All of lore.kernel.org
 help / color / mirror / Atom feed
From: John Hubbard <jhubbard@nvidia.com>
To: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: "DRI Development" <dri-devel@lists.freedesktop.org>,
	LKML <linux-kernel@vger.kernel.org>,
	kvm@vger.kernel.org, "Linux MM" <linux-mm@kvack.org>,
	"Linux ARM" <linux-arm-kernel@lists.infradead.org>,
	linux-samsung-soc <linux-samsung-soc@vger.kernel.org>,
	"open list:DMA BUFFER SHARING FRAMEWORK"
	<linux-media@vger.kernel.org>,
	linux-s390@vger.kernel.org,
	"Daniel Vetter" <daniel.vetter@intel.com>,
	"Jason Gunthorpe" <jgg@ziepe.ca>,
	"Inki Dae" <inki.dae@samsung.com>,
	"Joonyoung Shim" <jy0922.shim@samsung.com>,
	"Seung-Woo Kim" <sw0312.kim@samsung.com>,
	"Kyungmin Park" <kyungmin.park@samsung.com>,
	"Kukjin Kim" <kgene@kernel.org>,
	"Krzysztof Kozlowski" <krzk@kernel.org>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	"Jérôme Glisse" <jglisse@redhat.com>, "Jan Kara" <jack@suse.cz>,
	"Dan Williams" <dan.j.williams@intel.com>
Subject: Re: [PATCH 01/13] drm/exynos: Stop using frame_vector helpers
Date: Wed, 7 Oct 2020 14:36:52 -0700	[thread overview]
Message-ID: <895eb064-1c8f-ecfc-0a98-1fbe40cb3161@nvidia.com> (raw)
In-Reply-To: <CAKMK7uEuwmCkpSeKxjDgHXWrE9P-TbawLoQ4dpu0nc4MYykScA@mail.gmail.com>

On 10/7/20 2:32 PM, Daniel Vetter wrote:
> On Wed, Oct 7, 2020 at 10:33 PM John Hubbard <jhubbard@nvidia.com> wrote:
>>
>> On 10/7/20 9:44 AM, Daniel Vetter wrote:
...
>>> @@ -398,15 +399,11 @@ static void g2d_userptr_put_dma_addr(struct g2d_data *g2d,
>>>        dma_unmap_sgtable(to_dma_dev(g2d->drm_dev), g2d_userptr->sgt,
>>>                          DMA_BIDIRECTIONAL, 0);
>>>
>>> -     pages = frame_vector_pages(g2d_userptr->vec);
>>> -     if (!IS_ERR(pages)) {
>>> -             int i;
>>> +     for (i = 0; i < g2d_userptr->npages; i++)
>>> +             set_page_dirty_lock(g2d_userptr->pages[i]);
>>>
>>> -             for (i = 0; i < frame_vector_count(g2d_userptr->vec); i++)
>>> -                     set_page_dirty_lock(pages[i]);
>>> -     }
>>> -     put_vaddr_frames(g2d_userptr->vec);
>>> -     frame_vector_destroy(g2d_userptr->vec);
>>> +     unpin_user_pages(g2d_userptr->pages, g2d_userptr->npages);
>>> +     kvfree(g2d_userptr->pages);
>>
>> You can avoid writing your own loop, and just simplify the whole thing down to
>> two lines:
>>
>>          unpin_user_pages_dirty_lock(g2d_userptr->pages, g2d_userptr->npages,
>>                                      true);
>>          kvfree(g2d_userptr->pages);
> 
> Oh nice, this is neat. I'll also roll it out in the habanalabs patch,
> that has the same thing. Well almost, it only uses set_page_dirty, not
> the _lock variant. But I have no idea whether that matters or not?


It matters. And invariably, call sites that use set_page_dirty() instead
of set_page_dirty_lock() were already wrong.  Which is why I never had to
provide anything like "unpin_user_pages_dirty (not locked)".

Although in habanalabs case, I just reviewed patch 3 and I think they *were*
correctly using set_page_dirty_lock()...

thanks,
-- 
John Hubbard
NVIDIA

WARNING: multiple messages have this Message-ID
From: John Hubbard <jhubbard@nvidia.com>
To: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: "Jérôme Glisse" <jglisse@redhat.com>,
	linux-s390@vger.kernel.org,
	linux-samsung-soc <linux-samsung-soc@vger.kernel.org>,
	"Jan Kara" <jack@suse.cz>,
	"Joonyoung Shim" <jy0922.shim@samsung.com>,
	kvm@vger.kernel.org, "Jason Gunthorpe" <jgg@ziepe.ca>,
	"Seung-Woo Kim" <sw0312.kim@samsung.com>,
	LKML <linux-kernel@vger.kernel.org>,
	"DRI Development" <dri-devel@lists.freedesktop.org>,
	"Inki Dae" <inki.dae@samsung.com>,
	"Linux MM" <linux-mm@kvack.org>,
	"Kyungmin Park" <kyungmin.park@samsung.com>,
	"Kukjin Kim" <kgene@kernel.org>,
	"Krzysztof Kozlowski" <krzk@kernel.org>,
	"Daniel Vetter" <daniel.vetter@intel.com>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	"Dan Williams" <dan.j.williams@intel.com>,
	"Linux ARM" <linux-arm-kernel@lists.infradead.org>,
	"open list:DMA BUFFER SHARING FRAMEWORK"
	<linux-media@vger.kernel.org>
Subject: Re: [PATCH 01/13] drm/exynos: Stop using frame_vector helpers
Date: Wed, 7 Oct 2020 14:36:52 -0700	[thread overview]
Message-ID: <895eb064-1c8f-ecfc-0a98-1fbe40cb3161@nvidia.com> (raw)
In-Reply-To: <CAKMK7uEuwmCkpSeKxjDgHXWrE9P-TbawLoQ4dpu0nc4MYykScA@mail.gmail.com>

On 10/7/20 2:32 PM, Daniel Vetter wrote:
> On Wed, Oct 7, 2020 at 10:33 PM John Hubbard <jhubbard@nvidia.com> wrote:
>>
>> On 10/7/20 9:44 AM, Daniel Vetter wrote:
...
>>> @@ -398,15 +399,11 @@ static void g2d_userptr_put_dma_addr(struct g2d_data *g2d,
>>>        dma_unmap_sgtable(to_dma_dev(g2d->drm_dev), g2d_userptr->sgt,
>>>                          DMA_BIDIRECTIONAL, 0);
>>>
>>> -     pages = frame_vector_pages(g2d_userptr->vec);
>>> -     if (!IS_ERR(pages)) {
>>> -             int i;
>>> +     for (i = 0; i < g2d_userptr->npages; i++)
>>> +             set_page_dirty_lock(g2d_userptr->pages[i]);
>>>
>>> -             for (i = 0; i < frame_vector_count(g2d_userptr->vec); i++)
>>> -                     set_page_dirty_lock(pages[i]);
>>> -     }
>>> -     put_vaddr_frames(g2d_userptr->vec);
>>> -     frame_vector_destroy(g2d_userptr->vec);
>>> +     unpin_user_pages(g2d_userptr->pages, g2d_userptr->npages);
>>> +     kvfree(g2d_userptr->pages);
>>
>> You can avoid writing your own loop, and just simplify the whole thing down to
>> two lines:
>>
>>          unpin_user_pages_dirty_lock(g2d_userptr->pages, g2d_userptr->npages,
>>                                      true);
>>          kvfree(g2d_userptr->pages);
> 
> Oh nice, this is neat. I'll also roll it out in the habanalabs patch,
> that has the same thing. Well almost, it only uses set_page_dirty, not
> the _lock variant. But I have no idea whether that matters or not?


It matters. And invariably, call sites that use set_page_dirty() instead
of set_page_dirty_lock() were already wrong.  Which is why I never had to
provide anything like "unpin_user_pages_dirty (not locked)".

Although in habanalabs case, I just reviewed patch 3 and I think they *were*
correctly using set_page_dirty_lock()...

thanks,
-- 
John Hubbard
NVIDIA

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

WARNING: multiple messages have this Message-ID
From: John Hubbard <jhubbard@nvidia.com>
To: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: "Jérôme Glisse" <jglisse@redhat.com>,
	linux-s390@vger.kernel.org,
	linux-samsung-soc <linux-samsung-soc@vger.kernel.org>,
	"Jan Kara" <jack@suse.cz>,
	"Joonyoung Shim" <jy0922.shim@samsung.com>,
	kvm@vger.kernel.org, "Jason Gunthorpe" <jgg@ziepe.ca>,
	"Seung-Woo Kim" <sw0312.kim@samsung.com>,
	LKML <linux-kernel@vger.kernel.org>,
	"DRI Development" <dri-devel@lists.freedesktop.org>,
	"Linux MM" <linux-mm@kvack.org>,
	"Kyungmin Park" <kyungmin.park@samsung.com>,
	"Kukjin Kim" <kgene@kernel.org>,
	"Krzysztof Kozlowski" <krzk@kernel.org>,
	"Daniel Vetter" <daniel.vetter@intel.com>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	"Dan Williams" <dan.j.williams@intel.com>,
	"Linux ARM" <linux-arm-kernel@lists.infradead.org>,
	"open list:DMA BUFFER SHARING FRAMEWORK"
	<linux-media@vger.kernel.org>
Subject: Re: [PATCH 01/13] drm/exynos: Stop using frame_vector helpers
Date: Wed, 7 Oct 2020 14:36:52 -0700	[thread overview]
Message-ID: <895eb064-1c8f-ecfc-0a98-1fbe40cb3161@nvidia.com> (raw)
In-Reply-To: <CAKMK7uEuwmCkpSeKxjDgHXWrE9P-TbawLoQ4dpu0nc4MYykScA@mail.gmail.com>

On 10/7/20 2:32 PM, Daniel Vetter wrote:
> On Wed, Oct 7, 2020 at 10:33 PM John Hubbard <jhubbard@nvidia.com> wrote:
>>
>> On 10/7/20 9:44 AM, Daniel Vetter wrote:
...
>>> @@ -398,15 +399,11 @@ static void g2d_userptr_put_dma_addr(struct g2d_data *g2d,
>>>        dma_unmap_sgtable(to_dma_dev(g2d->drm_dev), g2d_userptr->sgt,
>>>                          DMA_BIDIRECTIONAL, 0);
>>>
>>> -     pages = frame_vector_pages(g2d_userptr->vec);
>>> -     if (!IS_ERR(pages)) {
>>> -             int i;
>>> +     for (i = 0; i < g2d_userptr->npages; i++)
>>> +             set_page_dirty_lock(g2d_userptr->pages[i]);
>>>
>>> -             for (i = 0; i < frame_vector_count(g2d_userptr->vec); i++)
>>> -                     set_page_dirty_lock(pages[i]);
>>> -     }
>>> -     put_vaddr_frames(g2d_userptr->vec);
>>> -     frame_vector_destroy(g2d_userptr->vec);
>>> +     unpin_user_pages(g2d_userptr->pages, g2d_userptr->npages);
>>> +     kvfree(g2d_userptr->pages);
>>
>> You can avoid writing your own loop, and just simplify the whole thing down to
>> two lines:
>>
>>          unpin_user_pages_dirty_lock(g2d_userptr->pages, g2d_userptr->npages,
>>                                      true);
>>          kvfree(g2d_userptr->pages);
> 
> Oh nice, this is neat. I'll also roll it out in the habanalabs patch,
> that has the same thing. Well almost, it only uses set_page_dirty, not
> the _lock variant. But I have no idea whether that matters or not?


It matters. And invariably, call sites that use set_page_dirty() instead
of set_page_dirty_lock() were already wrong.  Which is why I never had to
provide anything like "unpin_user_pages_dirty (not locked)".

Although in habanalabs case, I just reviewed patch 3 and I think they *were*
correctly using set_page_dirty_lock()...

thanks,
-- 
John Hubbard
NVIDIA
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  reply	other threads:[~2020-10-07 21:37 UTC|newest]

Thread overview: 166+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-07 16:44 [PATCH 00/13] follow_pfn and other iomap races Daniel Vetter
2020-10-07 16:44 ` Daniel Vetter
2020-10-07 16:44 ` Daniel Vetter
2020-10-07 16:44 ` [PATCH 01/13] drm/exynos: Stop using frame_vector helpers Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 20:32   ` John Hubbard
2020-10-07 20:32     ` John Hubbard
2020-10-07 20:32     ` John Hubbard
2020-10-07 21:32     ` Daniel Vetter
2020-10-07 21:32       ` Daniel Vetter
2020-10-07 21:32       ` Daniel Vetter
2020-10-07 21:36       ` John Hubbard [this message]
2020-10-07 21:36         ` John Hubbard
2020-10-07 21:36         ` John Hubbard
2020-10-07 21:50         ` Daniel Vetter
2020-10-07 21:50           ` Daniel Vetter
2020-10-07 21:50           ` Daniel Vetter
2020-10-07 16:44 ` [PATCH 02/13] drm/exynos: Use FOLL_LONGTERM for g2d cmdlists Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 20:43   ` John Hubbard
2020-10-07 20:43     ` John Hubbard
2020-10-07 20:43     ` John Hubbard
2020-10-07 16:44 ` [PATCH 03/13] misc/habana: Stop using frame_vector helpers Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 20:38   ` John Hubbard
2020-10-07 20:38     ` John Hubbard
2020-10-07 20:38     ` John Hubbard
2020-10-07 16:44 ` [PATCH 04/13] misc/habana: Use FOLL_LONGTERM for userptr Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 20:46   ` John Hubbard
2020-10-07 20:46     ` John Hubbard
2020-10-07 20:46     ` John Hubbard
2020-10-07 16:44 ` [PATCH 05/13] mm/frame-vector: Use FOLL_LONGTERM Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:53   ` Jason Gunthorpe
2020-10-07 16:53     ` Jason Gunthorpe
2020-10-07 16:53     ` Jason Gunthorpe
2020-10-07 17:12     ` Daniel Vetter
2020-10-07 17:12       ` Daniel Vetter
2020-10-07 17:12       ` Daniel Vetter
2020-10-07 17:33       ` Jason Gunthorpe
2020-10-07 17:33         ` Jason Gunthorpe
2020-10-07 17:33         ` Jason Gunthorpe
2020-10-07 21:13   ` John Hubbard
2020-10-07 21:13     ` John Hubbard
2020-10-07 21:13     ` John Hubbard
2020-10-07 21:30     ` Daniel Vetter
2020-10-07 21:30       ` Daniel Vetter
2020-10-07 21:30       ` Daniel Vetter
2020-10-07 16:44 ` [PATCH 06/13] media: videobuf2: Move frame_vector into media subsystem Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 22:18   ` John Hubbard
2020-10-07 22:18     ` John Hubbard
2020-10-07 22:18     ` John Hubbard
2020-10-07 16:44 ` [PATCH 07/13] mm: close race in generic_access_phys Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 17:27   ` Jason Gunthorpe
2020-10-07 17:27     ` Jason Gunthorpe
2020-10-07 17:27     ` Jason Gunthorpe
2020-10-07 18:01     ` Daniel Vetter
2020-10-07 18:01       ` Daniel Vetter
2020-10-07 18:01       ` Daniel Vetter
2020-10-07 23:21       ` Jason Gunthorpe
2020-10-07 23:21         ` Jason Gunthorpe
2020-10-07 23:21         ` Jason Gunthorpe
2020-10-08  0:44   ` John Hubbard
2020-10-08  0:44     ` John Hubbard
2020-10-08  0:44     ` John Hubbard
2020-10-08  7:23     ` Daniel Vetter
2020-10-08  7:23       ` Daniel Vetter
2020-10-08  7:23       ` Daniel Vetter
2020-10-07 16:44 ` [PATCH 08/13] s390/pci: Remove races against pte updates Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-08 16:44   ` Gerald Schaefer
2020-10-08 16:44     ` Gerald Schaefer
2020-10-08 16:44     ` Gerald Schaefer
2020-10-08 17:16     ` Daniel Vetter
2020-10-08 17:16       ` Daniel Vetter
2020-10-08 17:16       ` Daniel Vetter
2020-10-07 16:44 ` [PATCH 09/13] PCI: obey iomem restrictions for procfs mmap Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 18:46   ` Bjorn Helgaas
2020-10-07 18:46     ` Bjorn Helgaas
2020-10-07 18:46     ` Bjorn Helgaas
2020-10-07 18:46     ` Bjorn Helgaas
2020-10-07 16:44 ` [PATCH 10/13] PCI: revoke mappings like devmem Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 18:41   ` Bjorn Helgaas
2020-10-07 18:41     ` Bjorn Helgaas
2020-10-07 18:41     ` Bjorn Helgaas
2020-10-07 19:24     ` Daniel Vetter
2020-10-07 19:24       ` Daniel Vetter
2020-10-07 19:24       ` Daniel Vetter
2020-10-07 19:33   ` Dan Williams
2020-10-07 19:33     ` Dan Williams
2020-10-07 19:33     ` Dan Williams
2020-10-07 19:47     ` Daniel Vetter
2020-10-07 19:47       ` Daniel Vetter
2020-10-07 19:47       ` Daniel Vetter
2020-10-07 22:23       ` Dan Williams
2020-10-07 22:23         ` Dan Williams
2020-10-07 22:23         ` Dan Williams
2020-10-07 22:29         ` Dan Williams
2020-10-07 22:29           ` Dan Williams
2020-10-07 22:29           ` Dan Williams
2020-10-08  8:09           ` Daniel Vetter
2020-10-08  8:09             ` Daniel Vetter
2020-10-08  8:09             ` Daniel Vetter
2020-10-07 23:24     ` Jason Gunthorpe
2020-10-07 23:24       ` Jason Gunthorpe
2020-10-07 23:24       ` Jason Gunthorpe
2020-10-08  7:31       ` Daniel Vetter
2020-10-08  7:31         ` Daniel Vetter
2020-10-08  7:31         ` Daniel Vetter
2020-10-08  7:49       ` Dan Williams
2020-10-08  7:49         ` Dan Williams
2020-10-08  7:49         ` Dan Williams
2020-10-08  8:13         ` Daniel Vetter
2020-10-08  8:13           ` Daniel Vetter
2020-10-08  8:13           ` Daniel Vetter
2020-10-08  8:35           ` Dan Williams
2020-10-08  8:35             ` Dan Williams
2020-10-08  8:35             ` Dan Williams
2020-10-08 12:41         ` Jason Gunthorpe
2020-10-08 12:41           ` Jason Gunthorpe
2020-10-08 12:41           ` Jason Gunthorpe
2020-10-07 16:44 ` [PATCH 11/13] mm: add unsafe_follow_pfn Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 17:36   ` Jason Gunthorpe
2020-10-07 17:36     ` Jason Gunthorpe
2020-10-07 17:36     ` Jason Gunthorpe
2020-10-07 18:10     ` Daniel Vetter
2020-10-07 18:10       ` Daniel Vetter
2020-10-07 18:10       ` Daniel Vetter
2020-10-07 19:00       ` Jason Gunthorpe
2020-10-07 19:00         ` Jason Gunthorpe
2020-10-07 19:00         ` Jason Gunthorpe
2020-10-07 19:38         ` Daniel Vetter
2020-10-07 19:38           ` Daniel Vetter
2020-10-07 19:38           ` Daniel Vetter
2020-10-07 16:44 ` [PATCH 12/13] media/videbuf1|2: Mark follow_pfn usage as unsafe Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44 ` [PATCH 13/13] vfio/type1: Mark follow_pfn " Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 16:44   ` Daniel Vetter
2020-10-07 17:39   ` Jason Gunthorpe
2020-10-07 17:39     ` Jason Gunthorpe
2020-10-07 17:39     ` Jason Gunthorpe
2020-10-07 18:14     ` Daniel Vetter
2020-10-07 18:14       ` Daniel Vetter
2020-10-07 18:14       ` Daniel Vetter
2020-10-07 18:47       ` Jason Gunthorpe
2020-10-07 18:47         ` Jason Gunthorpe
2020-10-07 18:47         ` Jason Gunthorpe

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=895eb064-1c8f-ecfc-0a98-1fbe40cb3161@nvidia.com \
    --to=jhubbard@nvidia.com \
    --cc=akpm@linux-foundation.org \
    --cc=dan.j.williams@intel.com \
    --cc=daniel.vetter@ffwll.ch \
    --cc=daniel.vetter@intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=inki.dae@samsung.com \
    --cc=jack@suse.cz \
    --cc=jgg@ziepe.ca \
    --cc=jglisse@redhat.com \
    --cc=jy0922.shim@samsung.com \
    --cc=kgene@kernel.org \
    --cc=krzk@kernel.org \
    --cc=kvm@vger.kernel.org \
    --cc=kyungmin.park@samsung.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-s390@vger.kernel.org \
    --cc=linux-samsung-soc@vger.kernel.org \
    --cc=sw0312.kim@samsung.com \
    --subject='Re: [PATCH 01/13] drm/exynos: Stop using frame_vector helpers' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.