Received-SPF: pass (google.com: domain of linux-kernel+bounces-156152-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Precedence: bulk
MIME-Version: 1.0
References: <20240407090558.3395-1-jiangshanlai@gmail.com> <20240407090558.3395-12-jiangshanlai@gmail.com>
 <CAEXW_YTin=EgjoypCD=eYh-NLRwxbhdr+B-jBO5TpC4qcef2_Q@mail.gmail.com>
In-Reply-To: <CAEXW_YTin=EgjoypCD=eYh-NLRwxbhdr+B-jBO5TpC4qcef2_Q@mail.gmail.com>
From: Lai Jiangshan <jiangshanlai@gmail.com>
Date: Wed, 24 Apr 2024 10:43:05 +0800
Message-ID: <CAJhGHyBt8y46i9fkpCDS720khgZQhRD5NTfsW2X0U41Edcc=FA@mail.gmail.com>
Subject: Re: [PATCH V2 11/11] x86/rcu: Add THUNK rcu_read_unlock_special_thunk
To: Joel Fernandes <joel@joelfernandes.org>
Cc: linux-kernel@vger.kernel.org, rcu@vger.kernel.org, x86@kernel.org, 
	Lai Jiangshan <jiangshan.ljs@antgroup.com>, "Paul E. McKenney" <paulmck@kernel.org>, 
	Peter Zijlstra <peterz@infradead.org>, Frederic Weisbecker <frederic@kernel.org>, 
	Andy Lutomirski <luto@kernel.org>, Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@redhat.com>, 
	Borislav Petkov <bp@alien8.de>, Dave Hansen <dave.hansen@linux.intel.com>, 
	"H. Peter Anvin" <hpa@zytor.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Wed, Apr 24, 2024 at 1:26=E2=80=AFAM Joel Fernandes <joel@joelfernandes.=
org> wrote:
>
> Hello Lai,
>
> On Sun, Apr 7, 2024 at 5:07=E2=80=AFAM Lai Jiangshan <jiangshanlai@gmail.=
com> wrote:
> >
> > From: Lai Jiangshan <jiangshan.ljs@antgroup.com>
> >
> > Add rcu_read_unlock_special_thunk(), so that the inlined rcu_read_unloc=
k()
> > doesn't need any code to save the caller-saved registers.
> >
> > Make rcu_read_unlock() only two instructions in the slow path at the
> > caller site.
> >
> > Cc: "Paul E. McKenney" <paulmck@kernel.org>
> > Cc: Peter Zijlstra <peterz@infradead.org>
> > Cc: Frederic Weisbecker <frederic@kernel.org>
> > Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
> > ---
> >  arch/x86/entry/thunk.S             | 5 +++++
> >  arch/x86/include/asm/rcu_preempt.h | 4 +++-
> >  2 files changed, 8 insertions(+), 1 deletion(-)
> >
> > diff --git a/arch/x86/entry/thunk.S b/arch/x86/entry/thunk.S
> > index 119ebdc3d362..10c60369a67c 100644
> > --- a/arch/x86/entry/thunk.S
> > +++ b/arch/x86/entry/thunk.S
> > @@ -13,3 +13,8 @@ THUNK preempt_schedule_thunk, preempt_schedule
> >  THUNK preempt_schedule_notrace_thunk, preempt_schedule_notrace
> >  EXPORT_SYMBOL(preempt_schedule_thunk)
> >  EXPORT_SYMBOL(preempt_schedule_notrace_thunk)
> > +
> > +#ifdef CONFIG_PCPU_RCU_PREEMPT_COUNT
> > +THUNK rcu_read_unlock_special_thunk, rcu_read_unlock_special
> > +EXPORT_SYMBOL_GPL(rcu_read_unlock_special_thunk)
> > +#endif /* #ifdef CONFIG_PCPU_RCU_PREEMPT_COUNT */
> > diff --git a/arch/x86/include/asm/rcu_preempt.h b/arch/x86/include/asm/=
rcu_preempt.h
> > index cb25ebe038a5..acdd73b74c05 100644
> > --- a/arch/x86/include/asm/rcu_preempt.h
> > +++ b/arch/x86/include/asm/rcu_preempt.h
> > @@ -97,9 +97,11 @@ static __always_inline bool pcpu_rcu_preempt_count_d=
ec_and_test(void)
> >                                __percpu_arg([var]));
> >  }
> >
> > +extern asmlinkage void rcu_read_unlock_special_thunk(void);
> > +
> >  #define pcpu_rcu_read_unlock_special()                                =
         \
> >  do {                                                                  =
         \
> > -       rcu_read_unlock_special();
>
> Instead, can you not use __no_caller_saved_registers attribute for
> definition of rcu_read_unlock_special() or does that not work for what
> you're trying to do here?
>

I think it is paramount to make it the same as preempt_schedule[_thunk]()
when it comes to avoiding the caller-saving-registers-procedures in the
unlock paths.

I had no idea of 'no_caller_saved_registers' before, so I haven't tried it.

And there are limits to 'no_caller_saved_registers' either:

https://clang.llvm.org/docs/AttributeReference.html#no-caller-saved-registe=
rs:

Functions specified with the =E2=80=98no_caller_saved_registers=E2=80=99 at=
tribute
should only call other functions with the =E2=80=98no_caller_saved_register=
s=E2=80=99
attribute, or should be compiled with the =E2=80=98-mgeneral-regs-only=E2=
=80=99 flag
to avoid saving unused non-GPR registers.

https://gcc.gnu.org/onlinedocs/gcc/x86-Function-Attributes.html#index-no_00=
5fcaller_005fsaved_005fregisters-function-attribute_002c-x86:

Since GCC doesn=E2=80=99t preserve SSE, MMX nor x87 states, the GCC option
-mgeneral-regs-only should be used to compile functions with
no_caller_saved_registers attribute.

And I don't think =E2=80=98-mgeneral-regs-only=E2=80=99 is being used in th=
e kernel for x86.

Thanks
Lai