Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Nirmoy Das <nirmoy.das@intel.com>
To: "Christian König" <christian.koenig@amd.com>,
	dri-devel@lists.freedesktop.org
Cc: intel-xe@lists.freedesktop.org,
	"Matthew Auld" <matthew.auld@intel.com>,
	"Thomas Hellström" <thomas.hellstrom@linux.intel.com>
Subject: Re: [PATCH] drm/ttm: Add a flag to allow drivers to skip clear-on-free
Date: Mon, 24 Jun 2024 14:19:45 +0200	[thread overview]
Message-ID: <dd330f73-d83f-4894-a48b-aa2ad004a82e@intel.com> (raw)
In-Reply-To: <7d576b58-81f7-4f2c-8932-70b96764c634@amd.com>

Hi Christian,

On 6/24/2024 1:39 PM, Christian König wrote:
> Am 24.06.24 um 12:07 schrieb Nirmoy Das:
>> Add TTM_TT_FLAG_CLEARED_ON_FREE, which DRM drivers can set before
>> releasing backing stores if they want to skip clear-on-free.
>>
>> Cc: Matthew Auld <matthew.auld@intel.com>
>> Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
>> Suggested-by: Christian König <christian.koenig@amd.com>
>> Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
>
> Reviewed-by: Christian König <christian.koenig@amd.com>

Thanks a lot. Could you also please help get this merge ?


Regards,

Nirmoy

>
>> ---
>>   drivers/gpu/drm/ttm/ttm_pool.c | 18 +++++++++++-------
>>   include/drm/ttm/ttm_tt.h       |  6 +++++-
>>   2 files changed, 16 insertions(+), 8 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/ttm/ttm_pool.c 
>> b/drivers/gpu/drm/ttm/ttm_pool.c
>> index 6e1fd6985ffc..b78ee7524bcf 100644
>> --- a/drivers/gpu/drm/ttm/ttm_pool.c
>> +++ b/drivers/gpu/drm/ttm/ttm_pool.c
>> @@ -222,15 +222,18 @@ static void ttm_pool_unmap(struct ttm_pool 
>> *pool, dma_addr_t dma_addr,
>>   }
>>     /* Give pages into a specific pool_type */
>> -static void ttm_pool_type_give(struct ttm_pool_type *pt, struct page 
>> *p)
>> +static void ttm_pool_type_give(struct ttm_pool_type *pt, struct page 
>> *p,
>> +                   bool cleared)
>>   {
>>       unsigned int i, num_pages = 1 << pt->order;
>>   -    for (i = 0; i < num_pages; ++i) {
>> -        if (PageHighMem(p))
>> -            clear_highpage(p + i);
>> -        else
>> -            clear_page(page_address(p + i));
>> +    if (!cleared) {
>> +        for (i = 0; i < num_pages; ++i) {
>> +            if (PageHighMem(p))
>> +                clear_highpage(p + i);
>> +            else
>> +                clear_page(page_address(p + i));
>> +        }
>>       }
>>         spin_lock(&pt->lock);
>> @@ -394,6 +397,7 @@ static void ttm_pool_free_range(struct ttm_pool 
>> *pool, struct ttm_tt *tt,
>>                   pgoff_t start_page, pgoff_t end_page)
>>   {
>>       struct page **pages = &tt->pages[start_page];
>> +    bool cleared = tt->page_flags & TTM_TT_FLAG_CLEARED_ON_FREE;
>>       unsigned int order;
>>       pgoff_t i, nr;
>>   @@ -407,7 +411,7 @@ static void ttm_pool_free_range(struct ttm_pool 
>> *pool, struct ttm_tt *tt,
>>             pt = ttm_pool_select_type(pool, caching, order);
>>           if (pt)
>> -            ttm_pool_type_give(pt, *pages);
>> +            ttm_pool_type_give(pt, *pages, cleared);
>>           else
>>               ttm_pool_free_page(pool, caching, order, *pages);
>>       }
>> diff --git a/include/drm/ttm/ttm_tt.h b/include/drm/ttm/ttm_tt.h
>> index 2b9d856ff388..cfaf49de2419 100644
>> --- a/include/drm/ttm/ttm_tt.h
>> +++ b/include/drm/ttm/ttm_tt.h
>> @@ -85,6 +85,9 @@ struct ttm_tt {
>>        * fault handling abuses the DMA api a bit and dma_map_attrs 
>> can't be
>>        * used to assure pgprot always matches.
>>        *
>> +     * TTM_TT_FLAG_CLEARED_ON_FREE: Set this if a drm driver handles
>> +     * clearing backing store
>> +     *
>>        * TTM_TT_FLAG_PRIV_POPULATED: TTM internal only. DO NOT USE. 
>> This is
>>        * set by TTM after ttm_tt_populate() has successfully 
>> returned, and is
>>        * then unset when TTM calls ttm_tt_unpopulate().
>> @@ -94,8 +97,9 @@ struct ttm_tt {
>>   #define TTM_TT_FLAG_EXTERNAL        BIT(2)
>>   #define TTM_TT_FLAG_EXTERNAL_MAPPABLE    BIT(3)
>>   #define TTM_TT_FLAG_DECRYPTED        BIT(4)
>> +#define TTM_TT_FLAG_CLEARED_ON_FREE    BIT(5)
>>   -#define TTM_TT_FLAG_PRIV_POPULATED    BIT(5)
>> +#define TTM_TT_FLAG_PRIV_POPULATED    BIT(6)
>>       uint32_t page_flags;
>>       /** @num_pages: Number of pages in the page array. */
>>       uint32_t num_pages;
>

  reply	other threads:[~2024-06-24 12:20 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-24 10:07 [PATCH] drm/ttm: Add a flag to allow drivers to skip clear-on-free Nirmoy Das
2024-06-24 10:26 ` ✓ CI.Patch_applied: success for " Patchwork
2024-06-24 10:27 ` ✓ CI.checkpatch: " Patchwork
2024-06-24 10:28 ` ✓ CI.KUnit: " Patchwork
2024-06-24 10:40 ` ✓ CI.Build: " Patchwork
2024-06-24 10:42 ` ✗ CI.Hooks: failure " Patchwork
2024-06-24 10:43 ` ✗ CI.checksparse: warning " Patchwork
2024-06-24 11:06 ` ✓ CI.BAT: success " Patchwork
2024-06-24 11:39 ` [PATCH] " Christian König
2024-06-24 12:19   ` Nirmoy Das [this message]
2024-06-24 12:21     ` Christian König
2024-06-24 12:24       ` Nirmoy Das
2024-06-24 12:14 ` ✓ CI.FULL: success for " 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=dd330f73-d83f-4894-a48b-aa2ad004a82e@intel.com \
    --to=nirmoy.das@intel.com \
    --cc=christian.koenig@amd.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=matthew.auld@intel.com \
    --cc=thomas.hellstrom@linux.intel.com \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox