Attention is currently required from: laforge, pespin.
jolly has posted comments on this change by jolly. ( https://gerrit.osmocom.org/c/libosmocore/+/40725?usp=email )
Change subject: Automatically increase io_uring, if too small. ......................................................................
Patch Set 4:
(6 comments)
File src/core/osmo_io_internal.h:
https://gerrit.osmocom.org/c/libosmocore/+/40725/comment/88d1e846_2dae4314?u... : PS2, Line 110: void *read_ring[IOFD_MSGHDR_READ_SQES];
some comments here would be useful to explan what those fields are and how hey are used (not just th […]
Done
File src/core/osmo_io_uring.c:
https://gerrit.osmocom.org/c/libosmocore/+/40725/comment/71b86426_f122d2f5?u... : PS2, Line 67: g_io_uring_queue
pleaes use a more descriptive name. The variable is not a qeuue, but it is the size of the ring. […]
Done
https://gerrit.osmocom.org/c/libosmocore/+/40725/comment/e38bd57a_d54e1f06?u... : PS2, Line 113: OSMO_IO_URING_QUEUE
like the variable name in the code, the environment variable name shold be self-descriptive. […]
Done
https://gerrit.osmocom.org/c/libosmocore/+/40725/comment/720762d0_8db9c7c1?u... : PS2, Line 163: osmo_iofd_uring_get_sqe
the 'osmo_' prefix is reserverd for exported functions. […]
Done
https://gerrit.osmocom.org/c/libosmocore/+/40725/comment/86884b81_0162c7d1?u... : PS2, Line 172: LOGP(DLIO, LOGL_NOTICE, "io_uring too small to handle all SQEs, increasing size to %d.\n", g_io_uring_queue);
minor: while the current implementation alsways doubles the size, it might still be useful to print […]
Done
https://gerrit.osmocom.org/c/libosmocore/+/40725/comment/13d47890_d1b52789?u... : PS2, Line 175:
is there code to clean up the old ring after completion of the last cqe of the old ring?
No, it doesn't, because I don't track the number of submissions and the number of completions. I could do that. This would make it more complex, especially during cancellation process. I can add this if you want and test with the ttcn3 tests.
-
jolly