qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Qemu-devel] [PATCH v2 00/18] Support mismatched host and guest logical


From: Paolo Bonzini
Subject: [Qemu-devel] [PATCH v2 00/18] Support mismatched host and guest logical block sizes
Date: Thu, 26 Jan 2012 18:22:31 +0100

As in v1, the series is logically split in two parts.

The first six patches ensure that BDRV_O_NO_CACHE is set for all non-file
protocols.  This makes sure that large sectors are handled correctly
for the libiscsi backend.

The next patches add support for arbitrary combinations of host/guest
logical block size even for cache=none.  Having guest_block_size >
host_block_size is unsafe in the presence of power failures, while
having guest_block_size < host_block_size is just (possibly) slower.

Physical block size and minimum I/O size are set to the host block
size in order to minimize ill performance behavior.  This is a small
guest-visible change, but only if you were hosting your images on a
4k-sector partition _and_ using cache != none:

- 4k-sectors and cache=none used to fail horribly

- filesystem images with cache != none cannot detect they are on a
  4k-sector disk, because BLKSSZGET fails

This should be rare enough that it's bearable to trade this for better
defaults.

v1->v2: new patch to set min_io_size and please XFS;
    aesthetic changes to patches 2 and 3 suggested by Zhi;
    rebase around copy-on-read changes.

Paolo Bonzini (18):
  block: do not rely on open_flags for bdrv_is_snapshot
  block: store actual flags in bs->open_flags
  block: pass protocol flags up to the format
  block: non-raw protocols never cache
  block: remove enable_write_cache
  block: move flag bits together
  raw: remove the aligned_buf
  block: rename buffer_alignment to guest_block_size
  block: add host_block_size
  raw: probe host_block_size
  iscsi: save host block size
  block: allow waiting only for overlapping writes
  block: allow waiting at arbitrary granularity
  block: protect against "torn reads" for guest_block_size > host_block_size
  block: align and serialize I/O when guest_block_size < host_block_size
  block: default physical block size to host block size
  qemu-io: add blocksize argument to open
  block: default min_io_size to host block size when doing rmw

 Makefile.objs     |    4 +-
 block.c           |  336 ++++++++++++++++++++++++++++++++++++++++++++++-------
 block.h           |   18 +---
 block/curl.c      |    1 +
 block/iscsi.c     |    3 +
 block/nbd.c       |    1 +
 block/raw-posix.c |   97 +++++++++------
 block/raw-win32.c |   42 +++++++
 block/rbd.c       |    1 +
 block/sheepdog.c  |    1 +
 block/vdi.c       |    1 +
 block_int.h       |   25 ++--
 hw/ide/core.c     |    2 +-
 hw/scsi-disk.c    |    4 +-
 hw/scsi-generic.c |    2 +-
 hw/virtio-blk.c   |    4 +-
 qemu-io.c         |   33 +++++-
 trace-events      |    1 +
 18 files changed, 452 insertions(+), 124 deletions(-)

-- 
1.7.7.6




reply via email to

[Prev in Thread] Current Thread [Next in Thread]