On 12/8/25 3:12 AM, Henson Choi wrote:
The patch is minimal (adds one function, modifies one condition) and
follows the existing pattern used for injection points testing.
Thoughts?
I feel this patch is also related to the the current proposed patch[1]
for giving a proper error on kernel versions older than 5.6. If we
actually want to automatically skip tests if we cannot run io_uring on a
machine it feels more important to skip them on a kernel which does not
support all io_uring ops that we need.
I have not yet made up my mind about automatic skipping when we lack
permissions. Automatically skipping tests can hide issues since you may
not even notice that it happened. Maybe we should call skip if we lack
permissions so it is visible?
1.
https://www.postgresql.org/message-id/17879f12-609b-4730-bfff-6dd11a46d5e6%40gmail.com
Andreas
