From: "Morten Brørup" <mb@smartsharesystems.com>
To: "Wathsala Vithanage" <wathsala.vithanage@arm.com>,
<stephen@networkplumber.org>
Cc: <dev@dpdk.org>
Subject: RE: [PATCH] mbuf: optimize segment prefree
Date: Mon, 6 Oct 2025 20:26:07 +0200 [thread overview]
Message-ID: <98CBD80474FA8B44BF855DF32C47DC35F654A4@smartserver.smartshare.dk> (raw)
In-Reply-To: <9428c18d-fec7-4400-a4f3-b2614b7fb058@arm.com>
> From: Wathsala Vithanage [mailto:wathsala.vithanage@arm.com]
> Sent: Monday, 6 October 2025 19.46
>
> On 8/27/25 18:17, Stephen Hemminger wrote:
> > On Wed, 27 Aug 2025 21:35:34 +0000
> > Morten Brørup <mb@smartsharesystems.com> wrote:
> >
> >>
> >> +/* GCC only optimizes single-bit MSB tests this way, so do it by
> hand with multi-bit. */
> >> +#if defined(RTE_TOOLCHAIN_GCC) && defined(RTE_ARCH_X86)
> >> +#undef RTE_MBUF_DIRECT
> >> +#define RTE_MBUF_DIRECT(mb) \
> >> + (!(((const uint8_t *)(mb))[offsetof(struct rte_mbuf, ol_flags) +
> 7] & \
> >> + (uint8_t)((RTE_MBUF_F_INDIRECT | RTE_MBUF_F_EXTERNAL) >> 56)))
> >> +#endif
> > Complex enough expression that I would prefer this be an inline
> function
> > with some more temporary variables and more comments.
> > Like the magic 7 for mask??
>
> +1
So, instead of overriding the macro definition in the GCC exception case, you prefer something like:
#if defined(RTE_TOOLCHAIN_GCC) && defined(RTE_ARCH_X86)
static __rte_always_inline
RTE_MBUF_DIRECT(const struct rte_mbuf * const mb)
{...}
#else
#define RTE_MBUF_DIRECT(mb) \
(!((mb)->ol_flags & (RTE_MBUF_F_INDIRECT | RTE_MBUF_F_EXTERNAL)))
#endif
which would give us a function with a macro-like (upper case) name.
How about I just add a more detailed description to this macro?
After all, formally, it's an exception to the simple default macro.
next prev parent reply other threads:[~2025-10-06 18:26 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-08-27 21:35 Morten Brørup
2025-08-27 23:17 ` Stephen Hemminger
2025-10-06 17:46 ` Wathsala Vithanage
2025-10-06 18:26 ` Morten Brørup [this message]
2025-10-06 14:49 ` Morten Brørup
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=98CBD80474FA8B44BF855DF32C47DC35F654A4@smartserver.smartshare.dk \
--to=mb@smartsharesystems.com \
--cc=dev@dpdk.org \
--cc=stephen@networkplumber.org \
--cc=wathsala.vithanage@arm.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).