Re: [PATCH] Only advertise aio=io_uring if support is actually available

2022-04-25 Thread Dirk Müller
On Mittwoch, 20. April 2022 10:33:38 CEST Daniel P. Berrangé wrote: Hi Daniel, > If we're going to conditionalize this, then we really ought to be > address it fully, because 'native' is also platform specific. Good point. I can do that as well. > IOW, we would end up needing something more li

Re: [PATCH] Only advertise aio=io_uring if support is actually available

2022-04-21 Thread Dirk Müller
On Mittwoch, 20. April 2022 10:33:38 CEST Daniel P. Berrangé wrote: Hi Daniel, > That all said, the patch itself is OK, because for human targetted > interactive usage, it is desirable for --help to be representative > of what's available. > > IOW, I'm just complaining about the commit message j

Re: [PATCH] Only advertise aio=io_uring if support is actually available

2022-04-20 Thread Daniel P . Berrangé
On Tue, Apr 19, 2022 at 07:19:31PM +0200, Dirk Müller wrote: > This allows $qemu --help runtime configure checks for detecting > the host support. Note: detecting features by parsing --help output is something that is explicitly a non-goal for QEMU. The only supported interface for detecting featu

Re: [PATCH] Only advertise aio=io_uring if support is actually available

2022-04-19 Thread Eric Blake
On Tue, Apr 19, 2022 at 07:19:31PM +0200, Dirk Müller wrote: > This allows $qemu --help runtime configure checks for detecting > the host support. --help is human-parseable, and thus not stable text. Relying on it for detecting host support is not always ideal, but anything is better than nothing

[PATCH] Only advertise aio=io_uring if support is actually available

2022-04-19 Thread Dirk Müller
This allows $qemu --help runtime configure checks for detecting the host support. Signed-off-by: Dirk Müller --- block/file-posix.c | 4 qemu-nbd.c | 4 qemu-options.hx| 6 +- 3 files changed, 13 insertions(+), 1 deletion(-) diff --git a/block/file-posix.c b/block/file