All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
To: Jan Kara <jack@suse.cz>
Cc: cmm@us.ibm.com, linux-ext4@vger.kernel.org
Subject: Re: [PATCH] ext4: Fix delalloc sync hang with journal lock inversion
Date: Mon, 2 Jun 2008 15:29:56 +0530	[thread overview]
Message-ID: <20080602095956.GB9225@skywalker> (raw)
In-Reply-To: <20080602093459.GC30613@duck.suse.cz>

On Mon, Jun 02, 2008 at 11:35:00AM +0200, Jan Kara wrote:
> > @@ -1052,6 +1051,7 @@ static int __mpage_da_writepage(struct page *page,
> >  		head = page_buffers(page);
> >  		bh = head;
> >  		do {
> > +
>   I guess this line is a typo.
> 

Yes, Mostly some debug lines I removed, but missed the newline added.


> >  			BUG_ON(buffer_locked(bh));
> >  			if (buffer_dirty(bh))
> >  				mpage_add_bh_to_extent(mpd, logical, bh);
> > diff --git a/mm/page-writeback.c b/mm/page-writeback.c
> > index 789b6ad..655b8bf 100644
> > --- a/mm/page-writeback.c
> > +++ b/mm/page-writeback.c
> > @@ -881,7 +881,12 @@ int write_cache_pages(struct address_space *mapping,
> >  	pagevec_init(&pvec, 0);
> >  	if (wbc->range_cyclic) {
> >  		index = mapping->writeback_index; /* Start from prev offset */
> > -		end = -1;
> > +		/*
> > +		 * write only till the specified range_end even in cyclic mode
> > +		 */
> > +		end = wbc->range_end >> PAGE_CACHE_SHIFT;
> > +		if (!end)
> > +			end = -1;
> >  	} else {
> >  		index = wbc->range_start >> PAGE_CACHE_SHIFT;
> >  		end = wbc->range_end >> PAGE_CACHE_SHIFT;
>   Are you sure you won't break other users of range_cyclic with this
> change?
>

I haven't run any specific test to verify that. The concern was that if
we force cyclic mode for writeout in delalloc we may be starting the
writeout from a different offset than specified and would be writing
more. So the changes was to use the offset specified. A quick look at
the kernel suggested most of them had range_end as 0 with cyclic_mode.
I haven't audited the full kernel. I will do that. Meanwhile if you
think it is risky to make this changes i guess we should drop this
part. But i guess we can keep the below change

+       index =  mapping->writeback_index;
+       if (!range_cyclic) {
+               /*
+                * We force cyclic write out of pages. If the
+                * caller didn't request for range_cyclic update
+                * set the writeback_index to what the caller requested.
+                */
+               mapping->writeback_index = wbc->range_start >> PAGE_CACHE_SHIFT;
+       }


-aneesh

  reply	other threads:[~2008-06-02 10:00 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-05-30 13:39 [PATCH -v2] delalloc and journal locking order inversion fixes Aneesh Kumar K.V
2008-05-30 13:39 ` [PATCH] ext4: Use page_mkwrite vma_operations to get mmap write notification Aneesh Kumar K.V
2008-05-30 13:39   ` [PATCH] ext4: Inverse locking order of page_lock and transaction start Aneesh Kumar K.V
2008-05-30 13:39     ` [PATCH] vfs: Move mark_inode_dirty() from under page lock in generic_write_end() Aneesh Kumar K.V
2008-05-30 13:39       ` [PATCH] ext4: Add validation to jbd lock inversion patch and split and writepage Aneesh Kumar K.V
2008-05-30 13:39         ` [PATCH] ext4: inverse locking ordering of page_lock and transaction start in delalloc Aneesh Kumar K.V
2008-05-30 13:39           ` [PATCH] ext4: Fix delalloc sync hang with journal lock inversion Aneesh Kumar K.V
2008-06-02  9:35             ` Jan Kara
2008-06-02  9:59               ` Aneesh Kumar K.V [this message]
2008-06-02 10:27                 ` Jan Kara
2008-06-05 13:54                   ` Aneesh Kumar K.V
2008-06-05 16:22                     ` Jan Kara
2008-06-05 19:19                       ` Aneesh Kumar K.V
2008-06-11 12:41                         ` Jan Kara
2008-06-11 13:56                           ` Aneesh Kumar K.V
2008-06-11 17:48                             ` Jan Kara
2008-06-12 23:10                             ` Mingming Cao
2008-06-02  9:31         ` [PATCH] ext4: Add validation to jbd lock inversion patch and split and writepage Jan Kara
2008-06-02  9:52           ` Aneesh Kumar K.V
2008-06-02 10:40             ` Jan Kara
2008-05-30 17:51 ` [PATCH -v2] delalloc and journal locking order inversion fixes Mingming
2008-06-01 21:10 ` [PATCH] ext4: Need clear buffer_delay after page writeout for delayed allocation Mingming Cao
2008-06-02  3:14   ` Aneesh Kumar K.V
2008-06-02  3:50     ` Mingming Cao
2008-06-02  4:09       ` Aneesh Kumar K.V
2008-06-02  5:38         ` Mingming Cao
2008-06-02  6:35           ` Aneesh Kumar K.V
2008-06-02  7:04             ` Mingming Cao
2008-06-02  8:05               ` Aneesh Kumar K.V
2008-06-03  4:43                 ` Mingming Cao
2008-06-03 10:07                   ` Aneesh Kumar K.V
  -- strict thread matches above, loose matches on Subject: below --
2008-06-06 18:24 Patches for the patchqueue Aneesh Kumar K.V
2008-06-06 18:24 ` [PATCH] ext4: cleanup blockallocator Aneesh Kumar K.V
2008-06-06 18:24   ` [PATCH] ext2: Use page_mkwrite vma_operations to get mmap write notification Aneesh Kumar K.V
2008-06-06 18:24     ` [PATCH] ext3: " Aneesh Kumar K.V
2008-06-06 18:24       ` [PATCH] vfs: Don't flush delay buffer to disk Aneesh Kumar K.V
2008-06-06 18:24         ` [PATCH] mm: Add range_cont mode for writeback Aneesh Kumar K.V
2008-06-06 18:24           ` [PATCH] ext4: Fix delalloc sync hang with journal lock inversion Aneesh Kumar K.V
2008-05-21 17:44 delalloc and journal locking order inversion fixes Aneesh Kumar K.V
2008-05-21 17:44 ` [PATCH] ext4: Add validation to jbd lock inversion patch and split and writepage Aneesh Kumar K.V
2008-05-21 17:44   ` [PATCH] ext4: inverse locking ordering of page_lock and transaction start in delalloc Aneesh Kumar K.V
2008-05-21 17:44     ` [PATCH] ext4: Fix delalloc sync hang with journal lock inversion Aneesh Kumar K.V
2008-05-22 10:25       ` Aneesh Kumar K.V
2008-05-22 17:58         ` Mingming
2008-05-22 18:23           ` Aneesh Kumar K.V
2008-05-22 19:45             ` Mingming
2008-05-22 18:10       ` Mingming
2008-05-22 18:26         ` Aneesh Kumar K.V
2008-05-22 19:26           ` Mingming

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20080602095956.GB9225@skywalker \
    --to=aneesh.kumar@linux.vnet.ibm.com \
    --cc=cmm@us.ibm.com \
    --cc=jack@suse.cz \
    --cc=linux-ext4@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.