernel.org List-ID: On Thu, 2007-04-12 at 00:55 -0700, Andrew Morton wrote: > On Thu, 12 Apr 2007 09:39:25 +0200 Peter Zijlstra wrote: > > > On Wed, 2007-04-11 at 15:30 -0700, Andrew Morton wrote: > > > > > There used to be a cond_resched() in invalidate_mapping_pages() which would > > > have prevented this, but I rudely removed it to support > > > /proc/sys/vm/drop_caches (which needs to call invalidate_inode_pages() > > > under spinlock). > > > > > > We could resurrect that cond_resched() by passing in some flag, I guess. > > > Or change the code to poke the softlockup detector. The former would be > > > better. > > > > cond_resched() is conditional on __resched_legal(0) > > What's __resched_legal()? Hmm, I was looking at the RHEL-5 sources, it seems this function was removed later (commmit 9414232fa0cc28e2f51b8c76d260f2748f7953fc). > > which should take > > care of being called under a spinlock. > > We only increment preempt_count() in spin_lock() if CONFIG_PREEMPT. Right, ok, that makes the whole thing fall apart indeed. > > Index: linux-2.6-mm/mm/truncate.c > > =================================================================== > > --- linux-2.6-mm.orig/mm/truncate.c > > +++ linux-2.6-mm/mm/truncate.c > > @@ -292,6 +292,8 @@ unsigned long invalidate_mapping_pages(s > > pgoff_t index; > > int lock_failed; > > > > + cond_resched(); > > + > > lock_failed = TestSetPageLocked(page); > > Is deadlocky on the drop_caches path and if CONFIG_PREEMPT we'll get > scheduling-in-spinlock warnings. *nod* > For the blkdev_close() path the change is unneeded if CONFIG_PREEMPT and > will fix things if !CONFIG_PREEMPT. > > We can presumably just remove the invalidate_mapping_pages() call from the > kill_bdev() path (at least) - kill_bdev()'s truncate_inode_pages() will do > the same thing. Certainly looks that way, I'll see if I can spot a hole in that. > It might be time to remove that unused-for-six-years destroy_dirty_buffers > too. ok. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org