On 07/23/2015 01:09 PM, Christian Borntraeger wrote:
Am 17.07.2015 um 16:23 schrieb Stefan Hajnoczi:
raw-posix.c silently ignores BDRV_O_NATIVE_AIO if libaio is unavailable.
It is confusing when aio=native performance is identical to aio=threads
because the binary was accidentally built without libaio.

Print a deprecation warning if -drive aio=native is used with a binary
that does not support libaio.  There are probably users using aio=native
who would be inconvenienced if QEMU suddenly refused to start their
guests.  In the future this will become an error.

Signed-off-by: Stefan Hajnoczi <stefa...@redhat.com>
had that myself on a freshly installed system without libaio-devel.
Acked-by: Christian Borntraeger <borntrae...@de.ibm.com>


Another thing. Would it make sense to change the default to aio=native somewhen?
 From what I can tell this seems to outperform aio=threads in most cases.


this seems a good idea to me, we are always changing
from threads to native in our installations

---
  block/raw-posix.c | 11 ++++++++++-
  1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/block/raw-posix.c b/block/raw-posix.c
index 855febe..e09019c 100644
--- a/block/raw-posix.c
+++ b/block/raw-posix.c
@@ -519,7 +519,16 @@ static int raw_open_common(BlockDriverState *bs, QDict 
*options,
                       "future QEMU versions.\n",
                       bs->filename);
      }
-#endif
+#else
+    if (bdrv_flags & BDRV_O_NATIVE_AIO) {
+        error_printf("WARNING: aio=native was specified for '%s', but "
+                     "is not supported in this build. Falling back to "
+                     "aio=threads.\n"
+                     "         This will become an error condition in "
+                     "future QEMU versions.\n",
+                     bs->filename);
+    }
+#endif /* !defined(CONFIG_LINUX_AIO) */

      s->has_discard = true;
      s->has_write_zeroes = true;



Reply via email to