Re: [PATCH v2 2/4] mm: rmap: Allow platforms without mm_cpumask to defer TLB flush

From: Barry Song <21cnbao@gmail.com>
To: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: "Andrew Morton" <akpm@linux-foundation.org>,
	Linux-MM <linux-mm@kvack.org>,
	LAK <linux-arm-kernel@lists.infradead.org>, x86 <x86@kernel.org>,
	"Catalin Marinas" <catalin.marinas@arm.com>,
	"Will Deacon" <will@kernel.org>,
	"Linux Doc Mailing List" <linux-doc@vger.kernel.org>,
	"Jonathan Corbet" <corbet@lwn.net>,
	"Arnd Bergmann" <arnd@arndb.de>,
	LKML <linux-kernel@vger.kernel.org>,
	"Darren Hart" <darren@os.amperecomputing.com>,
	"Yicong Yang" <yangyicong@hisilicon.com>,
	huzhanyuan@oppo.com, "李培锋(wink)" <lipeifeng@oppo.com>,
	"张诗明(Simon Zhang)" <zhangshiming@oppo.com>, 郭健 <guojian@oppo.com>,
	"real mz" <realmz6@gmail.com>,
	linux-mips@vger.kernel.org, openrisc@lists.librecores.org,
	linuxppc-dev@lists.ozlabs.org, linux-riscv@lists.infradead.org,
	linux-s390@vger.kernel.org, "Barry Song" <v-songbaohua@oppo.com>
Subject: Re: [PATCH v2 2/4] mm: rmap: Allow platforms without mm_cpumask to defer TLB flush
Date: Tue, 12 Jul 2022 10:52:24 +1200	[thread overview]
Message-ID: <CAGsJ_4wW=3xsmB1jzcv=sD9OhhPc0v=+odp=JY80vEbMY6OZ_g@mail.gmail.com> (raw)
In-Reply-To: <13b283fe-10f7-376f-9b8e-856e4d1e0ede@huawei.com>

On Tue, Jul 12, 2022 at 1:35 AM Kefeng Wang <wangkefeng.wang@huawei.com> wrote:
>
> Hi Barry，
>
> On 2022/7/11 11:46, Barry Song wrote:
> > From: Barry Song <v-songbaohua@oppo.com>
> >
> > Platforms like ARM64 have hareware TLB shootdown broadcast. They
> > don't maintain mm_cpumask but just send tlbi and related sync
> > instructions for TLB flush. task's mm_cpumask is normally empty
> > in this case. We also allow deferred TLB flush on this kind of
> > platforms.
> >
> > Signed-off-by: Barry Song <v-songbaohua@oppo.com>>
> > ---
> ...
> > diff --git a/mm/Kconfig b/mm/Kconfig
> > index 169e64192e48..7bf54f57ca01 100644
> > --- a/mm/Kconfig
> > +++ b/mm/Kconfig
> > @@ -951,6 +951,9 @@ config ARCH_HAS_CURRENT_STACK_POINTER
> >         register alias named "current_stack_pointer", this config can be
> >         selected.
> >
> > +config ARCH_HAS_MM_CPUMASK
> > +     bool
> > +
> >   config ARCH_HAS_VM_GET_PAGE_PROT
> >       bool
> >
> > diff --git a/mm/rmap.c b/mm/rmap.c
> > index 5bcb334cd6f2..13d4f9a1d4f1 100644
> > --- a/mm/rmap.c
> > +++ b/mm/rmap.c
> > @@ -692,6 +692,10 @@ static bool should_defer_flush(struct mm_struct *mm, enum ttu_flags flags)
> >       if (!(flags & TTU_BATCH_FLUSH))
> >               return false;
> >
> > +#ifndef CONFIG_ARCH_HAS_MM_CPUMASK
> > +     return true;
> > +#endif
> > +
>
> Here is another option to enable arch's tlbbatch defer
>

This option is even better than simply having ARCH_HAS_MM_CPUMASK
since arch might make decisions based on specific hardware characters.
for example,
https://lists.ozlabs.org/pipermail/linuxppc-dev/2017-November/165468.html

+bool arch_tlbbatch_should_defer(struct mm_struct *mm)
+{
+     if (!radix_enabled() || cpu_has_feature(CPU_FTR_POWER9_DD1))
+         return false;
+
+     if (!mm_is_thread_local(mm))
+         return true;
+
+     return false;
+}

In this case, having MM_CPUMASK doesn't necessarily mean tlbbatch is needed.

> [1]
> https://patchwork.ozlabs.org/project/linuxppc-dev/patch/20171101101735.2318-2-khandual@linux.vnet.ibm.com/
>
> >       /* If remote CPUs need to be flushed then defer batch the flush */
> >       if (cpumask_any_but(mm_cpumask(mm), get_cpu()) < nr_cpu_ids)
> >               should_defer = true;

Thanks
Barry