qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Kevin Wolf <kwolf@redhat.com>
To: qemu-block@nongnu.org
Cc: kwolf@redhat.com, stefanha@redhat.com, qemu-devel@nongnu.org
Subject: [PULL 04/33] block/file-posix: set up Linux AIO and io_uring in the current thread
Date: Thu, 21 Dec 2023 22:23:09 +0100	[thread overview]
Message-ID: <20231221212339.164439-5-kwolf@redhat.com> (raw)
In-Reply-To: <20231221212339.164439-1-kwolf@redhat.com>

From: Stefan Hajnoczi <stefanha@redhat.com>

The file-posix block driver currently only sets up Linux AIO and
io_uring in the BDS's AioContext. In the multi-queue block layer we must
be able to submit I/O requests in AioContexts that do not have Linux AIO
and io_uring set up yet since any thread can call into the block driver.

Set up Linux AIO and io_uring for the current AioContext during request
submission. We lose the ability to return an error from
.bdrv_file_open() when Linux AIO and io_uring setup fails (e.g. due to
resource limits). Instead the user only gets warnings and we fall back
to aio=threads. This is still better than a fatal error after startup.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230914140101.1065008-2-stefanha@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
 block/file-posix.c | 99 ++++++++++++++++++++++------------------------
 1 file changed, 47 insertions(+), 52 deletions(-)

diff --git a/block/file-posix.c b/block/file-posix.c
index b862406c71..0b9d4679c4 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -712,17 +712,11 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
 
 #ifdef CONFIG_LINUX_AIO
      /* Currently Linux does AIO only for files opened with O_DIRECT */
-    if (s->use_linux_aio) {
-        if (!(s->open_flags & O_DIRECT)) {
-            error_setg(errp, "aio=native was specified, but it requires "
-                             "cache.direct=on, which was not specified.");
-            ret = -EINVAL;
-            goto fail;
-        }
-        if (!aio_setup_linux_aio(bdrv_get_aio_context(bs), errp)) {
-            error_prepend(errp, "Unable to use native AIO: ");
-            goto fail;
-        }
+    if (s->use_linux_aio && !(s->open_flags & O_DIRECT)) {
+        error_setg(errp, "aio=native was specified, but it requires "
+                         "cache.direct=on, which was not specified.");
+        ret = -EINVAL;
+        goto fail;
     }
 #else
     if (s->use_linux_aio) {
@@ -733,14 +727,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
     }
 #endif /* !defined(CONFIG_LINUX_AIO) */
 
-#ifdef CONFIG_LINUX_IO_URING
-    if (s->use_linux_io_uring) {
-        if (!aio_setup_linux_io_uring(bdrv_get_aio_context(bs), errp)) {
-            error_prepend(errp, "Unable to use io_uring: ");
-            goto fail;
-        }
-    }
-#else
+#ifndef CONFIG_LINUX_IO_URING
     if (s->use_linux_io_uring) {
         error_setg(errp, "aio=io_uring was specified, but is not supported "
                          "in this build.");
@@ -2444,6 +2431,44 @@ static bool bdrv_qiov_is_aligned(BlockDriverState *bs, QEMUIOVector *qiov)
     return true;
 }
 
+static inline bool raw_check_linux_io_uring(BDRVRawState *s)
+{
+    Error *local_err = NULL;
+    AioContext *ctx;
+
+    if (!s->use_linux_io_uring) {
+        return false;
+    }
+
+    ctx = qemu_get_current_aio_context();
+    if (unlikely(!aio_setup_linux_io_uring(ctx, &local_err))) {
+        error_reportf_err(local_err, "Unable to use linux io_uring, "
+                                     "falling back to thread pool: ");
+        s->use_linux_io_uring = false;
+        return false;
+    }
+    return true;
+}
+
+static inline bool raw_check_linux_aio(BDRVRawState *s)
+{
+    Error *local_err = NULL;
+    AioContext *ctx;
+
+    if (!s->use_linux_aio) {
+        return false;
+    }
+
+    ctx = qemu_get_current_aio_context();
+    if (unlikely(!aio_setup_linux_aio(ctx, &local_err))) {
+        error_reportf_err(local_err, "Unable to use Linux AIO, "
+                                     "falling back to thread pool: ");
+        s->use_linux_aio = false;
+        return false;
+    }
+    return true;
+}
+
 static int coroutine_fn raw_co_prw(BlockDriverState *bs, int64_t *offset_ptr,
                                    uint64_t bytes, QEMUIOVector *qiov, int type)
 {
@@ -2474,13 +2499,13 @@ static int coroutine_fn raw_co_prw(BlockDriverState *bs, int64_t *offset_ptr,
     if (s->needs_alignment && !bdrv_qiov_is_aligned(bs, qiov)) {
         type |= QEMU_AIO_MISALIGNED;
 #ifdef CONFIG_LINUX_IO_URING
-    } else if (s->use_linux_io_uring) {
+    } else if (raw_check_linux_io_uring(s)) {
         assert(qiov->size == bytes);
         ret = luring_co_submit(bs, s->fd, offset, qiov, type);
         goto out;
 #endif
 #ifdef CONFIG_LINUX_AIO
-    } else if (s->use_linux_aio) {
+    } else if (raw_check_linux_aio(s)) {
         assert(qiov->size == bytes);
         ret = laio_co_submit(s->fd, offset, qiov, type,
                               s->aio_max_batch);
@@ -2567,39 +2592,13 @@ static int coroutine_fn raw_co_flush_to_disk(BlockDriverState *bs)
     };
 
 #ifdef CONFIG_LINUX_IO_URING
-    if (s->use_linux_io_uring) {
+    if (raw_check_linux_io_uring(s)) {
         return luring_co_submit(bs, s->fd, 0, NULL, QEMU_AIO_FLUSH);
     }
 #endif
     return raw_thread_pool_submit(handle_aiocb_flush, &acb);
 }
 
-static void raw_aio_attach_aio_context(BlockDriverState *bs,
-                                       AioContext *new_context)
-{
-    BDRVRawState __attribute__((unused)) *s = bs->opaque;
-#ifdef CONFIG_LINUX_AIO
-    if (s->use_linux_aio) {
-        Error *local_err = NULL;
-        if (!aio_setup_linux_aio(new_context, &local_err)) {
-            error_reportf_err(local_err, "Unable to use native AIO, "
-                                         "falling back to thread pool: ");
-            s->use_linux_aio = false;
-        }
-    }
-#endif
-#ifdef CONFIG_LINUX_IO_URING
-    if (s->use_linux_io_uring) {
-        Error *local_err = NULL;
-        if (!aio_setup_linux_io_uring(new_context, &local_err)) {
-            error_reportf_err(local_err, "Unable to use linux io_uring, "
-                                         "falling back to thread pool: ");
-            s->use_linux_io_uring = false;
-        }
-    }
-#endif
-}
-
 static void raw_close(BlockDriverState *bs)
 {
     BDRVRawState *s = bs->opaque;
@@ -3896,7 +3895,6 @@ BlockDriver bdrv_file = {
     .bdrv_co_copy_range_from = raw_co_copy_range_from,
     .bdrv_co_copy_range_to  = raw_co_copy_range_to,
     .bdrv_refresh_limits = raw_refresh_limits,
-    .bdrv_attach_aio_context = raw_aio_attach_aio_context,
 
     .bdrv_co_truncate                   = raw_co_truncate,
     .bdrv_co_getlength                  = raw_co_getlength,
@@ -4266,7 +4264,6 @@ static BlockDriver bdrv_host_device = {
     .bdrv_co_copy_range_from = raw_co_copy_range_from,
     .bdrv_co_copy_range_to  = raw_co_copy_range_to,
     .bdrv_refresh_limits = raw_refresh_limits,
-    .bdrv_attach_aio_context = raw_aio_attach_aio_context,
 
     .bdrv_co_truncate                   = raw_co_truncate,
     .bdrv_co_getlength                  = raw_co_getlength,
@@ -4402,7 +4399,6 @@ static BlockDriver bdrv_host_cdrom = {
     .bdrv_co_pwritev        = raw_co_pwritev,
     .bdrv_co_flush_to_disk  = raw_co_flush_to_disk,
     .bdrv_refresh_limits    = cdrom_refresh_limits,
-    .bdrv_attach_aio_context = raw_aio_attach_aio_context,
 
     .bdrv_co_truncate                   = raw_co_truncate,
     .bdrv_co_getlength                  = raw_co_getlength,
@@ -4528,7 +4524,6 @@ static BlockDriver bdrv_host_cdrom = {
     .bdrv_co_pwritev        = raw_co_pwritev,
     .bdrv_co_flush_to_disk  = raw_co_flush_to_disk,
     .bdrv_refresh_limits    = cdrom_refresh_limits,
-    .bdrv_attach_aio_context = raw_aio_attach_aio_context,
 
     .bdrv_co_truncate                   = raw_co_truncate,
     .bdrv_co_getlength                  = raw_co_getlength,
-- 
2.43.0



  parent reply	other threads:[~2023-12-21 21:24 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-21 21:23 [PULL 00/33] Block layer patches Kevin Wolf
2023-12-21 21:23 ` [PULL 01/33] nbd/server: avoid per-NBDRequest nbd_client_get/put() Kevin Wolf
2023-12-21 21:23 ` [PULL 02/33] nbd/server: only traverse NBDExport->clients from main loop thread Kevin Wolf
2023-12-21 21:23 ` [PULL 03/33] nbd/server: introduce NBDClient->lock to protect fields Kevin Wolf
2023-12-21 21:23 ` Kevin Wolf [this message]
2023-12-21 21:23 ` [PULL 05/33] virtio-blk: add lock to protect s->rq Kevin Wolf
2023-12-21 21:23 ` [PULL 06/33] virtio-blk: don't lock AioContext in the completion code path Kevin Wolf
2023-12-21 21:23 ` [PULL 07/33] virtio-blk: don't lock AioContext in the submission " Kevin Wolf
2023-12-21 21:23 ` [PULL 08/33] block: Fix crash when loading snapshot on inactive node Kevin Wolf
2023-12-21 21:23 ` [PULL 09/33] vl: Improve error message for conflicting -incoming and -loadvm Kevin Wolf
2023-12-21 21:23 ` [PULL 10/33] iotests: Basic tests for internal snapshots Kevin Wolf
2023-12-21 21:23 ` [PULL 11/33] scsi: only access SCSIDevice->requests from one thread Kevin Wolf
2024-01-23 16:40   ` Hanna Czenczek
2024-01-23 17:10     ` Kevin Wolf
2024-01-23 17:23       ` Hanna Czenczek
2024-01-24 12:12       ` Hanna Czenczek
2024-01-24 21:53         ` Stefan Hajnoczi
2024-01-25  9:06           ` Hanna Czenczek
2024-01-25 13:25             ` Stefan Hajnoczi
2024-01-25 17:32       ` Hanna Czenczek
2024-01-26 13:18         ` Kevin Wolf
2024-01-26 15:24           ` Hanna Czenczek
2024-01-31 20:35             ` Stefan Hajnoczi
2024-02-01 14:10               ` Hanna Czenczek
2024-02-01 14:28                 ` Stefan Hajnoczi
2024-02-01 15:25                   ` Hanna Czenczek
2024-02-01 15:49                     ` Hanna Czenczek
2024-02-02 12:32                     ` Hanna Czenczek
2024-02-06 19:32                       ` Stefan Hajnoczi
2024-01-29 16:30       ` Hanna Czenczek
2024-01-31 10:17         ` Kevin Wolf
2024-02-01  9:43           ` Hanna Czenczek
2024-02-01 10:21             ` Kevin Wolf
2024-02-01 10:35               ` Hanna Czenczek
2024-01-23 17:21     ` Hanna Czenczek
2023-12-21 21:23 ` [PULL 12/33] virtio-scsi: don't lock AioContext around virtio_queue_aio_attach_host_notifier() Kevin Wolf
2023-12-21 21:23 ` [PULL 13/33] scsi: don't lock AioContext in I/O code path Kevin Wolf
2023-12-21 21:23 ` [PULL 14/33] dma-helpers: don't lock AioContext in dma_blk_cb() Kevin Wolf
2023-12-21 21:23 ` [PULL 15/33] virtio-scsi: replace AioContext lock with tmf_bh_lock Kevin Wolf
2023-12-21 21:23 ` [PULL 16/33] scsi: assert that callbacks run in the correct AioContext Kevin Wolf
2023-12-21 21:23 ` [PULL 17/33] tests: remove aio_context_acquire() tests Kevin Wolf
2023-12-21 21:23 ` [PULL 18/33] aio: make aio_context_acquire()/aio_context_release() a no-op Kevin Wolf
2023-12-21 21:23 ` [PULL 19/33] graph-lock: remove AioContext locking Kevin Wolf
2023-12-21 21:23 ` [PULL 20/33] block: " Kevin Wolf
2023-12-21 21:23 ` [PULL 21/33] block: remove bdrv_co_lock() Kevin Wolf
2023-12-21 21:23 ` [PULL 22/33] scsi: remove AioContext locking Kevin Wolf
2023-12-21 21:23 ` [PULL 23/33] aio-wait: draw equivalence between AIO_WAIT_WHILE() and AIO_WAIT_WHILE_UNLOCKED() Kevin Wolf
2023-12-21 21:23 ` [PULL 24/33] aio: remove aio_context_acquire()/aio_context_release() API Kevin Wolf
2023-12-21 21:23 ` [PULL 25/33] docs: remove AioContext lock from IOThread docs Kevin Wolf
2023-12-21 21:23 ` [PULL 26/33] scsi: remove outdated AioContext lock comment Kevin Wolf
2023-12-21 21:23 ` [PULL 27/33] job: remove outdated AioContext locking comments Kevin Wolf
2023-12-21 21:23 ` [PULL 28/33] block: " Kevin Wolf
2023-12-21 21:23 ` [PULL 29/33] block-coroutine-wrapper: use qemu_get_current_aio_context() Kevin Wolf
2023-12-21 21:23 ` [PULL 30/33] string-output-visitor: show structs as "<omitted>" Kevin Wolf
2023-12-21 21:23 ` [PULL 31/33] qdev-properties: alias all object class properties Kevin Wolf
2023-12-21 21:23 ` [PULL 32/33] qdev: add IOThreadVirtQueueMappingList property type Kevin Wolf
2023-12-21 21:23 ` [PULL 33/33] virtio-blk: add iothread-vq-mapping parameter Kevin Wolf

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=20231221212339.164439-5-kwolf@redhat.com \
    --to=kwolf@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.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).