Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20220126060514.1574935-1-pasha.tatashin@soleen.com>
 <20220126060514.1574935-4-pasha.tatashin@soleen.com> <edd2d5d0-998a-633c-fc60-4f31ae768d3e@google.com>
In-Reply-To: <edd2d5d0-998a-633c-fc60-4f31ae768d3e@google.com>
From:   Pasha Tatashin <pasha.tatashin@soleen.com>
Date:   Wed, 26 Jan 2022 07:34:19 -0500
Message-ID: <CA+CK2bB_DwxGCD0TOJ7B1gN4sgXg5Bptw2LQbuD7Q4gQFm3vCg@mail.gmail.com>
Subject: Re: [PATCH v3 3/4] mm/khugepaged: unify collapse pmd clear, flush and free
To:     David Rientjes <rientjes@google.com>
Cc:     LKML <linux-kernel@vger.kernel.org>, linux-mm <linux-mm@kvack.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Paul Turner <pjt@google.com>, Wei Xu <weixugc@google.com>,
        Greg Thelen <gthelen@google.com>,
        Ingo Molnar <mingo@redhat.com>, Will Deacon <will@kernel.org>,
        Mike Rapoport <rppt@kernel.org>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        "H. Peter Anvin" <hpa@zytor.com>,
        "Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>,
        Jiri Slaby <jirislaby@kernel.org>,
        Muchun Song <songmuchun@bytedance.com>,
        Fusion Future <qydwhotmail@gmail.com>,
        Hugh Dickins <hughd@google.com>, Zi Yan <ziy@nvidia.com>,
        Anshuman Khandual <anshuman.khandual@arm.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, Jan 26, 2022 at 1:34 AM David Rientjes <rientjes@google.com> wrote:
>
> On Wed, 26 Jan 2022, Pasha Tatashin wrote:
>
> > Unify the code that flushes, clears pmd entry, and frees the PTE table
> > level into a new function collapse_and_free_pmd().
> >
> > This clean-up is useful as in the next patch we will add another call to
> > this function to iterate through PTE prior to freeing the level for page
> > table check.
> >
> > Signed-off-by: Pasha Tatashin <pasha.tatashin@soleen.com>
>
> Acked-by: David Rientjes <rientjes@google.com>

Thank you, David.

>
> One nit below.
>
> > ---
> >  mm/khugepaged.c | 32 ++++++++++++++++----------------
> >  1 file changed, 16 insertions(+), 16 deletions(-)
> >
> > diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> > index 35f14d0a00a6..440112355ffe 100644
> > --- a/mm/khugepaged.c
> > +++ b/mm/khugepaged.c
> > @@ -1416,6 +1416,17 @@ static int khugepaged_add_pte_mapped_thp(struct mm_struct *mm,
> >       return 0;
> >  }
> >
> > +static void collapse_and_free_pmd(struct mm_struct *mm, struct vm_area_struct *vma,
> > +                               unsigned long addr, pmd_t *pmdp)
> > +{
> > +     spinlock_t *ptl = pmd_lock(vma->vm_mm, pmdp);
> > +     pmd_t pmd = pmdp_collapse_flush(vma, addr, pmdp);
> > +
> > +     spin_unlock(ptl);
>
> No strong objection, but I think the typical style would be to declare the
> local variables separately and avoid mixing the code, especially in cases
> when taking locks (and not just initializing the local variables).

I will address this in the next revision.

Thank you,
Pasha