* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 2:42 ` Minchan Kim
@ 2009-04-20 3:28 ` Huang Shijie
2009-04-20 3:42 ` KOSAKI Motohiro
2009-04-20 4:53 ` Minchan Kim
2009-04-20 3:57 ` Huang Shijie
2009-04-22 6:08 ` Huang Shijie
2 siblings, 2 replies; 25+ messages in thread
From: Huang Shijie @ 2009-04-20 3:28 UTC (permalink / raw)
To: Minchan Kim; +Cc: linux-mm
Minchan Kim a??e??:
> On Mon, 20 Apr 2009 10:15:55 +0800
> Huang Shijie <shijie8@gmail.com> wrote:
>
>
>> Minchan Kim a??e??:
>>
>>> On Fri, 17 Apr 2009 15:01:01 +0800
>>> Huang Shijie <shijie8@gmail.com> wrote:
>>>
>>>
>>>
>>>> I'm writting a driver for a video card with the V4L2 interface .
>>>> V4L2 interface supports the USER-POINTER method for the video frame
>>>> handling.
>>>>
>>>> VLC player supports the USER-POINTER method,while MPALYER does not.
>>>>
>>>> In the USER-POINTER method, VLC will call the posix_memalign() to
>>>> allocate
>>>> 203 pages in certain PAL mode (that is 720*576*2) for a single frame.
>>>> In my driver , I call the get_user_pages_fast() to obtain the pages
>>>> array,and then call
>>>> the vmap() to map the pages to VMALLOC space for the memcpy().The code
>>>> shows below:
>>>> ....................
>>>> get_user_pages_fast();
>>>> ...
>>>> f->data = vmap();
>>>> .......................
>>>>
>>>>
>>> What I understand is that you get the pages of posix_memalign by get_user_pages_fast
>>> and then that pages are mapped at kernel vmalloc space by vmap.
>>>
>>> Is it for removing copy overhead from kernel to user ?
>>>
>>>
>>>
>> I need a large range of virtual contigous memory to store my video
>> frame(about 203 pages). When I received a full frame ,I will queue the
>> buffer in
>> a VIDIOC queue,which will be remove by the VIDIOC_DQBUF.'
>>
>
> I can't understand your point.
> Sorry for that.
>
> Could you explain more detail relation (user buffer which is allocated by posix_memalign) and (kernel buffer which is mapped by vmap) ?
>
>
:) sorry for my poor english.
[1] VLC uses the posix_memalign to allocate a big buffer for a single
frame(203 pages).
[2] vmap sets up the mapping of virtual contigous address for gup()'s
pages array(the pages are not consecutive).
memcpy() needs a contiguous address to copy in kernel mode.
[3] my driver do some specail operations to received data, then
memcopy the data to the buffer get in step [2].
[4] when the buffer is full ,I will give the the user process (VLC).
That's all.
>>>> In comments, it said :
>>>> "
>>>> +/**
>>>> + * get_user_pages_fast() - pin user pages in memory
>>>> + * @start: starting user address
>>>> + * @nr_pages: number of pages from start to pin
>>>> + * @write: whether pages will be written to
>>>> + * @pages: array that receives pointers to the pages pinned.
>>>> + * Should be at least nr_pages long.
>>>> "
>>>>
>>>> But after I digged the code of kswap and the get_user_pages(called by
>>>> get_user_pages_fast),
>>>> I did not find how the pages pinned in memory.I really need the pages
>>>> pinned in memory.
>>>>
>>>> Assume page A is one of the pages obtained by get_user_pages_fast()
>>>> during page-fault.
>>>>
>>>> [1] page A will on the LRU_ACTIVE_ANON list;
>>>> the _count of page A increment by one;
>>>> PTE for page A will be set ACCESSED.
>>>>
>>>> [2] kswapd will scan the lru list,and move page A from LRU_ACTIVE_ANON
>>>> to LRU_INACTIVE_ANON.
>>>> In the shrink_page_list(), there is nothing can stop page A been
>>>> swapped out.
>>>> I don't think the page_reference() can move page A back to
>>>> LRU_ACTIVE_ANON.In my driver,
>>>> I am not sure if the VLC can access the page A.
>>>>
>>>> Is this a bug? or I miss something?
>>>> Thanks .
>>>>
>>>>
>>> If above my assumption is right, It's not a BUG.
>>> You get the application's pages by get_user_pages_fast.
>>> 'Page pinning' means it shouldn't be freed.
>>> Application's pages always can be swapped out.
>>> If you don't want to swap out the page, you should use mlock.
>>> If you use mlock, kernel won't insert the page to lru [in]active list.
>>> So the page never can be swapped out.
>>>
>>>
>>>
>> Yes, it not a bug .
>>
>> I read the kernel code again. In my case ,the kernel will pin the pages
>> in memory.
>> I missed function is_page_cache_freeable() in the pageout().
>>
>> In my case, is_page_cache_freeable()will return false ,for
>> page_count(page) is 3 now:
>> <1> one is from alloc_page_* in page fault.
>> <2> one is from get_usr_pages()
>> <3> one is from add_to_swap() in shrink_page_list()
>>
>
> One more, try_to_unmap will call page_cache_release.
> So, count is 2.
>
>
Yes, you are right. I missed the page_cache_release() in try_to_unmap().
:(
>> So ,there is no need to use the mlock, it will mess my driver.
>> is_page_cache_freeable()will return PAGE_KEEP, and page is locked in
>> swap cache.
>>
>>
>
> I can't understand your point exactly yet.
> But what I mean is following as in user mode
>
> posix_memalignq(&buffer);
> mlock(buffer, buffer_len);
>
>
I also wish the VLC use the mlock,but it does not.If it uses mlock(),
the pages will be put in LRU_UNEVICETABL LIST.
Maybe the programmer of VLC thinks: Why i add mlock, for the kernel has the
gup() which could pin the pages in memory?
> I will not dirty your driver.
> Do I miss something ?
>
>
I did add the Mlock bit to the VMA->vm_flags in my driver before,but I
think that's ugly.
>> Unfortunately, the page is unmaped, and the PTE of the page has been
>> replaced by a swp_entry_t .
>> When the process read the page ,it will raise a page fault again, the
>> kernel will find the page in the
>> swap cache, and requeue the page in LRU_ACTIVE_ANON, ---I think it is a
>> vicious circle for the kernel.
>>
>> I think there two places to put back the gup() pages.
>> <1> isolate_page_glable()
>> <2> in the shrink_page_list(), before called the try_to_unmap().
>> KOSAKI Motohiro 's patch takes effect in the second place.
>> I think the first place is better.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 3:28 ` Huang Shijie
@ 2009-04-20 3:42 ` KOSAKI Motohiro
2009-04-20 4:53 ` Minchan Kim
1 sibling, 0 replies; 25+ messages in thread
From: KOSAKI Motohiro @ 2009-04-20 3:42 UTC (permalink / raw)
To: Huang Shijie; +Cc: kosaki.motohiro, Minchan Kim, linux-mm
> > I can't understand your point exactly yet.
> > But what I mean is following as in user mode
> >
> > posix_memalignq(&buffer);
> > mlock(buffer, buffer_len);
> >
> >
> I also wish the VLC use the mlock,but it does not.If it uses mlock(),
> the pages will be put in LRU_UNEVICETABL LIST.
>
> Maybe the programmer of VLC thinks: Why i add mlock, for the kernel has the
> gup() which could pin the pages in memory?
more weakness.
mlock() only gurantee the address range is memory-resident, not gurantee to no change
virtual-physical mappings.
There are different operation.
example, gup() prevent page migration but mlock doesn't.
> > I will not dirty your driver.
> > Do I miss something ?
> >
> I did add the Mlock bit to the VMA->vm_flags in my driver before,but I
> think that's ugly.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 3:28 ` Huang Shijie
2009-04-20 3:42 ` KOSAKI Motohiro
@ 2009-04-20 4:53 ` Minchan Kim
2009-04-20 5:05 ` KOSAKI Motohiro
2009-04-20 5:05 ` Huang Shijie
1 sibling, 2 replies; 25+ messages in thread
From: Minchan Kim @ 2009-04-20 4:53 UTC (permalink / raw)
To: Huang Shijie; +Cc: Minchan Kim, linux-mm
On Mon, 20 Apr 2009 11:28:00 +0800
Huang Shijie <shijie8@gmail.com> wrote:
I will summarize your method.
Is right ?
kernel(driver) application
posix_memalign(buffer)
ioctl(buffer)
ioctl handler
get_user_pages(pages);
/* This pages are mapped at user's vma'
address space */
vaddr = vmap(pages);
/* This pages are mapped at vmalloc space */
.
.
<after sometime,
It may change to other process context>
.
.
interrupt handler in your driver
memcpy(vaddr, src, len);
notify_user();
processing(buffer);
It's rather awkward use case of get_user_pages.
If you want to share one big buffer between kernel and user,
You can vmalloc and remap_pfn_range.
You can refer cpia_mmap in drivers/media/video/cpia.c
> Minchan Kim a??e??:
> > On Mon, 20 Apr 2009 10:15:55 +0800
> > Huang Shijie <shijie8@gmail.com> wrote:
> >
> >
> >> Minchan Kim a??e??:
> >>
> >>> On Fri, 17 Apr 2009 15:01:01 +0800
> >>> Huang Shijie <shijie8@gmail.com> wrote:
> >>>
> >>>
> >>>
> >>>> I'm writting a driver for a video card with the V4L2 interface .
> >>>> V4L2 interface supports the USER-POINTER method for the video frame
> >>>> handling.
> >>>>
> >>>> VLC player supports the USER-POINTER method,while MPALYER does not.
> >>>>
> >>>> In the USER-POINTER method, VLC will call the posix_memalign() to
> >>>> allocate
> >>>> 203 pages in certain PAL mode (that is 720*576*2) for a single frame.
> >>>> In my driver , I call the get_user_pages_fast() to obtain the pages
> >>>> array,and then call
> >>>> the vmap() to map the pages to VMALLOC space for the memcpy().The code
> >>>> shows below:
> >>>> ....................
> >>>> get_user_pages_fast();
> >>>> ...
> >>>> f->data = vmap();
> >>>> .......................
> >>>>
> >>>>
> >>> What I understand is that you get the pages of posix_memalign by get_user_pages_fast
> >>> and then that pages are mapped at kernel vmalloc space by vmap.
> >>>
> >>> Is it for removing copy overhead from kernel to user ?
> >>>
> >>>
> >>>
> >> I need a large range of virtual contigous memory to store my video
> >> frame(about 203 pages). When I received a full frame ,I will queue the
> >> buffer in
> >> a VIDIOC queue,which will be remove by the VIDIOC_DQBUF.'
> >>
> >
> > I can't understand your point.
> > Sorry for that.
> >
> > Could you explain more detail relation (user buffer which is allocated by posix_memalign) and (kernel buffer which is mapped by vmap) ?
> >
> >
> :) sorry for my poor english.
> [1] VLC uses the posix_memalign to allocate a big buffer for a single
> frame(203 pages).
> [2] vmap sets up the mapping of virtual contigous address for gup()'s
> pages array(the pages are not consecutive).
> memcpy() needs a contiguous address to copy in kernel mode.
> [3] my driver do some specail operations to received data, then
> memcopy the data to the buffer get in step [2].
> [4] when the buffer is full ,I will give the the user process (VLC).
>
> That's all.
--
Kinds Regards
Minchan Kim
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 4:53 ` Minchan Kim
@ 2009-04-20 5:05 ` KOSAKI Motohiro
2009-04-20 5:05 ` Huang Shijie
1 sibling, 0 replies; 25+ messages in thread
From: KOSAKI Motohiro @ 2009-04-20 5:05 UTC (permalink / raw)
To: Minchan Kim; +Cc: kosaki.motohiro, Huang Shijie, linux-mm
> It's rather awkward use case of get_user_pages.
>
> If you want to share one big buffer between kernel and user,
> You can vmalloc and remap_pfn_range.
> You can refer cpia_mmap in drivers/media/video/cpia.c
Hm, good opinion.
gup()ed page stay in lru, but remap_pfn_range() page doesn't.
it cause
gup() pinning:
merit
- Can processing any user process patch
(DirectIO need it)
demerit
- introduce reclaim slowdown
remap
merit
- Don't cause any slowdown
demerit
- can be used on some special situation only.
Then, driver can use special memory and need long time pinning.
remap_pfn_range() is better.
but, DirectIO liked general pinning need gup() pinning.
I think.
but I'm not remap_pfn_range() specialist. perhaps I can talk about
incorrect thing ...
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 4:53 ` Minchan Kim
2009-04-20 5:05 ` KOSAKI Motohiro
@ 2009-04-20 5:05 ` Huang Shijie
2009-04-20 5:19 ` KOSAKI Motohiro
2009-04-20 5:24 ` Minchan Kim
1 sibling, 2 replies; 25+ messages in thread
From: Huang Shijie @ 2009-04-20 5:05 UTC (permalink / raw)
To: Minchan Kim; +Cc: linux-mm
Minchan Kim a??e??:
> On Mon, 20 Apr 2009 11:28:00 +0800
> Huang Shijie <shijie8@gmail.com> wrote:
>
> I will summarize your method.
> Is right ?
>
>
> kernel(driver) application
>
> posix_memalign(buffer)
> ioctl(buffer)
>
> ioctl handler
> get_user_pages(pages);
> /* This pages are mapped at user's vma'
> address space */
> vaddr = vmap(pages);
> /* This pages are mapped at vmalloc space */
> .
> .
> <after sometime,
> It may change to other process context>
> .
> .
> interrupt handler in your driver
> memcpy(vaddr, src, len);
> notify_user();
>
> processing(buffer);
>
> It's rather awkward use case of get_user_pages.
>
> If you want to share one big buffer between kernel and user,
> You can vmalloc and remap_pfn_range.
>
The v4l2 method IO_METHOD_MMAP does use the vmaloc() method you told above ,
our driver also support this method,we user vmalloc /remap_vmalloc_range().
But the v4l2 method IO_METHOD_USERPTR must use the method I told above.
> You can refer cpia_mmap in drivers/media/video/cpia.c
>
>
>
>
>> Minchan Kim a??e??:
>>
>>> On Mon, 20 Apr 2009 10:15:55 +0800
>>> Huang Shijie <shijie8@gmail.com> wrote:
>>>
>>>
>>>
>>>> Minchan Kim a??e??:
>>>>
>>>>
>>>>> On Fri, 17 Apr 2009 15:01:01 +0800
>>>>> Huang Shijie <shijie8@gmail.com> wrote:
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>> I'm writting a driver for a video card with the V4L2 interface .
>>>>>> V4L2 interface supports the USER-POINTER method for the video frame
>>>>>> handling.
>>>>>>
>>>>>> VLC player supports the USER-POINTER method,while MPALYER does not.
>>>>>>
>>>>>> In the USER-POINTER method, VLC will call the posix_memalign() to
>>>>>> allocate
>>>>>> 203 pages in certain PAL mode (that is 720*576*2) for a single frame.
>>>>>> In my driver , I call the get_user_pages_fast() to obtain the pages
>>>>>> array,and then call
>>>>>> the vmap() to map the pages to VMALLOC space for the memcpy().The code
>>>>>> shows below:
>>>>>> ....................
>>>>>> get_user_pages_fast();
>>>>>> ...
>>>>>> f->data = vmap();
>>>>>> .......................
>>>>>>
>>>>>>
>>>>>>
>>>>> What I understand is that you get the pages of posix_memalign by get_user_pages_fast
>>>>> and then that pages are mapped at kernel vmalloc space by vmap.
>>>>>
>>>>> Is it for removing copy overhead from kernel to user ?
>>>>>
>>>>>
>>>>>
>>>>>
>>>> I need a large range of virtual contigous memory to store my video
>>>> frame(about 203 pages). When I received a full frame ,I will queue the
>>>> buffer in
>>>> a VIDIOC queue,which will be remove by the VIDIOC_DQBUF.'
>>>>
>>>>
>>> I can't understand your point.
>>> Sorry for that.
>>>
>>> Could you explain more detail relation (user buffer which is allocated by posix_memalign) and (kernel buffer which is mapped by vmap) ?
>>>
>>>
>>>
>> :) sorry for my poor english.
>> [1] VLC uses the posix_memalign to allocate a big buffer for a single
>> frame(203 pages).
>> [2] vmap sets up the mapping of virtual contigous address for gup()'s
>> pages array(the pages are not consecutive).
>> memcpy() needs a contiguous address to copy in kernel mode.
>> [3] my driver do some specail operations to received data, then
>> memcopy the data to the buffer get in step [2].
>> [4] when the buffer is full ,I will give the the user process (VLC).
>>
>> That's all.
>>
>
>
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 5:05 ` Huang Shijie
@ 2009-04-20 5:19 ` KOSAKI Motohiro
2009-04-20 5:37 ` Huang Shijie
2009-04-20 5:24 ` Minchan Kim
1 sibling, 1 reply; 25+ messages in thread
From: KOSAKI Motohiro @ 2009-04-20 5:19 UTC (permalink / raw)
To: Huang Shijie; +Cc: kosaki.motohiro, Minchan Kim, linux-mm
> The v4l2 method IO_METHOD_MMAP does use the vmaloc() method you told above ,
> our driver also support this method,we user vmalloc /remap_vmalloc_range().
>
> But the v4l2 method IO_METHOD_USERPTR must use the method I told above.
I guess you mean IO_METHOD_USERPTR can't use remap_vmalloc_range, right?
we need explanation of v4l2 requirement.
Can you explain why v4l2 use two different way? Why application developer
need two way?
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 5:19 ` KOSAKI Motohiro
@ 2009-04-20 5:37 ` Huang Shijie
2009-04-20 7:59 ` KOSAKI Motohiro
0 siblings, 1 reply; 25+ messages in thread
From: Huang Shijie @ 2009-04-20 5:37 UTC (permalink / raw)
To: KOSAKI Motohiro; +Cc: Minchan Kim, linux-mm
KOSAKI Motohiro a??e??:
>> The v4l2 method IO_METHOD_MMAP does use the vmaloc() method you told above ,
>> our driver also support this method,we user vmalloc /remap_vmalloc_range().
>>
>> But the v4l2 method IO_METHOD_USERPTR must use the method I told above.
>>
>
> I guess you mean IO_METHOD_USERPTR can't use remap_vmalloc_range, right?
>
Yes.
IO_METHOD_USERPTR method uses the anonymous pages allocated by posix_memalign(),
while the remap_vmalloc_range() use the pages alloced by vmalloc().
> we need explanation of v4l2 requirement.
>
> Can you explain why v4l2 use two different way? Why application developer
> need two way?
>
>
>
pleasure :)
http://v4l2spec.bytesex.org/spec/r13696.htm
shows the vidioc_reqbufs(). It determines the method of IO : "Memory
Mapping or User Pointer I/O"
The application developers can support any methodes of the Two, there is
no mandatory request to realize
both methods. For example, the Mplayer only support the "memory
maping" method ,and it does't support the "user pointer",
while the VLC supports both.
The full spec is below.
http://v4l2spec.bytesex.org/spec/
>
>
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 5:37 ` Huang Shijie
@ 2009-04-20 7:59 ` KOSAKI Motohiro
2009-04-20 8:23 ` Huang Shijie
0 siblings, 1 reply; 25+ messages in thread
From: KOSAKI Motohiro @ 2009-04-20 7:59 UTC (permalink / raw)
To: Huang Shijie; +Cc: kosaki.motohiro, Minchan Kim, linux-mm
> http://v4l2spec.bytesex.org/spec/r13696.htm
> shows the vidioc_reqbufs(). It determines the method of IO : "Memory
> Mapping or User Pointer I/O"
>
> The application developers can support any methodes of the Two, there is
> no mandatory request to realize
> both methods. For example, the Mplayer only support the "memory
> maping" method ,and it does't support the "user pointer",
> while the VLC supports both.
I greped VIDIOC_REQBUFS on current tree.
Almost driver has following check.
if (rb->memory != V4L2_MEMORY_MMAP)
return -EINVAL;
IOW, almost one don't provide V4L2_MEMORY_USERPTR method.
Thus, I think any userland application don't want use V4L2_MEMORY_USERPTR.
I recommend you also return -EINVAL.
I think we can't implement V4L2_MEMORY_USERPTR properly.
it is mistake by specification.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 7:59 ` KOSAKI Motohiro
@ 2009-04-20 8:23 ` Huang Shijie
2009-04-20 9:24 ` KOSAKI Motohiro
0 siblings, 1 reply; 25+ messages in thread
From: Huang Shijie @ 2009-04-20 8:23 UTC (permalink / raw)
To: KOSAKI Motohiro; +Cc: Minchan Kim, linux-mm, Huang Shijie
KOSAKI Motohiro a??e??:
>> http://v4l2spec.bytesex.org/spec/r13696.htm
>> shows the vidioc_reqbufs(). It determines the method of IO : "Memory
>> Mapping or User Pointer I/O"
>>
>> The application developers can support any methodes of the Two, there is
>> no mandatory request to realize
>> both methods. For example, the Mplayer only support the "memory
>> maping" method ,and it does't support the "user pointer",
>> while the VLC supports both.
>>
>
> I greped VIDIOC_REQBUFS on current tree.
> Almost driver has following check.
>
> if (rb->memory != V4L2_MEMORY_MMAP)
> return -EINVAL;
>
> IOW, almost one don't provide V4L2_MEMORY_USERPTR method.
> Thus, I think any userland application don't want use V4L2_MEMORY_USERPTR.
> I recommend you also return -EINVAL.
>
>
Thanks.
In the V4L2_MEMORY_USERPTR method, what I want to do is pin the
anonymous pages in memory.
I used to add the VM_LOCKED to vma associated with the pages.In my
opinion, the pages will:
LRU_ACTIVE_ANON ---> LRU_INACTIVE_ANON---> LRU_UNEVICTABLE
so the pages are pinned in memory.It was ugly, but it works I think.
Do you have any suggestions about this method?
> I think we can't implement V4L2_MEMORY_USERPTR properly.
> it is mistake by specification.
>
>
>
>
>
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 8:23 ` Huang Shijie
@ 2009-04-20 9:24 ` KOSAKI Motohiro
2009-04-20 9:47 ` Huang Shijie
0 siblings, 1 reply; 25+ messages in thread
From: KOSAKI Motohiro @ 2009-04-20 9:24 UTC (permalink / raw)
To: Huang Shijie; +Cc: kosaki.motohiro, Minchan Kim, linux-mm
> In the V4L2_MEMORY_USERPTR method, what I want to do is pin the
> anonymous pages in memory.
>
> I used to add the VM_LOCKED to vma associated with the pages.In my
> opinion, the pages will:
> LRU_ACTIVE_ANON ---> LRU_INACTIVE_ANON---> LRU_UNEVICTABLE
>
> so the pages are pinned in memory.It was ugly, but it works I think.
> Do you have any suggestions about this method?
page migration (e.g. move_pages) ignore MLOCK.
maybe, VM_LOCKED + gut()ed solved it partially :)
but, user process still can call munlock. it cause disaster.
I still think -EINVAL is better.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 9:24 ` KOSAKI Motohiro
@ 2009-04-20 9:47 ` Huang Shijie
2009-04-21 1:43 ` KOSAKI Motohiro
2009-04-21 19:47 ` Christoph Lameter
0 siblings, 2 replies; 25+ messages in thread
From: Huang Shijie @ 2009-04-20 9:47 UTC (permalink / raw)
To: KOSAKI Motohiro; +Cc: Minchan Kim, linux-mm, Huang Shijie
KOSAKI Motohiro a??e??:
>> In the V4L2_MEMORY_USERPTR method, what I want to do is pin the
>> anonymous pages in memory.
>>
>> I used to add the VM_LOCKED to vma associated with the pages.In my
>> opinion, the pages will:
>> LRU_ACTIVE_ANON ---> LRU_INACTIVE_ANON---> LRU_UNEVICTABLE
>>
>> so the pages are pinned in memory.It was ugly, but it works I think.
>> Do you have any suggestions about this method?
>>
>
> page migration (e.g. move_pages) ignore MLOCK.
> maybe, VM_LOCKED + gut()ed solved it partially :)
>
>
My old codes used the get_user_pages()/VM_LOCKED just as you said.
I will read the migration code, I am not clear about why the gup() can
stop the migraion.
> but, user process still can call munlock. it cause disaster.
> I still think -EINVAL is better.
>
>
>
Why the user process call munlock? VLC or Mplayer do not call it, so I
don't worry about that.
Our video card is still not on sale.So I can wait until the bug is fixed. :)
If there is no method to bypass the problem in future,I will return -EINVAL.
thanks
>
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 9:47 ` Huang Shijie
@ 2009-04-21 1:43 ` KOSAKI Motohiro
2009-04-21 19:47 ` Christoph Lameter
1 sibling, 0 replies; 25+ messages in thread
From: KOSAKI Motohiro @ 2009-04-21 1:43 UTC (permalink / raw)
To: Huang Shijie; +Cc: kosaki.motohiro, Minchan Kim, linux-mm
> > but, user process still can call munlock. it cause disaster.
> > I still think -EINVAL is better.
> >
> Why the user process call munlock? VLC or Mplayer do not call it, so I
> don't worry about that.
>
> Our video card is still not on sale.So I can wait until the bug is fixed. :)
> If there is no method to bypass the problem in future,I will return -EINVAL.
We don't assume any userspace behavior in kernel. but you can ignore
our recommendation, of cource :)
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 9:47 ` Huang Shijie
2009-04-21 1:43 ` KOSAKI Motohiro
@ 2009-04-21 19:47 ` Christoph Lameter
1 sibling, 0 replies; 25+ messages in thread
From: Christoph Lameter @ 2009-04-21 19:47 UTC (permalink / raw)
To: Huang Shijie; +Cc: KOSAKI Motohiro, Minchan Kim, linux-mm
On Mon, 20 Apr 2009, Huang Shijie wrote:
> I will read the migration code, I am not clear about why the gup() can stop
> the migraion.
Because it increases the refcount of the page. Page migration is then
unable to account for all the references to a page and therefore the page
cannot be migrated.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 5:05 ` Huang Shijie
2009-04-20 5:19 ` KOSAKI Motohiro
@ 2009-04-20 5:24 ` Minchan Kim
2009-04-20 5:42 ` Huang Shijie
1 sibling, 1 reply; 25+ messages in thread
From: Minchan Kim @ 2009-04-20 5:24 UTC (permalink / raw)
To: Huang Shijie; +Cc: Minchan Kim, linux-mm
On Mon, 20 Apr 2009 13:05:33 +0800
Huang Shijie <shijie8@gmail.com> wrote:
> Minchan Kim a??e??:
> > On Mon, 20 Apr 2009 11:28:00 +0800
> > Huang Shijie <shijie8@gmail.com> wrote:
> >
> > I will summarize your method.
> > Is right ?
> >
> >
> > kernel(driver) application
> >
> > posix_memalign(buffer)
> > ioctl(buffer)
> >
> > ioctl handler
> > get_user_pages(pages);
> > /* This pages are mapped at user's vma'
> > address space */
> > vaddr = vmap(pages);
> > /* This pages are mapped at vmalloc space */
> > .
> > .
> > <after sometime,
> > It may change to other process context>
> > .
> > .
> > interrupt handler in your driver
> > memcpy(vaddr, src, len);
> > notify_user();
> >
> > processing(buffer);
> >
> > It's rather awkward use case of get_user_pages.
> >
> > If you want to share one big buffer between kernel and user,
> > You can vmalloc and remap_pfn_range.
> >
> The v4l2 method IO_METHOD_MMAP does use the vmaloc() method you told above ,
> our driver also support this method,we user vmalloc /remap_vmalloc_range().
>
> But the v4l2 method IO_METHOD_USERPTR must use the method I told above.
I can't understand IO_METHOD_USERPTR's benefit compared with IO_METHOD_MMAP.
I think both solution can support that application programmer can handle buffer as like pointer and kernel can reduce copy overhead from kernel to user.
Why do you have to support IO_METHOD_USERPTR?
If you can justify your goal, we can add locked GUP.
--
Kinds Regards
Minchan Kim
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 5:24 ` Minchan Kim
@ 2009-04-20 5:42 ` Huang Shijie
0 siblings, 0 replies; 25+ messages in thread
From: Huang Shijie @ 2009-04-20 5:42 UTC (permalink / raw)
To: Minchan Kim; +Cc: linux-mm
Minchan Kim a??e??:
> On Mon, 20 Apr 2009 13:05:33 +0800
> Huang Shijie <shijie8@gmail.com> wrote:
>
>
>> Minchan Kim a??e??:
>>
>>> On Mon, 20 Apr 2009 11:28:00 +0800
>>> Huang Shijie <shijie8@gmail.com> wrote:
>>>
>>> I will summarize your method.
>>> Is right ?
>>>
>>>
>>> kernel(driver) application
>>>
>>> posix_memalign(buffer)
>>> ioctl(buffer)
>>>
>>> ioctl handler
>>> get_user_pages(pages);
>>> /* This pages are mapped at user's vma'
>>> address space */
>>> vaddr = vmap(pages);
>>> /* This pages are mapped at vmalloc space */
>>> .
>>> .
>>> <after sometime,
>>> It may change to other process context>
>>> .
>>> .
>>> interrupt handler in your driver
>>> memcpy(vaddr, src, len);
>>> notify_user();
>>>
>>> processing(buffer);
>>>
>>> It's rather awkward use case of get_user_pages.
>>>
>>> If you want to share one big buffer between kernel and user,
>>> You can vmalloc and remap_pfn_range.
>>>
>>>
>> The v4l2 method IO_METHOD_MMAP does use the vmaloc() method you told above ,
>> our driver also support this method,we user vmalloc /remap_vmalloc_range().
>>
>> But the v4l2 method IO_METHOD_USERPTR must use the method I told above.
>>
>
> I can't understand IO_METHOD_USERPTR's benefit compared with IO_METHOD_MMAP.
> I think both solution can support that application programmer can handle buffer as like pointer and kernel can reduce copy overhead from kernel to user.
>
>
yes ,I agree with you .
But the application programmers do not know which method is more efficient.
> Why do you have to support IO_METHOD_USERPTR?
>
just for fun. For the v4l2 spec has the method ,why I don't realize it?
> If you can justify your goal, we can add locked GUP.
>
>
I can't .
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 2:42 ` Minchan Kim
2009-04-20 3:28 ` Huang Shijie
@ 2009-04-20 3:57 ` Huang Shijie
2009-04-22 6:08 ` Huang Shijie
2 siblings, 0 replies; 25+ messages in thread
From: Huang Shijie @ 2009-04-20 3:57 UTC (permalink / raw)
To: Minchan Kim; +Cc: linux-mm
Minchan Kim a??e??:
> On Mon, 20 Apr 2009 10:15:55 +0800
> Huang Shijie <shijie8@gmail.com> wrote:
>
>
>> Minchan Kim a??e??:
>>
>>> On Fri, 17 Apr 2009 15:01:01 +0800
>>> Huang Shijie <shijie8@gmail.com> wrote:
>>>
>>>
>>>
>>>> I'm writting a driver for a video card with the V4L2 interface .
>>>> V4L2 interface supports the USER-POINTER method for the video frame
>>>> handling.
>>>>
>>>> VLC player supports the USER-POINTER method,while MPALYER does not.
>>>>
>>>> In the USER-POINTER method, VLC will call the posix_memalign() to
>>>> allocate
>>>> 203 pages in certain PAL mode (that is 720*576*2) for a single frame.
>>>> In my driver , I call the get_user_pages_fast() to obtain the pages
>>>> array,and then call
>>>> the vmap() to map the pages to VMALLOC space for the memcpy().The code
>>>> shows below:
>>>> ....................
>>>> get_user_pages_fast();
>>>> ...
>>>> f->data = vmap();
>>>> .......................
>>>>
>>>>
>>> What I understand is that you get the pages of posix_memalign by get_user_pages_fast
>>> and then that pages are mapped at kernel vmalloc space by vmap.
>>>
>>> Is it for removing copy overhead from kernel to user ?
>>>
>>>
>>>
>> I need a large range of virtual contigous memory to store my video
>> frame(about 203 pages). When I received a full frame ,I will queue the
>> buffer in
>> a VIDIOC queue,which will be remove by the VIDIOC_DQBUF.'
>>
>
> I can't understand your point.
> Sorry for that.
>
> Could you explain more detail relation (user buffer which is allocated by posix_memalign) and (kernel buffer which is mapped by vmap) ?
>
>
[1] fork() will copy the range of 768-1024 index of swapper_pg_dir to
mm->pgd.
[2] vmap setup mapping in the swapper_pg_dir.
[3] my driver will use the mm->pgd in the ioctl context.for the reason
of [1],driver could see
the address setup by vmap.
>>>> In comments, it said :
>>>> "
>>>> +/**
>>>> + * get_user_pages_fast() - pin user pages in memory
>>>> + * @start: starting user address
>>>> + * @nr_pages: number of pages from start to pin
>>>> + * @write: whether pages will be written to
>>>> + * @pages: array that receives pointers to the pages pinned.
>>>> + * Should be at least nr_pages long.
>>>> "
>>>>
>>>> But after I digged the code of kswap and the get_user_pages(called by
>>>> get_user_pages_fast),
>>>> I did not find how the pages pinned in memory.I really need the pages
>>>> pinned in memory.
>>>>
>>>> Assume page A is one of the pages obtained by get_user_pages_fast()
>>>> during page-fault.
>>>>
>>>> [1] page A will on the LRU_ACTIVE_ANON list;
>>>> the _count of page A increment by one;
>>>> PTE for page A will be set ACCESSED.
>>>>
>>>> [2] kswapd will scan the lru list,and move page A from LRU_ACTIVE_ANON
>>>> to LRU_INACTIVE_ANON.
>>>> In the shrink_page_list(), there is nothing can stop page A been
>>>> swapped out.
>>>> I don't think the page_reference() can move page A back to
>>>> LRU_ACTIVE_ANON.In my driver,
>>>> I am not sure if the VLC can access the page A.
>>>>
>>>> Is this a bug? or I miss something?
>>>> Thanks .
>>>>
>>>>
>>> If above my assumption is right, It's not a BUG.
>>> You get the application's pages by get_user_pages_fast.
>>> 'Page pinning' means it shouldn't be freed.
>>> Application's pages always can be swapped out.
>>> If you don't want to swap out the page, you should use mlock.
>>> If you use mlock, kernel won't insert the page to lru [in]active list.
>>> So the page never can be swapped out.
>>>
>>>
>>>
>> Yes, it not a bug .
>>
>> I read the kernel code again. In my case ,the kernel will pin the pages
>> in memory.
>> I missed function is_page_cache_freeable() in the pageout().
>>
>> In my case, is_page_cache_freeable()will return false ,for
>> page_count(page) is 3 now:
>> <1> one is from alloc_page_* in page fault.
>> <2> one is from get_usr_pages()
>> <3> one is from add_to_swap() in shrink_page_list()
>>
>
> One more, try_to_unmap will call page_cache_release.
> So, count is 2.
>
>
>> So ,there is no need to use the mlock, it will mess my driver.
>> is_page_cache_freeable()will return PAGE_KEEP, and page is locked in
>> swap cache.
>>
>>
>
> I can't understand your point exactly yet.
> But what I mean is following as in user mode
>
> posix_memalignq(&buffer);
> mlock(buffer, buffer_len);
>
> I will not dirty your driver.
> Do I miss something ?
>
>
>> Unfortunately, the page is unmaped, and the PTE of the page has been
>> replaced by a swp_entry_t .
>> When the process read the page ,it will raise a page fault again, the
>> kernel will find the page in the
>> swap cache, and requeue the page in LRU_ACTIVE_ANON, ---I think it is a
>> vicious circle for the kernel.
>>
>> I think there two places to put back the gup() pages.
>> <1> isolate_page_glable()
>> <2> in the shrink_page_list(), before called the try_to_unmap().
>> KOSAKI Motohiro 's patch takes effect in the second place.
>> I think the first place is better.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-20 2:42 ` Minchan Kim
2009-04-20 3:28 ` Huang Shijie
2009-04-20 3:57 ` Huang Shijie
@ 2009-04-22 6:08 ` Huang Shijie
2009-04-22 9:46 ` Minchan Kim
2 siblings, 1 reply; 25+ messages in thread
From: Huang Shijie @ 2009-04-22 6:08 UTC (permalink / raw)
To: Minchan Kim; +Cc: linux-mm, KOSAKI Motohiro
>> I read the kernel code again. In my case ,the kernel will pin the pages
>> in memory.
>> I missed function is_page_cache_freeable() in the pageout().
>>
>> In my case, is_page_cache_freeable()will return false ,for
>> page_count(page) is 3 now:
>> <1> one is from alloc_page_* in page fault.
>> <2> one is from get_usr_pages()
>> <3> one is from add_to_swap() in shrink_page_list()
>>
>
> One more, try_to_unmap will call page_cache_release.
> So, count is 2.
>
>
I found I missed something.When code reachs is_page_cache_freeable().
page_count(page) is 3:
<1> alloc_page_* in page fault . [page count is 1]
<2> get_usr_pages(). [page count is 2]
<3> isolate_pages_global() [page count is 3]
<4> add_to_swap() [page count is 4]
<5> try_to_unmap() [page count is 3]
so it not a bug, just a vicious circle.
Do i miss something?
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: Does get_user_pages_fast lock the user pages in memory in my case?
2009-04-22 6:08 ` Huang Shijie
@ 2009-04-22 9:46 ` Minchan Kim
0 siblings, 0 replies; 25+ messages in thread
From: Minchan Kim @ 2009-04-22 9:46 UTC (permalink / raw)
To: Huang Shijie; +Cc: linux-mm, KOSAKI Motohiro
On Wed, Apr 22, 2009 at 3:08 PM, Huang Shijie <shijie8@gmail.com> wrote:
>
>>> I read the kernel code again. In my case ,the kernel will pin the pages
>>> in memory.
>>> I missed function is_page_cache_freeable() in the pageout().
>>>
>>> In my case, is_page_cache_freeable()will return false ,for
>>> page_count(page) is 3 now:
>>> <1> one is from alloc_page_* in page fault.
>>> <2> one is from get_usr_pages()
>>> <3> one is from add_to_swap() in shrink_page_list()
>>>
>>
>> One more, try_to_unmap will call page_cache_release. So, count is 2.
>>
>
> I found I missed something.When code reachs is_page_cache_freeable().
> page_count(page) is 3:
>
> <1> alloc_page_* in page fault . [page count is 1]
> <2> get_usr_pages(). [page count is 2]
> <3> isolate_pages_global() [page count is 3]
> <4> add_to_swap() [page count is 4]
> <5> try_to_unmap() [page count is 3]
>
Yes. It seems you're right.
I missed isolate. ;-;
Thanks for fixing me.
> so it not a bug, just a vicious circle.
>
> Do i miss something?
>
>
>
--
Kinds regards,
Minchan Kim
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 25+ messages in thread