qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Stefan Hajnoczi <stefanha@redhat.com>
To: qemu-devel@nongnu.org
Cc: Peter Maydell <peter.maydell@linaro.org>,
	Ekaterina Tumanova <tumanova@linux.vnet.ibm.com>,
	Stefan Hajnoczi <stefanha@redhat.com>
Subject: [Qemu-devel] [PULL 32/69] raw-posix: Factor block size detection out of raw_probe_alignment()
Date: Fri, 27 Feb 2015 18:18:30 +0000	[thread overview]
Message-ID: <1425061147-1411-33-git-send-email-stefanha@redhat.com> (raw)
In-Reply-To: <1425061147-1411-1-git-send-email-stefanha@redhat.com>

From: Ekaterina Tumanova <tumanova@linux.vnet.ibm.com>

Put it in new probe_logical_blocksize().

Signed-off-by: Ekaterina Tumanova <tumanova@linux.vnet.ibm.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 1424087278-49393-3-git-send-email-tumanova@linux.vnet.ibm.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 block/raw-posix.c | 51 +++++++++++++++++++++++++++++++++++----------------
 1 file changed, 35 insertions(+), 16 deletions(-)

diff --git a/block/raw-posix.c b/block/raw-posix.c
index c0b46ca..34d403d 100644
--- a/block/raw-posix.c
+++ b/block/raw-posix.c
@@ -218,39 +218,58 @@ static int raw_normalize_devicepath(const char **filename)
 }
 #endif
 
-static void raw_probe_alignment(BlockDriverState *bs, int fd, Error **errp)
+/*
+ * Get logical block size via ioctl. On success store it in @sector_size_p.
+ */
+static int probe_logical_blocksize(int fd, unsigned int *sector_size_p)
 {
-    BDRVRawState *s = bs->opaque;
-    char *buf;
     unsigned int sector_size;
+    bool success = false;
 
-    /* For /dev/sg devices the alignment is not really used.
-       With buffered I/O, we don't have any restrictions. */
-    if (bs->sg || !s->needs_alignment) {
-        bs->request_alignment = 1;
-        s->buf_align = 1;
-        return;
-    }
+    errno = ENOTSUP;
 
     /* Try a few ioctls to get the right size */
-    bs->request_alignment = 0;
-    s->buf_align = 0;
-
 #ifdef BLKSSZGET
     if (ioctl(fd, BLKSSZGET, &sector_size) >= 0) {
-        bs->request_alignment = sector_size;
+        *sector_size_p = sector_size;
+        success = true;
     }
 #endif
 #ifdef DKIOCGETBLOCKSIZE
     if (ioctl(fd, DKIOCGETBLOCKSIZE, &sector_size) >= 0) {
-        bs->request_alignment = sector_size;
+        *sector_size_p = sector_size;
+        success = true;
     }
 #endif
 #ifdef DIOCGSECTORSIZE
     if (ioctl(fd, DIOCGSECTORSIZE, &sector_size) >= 0) {
-        bs->request_alignment = sector_size;
+        *sector_size_p = sector_size;
+        success = true;
     }
 #endif
+
+    return success ? 0 : -errno;
+}
+
+static void raw_probe_alignment(BlockDriverState *bs, int fd, Error **errp)
+{
+    BDRVRawState *s = bs->opaque;
+    char *buf;
+
+    /* For /dev/sg devices the alignment is not really used.
+       With buffered I/O, we don't have any restrictions. */
+    if (bs->sg || !s->needs_alignment) {
+        bs->request_alignment = 1;
+        s->buf_align = 1;
+        return;
+    }
+
+    bs->request_alignment = 0;
+    s->buf_align = 0;
+    /* Let's try to use the logical blocksize for the alignment. */
+    if (probe_logical_blocksize(fd, &bs->request_alignment) < 0) {
+        bs->request_alignment = 0;
+    }
 #ifdef CONFIG_XFS
     if (s->is_xfs) {
         struct dioattr da;
-- 
2.1.0

  parent reply	other threads:[~2015-02-27 18:20 UTC|newest]

Thread overview: 78+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-27 18:17 [Qemu-devel] [PULL 00/69] Block patches Stefan Hajnoczi
2015-02-27 18:17 ` [Qemu-devel] [PULL 01/69] coroutine: Fix use after free with qemu_coroutine_yield() Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 02/69] coroutine: Clean up qemu_coroutine_enter() Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 03/69] test-coroutine: Regression test for yield bug Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 04/69] vpc: Fix size in fixed image creation Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 05/69] vpc: Implement bdrv_co_get_block_status() Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 06/69] sheepdog: selectable object size support Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 07/69] block/raw-posix: fix compilation warning on OSX Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 08/69] qcow2: Remove unused struct QCowCreateState Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 09/69] virtio-blk: Check return value of blk_aio_ioctl Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 10/69] libqos: Change use of pointers to uint64_t in virtio Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 11/69] tests: Prepare virtio-blk-test for multi-arch implementation Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 12/69] libqos: Remove PCI assumptions in constants of virtio driver Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 13/69] libqos: Add malloc generic Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 14/69] libqos: Add virtio MMIO support Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 15/69] qcow2: Add two new fields to BDRVQcowState Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 16/69] qcow2: Add refcount_bits to format-specific info Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 17/69] qcow2: Do not return new value after refcount update Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 18/69] qcow2: Only return status from qcow2_get_refcount Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 19/69] qcow2: Use unsigned addend for update_refcount() Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 20/69] qcow2: Use 64 bits for refcount values Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 21/69] qcow2: Helper for refcount array reallocation Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 22/69] qcow2: Helper function for refcount modification Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 23/69] qcow2: More helpers " Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 24/69] qcow2: Open images with refcount order != 4 Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 25/69] qcow2: refcount_order parameter for qcow2_create2 Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 26/69] qcow2: Use symbolic macros in qcow2_amend_options Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 27/69] iotests: Prepare for refcount_bits option Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 28/69] qcow2: Allow creation with refcount order != 4 Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 29/69] iotests: Add test for different refcount widths Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 30/69] blkdebug: fix "once" rule Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 31/69] block: add bdrv functions for geometry and blocksize Stefan Hajnoczi
2015-02-27 18:18 ` Stefan Hajnoczi [this message]
2015-02-27 18:18 ` [Qemu-devel] [PULL 33/69] block: Add driver methods to probe blocksizes and geometry Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 34/69] block-backend: Add wrappers for blocksizes and geometry probing Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 35/69] BlockConf: Call backend functions to detect geometry and blocksizes Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 36/69] ide: start extracting ide_restart_dma out of bmdma_restart_dma Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 37/69] ide: prepare to move restart to common code Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 38/69] ide: introduce ide_register_restart_cb Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 39/69] ide: do not use BMDMA in restart callback Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 40/69] ide: pass IDEBus to the restart_cb Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 41/69] ide: move restart callback to common code Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 42/69] ide: remove restart_cb callback Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 43/69] ide: replace set_unit callback with more IDEBus state Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 44/69] ide: place initial state of the current request to IDEBus Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 45/69] ide: migrate initial request state via IDEBus Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 46/69] ide: commonize io_buffer_index initialization Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 47/69] ide: make more functions static Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 48/69] ide: support PIO restart for the ISA controller Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 49/69] ahci: Migrate IDEStatus Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 50/69] ahci: add support for restarting non-queued commands Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 51/69] ahci: Recompute cur_cmd on migrate post load Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 52/69] qtest/ide: Test flush / retry for ISA and PCI Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 53/69] libqos/ahci: Zero-fill AHCI headers Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 54/69] qtest/ahci: Add a macro bootup routine Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 55/69] libqos/ahci: add ahci command helpers Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 56/69] qtest/ahci: Add DMA test variants Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 57/69] qtest/ahci: Add PIO and LBA48 tests Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 58/69] qtest/ahci: add fragmented dma test Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 59/69] qtest/ahci: add qcow2 support to ahci-test Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 60/69] qtest/ahci: test different disk sectors Stefan Hajnoczi
2015-02-27 18:18 ` [Qemu-devel] [PULL 61/69] qtest/ahci: Add simple flush test Stefan Hajnoczi
2015-02-27 18:19 ` [Qemu-devel] [PULL 62/69] qtest/ahci: Allow override of default CLI options Stefan Hajnoczi
2015-02-27 18:19 ` [Qemu-devel] [PULL 63/69] libqtest: add qmp_eventwait Stefan Hajnoczi
2015-02-27 18:19 ` [Qemu-devel] [PULL 64/69] libqtest: add qmp_async Stefan Hajnoczi
2015-02-27 18:19 ` [Qemu-devel] [PULL 65/69] libqos: add blkdebug_prepare_script Stefan Hajnoczi
2015-02-27 18:19 ` [Qemu-devel] [PULL 66/69] qtest/ahci: add flush retry test Stefan Hajnoczi
2015-02-27 18:19 ` [Qemu-devel] [PULL 67/69] sheepdog: fix confused return values Stefan Hajnoczi
2015-02-27 18:19 ` [Qemu-devel] [PULL 68/69] libqos: Solve bug in interrupt checking when using MSIX in virtio-pci.c Stefan Hajnoczi
2015-02-27 18:19 ` [Qemu-devel] [PULL 69/69] tests: Check QVIRTIO_F_ANY_LAYOUT flag in virtio-blk test Stefan Hajnoczi
2015-02-27 18:57 ` [Qemu-devel] [PULL 00/69] Block patches Stefan Hajnoczi
2015-03-03 14:52   ` Peter Maydell
2015-03-03 17:29     ` Christian Borntraeger
2015-03-03 17:33       ` Max Reitz
2015-03-04  8:28         ` Christian Borntraeger
2015-03-04 12:03           ` Kevin Wolf
2015-03-04 16:32     ` Stefan Hajnoczi
2015-03-08  8:49 ` Peter Maydell

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=1425061147-1411-33-git-send-email-stefanha@redhat.com \
    --to=stefanha@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=tumanova@linux.vnet.ibm.com \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).