dri-devel.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: "Hellstrom, Thomas" <thomas.hellstrom@intel.com>
To: "daniel@ffwll.ch" <daniel@ffwll.ch>,
	"christian.koenig@amd.com" <christian.koenig@amd.com>
Cc: "dri-devel@lists.freedesktop.org" <dri-devel@lists.freedesktop.org>
Subject: Re: [PATCH 1/9] drm/ttm: initialize the system domain with defaults
Date: Mon, 27 Jul 2020 13:01:44 +0000	[thread overview]
Message-ID: <6cf8d9c160517eb1991e471cfc601e380c6fb127.camel@intel.com> (raw)
In-Reply-To: <419b0a05-63f7-3daa-8ebc-cc6636d8837b@amd.com>

On Mon, 2020-07-27 at 14:23 +0200, Christian König wrote:
> Am 27.07.20 um 12:50 schrieb Hellstrom, Thomas:
> > On Mon, 2020-07-27 at 12:39 +0200, Christian König wrote:
> > > Am 27.07.20 um 11:42 schrieb daniel@ffwll.ch:
> > > > On Thu, Jul 23, 2020 at 05:16:13PM +0200, Christian König
> > > > wrote:
> > > > > Instead of repeating that in each driver.
> > > > > 
> > > > > Signed-off-by: Christian König <christian.koenig@amd.com>
> > > > > ---
> > > > >    drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c    | 3 ---
> > > > >    drivers/gpu/drm/drm_gem_vram_helper.c      | 3 ---
> > > > >    drivers/gpu/drm/nouveau/nouveau_bo.c       | 3 ---
> > > > >    drivers/gpu/drm/qxl/qxl_ttm.c              | 3 ---
> > > > >    drivers/gpu/drm/radeon/radeon_ttm.c        | 3 ---
> > > > >    drivers/gpu/drm/ttm/ttm_bo.c               | 2 ++
> > > > >    drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c | 2 --
> > > > >    7 files changed, 2 insertions(+), 17 deletions(-)
> > > > > 
> > > > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> > > > > b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> > > > > index 0dd5e802091d..e57c49a91b73 100644
> > > > > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> > > > > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> > > > > @@ -84,9 +84,6 @@ static int amdgpu_init_mem_type(struct
> > > > > ttm_bo_device *bdev, uint32_t type,
> > > > >    	switch (type) {
> > > > >    	case TTM_PL_SYSTEM:
> > > > >    		/* System memory */
> > > > > -		man->flags = 0;
> > > > > -		man->available_caching = TTM_PL_MASK_CACHING;
> > > > > -		man->default_caching = TTM_PL_FLAG_CACHED;
> > > > >    		break;
> > > > >    	case TTM_PL_TT:
> > > > >    		/* GTT memory  */
> > > > > diff --git a/drivers/gpu/drm/drm_gem_vram_helper.c
> > > > > b/drivers/gpu/drm/drm_gem_vram_helper.c
> > > > > index 3296ed3df358..be177afdeb9a 100644
> > > > > --- a/drivers/gpu/drm/drm_gem_vram_helper.c
> > > > > +++ b/drivers/gpu/drm/drm_gem_vram_helper.c
> > > > > @@ -1009,9 +1009,6 @@ static int
> > > > > bo_driver_init_mem_type(struct
> > > > > ttm_bo_device *bdev, uint32_t type,
> > > > >    {
> > > > >    	switch (type) {
> > > > >    	case TTM_PL_SYSTEM:
> > > > > -		man->flags = 0;
> > > > > -		man->available_caching = TTM_PL_MASK_CACHING;
> > > > > -		man->default_caching = TTM_PL_FLAG_CACHED;
> > > > >    		break;
> > > > >    	case TTM_PL_VRAM:
> > > > >    		man->func = &ttm_bo_manager_func;
> > > > > diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c
> > > > > b/drivers/gpu/drm/nouveau/nouveau_bo.c
> > > > > index 4ccf937df0d0..53af25020bb2 100644
> > > > > --- a/drivers/gpu/drm/nouveau/nouveau_bo.c
> > > > > +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
> > > > > @@ -655,9 +655,6 @@ nouveau_bo_init_mem_type(struct
> > > > > ttm_bo_device
> > > > > *bdev, uint32_t type,
> > > > >    
> > > > >    	switch (type) {
> > > > >    	case TTM_PL_SYSTEM:
> > > > > -		man->flags = 0;
> > > > > -		man->available_caching = TTM_PL_MASK_CACHING;
> > > > > -		man->default_caching = TTM_PL_FLAG_CACHED;
> > > > >    		break;
> > > > >    	case TTM_PL_VRAM:
> > > > >    		man->flags = TTM_MEMTYPE_FLAG_FIXED;
> > > > > diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c
> > > > > b/drivers/gpu/drm/qxl/qxl_ttm.c
> > > > > index 1d8e07b8b19e..e9b8c921c1f0 100644
> > > > > --- a/drivers/gpu/drm/qxl/qxl_ttm.c
> > > > > +++ b/drivers/gpu/drm/qxl/qxl_ttm.c
> > > > > @@ -54,9 +54,6 @@ static int qxl_init_mem_type(struct
> > > > > ttm_bo_device *bdev, uint32_t type,
> > > > >    	switch (type) {
> > > > >    	case TTM_PL_SYSTEM:
> > > > >    		/* System memory */
> > > > > -		man->flags = 0;
> > > > > -		man->available_caching = TTM_PL_MASK_CACHING;
> > > > > -		man->default_caching = TTM_PL_FLAG_CACHED;
> > > > >    		break;
> > > > >    	case TTM_PL_VRAM:
> > > > >    	case TTM_PL_PRIV:
> > > > > diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c
> > > > > b/drivers/gpu/drm/radeon/radeon_ttm.c
> > > > > index b474781a0920..b4cb75361577 100644
> > > > > --- a/drivers/gpu/drm/radeon/radeon_ttm.c
> > > > > +++ b/drivers/gpu/drm/radeon/radeon_ttm.c
> > > > > @@ -76,9 +76,6 @@ static int radeon_init_mem_type(struct
> > > > > ttm_bo_device *bdev, uint32_t type,
> > > > >    	switch (type) {
> > > > >    	case TTM_PL_SYSTEM:
> > > > >    		/* System memory */
> > > > > -		man->flags = 0;
> > > > > -		man->available_caching = TTM_PL_MASK_CACHING;
> > > > > -		man->default_caching = TTM_PL_FLAG_CACHED;
> > > > >    		break;
> > > > >    	case TTM_PL_TT:
> > > > >    		man->func = &ttm_bo_manager_func;
> > > > > diff --git a/drivers/gpu/drm/ttm/ttm_bo.c
> > > > > b/drivers/gpu/drm/ttm/ttm_bo.c
> > > > > index 7c02ce784805..1f1f9e463265 100644
> > > > > --- a/drivers/gpu/drm/ttm/ttm_bo.c
> > > > > +++ b/drivers/gpu/drm/ttm/ttm_bo.c
> > > > > @@ -1677,6 +1677,8 @@ int ttm_bo_device_init(struct
> > > > > ttm_bo_device
> > > > > *bdev,
> > > > >    	 * Initialize the system memory buffer type.
> > > > >    	 * Other types need to be driver / IOCTL initialized.
> > > > >    	 */
> > > > > +	bdev->man[TTM_PL_SYSTEM].available_caching =
> > > > > TTM_PL_MASK_CACHING;
> > > > > +	bdev->man[TTM_PL_SYSTEM].default_caching =
> > > > > TTM_PL_FLAG_CACHED;
> > > > >    	ret = ttm_bo_init_mm(bdev, TTM_PL_SYSTEM, 0);
> > > > >    	if (unlikely(ret != 0))
> > > > >    		goto out_no_sys;
> > > > > diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
> > > > > b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
> > > > > index 1d78187eaba6..00cef1a3a178 100644
> > > > > --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
> > > > > +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
> > > > > @@ -742,8 +742,6 @@ static int vmw_init_mem_type(struct
> > > > > ttm_bo_device *bdev, uint32_t type,
> > > > >    	switch (type) {
> > > > >    	case TTM_PL_SYSTEM:
> > > > >    		/* System memory */
> > > > > -		man->available_caching = TTM_PL_FLAG_CACHED;
> > > > Above is CACHED, not CACHING, so needs to stay to overwrite the
> > > > default.
> > > Crap I missed that. Problem is that I wanted to remove the
> > > possibility
> > > to init the system domain with different caching attributes.
> > > 
> > > I don't see how vmwgfx is every going to use this? Thomas do you
> > > have
> > > any idea what this was good for?
> > I interpret that like vmwgfx can't handle any other caching
> > attribute
> > than 'cached'. But I can't see anyone feeding vmwgfx system memory
> > buffers with other caching attributes. In that case, those would be
> > ignored.
> 
> Yeah, agree. So any objections to change that setting in vmwgfx?
> 
> This way we could have the same settings for all drivers in the
> kernel 
> and I don't see why any driver should have something special here.
> 
> Christian.
> 

No objections from my side, although I figure Roland has the last
saying.

/Thomas



----------------------------------------------------------------------
Intel Sweden AB
Registered Office: Isafjordsgatan 30B, 164 40 Kista, Stockholm, Sweden
Registration Number: 556189-6027

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  reply	other threads:[~2020-07-27 13:01 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-23 15:16 [PATCH 1/9] drm/ttm: initialize the system domain with defaults Christian König
2020-07-23 15:16 ` [PATCH 2/9] drm/ttm: remove TTM_MEMTYPE_FLAG_FIXED Christian König
2020-07-27  9:48   ` daniel
2020-07-27  9:54     ` Christian König
2020-07-27 10:27       ` daniel
2020-07-23 15:16 ` [PATCH 3/9] drm/radeon: stop implementing init_mem_type Christian König
2020-07-27 10:30   ` daniel
2020-07-27 10:30     ` daniel
2020-07-27 10:54     ` Christian König
2020-07-27  9:42 ` [PATCH 1/9] drm/ttm: initialize the system domain with defaults daniel
2020-07-27 10:39   ` Christian König
2020-07-27 10:50     ` Hellstrom, Thomas
2020-07-27 12:23       ` Christian König
2020-07-27 13:01         ` Hellstrom, Thomas [this message]
2020-07-23 15:17 More TTM cleanups Christian König
2020-07-23 15:17 ` [PATCH 1/9] drm/ttm: initialize the system domain with defaults Christian König
2020-07-24  6:43   ` Thomas Zimmermann
2020-07-29  6:21     ` Dave Airlie
2020-07-29  6:23       ` Dave Airlie
2020-07-29  9:47         ` Christian König

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=6cf8d9c160517eb1991e471cfc601e380c6fb127.camel@intel.com \
    --to=thomas.hellstrom@intel.com \
    --cc=christian.koenig@amd.com \
    --cc=daniel@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.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
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;
as well as URLs for NNTP newsgroup(s).