From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from userp1040.oracle.com ([156.151.31.81]:34941 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751123AbdBJW6t (ORCPT ); Fri, 10 Feb 2017 17:58:49 -0500 Date: Fri, 10 Feb 2017 14:57:52 -0800 From: Liu Bo To: Omar Sandoval Cc: Chris Mason , linux-btrfs@vger.kernel.org, Christoph Hellwig , Pat Erley , kernel-team@fb.com Subject: Re: [PATCH] Btrfs: fix btrfs_decompress_buf2page() Message-ID: <20170210225752.GC30334@localhost.localdomain> Reply-To: bo.li.liu@oracle.com References: <20170210224609.GB30334@localhost.localdomain> <20170210225211.GA20287@vader.DHCP.thefacebook.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20170210225211.GA20287@vader.DHCP.thefacebook.com> Sender: linux-btrfs-owner@vger.kernel.org List-ID: On Fri, Feb 10, 2017 at 02:52:11PM -0800, Omar Sandoval wrote: > On Fri, Feb 10, 2017 at 02:46:09PM -0800, Liu Bo wrote: > > On Fri, Feb 10, 2017 at 12:15:11PM -0800, Omar Sandoval wrote: > > > From: Omar Sandoval > > > > > > If btrfs_decompress_buf2page() is handed a bio with its page in the > > > middle of the working buffer, then we adjust the offset into the working > > > buffer. After we copy into the bio, we advance the iterator by the > > > number of bytes we copied. Then, we have some logic to handle the case > > > of discontiguous pages and adjust the offset into the working buffer > > > again. However, if we didn't advance the bio to a new page, we may enter > > > this case in error, essentially repeating the adjustment that we already > > > made when we entered the function. The end result is bogus data in the > > > bio. > > > > > > Previously, we only checked for this case when we advanced to a new > > > page, but the conversion to bio iterators changed that. This restores > > > the old, correct behavior. > > > > The fix looks good to me, just one comment below. > > > > > > > > Fixes: 974b1adc3b10 ("btrfs: use bio iterators for the decompression handlers") > > > Reported-by: Pat Erley > > > Signed-off-by: Omar Sandoval > > > --- > > > A case I saw when testing with zlib was: > > > > > > buf_start = 42769 > > > total_out = 46865 > > > working_bytes = total_out - buf_start = 4096 > > > start_byte = 45056 > > > > > > > > > The condition (total_out > start_byte && buf_start < start_byte) is > > > true, so we adjust the offset: > > > > > > buf_offset = start_byte - buf_start = 2287 > > > working_bytes -= buf_offset = 1809 > > > current_buf_start = buf_start = 42769 > > > > > > Then, we copy > > > > > > bytes = min(bvec.bv_len, PAGE_SIZE - buf_offset, working_bytes) = 1809 > > > buf_offset += bytes = 4096 > > > working_bytes -= bytes = 0 > > > current_buf_start += bytes = 44578 > > > > > > After bio_advance(), we are still in the same page, so start_byte is the > > > same. Then, we check (total_out > start_byte && current_buf_start < start_byte), > > > which is true! So, we adjust the values again: > > > > > > buf_offset = start_byte - buf_start = 2287 > > > working_bytes = total_out - start_byte = 1809 > > > current_buf_start = buf_start + buf_offset = 45056 > > > > > > But note that working_bytes was already zero before this, so we should > > > have stopped copying. > > > > > > fs/btrfs/compression.c | 36 +++++++++++++++++++----------------- > > > 1 file changed, 19 insertions(+), 17 deletions(-) > > > > > > diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c > > > index 7f390849343b..f9f22976d77d 100644 > > > --- a/fs/btrfs/compression.c > > > +++ b/fs/btrfs/compression.c > > > @@ -1072,25 +1072,27 @@ int btrfs_decompress_buf2page(char *buf, unsigned long buf_start, > > > return 0; > > > bvec = bio_iter_iovec(bio, bio->bi_iter); > > > > > > - start_byte = page_offset(bvec.bv_page) - disk_start; > > > + if (bvec.bv_offset == 0) { > > > + start_byte = page_offset(bvec.bv_page) - disk_start; > > > > I'm not fully convinced that the next bvec's bv_offset is always > > zero, since the pages are all locked, can we keep a orig_page and > > check if (orig_page == bvec.bv_page)? > > That's a good point, that's more foolproof. I'll send a v2. With that, you can have Reviewed-by: Liu Bo Tested-by: Liu Bo We may add ASSERT(bvec.bv_offset == 0) in the if statement of (orig_page == bvec.bv_page), then we could know whether it's true :) Thanks, -liubo