Re: [PATCH V2 3/4] mm/sparse-vmemmap: Generalise vmemmap_populate_hugepages()

From: Huacai Chen <chenhuacai@kernel.org>
To: Arnd Bergmann <arnd@arndb.de>
Cc: Huacai Chen <chenhuacai@loongson.cn>,
	Thomas Bogendoerfer <tsbogend@alpha.franken.de>,
	 Dave Hansen <dave.hansen@linux.intel.com>,
	Andy Lutomirski <luto@kernel.org>,
	 Peter Zijlstra <peterz@infradead.org>,
	Catalin Marinas <catalin.marinas@arm.com>,
	 Will Deacon <will@kernel.org>,
	loongarch@lists.linux.dev,
	 linux-arch <linux-arch@vger.kernel.org>,
	Xuefeng Li <lixuefeng@loongson.cn>,  Guo Ren <guoren@kernel.org>,
	Xuerui Wang <kernel@xen0n.name>,
	 Jiaxun Yang <jiaxun.yang@flygoat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	 Linux-MM <linux-mm@kvack.org>,
	 "open list:BROADCOM NVRAM DRIVER" <linux-mips@vger.kernel.org>,
	 Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	 Linux ARM <linux-arm-kernel@lists.infradead.org>,
	 Feiyang Chen <chenfeiyang@loongson.cn>
Subject: Re: [PATCH V2 3/4] mm/sparse-vmemmap: Generalise vmemmap_populate_hugepages()
Date: Thu, 30 Jun 2022 16:27:53 +0800	[thread overview]
Message-ID: <CAAhV-H5kywXfSLaKRhobqzozOuU0UyEKcOApu3Abz+csCgJPgg@mail.gmail.com> (raw)
In-Reply-To: <CAK8P3a1f94z4oSnMr73PuiXkMR7uGhthzY_EWVniB+G4KXBcBQ@mail.gmail.com>

Hi, Arnd,

On Thu, Jun 30, 2022 at 2:05 PM Arnd Bergmann <arnd@arndb.de> wrote:
>
> On Thu, Jun 30, 2022 at 6:32 AM Huacai Chen <chenhuacai@loongson.cn> wrote:
> >
> > From: Feiyang Chen <chenfeiyang@loongson.cn>
> >
> > Generalise vmemmap_populate_hugepages() so ARM64 & X86 & LoongArch can
> > share its implementation.
>
> Sharing this function is good, thanks for consolidating this
>
> > Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
> > Signed-off-by: Feiyang Chen <chenfeiyang@loongson.cn>
>
> The Signed-off-by lines are in the wrong order, it should start with the author
> and end with the final submitter.
OK,  I will change the order.

>
> > index 33e2a1ceee72..6f2e40bb695d 100644
> > --- a/mm/sparse-vmemmap.c
> > +++ b/mm/sparse-vmemmap.c
> > @@ -686,6 +686,60 @@ int __meminit vmemmap_populate_basepages(unsigned long start, unsigned long end,
> >         return vmemmap_populate_range(start, end, node, altmap, NULL);
> >  }
> >
> > +void __weak __meminit vmemmap_set_pmd(pmd_t *pmd, void *p, int node,
> > +                                     unsigned long addr, unsigned long next)
> > +{
> > +}
> > +
> > +int __weak __meminit vmemmap_check_pmd(pmd_t *pmd, int node, unsigned long addr,
> > +                                      unsigned long next)
> > +{
> > +       return 0;
> > +}
> > +
>
> I think inline functions would be better here, both for compiler optimization
> and to make it easier to track the code flow. The normal way we do these
> in architecture specific headers is to override the functions by defining a
> macro of the same name.
In my opinion, weak functions are suitable for overriding if they are
only used in a single .c file (this case). If we don't use weak
functions, this series needs as many as 4 #ifdefs, for pud_init(),
pmd_init(), vmemmap_set_pmd() and  vmemmap_check_pmd() respectively,
which increase the difficulty of maintain (just my own opinion, maybe
not a objective fact).

Huacai
>
>
>         Arnd