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;
Content-Type: text/plain;
        charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 14.0 \(3654.120.0.1.13\))
Subject: Re: [PATCH 1/2] mm/mprotect: use mmu_gather
From:   Nadav Amit <nadav.amit@gmail.com>
In-Reply-To: <20211003121019.GF4323@worktop.programming.kicks-ass.net>
Date:   Mon, 4 Oct 2021 12:24:14 -0700
Cc:     Andrew Morton <akpm@linux-foundation.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Linux-MM <linux-mm@kvack.org>, Peter Xu <peterx@redhat.com>,
        Andrea Arcangeli <aarcange@redhat.com>,
        Andrew Cooper <andrew.cooper3@citrix.com>,
        Andy Lutomirski <luto@kernel.org>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Will Deacon <will@kernel.org>, Yu Zhao <yuzhao@google.com>,
        Nick Piggin <npiggin@gmail.com>, x86@kernel.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <620B1A38-1457-4F77-8666-E73A318392B6@gmail.com>
References: <20210925205423.168858-1-namit@vmware.com>
 <20210925205423.168858-2-namit@vmware.com>
 <20211003121019.GF4323@worktop.programming.kicks-ass.net>
To:     Peter Zijlstra <peterz@infradead.org>
Precedence: bulk


> On Oct 3, 2021, at 5:10 AM, Peter Zijlstra <peterz@infradead.org> =
wrote:
>=20
> On Sat, Sep 25, 2021 at 01:54:22PM -0700, Nadav Amit wrote:
>=20
>> @@ -338,25 +344,25 @@ static unsigned long =
change_protection_range(struct vm_area_struct *vma,
>> 	struct mm_struct *mm =3D vma->vm_mm;
>> 	pgd_t *pgd;
>> 	unsigned long next;
>> -	unsigned long start =3D addr;
>> 	unsigned long pages =3D 0;
>> +	struct mmu_gather tlb;
>>=20
>> 	BUG_ON(addr >=3D end);
>> 	pgd =3D pgd_offset(mm, addr);
>> 	flush_cache_range(vma, addr, end);
>> 	inc_tlb_flush_pending(mm);
>=20
> That seems unbalanced...

Bad rebase. Thanks for catching it!

>=20
>> +	tlb_gather_mmu(&tlb, mm);
>> +	tlb_start_vma(&tlb, vma);
>> 	do {
>> 		next =3D pgd_addr_end(addr, end);
>> 		if (pgd_none_or_clear_bad(pgd))
>> 			continue;
>> -		pages +=3D change_p4d_range(vma, pgd, addr, next, =
newprot,
>> +		pages +=3D change_p4d_range(&tlb, vma, pgd, addr, next, =
newprot,
>> 					  cp_flags);
>> 	} while (pgd++, addr =3D next, addr !=3D end);
>>=20
>> -	/* Only flush the TLB if we actually modified any entries: */
>> -	if (pages)
>> -		flush_tlb_range(vma, start, end);
>> -	dec_tlb_flush_pending(mm);
>=20
> ... seeing you do remove the extra decrement.

Is it really needed? We do not put this comment elsewhere for
tlb_finish_mmu(). But no problem, I=E2=80=99ll keep it.=