[PATCH v2 0/7] block device fixes for large block sizes, IOCB_NOWAIT, and direct I/O · Tal Zussman <hidden> · 2026-08-28
[PATCH v2 1/7] block: use iomap_dirty_folio for block devices · Tal Zussman <hidden> · 2026-08-28
Re: [PATCH v2 1/7] block: use iomap_dirty_folio for block devices · Hannes Reinecke <hare@suse.de> · 2026-09-02
[PATCH v2 2/7] block: take i_rwsem for the direct I/O write fallback · Tal Zussman <hidden> · 2026-08-28
Re: [PATCH v2 2/7] block: take i_rwsem for the direct I/O write fallback · Hannes Reinecke <hare@suse.de> · 2026-09-02
Re: [PATCH v2 2/7] block: take i_rwsem for the direct I/O write fallback · Tal Zussman <hidden> · 2026-09-04
Re: [PATCH v2 2/7] block: take i_rwsem for the direct I/O write fallback · Christoph Hellwig <hch@lst.de> · 2026-09-07
Re: [PATCH v2 2/7] block: take i_rwsem for the direct I/O write fallback · Tal Zussman <hidden> · 2026-09-07
[PATCH v2 3/7] block: take i_rwsem for the splice read path · Tal Zussman <hidden> · 2026-08-28
Re: [PATCH v2 3/7] block: take i_rwsem for the splice read path · Hannes Reinecke <hare@suse.de> · 2026-09-02
Re: [PATCH v2 3/7] block: take i_rwsem for the splice read path · Christoph Hellwig <hch@lst.de> · 2026-09-07
[PATCH v2 4/7] block: honor IOCB_NOWAIT in the block device buffered read path · Tal Zussman <hidden> · 2026-08-28
Re: [PATCH v2 4/7] block: honor IOCB_NOWAIT in the block device buffered read path · Hannes Reinecke <hare@suse.de> · 2026-09-02
Re: [PATCH v2 4/7] block: honor IOCB_NOWAIT in the block device buffered read path · Christoph Hellwig <hch@lst.de> · 2026-09-07
Re: [PATCH v2 4/7] block: honor IOCB_NOWAIT in the block device buffered read path · Tal Zussman <hidden> · 2026-09-08
[PATCH v2 5/7] block: fail atomic writes instead of falling back to buffered I/O · Tal Zussman <hidden> · 2026-08-28
Re: [PATCH v2 5/7] block: fail atomic writes instead of falling back to buffered I/O · Hannes Reinecke <hare@suse.de> · 2026-09-02
Re: [PATCH v2 5/7] block: fail atomic writes instead of falling back to buffered I/O · Christoph Hellwig <hch@lst.de> · 2026-09-07
Re: [PATCH v2 5/7] block: fail atomic writes instead of falling back to buffered I/O · John Garry <john.g.garry@oracle.com> · 2026-09-07
[PATCH v2 6/7] block: unpin all pages of a bvec in bio_iov_iter_align_down() · Tal Zussman <hidden> · 2026-08-28
Re: [PATCH v2 6/7] block: unpin all pages of a bvec in bio_iov_iter_align_down() · Tal Zussman <hidden> · 2026-08-28
Re: [PATCH v2 6/7] block: unpin all pages of a bvec in bio_iov_iter_align_down() · Christoph Hellwig <hch@lst.de> · 2026-09-07
Re: [PATCH v2 6/7] block: unpin all pages of a bvec in bio_iov_iter_align_down() · Tal Zussman <hidden> · 2026-09-08
Re: [PATCH v2 6/7] block: unpin all pages of a bvec in bio_iov_iter_align_down() · Hannes Reinecke <hare@suse.de> · 2026-09-02
[PATCH v2 7/7] block: remove dead metadata handling from the async direct I/O path · Tal Zussman <hidden> · 2026-08-28
Re: [PATCH v2 7/7] block: remove dead metadata handling from the async direct I/O path · Christoph Hellwig <hch@lst.de> · 2026-09-07
Re: [PATCH v2 0/7] block device fixes for large block sizes, IOCB_NOWAIT, and direct I/O · Tal Zussman <hidden> · 2026-08-28
Re: [PATCH v2 0/7] block device fixes for large block sizes, IOCB_NOWAIT, and direct I/O · Tal Zussman <hidden> · 2026-09-04

Re: [PATCH v2 3/7] block: take i_rwsem for the splice read path

From: Christoph Hellwig <hch@lst.de>
Date: 2026-09-07 07:10:04
Also in: lkml

On Fri, Aug 28, 2026 at 09:49:52AM -0400, Tal Zussman wrote:

Splicing from a block device while toggling the block size between 512
bytes and 64K with BLKBSZSET hits this within seconds on a
CONFIG_DEBUG_VM kernel:

Can you wire this up to blktests?

The fix looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

`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