fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-29
[PATCH 1/4] fs: replace FOP_DIO_PARALLEL_WRITE with a fmode bits · Christoph Hellwig <hch@lst.de> · 2025-10-29
Re: [PATCH 1/4] fs: replace FOP_DIO_PARALLEL_WRITE with a fmode bits · "Darrick J. Wong" <djwong@kernel.org> · 2025-10-29
Re: [PATCH 1/4] fs: replace FOP_DIO_PARALLEL_WRITE with a fmode bits · Nirjhar Roy (IBM) <hidden> · 2025-11-04
Re: [PATCH 1/4] fs: replace FOP_DIO_PARALLEL_WRITE with a fmode bits · Christoph Hellwig <hch@lst.de> · 2025-11-05
Re: [PATCH 1/4] fs: replace FOP_DIO_PARALLEL_WRITE with a fmode bits · Christian Brauner <brauner@kernel.org> · 2025-11-11
[PATCH 2/4] fs: return writeback errors for IOCB_DONTCACHE in generic_write_sync · Christoph Hellwig <hch@lst.de> · 2025-10-29
Re: [PATCH 2/4] fs: return writeback errors for IOCB_DONTCACHE in generic_write_sync · "Darrick J. Wong" <djwong@kernel.org> · 2025-10-29
Re: [PATCH 2/4] fs: return writeback errors for IOCB_DONTCACHE in generic_write_sync · Christoph Hellwig <hch@lst.de> · 2025-10-29
Re: [PATCH 2/4] fs: return writeback errors for IOCB_DONTCACHE in generic_write_sync · "Darrick J. Wong" <djwong@kernel.org> · 2025-10-29
Re: [PATCH 2/4] fs: return writeback errors for IOCB_DONTCACHE in generic_write_sync · Christoph Hellwig <hch@lst.de> · 2025-10-30
Re: [PATCH 2/4] fs: return writeback errors for IOCB_DONTCACHE in generic_write_sync · Nirjhar Roy (IBM) <hidden> · 2025-11-04
Re: [PATCH 2/4] fs: return writeback errors for IOCB_DONTCACHE in generic_write_sync · Christoph Hellwig <hch@lst.de> · 2025-11-04
[PATCH 3/4] xfs: use IOCB_DONTCACHE when falling back to buffered writes · Christoph Hellwig <hch@lst.de> · 2025-10-29
Re: [PATCH 3/4] xfs: use IOCB_DONTCACHE when falling back to buffered writes · "Darrick J. Wong" <djwong@kernel.org> · 2025-10-29
Re: [PATCH 3/4] xfs: use IOCB_DONTCACHE when falling back to buffered writes · Nirjhar Roy (IBM) <hidden> · 2025-11-04
Re: [PATCH 3/4] xfs: use IOCB_DONTCACHE when falling back to buffered writes · Christoph Hellwig <hch@lst.de> · 2025-11-04
[PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-29
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · "Darrick J. Wong" <djwong@kernel.org> · 2025-10-29
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-29
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Qu Wenruo <hidden> · 2025-10-29
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-30
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Qu Wenruo <hidden> · 2025-10-30
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-30
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Qu Wenruo <hidden> · 2025-10-30
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-30
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Qu Wenruo <hidden> · 2025-10-30
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-30
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Nirjhar Roy (IBM) <hidden> · 2025-11-10
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-11-10
Re: [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required · Nirjhar Roy (IBM) <hidden> · 2025-11-12
Re: fall back from direct to buffered I/O when stable writes are required · Bart Van Assche <hidden> · 2025-10-29
Re: fall back from direct to buffered I/O when stable writes are required · "Darrick J. Wong" <djwong@kernel.org> · 2025-10-29
Re: fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-29
Re: fall back from direct to buffered I/O when stable writes are required · Dave Chinner <david@fromorbit.com> · 2025-10-30
Re: fall back from direct to buffered I/O when stable writes are required · Geoff Back <hidden> · 2025-10-30
Re: fall back from direct to buffered I/O when stable writes are required · Jan Kara <jack@suse.cz> · 2025-10-30
Re: fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-30
Re: fall back from direct to buffered I/O when stable writes are required · Dave Chinner <david@fromorbit.com> · 2025-10-30
Re: fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-30
Re: fall back from direct to buffered I/O when stable writes are required · Dave Chinner <david@fromorbit.com> · 2025-10-30
Re: fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-31
Re: fall back from direct to buffered I/O when stable writes are required · Keith Busch <kbusch@kernel.org> · 2025-10-31
Re: fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-10-31
Re: fall back from direct to buffered I/O when stable writes are required · Jan Kara <jack@suse.cz> · 2025-11-03
Re: fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-11-03
Re: fall back from direct to buffered I/O when stable writes are required · Keith Busch <kbusch@kernel.org> · 2025-11-03
Re: fall back from direct to buffered I/O when stable writes are required · "Darrick J. Wong" <djwong@kernel.org> · 2025-11-04
Re: fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-11-05
Re: fall back from direct to buffered I/O when stable writes are required · "Darrick J. Wong" <djwong@kernel.org> · 2025-11-05
Re: fall back from direct to buffered I/O when stable writes are required · Johannes Thumshirn <hidden> · 2025-11-06
Re: fall back from direct to buffered I/O when stable writes are required · hch <hch@lst.de> · 2025-11-06
Re: fall back from direct to buffered I/O when stable writes are required · Ming Lei <hidden> · 2025-11-12
Re: fall back from direct to buffered I/O when stable writes are required · hch <hch@lst.de> · 2025-11-12
Re: fall back from direct to buffered I/O when stable writes are required · Kevin Wolf <hidden> · 2025-11-13
Re: fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-11-14
Re: fall back from direct to buffered I/O when stable writes are required · Kevin Wolf <hidden> · 2025-11-14
Re: fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-11-14
Re: fall back from direct to buffered I/O when stable writes are required · Kevin Wolf <hidden> · 2025-11-14
Re: fall back from direct to buffered I/O when stable writes are required · Christoph Hellwig <hch@lst.de> · 2025-11-14
Re: fall back from direct to buffered I/O when stable writes are required · Kevin Wolf <hidden> · 2025-11-14

Re: fall back from direct to buffered I/O when stable writes are required

From: hch <hch@lst.de>
Date: 2025-11-12 14:39:01
Also in: linux-block, linux-fsdevel, linux-xfs, lkml

On Wed, Nov 12, 2025 at 10:18:40PM +0800, Ming Lei wrote:

Looks buffer overwrite is actually done by buggy software in guest side,
why is qemu's trouble? Or will qemu IO emulator write to the IO buffer
when guest IO is inflight?

It is ultimately caused by the guest.  But just like local software it
is in no way buggy just because you don't like the pattern.

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help