[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87a5xwajn4.fsf@nvidia.com>
Date: Mon, 22 May 2023 17:45:59 +1000
From: Alistair Popple <apopple@...dia.com>
To: Qi Zheng <qi.zheng@...ux.dev>
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
robin.murphy@....com, will@...nel.org, nicolinc@...dia.com,
linux-arm-kernel@...ts.infradead.org, kvm@...r.kernel.org,
jgg@...dia.com, John Hubbard <jhubbard@...dia.com>,
Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH] mmu_notifiers: Notify on pte permission upgrades
Qi Zheng <qi.zheng@...ux.dev> writes:
> Hi Alistair,
>
> On 2023/5/22 14:37, Alistair Popple wrote:
[...]
>> + if (changed) {
>> + struct mmu_notifier_range range;
>> +
>> + mmu_notifier_range_init(&range, MMU_NOTIFY_PROTECTION_UPGRADE,
>> + 0, vmf->vma, vmf->vma->vm_mm,
>> + vmf->address & PAGE_MASK,
>> + (vmf->address & PAGE_MASK) + PAGE_SIZE);
>> + mmu_notifier_invalidate_range_start(&range);
>> + mmu_notifier_invalidate_range_end(&range);
>> + }
>
> There are four similar patterns, can we introduce a helper function to
> deduplicate them?
For sure. How about something like this?
void mmu_notifier_range_start_end(enum mmu_notifier_event event,
struct vm_area_struct *vma,
struct mm_struct *mm,
unsigned long start,
unsigned long end)
As an aside I didn't just use mmu_notifier_invalidate_range() as that
doesn't allow an event type to be set for interval notifiers which may
want to filter this.
>> +
>> return 0;
>> }
>>
Powered by blists - more mailing lists