Intel-GFX Archive on lore.kernel.org
 help / color / Atom feed
From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Christoph Hellwig <hch@lst.de>
Cc: Juergen Gross <jgross@suse.com>,
	Stefano Stabellini <sstabellini@kernel.org>,
	Minchan Kim <minchan@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	intel-gfx@lists.freedesktop.org, x86@kernel.org,
	linux-kernel@vger.kernel.org,
	Matthew Wilcox <willy@infradead.org>,
	Chris Wilson <chris@chris-wilson.co.uk>,
	linux-mm@kvack.org, dri-devel@lists.freedesktop.org,
	xen-devel@lists.xenproject.org,
	Boris Ostrovsky <boris.ostrovsky@oracle.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Nitin Gupta <ngupta@vflare.org>,
	Matthew Auld <matthew.auld@intel.com>
Subject: Re: [Intel-gfx] [PATCH 3/6] drm/i915: use vmap in shmem_pin_map
Date: Tue, 22 Sep 2020 18:04:37 +0100
Message-ID: <1b05b9d6-a14c-85cd-0728-d0d40c9ff84b@linux.intel.com> (raw)
In-Reply-To: <20200922163346.GA1701@lst.de>


On 22/09/2020 17:33, Christoph Hellwig wrote:
> On Tue, Sep 22, 2020 at 05:13:45PM +0100, Tvrtko Ursulin wrote:
>>>    void *shmem_pin_map(struct file *file)
>>>    {
>>> -	const size_t n_pte = shmem_npte(file);
>>> -	pte_t *stack[32], **ptes, **mem;
>>
>> Chris can comment how much he'd miss the 32 page stack shortcut.
> 
> I'd like to see a profile that claim that kmalloc matters in a
> path that does a vmap and reads pages through the page cache.
> Especially when the kmalloc saves doing another page cache lookup
> on the free side.

Only reason I can come up with now is if mapping side is on a latency 
sensitive path, while un-mapping is lazy/delayed so can be more costly. 
Then fast map and extra cost on unmap may make sense.

It more applies to the other i915 patch, which implements a much more 
used API, but whether or not we can demonstrate any difference in the 
perf profiles I couldn't tell you without trying to collect some.

>> Is there something in vmap() preventing us from freeing the pages array
>> here? I can't spot anything that is holding on to the pointer. Or it was
>> just a sketch before you realized we could walk the vm_area?
>>
>> Also, I may be totally misunderstanding something, but I think you need to
>> assign area->pages manually so shmem_unpin_map can access it below.
> 
> We need area->pages to hold the pages for the free side.  That being
> said the patch I posted is broken because it never assigned to that.
> As said it was a sketch.  This is the patch I just rebooted into on
> my Laptop:
> 
> http://git.infradead.org/users/hch/misc.git/commitdiff/048522dfa26b6667adfb0371ff530dc263abe829
> 
> it needs extra prep patches from the series:
> 
> http://git.infradead.org/users/hch/misc.git/shortlog/refs/heads/alloc_vm_area
> 
>>>    	mapping_clear_unevictable(file->f_mapping);
>>> -	__shmem_unpin_map(file, ptr, shmem_npte(file));
>>> +	for (i = 0; i < shmem_npages(file); i++)
>>> +		put_page(area->pages[i]);
>>> +	kvfree(area->pages);
>>> +	vunmap(ptr);
>>
>> Is the verdict from mm experts that we can't use vfree due __free_pages vs
>> put_page differences?
> 
> Switched to vfree now.
> 
>> Could we get from ptes to pages, so that we don't have to keep the
>> area->pages array allocated for the duration of the pin?
> 
> We could do vmalloc_to_page, but that is fairly expensive (not as bad
> as reading from the page cache..).  Are you really worried about the
> allocation?

Not so much given how we don't even use shmem_pin_map outside selftests.

If we start using it I expect it will be for tiny objects anyway. Only 
if they end up being pinned for the lifetime of the driver, it may be a 
pointless waste of memory compared to the downsides of vmalloc_to_page. 
But we can revisit this particular edge case optimization if the need 
arises.

I'll look at your other i915 patch tomorrow.

Regards,

Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  reply index

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-18 16:37 [Intel-gfx] remove alloc_vm_area Christoph Hellwig
2020-09-18 16:37 ` [Intel-gfx] [PATCH 1/6] zsmalloc: switch from alloc_vm_area to get_vm_area Christoph Hellwig
2020-09-21 17:42   ` Minchan Kim
2020-09-21 18:17     ` Christoph Hellwig
2020-09-21 18:42       ` Minchan Kim
2020-09-21 18:43         ` Christoph Hellwig
2020-09-18 16:37 ` [Intel-gfx] [PATCH 2/6] mm: add a vmap_pfn function Christoph Hellwig
2020-09-18 16:37 ` [Intel-gfx] [PATCH 3/6] drm/i915: use vmap in shmem_pin_map Christoph Hellwig
2020-09-21 19:11   ` Matthew Wilcox
2020-09-22  6:22     ` Christoph Hellwig
2020-09-22  8:23       ` Tvrtko Ursulin
2020-09-22 14:31         ` Christoph Hellwig
2020-09-22 16:13           ` Tvrtko Ursulin
2020-09-22 16:33             ` Christoph Hellwig
2020-09-22 17:04               ` Tvrtko Ursulin [this message]
2020-09-23  6:11                 ` Christoph Hellwig
2020-09-22 11:21       ` Matthew Wilcox
2020-09-22 14:39         ` Christoph Hellwig
2020-09-22 14:53           ` Matthew Wilcox
2020-09-18 16:37 ` [Intel-gfx] [PATCH 4/6] drm/i915: use vmap in i915_gem_object_map Christoph Hellwig
2020-09-23  9:52   ` Tvrtko Ursulin
2020-09-23 13:41     ` Christoph Hellwig
2020-09-23 13:58       ` Tvrtko Ursulin
2020-09-23 14:44         ` Christoph Hellwig
2020-09-24 12:22           ` Tvrtko Ursulin
2020-09-24 13:23             ` Christoph Hellwig
2020-09-18 16:37 ` [Intel-gfx] [PATCH 5/6] xen/xenbus: use apply_to_page_range directly in xenbus_map_ring_pv Christoph Hellwig
2020-09-18 16:37 ` [Intel-gfx] [PATCH 6/6] x86/xen: open code alloc_vm_area in arch_gnttab_valloc Christoph Hellwig
2020-09-21 20:44   ` boris.ostrovsky
2020-09-22 14:58     ` Christoph Hellwig
2020-09-22 15:24       ` boris.ostrovsky
2020-09-22 15:27         ` Christoph Hellwig
2020-09-22 15:34           ` boris.ostrovsky
2020-09-18 17:03 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for series starting with [1/6] zsmalloc: switch from alloc_vm_area to get_vm_area Patchwork
2020-09-21 17:50 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for series starting with [1/6] zsmalloc: switch from alloc_vm_area to get_vm_area (rev2) Patchwork
2020-09-21 18:47 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for series starting with [1/6] zsmalloc: switch from alloc_vm_area to get_vm_area (rev3) Patchwork
2020-09-22 14:44 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for series starting with [1/6] zsmalloc: switch from alloc_vm_area to get_vm_area (rev4) Patchwork
2020-09-22 15:01 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for series starting with [1/6] zsmalloc: switch from alloc_vm_area to get_vm_area (rev5) Patchwork

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=1b05b9d6-a14c-85cd-0728-d0d40c9ff84b@linux.intel.com \
    --to=tvrtko.ursulin@linux.intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=boris.ostrovsky@oracle.com \
    --cc=chris@chris-wilson.co.uk \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=hch@lst.de \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=jgross@suse.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=matthew.auld@intel.com \
    --cc=minchan@kernel.org \
    --cc=ngupta@vflare.org \
    --cc=peterz@infradead.org \
    --cc=sstabellini@kernel.org \
    --cc=willy@infradead.org \
    --cc=x86@kernel.org \
    --cc=xen-devel@lists.xenproject.org \
    /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

Intel-GFX Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/intel-gfx/0 intel-gfx/git/0.git
	git clone --mirror https://lore.kernel.org/intel-gfx/1 intel-gfx/git/1.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 intel-gfx intel-gfx/ https://lore.kernel.org/intel-gfx \
		intel-gfx@lists.freedesktop.org
	public-inbox-index intel-gfx

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.freedesktop.lists.intel-gfx


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git