Displaying 3 results from an estimated 3 matches for "821b28a".
2019 Jun 20
0
[libnbd PATCH 2/1] states: Avoid wasted send() when REPLY interrupts request
...IN failures (81%), while post-patch the run encountered
166066 pre-emptions and 552 EAGAIN failures (0.3%).
---
generator/states-issue-command.c | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/generator/states-issue-command.c b/generator/states-issue-command.c
index 821b28a..f746f80 100644
--- a/generator/states-issue-command.c
+++ b/generator/states-issue-command.c
@@ -25,12 +25,17 @@
assert (h->cmds_to_issue != NULL);
cmd = h->cmds_to_issue;
- /* Were we interrupted by reading a reply to an earlier command? */
+ /* Were we interrupted by reading a rep...
2019 Jun 25
2
Re: [libnbd PATCH 2/1] states: Avoid wasted send() when REPLY interrupts request
...the run encountered
> 166066 pre-emptions and 552 EAGAIN failures (0.3%).
> ---
> generator/states-issue-command.c | 11 ++++++++---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/generator/states-issue-command.c b/generator/states-issue-command.c
> index 821b28a..f746f80 100644
> --- a/generator/states-issue-command.c
> +++ b/generator/states-issue-command.c
> @@ -25,12 +25,17 @@
> assert (h->cmds_to_issue != NULL);
> cmd = h->cmds_to_issue;
>
> - /* Were we interrupted by reading a reply to an earlier command? */
> +...
2019 Jun 19
4
[libnbd PATCH] states: Never block state machine inside REPLY
When processing a server reply within the REPLY subgroup, we will
often hit a situation where recv() requires us to block until the next
NotifyRead. But since NotifyRead is the only permitted external action
while in this group, we are effectively blocking CmdIssue and
NotifyWrite events from happening until the server finishes the
in-progress reply, even though those events have no strict