I'm not sure how much testing is expected for "Tested-by".
If just checking my perspective is enough, yes. But I did not check that
this patch fixes the problem of excessive resource consumption.
On 2022/05/12 21:50, Philippe Mathieu-Daudé wrote:
Hi Hiroki,
On Thu, May 12, 2022 at 8:57 AM
Hi Hiroki,
On Thu, May 12, 2022 at 8:57 AM 成川 弘樹 wrote:
>
> Thank you for your fix.
>
> I confirmed that after applying this patch, my intended performance
> improvement by 4c41c69e is still kept in our environment.
Is that equivalent to a formal
Tested-by: Hiroki Narukawa
tag?
> On
Am 12.05.2022 um 08:56 hat 成川 弘樹 geschrieben:
> Thank you for your fix.
>
> I confirmed that after applying this patch, my intended performance
> improvement by 4c41c69e is still kept in our environment.
This is good news. Thank you for testing the patch!
Kevin
Thank you for your fix.
I confirmed that after applying this patch, my intended performance
improvement by 4c41c69e is still kept in our environment.
On 2022/05/11 0:10, Kevin Wolf wrote:
Commit 4c41c69e changed the way the coroutine pool is sized because for
virtio-blk devices with a large
Commit 4c41c69e changed the way the coroutine pool is sized because for
virtio-blk devices with a large queue size and heavy I/O, it was just
too small and caused coroutines to be deleted and reallocated soon
afterwards. The change made the size dynamic based on the number of
queues and the queue