qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v1 3/3] virtio-pci: defer to commit kvm irq routing when enab


From: Daniel P . Berrangé
Subject: Re: [PATCH v1 3/3] virtio-pci: defer to commit kvm irq routing when enable msi/msix
Date: Tue, 28 Feb 2023 11:07:21 +0000
User-agent: Mutt/2.2.9 (2022-11-12)

On Tue, Feb 28, 2023 at 05:40:33AM -0500, Michael S. Tsirkin wrote:
> On Tue, Feb 28, 2023 at 05:39:37PM +0800, Longpeng(Mike) wrote:
> > From: Longpeng <longpeng2@huawei.com>
> > 
> > All unmasked vectors will be setup in msix_set_vector_notifiers(), which
> > is a time-consuming operation because each vector need to be submit to
> > KVM once. It's even worse if the VM has several devices and each devices
> > has dozens of vectors.
> > 
> > We can defer and commit the vectors in batch, just like the commit 
> > dc580d51f7
> > ("vfio: defer to commit kvm irq routing when enable msi/msix"),
> > 
> > The can reduce 80% of the time spending on virtio_pci_set_guest_notifiers().
> 
> cover letter also refers to 80%. what about patch 1 then? does it 
> contribute some of this gain?
> 
> > Signed-off-by: Longpeng <longpeng2@huawei.com>
> 
> In the age of language models there's no longer any excuse to post
> agrammatical commit messages.

IMHO it is not appropriate to criticize the writing of people
who may not have English as a first language. 

>                               Please just give your text to one of these
> to correct.

I'd really rather we don't suggest our contributors feed stuff
through such systems. You might have an example where its output
made sense, but there's plenty of demonstrations of such tools
outputting obvious garbage, or worse stuff that superficially
looks OK but is subtly changing the meaning. The latter is going
to be especially hard to spot for contributors without English
as a first language. IMHO it is better to have explanations
directly written by the contributor so it accurately reflects
their own understanding of the patch, even if not grammatically
perfect.

And that's ignoring the unsettled legal questions around the
licensing of output from these systems. Personally I'd suggest
our coding guidelines should explicitly reject any usage of
so called "AI" systems for QEMU contributions because of legal
uncertainty alone.

With regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|




reply via email to

[Prev in Thread] Current Thread [Next in Thread]