RFC reduce CPU usage on exit for IOPOLL
Get series with b4
b4 am [email protected]Versions newest first
- v1 RFC Latest3 patches 1 replyIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/3 | io_uring: partially inline io_iopoll_getevents()fs/ | +3 | 0 replies | |
| 2/3 | io_uring: remove nr_events arg from iopoll_check()fs/ | +5 | 0 replies | |
| 3/3 | io_uring: don't burn CPU for iopoll on exitfs/ | +11 | 0 replies |
Reviews trailers per patch
No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.
Cover letter [RFC 0/3] reduce CPU usage on exit for IOPOLL
[1,2] are small preps for the future that should change nothing functionally.
[3] helps to reduce CPU usage on exit. We don't care about latency and CQEs when io_uring is going away, so checking it every HZ/20 should be enough
Please, check the assumptions, because I hope that nobody expects
->iopoll() to do real work but not just completing.
Pavel Begunkov (3):
io_uring: partially inline io_iopoll_getevents()
io_uring: remove nr_events arg from iopoll_check()
io_uring: don't burn CPU for iopoll on exit
fs/io_uring.c | 39 ++++++++++++++++++--------------------- 1 file changed, 18 insertions(+), 21 deletions(-)
Signature · 1 line
2.24.0