From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ralph Campbell Subject: Re: [PATCH v5 8/9] mm/mmu_notifier: mmu_notifier_range_update_to_read_only() helper Date: Fri, 22 Feb 2019 14:42:08 -0800 Message-ID: <9da5e3b9-7f4a-49ee-bf80-024e5aed20a1@nvidia.com> References: <20190219200430.11130-1-jglisse@redhat.com> <20190219200430.11130-9-jglisse@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8"; format=flowed Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <20190219200430.11130-9-jglisse@redhat.com> Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org To: jglisse@redhat.com, linux-mm@kvack.org, Andrew Morton Cc: linux-kernel@vger.kernel.org, =?UTF-8?Q?Christian_K=c3=b6nig?= , Joonas Lahtinen , Jani Nikula , Rodrigo Vivi , Jan Kara , Andrea Arcangeli , Peter Xu , Felix Kuehling , Jason Gunthorpe , Ross Zwisler , Dan Williams , Paolo Bonzini , =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= , Michal Hocko , John Hubbard , kvm@vger.kernel.org, dri-devel@lists.freedesktop.org, linux-rdma@vger.kernel.org, Arnd Bergmann List-Id: dri-devel@lists.freedesktop.org On 2/19/19 12:04 PM, jglisse@redhat.com wrote: > From: J=C3=A9r=C3=B4me Glisse >=20 > Helper to test if a range is updated to read only (it is still valid > to read from the range). This is useful for device driver or anyone > who wish to optimize out update when they know that they already have > the range map read only. >=20 > Signed-off-by: J=C3=A9r=C3=B4me Glisse > Cc: Christian K=C3=B6nig > Cc: Joonas Lahtinen > Cc: Jani Nikula > Cc: Rodrigo Vivi > Cc: Jan Kara > Cc: Andrea Arcangeli > Cc: Peter Xu > Cc: Felix Kuehling > Cc: Jason Gunthorpe > Cc: Ross Zwisler > Cc: Dan Williams > Cc: Paolo Bonzini > Cc: Radim Kr=C4=8Dm=C3=A1=C5=99 > Cc: Michal Hocko > Cc: Christian Koenig > Cc: Ralph Campbell > Cc: John Hubbard > Cc: kvm@vger.kernel.org > Cc: dri-devel@lists.freedesktop.org > Cc: linux-rdma@vger.kernel.org > Cc: Arnd Bergmann > --- > include/linux/mmu_notifier.h | 4 ++++ > mm/mmu_notifier.c | 10 ++++++++++ > 2 files changed, 14 insertions(+) >=20 > diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h > index 0379956fff23..b6c004bd9f6a 100644 > --- a/include/linux/mmu_notifier.h > +++ b/include/linux/mmu_notifier.h > @@ -259,6 +259,8 @@ extern void __mmu_notifier_invalidate_range_end(struc= t mmu_notifier_range *r, > bool only_end); > extern void __mmu_notifier_invalidate_range(struct mm_struct *mm, > unsigned long start, unsigned long end); > +extern bool > +mmu_notifier_range_update_to_read_only(const struct mmu_notifier_range *= range); > =20 > static inline bool > mmu_notifier_range_blockable(const struct mmu_notifier_range *range) > @@ -568,6 +570,8 @@ static inline void mmu_notifier_mm_destroy(struct mm_= struct *mm) > { > } > =20 > +#define mmu_notifier_range_update_to_read_only(r) false > + > #define ptep_clear_flush_young_notify ptep_clear_flush_young > #define pmdp_clear_flush_young_notify pmdp_clear_flush_young > #define ptep_clear_young_notify ptep_test_and_clear_young > diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c > index abd88c466eb2..ee36068077b6 100644 > --- a/mm/mmu_notifier.c > +++ b/mm/mmu_notifier.c > @@ -395,3 +395,13 @@ void mmu_notifier_unregister_no_release(struct mmu_n= otifier *mn, > mmdrop(mm); > } > EXPORT_SYMBOL_GPL(mmu_notifier_unregister_no_release); > + > +bool > +mmu_notifier_range_update_to_read_only(const struct mmu_notifier_range *= range) > +{ > + if (!range->vma || range->event !=3D MMU_NOTIFY_PROTECTION_VMA) > + return false; > + /* Return true if the vma still have the read flag set. */ > + return range->vma->vm_flags & VM_READ; > +} > +EXPORT_SYMBOL_GPL(mmu_notifier_range_update_to_read_only); >=20 Don't you have to check for !WRITE & READ? mprotect() can change the permissions from R/O to RW and end up calling mmu_notifier_range_init() and=20 mmu_notifier_invalidate_range_start()/end(). I'm not sure how useful this is since only applies to the MMU_NOTIFY_PROTECTION_VMA case. Anyway, you can add Reviewed-by: Ralph Campbell