man/io_uring_cq_has_overflow: mention IORING_FEAT_NODROP dependency

Only valid for kernels that suport FEAT_NODROP - which is basically anything that isn't 5.4 at this point. If you're running on a kernel without NODROP, you really should consider upgrading. Not necessarily because of the NODROP handling alone, but because newer kernels will provide much better performance. Link: #971 Signed-off-by: Jens Axboe <[email protected]>
axboe · Oct 18, 2023 · 7264f1f · 7264f1f
1 parent e307076
commit 7264f1f
Showing 1 changed file with 7 additions and 1 deletion.
diff --git a/man/io_uring_cq_has_overflow.3 b/man/io_uring_cq_has_overflow.3
@@ -18,7 +18,13 @@ The
 function informs the application if CQ entries have overflowed and are waiting to be flushed to
 the CQ ring. For example using
 .BR io_uring_get_events (3)
-.
+.SH NOTES
+Using this function is only valid if the ring has
+.B IORING_FEAT_NODROP
+set, as it's checking for a flag set by kernels supporting that feature. For
+really old kernels that don't support this feature, if CQE overflow is
+experienced the CQEs are lost. If that happens, the CQ ring overflow offset
+will get incremented.
 .SH RETURN VALUE
 True if there are CQ entries waiting to be flushed to the CQ ring.
 .SH SEE ALSO