qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Kevin Wolf <kwolf@redhat.com>
To: anthony@codemonkey.ws
Cc: kwolf@redhat.com, qemu-devel@nongnu.org
Subject: [Qemu-devel] [PATCH 19/24] linux-aio: Allow reads beyond the end of growable images
Date: Fri, 14 Oct 2011 18:49:14 +0200	[thread overview]
Message-ID: <1318610959-17971-20-git-send-email-kwolf@redhat.com> (raw)
In-Reply-To: <1318610959-17971-1-git-send-email-kwolf@redhat.com>

This is the linux-aio version of commits 22afa7b5 (raw-posix, synchronous) and
ba1d1afd (posix-aio-compat). Reads now produce zeros after the end of file
instead of failing or resulting in short reads, making linux-aio compatible
with the behaviour of synchronous raw-posix requests and posix-aio-compat.

The problem can be reproduced like this:

dd if=/dev/zero of=/tmp/test.raw bs=1 count=1234
./qemu-io -k -n -g -c 'read -p 1024 512' /tmp/test.raw

Previously, the result of this was 'read failed: Invalid argument', now the
read completes successfully.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
---
 linux-aio.c |   17 ++++++++++++++---
 1 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/linux-aio.c b/linux-aio.c
index 50da75d..1c635ef 100644
--- a/linux-aio.c
+++ b/linux-aio.c
@@ -31,6 +31,8 @@ struct qemu_laiocb {
     struct iocb iocb;
     ssize_t ret;
     size_t nbytes;
+    QEMUIOVector *qiov;
+    bool is_read;
     QLIST_ENTRY(qemu_laiocb) node;
 };
 
@@ -57,10 +59,17 @@ static void qemu_laio_process_completion(struct qemu_laio_state *s,
 
     ret = laiocb->ret;
     if (ret != -ECANCELED) {
-        if (ret == laiocb->nbytes)
+        if (ret == laiocb->nbytes) {
             ret = 0;
-        else if (ret >= 0)
-            ret = -EINVAL;
+        } else if (ret >= 0) {
+            /* Short reads mean EOF, pad with zeros. */
+            if (laiocb->is_read) {
+                qemu_iovec_memset_skip(laiocb->qiov, 0,
+                    laiocb->qiov->size - ret, ret);
+            } else {
+                ret = -EINVAL;
+            }
+        }
 
         laiocb->common.cb(laiocb->common.opaque, ret);
     }
@@ -162,6 +171,8 @@ BlockDriverAIOCB *laio_submit(BlockDriverState *bs, void *aio_ctx, int fd,
     laiocb->nbytes = nb_sectors * 512;
     laiocb->ctx = s;
     laiocb->ret = -EINPROGRESS;
+    laiocb->is_read = (type == QEMU_AIO_READ);
+    laiocb->qiov = qiov;
 
     iocbs = &laiocb->iocb;
 
-- 
1.7.6.4

  parent reply	other threads:[~2011-10-14 16:46 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-10-14 16:48 [Qemu-devel] [PULL 00/24] Block patches Kevin Wolf
2011-10-14 16:48 ` [Qemu-devel] [PATCH 01/24] block: allow resizing of images residing on host devices Kevin Wolf
2011-10-14 16:48 ` [Qemu-devel] [PATCH 02/24] linux-aio: Fix laio_submit error handling Kevin Wolf
2011-10-14 16:48 ` [Qemu-devel] [PATCH 03/24] block: Keep track of devices' I/O status Kevin Wolf
2011-10-14 16:48 ` [Qemu-devel] [PATCH 04/24] virtio: Support " Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 05/24] ide: " Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 06/24] scsi: " Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 07/24] QMP: query-status: Add 'io-status' key Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 08/24] HMP: Print 'io-status' information Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 09/24] block/vvfat: Fix potential memory leaks and other memory errors Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 10/24] block/vvfat: Remove unused code Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 11/24] vvfat: Fix potential buffer overflow Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 12/24] block: directly invoke .bdrv_aio_*() in bdrv_co_io_em() Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 13/24] block: directly invoke .bdrv_* from emulation functions Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 14/24] block: split out bdrv_co_do_readv() and bdrv_co_do_writev() Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 15/24] block: switch bdrv_read()/bdrv_write() to coroutines Kevin Wolf
2011-10-24 15:12   ` Pierre Riteau
2011-10-14 16:49 ` [Qemu-devel] [PATCH 16/24] block: switch bdrv_aio_readv() " Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 17/24] block: mark blocks dirty on coroutine write completion Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 18/24] block: switch bdrv_aio_writev() to coroutines Kevin Wolf
2011-10-14 16:49 ` Kevin Wolf [this message]
2011-10-14 16:49 ` [Qemu-devel] [PATCH 20/24] block: drop emulation functions that use coroutines Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 21/24] raw-posix: remove bdrv_read()/bdrv_write() Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 22/24] block: use coroutine interface for raw format Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 23/24] block: drop .bdrv_read()/.bdrv_write() emulation Kevin Wolf
2011-10-14 16:49 ` [Qemu-devel] [PATCH 24/24] block: drop bdrv_has_async_rw() Kevin Wolf
2011-10-14 17:48 ` [Qemu-devel] [PULL 00/24] Block patches Anthony Liguori

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=1318610959-17971-20-git-send-email-kwolf@redhat.com \
    --to=kwolf@redhat.com \
    --cc=anthony@codemonkey.ws \
    --cc=qemu-devel@nongnu.org \
    /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).