Linux recently added a new io_uring(7) optimization API that QEMU
doesn't take advantage of yet. The liburing library that QEMU uses
has added a corresponding new API calling io_uring_register_ring_fd().
When this API is called after creating the ring, the io_uring_submit()
library function passes a flag to the io_uring_enter(2) syscall
allowing it to skip the ring file descriptor fdget()/fdput()
operations. This saves some CPU cycles.

Signed-off-by: Sam Li <faithilikerun@gmail.com>
---
 block/io_uring.c | 12 +++++++++++-
 meson.build      |  1 +
 2 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/block/io_uring.c b/block/io_uring.c
index 0b401512b9..d48e472e74 100644
--- a/block/io_uring.c
+++ b/block/io_uring.c
@@ -18,6 +18,7 @@
 #include "qapi/error.h"
 #include "trace.h"
 
+
 /* io_uring ring size */
 #define MAX_ENTRIES 128
 
@@ -434,8 +435,17 @@ LuringState *luring_init(Error **errp)
     }
 
     ioq_init(&s->io_q);
-    return s;
+#ifdef CONFIG_LIBURING_REGISTER_RING_FD
+    if (io_uring_register_ring_fd(&s->ring) < 0) {
+        /*
+         * Only warn about this error: we will fallback to the non-optimized
+         * io_uring operations.
+         */
+        warn_report("failed to register linux io_uring ring file descriptor");
+    }
+#endif
 
+    return s;
 }
 
 void luring_cleanup(LuringState *s)
diff --git a/meson.build b/meson.build
index 9ebc00f032..9ffc74d8ed 100644
--- a/meson.build
+++ b/meson.build
@@ -1733,6 +1733,7 @@ config_host_data.set('CONFIG_LIBNFS', libnfs.found())
 config_host_data.set('CONFIG_LIBSSH', libssh.found())
 config_host_data.set('CONFIG_LINUX_AIO', libaio.found())
 config_host_data.set('CONFIG_LINUX_IO_URING', linux_io_uring.found())
+config_host_data.set('CONFIG_LIBURING_REGISTER_RING_FD', cc.has_function('io_uring_register_ring_fd', prefix: '#include <liburing.h>', dependencies:linux_io_uring))
 config_host_data.set('CONFIG_LIBPMEM', libpmem.found())
 config_host_data.set('CONFIG_NUMA', numa.found())
 config_host_data.set('CONFIG_OPENGL', opengl.found())
-- 
2.35.3

On Tue, May 31, 2022 at 06:50:11PM +0800, Sam Li wrote:
> Linux recently added a new io_uring(7) optimization API that QEMU
> doesn't take advantage of yet. The liburing library that QEMU uses
> has added a corresponding new API calling io_uring_register_ring_fd().
> When this API is called after creating the ring, the io_uring_submit()
> library function passes a flag to the io_uring_enter(2) syscall
> allowing it to skip the ring file descriptor fdget()/fdput()
> operations. This saves some CPU cycles.
> 
> Signed-off-by: Sam Li <faithilikerun@gmail.com>
> ---
>  block/io_uring.c | 12 +++++++++++-
>  meson.build      |  1 +
>  2 files changed, 12 insertions(+), 1 deletion(-)

Thanks, applied to my block tree:
https://gitlab.com/stefanha/qemu/commits/block

Stefan

Stefan Hajnoczi <stefanha@gmail.com> 于2022年5月31日周二 19:31写道：
>
> On Tue, May 31, 2022 at 06:50:11PM +0800, Sam Li wrote:
> > Linux recently added a new io_uring(7) optimization API that QEMU
> > doesn't take advantage of yet. The liburing library that QEMU uses
> > has added a corresponding new API calling io_uring_register_ring_fd().
> > When this API is called after creating the ring, the io_uring_submit()
> > library function passes a flag to the io_uring_enter(2) syscall
> > allowing it to skip the ring file descriptor fdget()/fdput()
> > operations. This saves some CPU cycles.
> >
> > Signed-off-by: Sam Li <faithilikerun@gmail.com>
> > ---
> >  block/io_uring.c | 12 +++++++++++-
> >  meson.build      |  1 +
> >  2 files changed, 12 insertions(+), 1 deletion(-)
>
> Thanks, applied to my block tree:
> https://gitlab.com/stefanha/qemu/commits/block
>

Great! Sorry about the mistake before. I'm flushed.


> Stefan

On Tue, 31 May 2022 at 12:34, Sam Li <faithilikerun@gmail.com> wrote:
>
> Stefan Hajnoczi <stefanha@gmail.com> 于2022年5月31日周二 19:31写道：
> >
> > On Tue, May 31, 2022 at 06:50:11PM +0800, Sam Li wrote:
> > > Linux recently added a new io_uring(7) optimization API that QEMU
> > > doesn't take advantage of yet. The liburing library that QEMU uses
> > > has added a corresponding new API calling io_uring_register_ring_fd().
> > > When this API is called after creating the ring, the io_uring_submit()
> > > library function passes a flag to the io_uring_enter(2) syscall
> > > allowing it to skip the ring file descriptor fdget()/fdput()
> > > operations. This saves some CPU cycles.
> > >
> > > Signed-off-by: Sam Li <faithilikerun@gmail.com>
> > > ---
> > >  block/io_uring.c | 12 +++++++++++-
> > >  meson.build      |  1 +
> > >  2 files changed, 12 insertions(+), 1 deletion(-)
> >
> > Thanks, applied to my block tree:
> > https://gitlab.com/stefanha/qemu/commits/block
> >
>
> Great! Sorry about the mistake before. I'm flushed.

Don't worry, it's easy for #ifdef issues to slip through. Two
techniques for testing #ifdefs:
1. To check whether compilation works both when the macro is defined
and also when it is undefined, edit the #ifdef to force the compiler
to take or not take the #ifdef. For example, if #ifdef FOO is compiled
but you want to test the case where FOO is undefined, temporarily
change it to #ifdef FOOX.
2. Add "#error foo" inside the #ifdef to test whether the compiler is
parsing the code. If you expect the #ifdef to be compiled but there is
no compilation error then there is a problem.

Stefan