[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250716184940.17b6b073@gandalf.local.home>
Date: Wed, 16 Jul 2025 18:49:40 -0400
From: Steven Rostedt <rostedt@...dmis.org>
To: Alexei Starovoitov <alexei.starovoitov@...il.com>
Cc: Peter Zijlstra <peterz@...radead.org>, Menglong Dong
<menglong.dong@...ux.dev>, Menglong Dong <menglong8.dong@...il.com>, Jiri
Olsa <jolsa@...nel.org>, bpf <bpf@...r.kernel.org>, Martin KaFai Lau
<martin.lau@...ux.dev>, Eduard Zingerman <eddyz87@...il.com>, LKML
<linux-kernel@...r.kernel.org>, Network Development
<netdev@...r.kernel.org>, "Jose E. Marchesi" <jemarch@....org>
Subject: Re: Inlining migrate_disable/enable. Was: [PATCH bpf-next v2 02/18]
x86,bpf: add bpf_global_caller for global trampoline
On Wed, 16 Jul 2025 15:35:16 -0700
Alexei Starovoitov <alexei.starovoitov@...il.com> wrote:
> 4.
> Maybe we should extend clang/gcc to support attr(preserve_access_index)
> on x86 and other architectures ;)
> We rely heavily on it in bpf backend.
> Then one can simply write:
>
> struct rq___my {
> unsigned int nr_pinned;
> } __attribute__((preserve_access_index));
>
> struct rq___my *rq;
>
> rq = this_rq();
> rq->nr_pinned++;
>
> and the compiler will do its magic of offset adjustment.
> That's how BPF CORE works.
GNU Cauldron in Porto, Portugal is having a kernel track (hopefully if it
gets accepted). I highly recommend you attending and recommending these
features. It's happening two days after Kernel Recipes (I already booked my
plane tickets).
https://gcc.gnu.org/wiki/cauldron2025
Peter, maybe you can attend too?
-- Steve
Powered by blists - more mailing lists