From: David Marchand <david.marchand@redhat.com>
To: Bruce Richardson <bruce.richardson@intel.com>
Cc: dev@dpdk.org, Maxime Coquelin <maxime.coquelin@redhat.com>
Subject: Re: [PATCH] build: remove obsolete compiler checks
Date: Fri, 18 Jul 2025 15:36:19 +0200 [thread overview]
Message-ID: <CAJFAV8wG+Y5f_G9t0jOGnyvq0PLwYLVWOHDFPXeQRHPF-tZ4OQ@mail.gmail.com> (raw)
In-Reply-To: <aHpGsX_il5_oRWBv@bricha3-mobl1.ger.corp.intel.com>
On Fri, Jul 18, 2025 at 3:06 PM Bruce Richardson
<bruce.richardson@intel.com> wrote:
>
> On Fri, Jul 18, 2025 at 02:45:48PM +0200, David Marchand wrote:
> > On Fri, Jul 18, 2025 at 2:14 PM Bruce Richardson
> > <bruce.richardson@intel.com> wrote:
> > > diff --git a/drivers/net/virtio/meson.build b/drivers/net/virtio/meson.build
> > > index d3caa3a3b4..a9ff87e863 100644
> > > --- a/drivers/net/virtio/meson.build
> > > +++ b/drivers/net/virtio/meson.build
> > > @@ -31,7 +31,7 @@ if arch_subdir == 'x86'
> > > sources_avx512 += files('virtio_rxtx_packed.c')
> > > if (toolchain == 'gcc' and cc.version().version_compare('>=8.3.0'))
> > > cflags += '-DVIRTIO_GCC_UNROLL_PRAGMA'
> > > - elif (toolchain == 'clang' and cc.version().version_compare('>=3.7.0'))
> > > + elif (toolchain == 'clang')
> > > cflags += '-DVIRTIO_CLANG_UNROLL_PRAGMA'
> > > endif
> > > endif
> >
> > One other nit, not blocking.
> >
> > There should be no need for special casing clang vs gcc, since clang
> > supports gcc syntax in general.
> > https://clang.llvm.org/docs/AttributeReference.html#pragma-unroll-pragma-nounroll
> >
> I was actually thinking of doing a follow-up patch to remove the "if" and
> instead do:
>
> cflags += '-DVIRTIO_' + toolchain.to_upper() + "_UNROLL_PRAGMA"
>
> on the basis that having a define for any other unknown compilers would be
> harmless. However, you are right that there seems to be little reason to do
> this in meson.build, and the code can do it directly itself. I'll leave
> this part as it is in the patch for now.
Cc: Maxime (who is off atm)
I wonder if this pragma stuff really helps..
One quick test on the loop in vhost_flush_enqueue_batch_packed() shows
the same generated code with gcc 10 and gcc 15 when compiling with -O3
(thanks to godbolt.org).
Let's leave it as is for now, but I would be for dropping this strange
construct, only used with the virtio ring packed layout iirc.
--
David Marchand
prev parent reply other threads:[~2025-07-18 13:36 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-07-18 12:14 Bruce Richardson
2025-07-18 12:21 ` David Marchand
2025-07-18 12:34 ` Bruce Richardson
2025-07-18 12:45 ` David Marchand
2025-07-18 13:05 ` Bruce Richardson
2025-07-18 13:36 ` David Marchand [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAJFAV8wG+Y5f_G9t0jOGnyvq0PLwYLVWOHDFPXeQRHPF-tZ4OQ@mail.gmail.com \
--to=david.marchand@redhat.com \
--cc=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=maxime.coquelin@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).