Remove the calls to pci_release_regions().
>
> Replace the call to sometimes-managed pci_request_regions() with one to
> always-managed pcim_request_all_regions().
>
> Signed-off-by: Philipp Stanner
> Reviewed-by: Zack Rusin
> ---
> Changes in v3:
> - Use the correct driv
pcim_request_all_regions().
> >
> > Signed-off-by: Philipp Stanner
>
> *PING*
Thanks, that looks great. I missed it because the driver's name is
vmwgfx. I'd be happy to fix the subject for you while pushing this to
drm-misc-fixes, if you're ok with it of course. Otherwise:
Reviewed-by: Zack Rusin
z
smime.p7s
Description: S/MIME Cryptographic Signature
u/drm/vmwgfx/vmwgfx_validation.c
> +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c
> @@ -262,9 +262,8 @@ int vmw_validation_add_bo(struct vmw_validation_context
> *ctx,
> bo_node->hash.key);
> }
> val_buf = &bo_node->base;
> - val_buf->bo = ttm_bo_get_unless_zero(&vbo->tbo);
> - if (!val_buf->bo)
> - return -ESRCH;
> + vmw_bo_reference(vbo);
> + val_buf->bo = &vbo->tbo;
> val_buf->num_shared = 0;
> list_add_tail(&val_buf->head, &ctx->bo_list);
> }
> @@ -656,7 +655,7 @@ void vmw_validation_unref_lists(struct
> vmw_validation_context *ctx)
> struct vmw_validation_res_node *val;
>
> list_for_each_entry(entry, &ctx->bo_list, base.head) {
> - ttm_bo_put(entry->base.bo);
> + drm_gem_object_put(&entry->base.bo->base);
> entry->base.bo = NULL;
> }
>
> --
> 2.45.2
>
Yea, that's probably a long time coming. As a followup it'd be good to
get rid of vmw_user_bo_ref/vmw_user_bo_unref now that they're exactly
the same as vmw_bo_reference/vmw_bo_unreference.
Reviewed-by: Zack Rusin
z
smime.p7s
Description: S/MIME Cryptographic Signature
>dumb_surface = vmw_res_to_srf(res);
> -
> + drm_gem_object_put(&vbo->tbo.base);
> + /*
> +* Unset the user surface dtor since this in not actually exposed
> +* to userspace. The suface is owned via the dumb_buffer's GEM handle
> +*/
> + st
Thomas Zimmermann
> Cc: Zack Rusin
> Cc: Broadcom internal kernel review list
>
> ---
> drivers/gpu/drm/vmwgfx/vmwgfx_gem.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c
> b/drivers/gpu
e import_attach optional.
>
> Signed-off-by: Thomas Zimmermann
> Cc: Zack Rusin
> Cc: Broadcom internal kernel review list
>
> ---
> drivers/gpu/drm/vmwgfx/vmwgfx_blit.c | 4 ++--
> drivers/gpu/drm/vmwgfx/vmwgfx_gem.c | 6 +++---
> 2 files changed, 5 insertions(+), 5 del
trackers in kms fb surface
cleanup code.
v3: Fix the copyrights as spotted by Maaz and fix the arm64 build as
spotted by the exynos kernel test robot
Zack Rusin (2):
drm/vmwgfx: Refactor cursor handling
drm/vmwgfx: Bump the minor version
drivers/gpu/drm/vmwgfx/Makefile | 2
properly updated in the guest.
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/Makefile | 2 +-
drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 6 +
drivers/gpu/drm/vmwgfx/vmwgfx_bo.h | 2 +
drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c | 844 ++
drivers
on vmwgfx
again.
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 6fc810632c98..0dfb88fb19e2 100644
--- a/drivers/gpu/drm/vmwgfx
trackers in kms fb surface
cleanup code.
Zack Rusin (2):
drm/vmwgfx: Refactor cursor handling
drm/vmwgfx: Bump the minor version
drivers/gpu/drm/vmwgfx/Makefile | 2 +-
drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 6 +
drivers/gpu/drm/vmwgfx/vmwgfx_bo.h | 2
on vmwgfx
again.
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 09ffb49970c9..a171379abf94 100644
--- a/drivers/gpu/drm/vmwgfx
properly updated in the guest.
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/Makefile | 2 +-
drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 6 +
drivers/gpu/drm/vmwgfx/vmwgfx_bo.h | 2 +
drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c | 844 ++
drivers
On Wed, Mar 5, 2025 at 2:10 PM kernel test robot wrote:
>
> Hi Zack,
>
> kernel test robot noticed the following build errors:
>
> [auto build test ERROR on drm-exynos/exynos-drm-next]
> [also build test ERROR on linus/master v6.14-rc5 next-20250305]
> [If your patch is applied to the wrong git tr
oadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 13 +
1 file changed, 13 insertions(+)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index d8937f3de51
r_srf->prime.base.handle;
> rep->backup_size = res->guest_memory_size;
> if (res->guest_memory_bo) {
> - vmw_bo_add_detached_resource(res->guest_memory_bo, res);
> rep->buffer_map_handle =
>
> drm_vma_node_offset_addr(&res->guest_memory_bo->tbo.base.vma_node);
> rep->buffer_size = res->guest_memory_bo->tbo.base.size;
> --
> 2.34.1
>
Looks good. Thank you for the patch! I'll push it to drm-misc-next for
now because I'd like it alongside the cursor rework.
Reviewed-by: Zack Rusin
z
smime.p7s
Description: S/MIME Cryptographic Signature
_priv = dev_priv;
> + vmw_seqno_waiter_add(dev_priv);
> + if (dma_fence_add_callback(&fence->base, &ctx->base,
> + seqno_waiter_rm_cb) < 0) {
> + vmw_seqno_waiter_remove(dev_priv);
> + kfree(ctx);
> + }
> }
> }
Great catch. Thanks!
Reviewed-by: Zack Rusin
z
smime.p7s
Description: S/MIME Cryptographic Signature
Sorry, I didn't see this before because the email formatting got
really butchered, I only found the underneath comment in it. Please
let me know if there were more.
On Tue, Feb 11, 2025 at 9:33 AM Martin Krastev
wrote:
> On Wed, Jan 15, 2025 at 6:50 AM Zack Rusin wrote:
...
> >
On Fri, Jan 24, 2025 at 11:48 AM Ian Forbes wrote:
>
> On Tue, Jan 14, 2025 at 10:50 PM Zack Rusin wrote:
> >
> > +int vmw_cursor_plane_prepare_fb(struct drm_plane *plane,
> > + struct drm_plane_state *new_state)
> > +{
> >
On Tue, Jan 21, 2025 at 2:11 PM Ian Forbes wrote:
>
> On Fri, Jan 17, 2025 at 1:20 PM Zack Rusin wrote:
> >
> > You're going to have to explain that one in the commit message a lot
> > better because as is it doesn't make sense to me. Especially the
> >
On Fri, Jan 17, 2025 at 2:14 PM Ian Forbes wrote:
>
> Fixes: d6667f0ddf46 ("drm/vmwgfx: Fix handling of dumb buffers")
> Signed-off-by: Ian Forbes
> ---
> drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 2 +-
> drivers/gpu/drm/vmwgfx/vmwgfx_surface.c | 1 +
> 2 files changed, 2 insertions(+), 1 deleti
vmwgfx had a number of cursor issues that related both to our handling
of dumb buffers and general detection when a cursor has actually changed.
Fix those issues and bump the kernel module version to allow userspace
to recognize fixed versions of the driver.
Zack Rusin (2):
drm/vmwgfx: Refactor
properly updated in the guest.
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/Makefile | 2 +-
drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 6 +
drivers/gpu/drm/vmwgfx/vmwgfx_bo.h | 2 +
drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c | 844 +++
drivers
on vmwgfx
again.
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 09ffb49970c9..a171379abf94 100644
--- a/drivers/gpu/drm/vmwgfx
.size = bo_size,
> - .pin = true
> + .pin = true,
> + .keep_resv = true,
> };
>
> ret = vmw_bo_create(dev_priv, &bo_params, &vbo);
> if (unlikely(ret != 0))
> return ret;
>
> - ret = ttm_bo_reserve(&vbo->tbo, false, true, NULL);
> - BUG_ON(ret != 0);
> ret = vmw_ttm_populate(vbo->tbo.bdev, vbo->tbo.ttm, &ctx);
> if (likely(ret == 0)) {
> struct vmw_ttm_tt *vmw_tt =
> --
> 2.34.1
>
That's a pretty nice cleanup and a fix.
Reviewed-by: Zack Rusin
z
smime.p7s
Description: S/MIME Cryptographic Signature
On Thu, Jan 9, 2025 at 10:03 AM Thomas Zimmermann wrote:
>
> Call drm_mode_size_dumb() to compute dumb-buffer scanline pitch
> and buffer size. No alignment required.
>
> Signed-off-by: Thomas Zimmermann
> Cc: Zack Rusin
> Cc: Broadcom internal kernel review list
>
On Tue, Dec 3, 2024 at 1:21 PM Jonas Ådahl wrote:
>
> On Tue, Dec 03, 2024 at 11:39:05AM -0500, Zack Rusin wrote:
> > On Tue, Dec 3, 2024 at 11:32 AM Jonas Ådahl wrote:
> > >
> > > On Tue, Dec 03, 2024 at 11:27:52AM -0500, Zack Rusin wrote:
> > > > O
On Tue, Dec 3, 2024 at 11:42 AM Thomas Zimmermann wrote:
>
> Hi
>
>
> Am 03.12.24 um 17:39 schrieb Zack Rusin:
> > On Tue, Dec 3, 2024 at 11:32 AM Jonas Ådahl wrote:
> >> On Tue, Dec 03, 2024 at 11:27:52AM -0500, Zack Rusin wrote:
> >>> On Tue,
On Tue, Dec 3, 2024 at 11:32 AM Jonas Ådahl wrote:
>
> On Tue, Dec 03, 2024 at 11:27:52AM -0500, Zack Rusin wrote:
> > On Tue, Dec 3, 2024 at 10:57 AM Jonas Ådahl wrote:
> > >
> > > On Wed, Nov 20, 2024 at 07:52:18AM -0500, Zack Rusin wrote:
> > > > O
On Tue, Dec 3, 2024 at 10:57 AM Jonas Ådahl wrote:
>
> On Wed, Nov 20, 2024 at 07:52:18AM -0500, Zack Rusin wrote:
> > On Wed, Nov 20, 2024 at 5:22 AM Jani Nikula
> > wrote:
> > >
> > > On Wed, 20 Nov 2024, Thomas Zimmermann wrote:
> > > >
> > }
> >
> > drm_connector_helper_add(connector, &vmw_stdu_connector_helper_funcs);
> > - connector->status = vmw_du_connector_detect(connector, false);
> >
> > ret = drm_encoder_init(dev, encoder, &vmw_stdu_encoder_funcs,
> > DRM_MODE_ENCODER_VIRTUAL, NULL);
>
Thank you. Looks good.
Reviewed-by: Zack Rusin
z
smime.p7s
Description: S/MIME Cryptographic Signature
On Thu, Nov 21, 2024 at 5:27 AM Christian König
wrote:
>
> Hey Zack,
>
> vmwgfx is the only driver left using this component.
>
> So I though I go ahead and complete switching over to drm_exec.
> The patches are only compile tested, but relatively staright forward.
Hi, Christian.
Thank you so mu
On Wed, Nov 20, 2024 at 5:22 AM Jani Nikula wrote:
>
> On Wed, 20 Nov 2024, Thomas Zimmermann wrote:
> > Hi
> >
> >
> > Am 19.11.24 um 20:40 schrieb Ian Forbes:
> >> Most compositors are using a change in EDID as an indicator to
> >> refresh their connector information on hotplug regardless of wh
On Fri, Oct 18, 2024 at 5:01 PM Maaz Mombasawala
wrote:
>
> A userspace may create a userspace managed surface but not destroy it,
> add hw_destroy function for userspace surfaces so that vmwgfx records the
> destroy command and submits it when the userspace context is destroyed.
>
> Signed-off-by
On Fri, Oct 18, 2024 at 5:01 PM Maaz Mombasawala
wrote:
>
> The kernel currently exposes both mobs and surfaces to userspace through
> ioctls. We would like to move to a model where kernel would only expose
> mobs and have userspace manage surfaces. This would simplify kernel paths
> for surfaces
On Tue, Oct 29, 2024 at 4:43 AM Chen Ridong wrote:
>
> From: Chen Ridong
>
> The 'vmw_user_object_buffer' function may return NULL with incorrect
> inputs. To avoid possible null pointer dereference, add a check whether
> the 'bo' is NULL in the vmw_framebuffer_surface_create_handle.
>
> Fixes: d
On Mon, Oct 28, 2024 at 4:14 PM Ian Forbes wrote:
>
> We replaced our custom list of resolutions with the noedid list, which is
> based on the VESA DMT standard, in the referenced fixes commit. The reason
> for this was that the user can technically set any resolution they want by
> using Autofit
On Mon, Oct 7, 2024 at 7:52 AM Thorsten Blum wrote:
>
> Since kvfree() already checks if its argument is NULL, an additional
> check before calling kvfree() is unnecessary and can be removed.
>
> Remove both and the following Coccinelle/coccicheck warnings reported by
> ifnullfree.cocci:
>
> WAR
a_v3(const struct vmw_private *dev)
> {
> + if (!IS_ENABLED(CONFIG_HAS_IOPORT))
> + return true;
> +
> return dev->pci_id == VMWGFX_PCI_ID_SVGA3;
> }
>
> --
> 2.39.2
>
Thanks! Looks good. Would you like me to push this through drm-misc-fixes?
Reviewed-by: Zack Rusin
z
On Wed, Oct 2, 2024 at 8:26 AM Nikolay Kuratov wrote:
>
> Currently if condition (!bo and !vmw_kms_srf_ok()) was met
> we go to err_out with ret == 0.
> err_out dereferences vfb if ret == 0, but in our case vfb is still NULL.
>
> Fix this by assigning sensible error to ret.
>
> Found by Linux Veri
n of adding yet another bool flag parameter to the
> function.
>
> Cc: Christian König
> Cc: amd-...@lists.freedesktop.org
> Cc: intel-...@lists.freedesktop.org
> Cc: nouv...@lists.freedesktop.org
> Cc: spice-de...@lists.freedesktop.org
> Cc: dri-devel@lists.freedesktop.org
>
On Mon, Sep 9, 2024 at 7:37 AM Thomas Zimmermann wrote:
>
> Call drm_client_setup() to run the kernel's default client setup
> for DRM. Set fbdev_probe in struct drm_driver, so that the client
> setup can start the common fbdev client.
>
> Signed-off-by: Thomas Zimmermann
On Fri, Sep 6, 2024 at 11:26 AM Ian Forbes wrote:
>
> On Thu, Sep 5, 2024 at 10:59 PM Zack Rusin wrote:
> >
> >
> > In general it looks good but what's the reason for the submit_time?
> >
> > z
>
> So you can get an approximate time of how long each
On Thu, Sep 5, 2024 at 2:17 PM Ian Forbes wrote:
>
> Adds the necessary files to create tracepoints for the vmwgfx driver.
>
> Adds a single tracepoint for command buffer completion. This tracepoint
> can be used to time command buffer execution time and to decode command
> buffer errors. The subm
abled
on configurations without 3d but with those changes the code actually
makes sense.
v2: Remove the now unused format variable
Signed-off-by: Zack Rusin
Fixes: d6667f0ddf46 ("drm/vmwgfx: Fix handling of dumb buffers")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.fre
abled
on configurations without 3d but with those changes the code actually
makes sense.
Signed-off-by: Zack Rusin
Fixes: d6667f0ddf46 ("drm/vmwgfx: Fix handling of dumb buffers")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Cc: # v6.9+
Cc: Maaz Mombasaw
On Thu, Aug 15, 2024 at 4:30 PM Andreas Piesk wrote:
>
> Hello,
>
> the bug was first reported on VMware Workstation by rdkehn.
>
> On my setup (archlinux text mode only VM on ESXi 8.0U3 latest) the kernel
> does NOT crash, the screen just goes dark after switching the console from
>
> [2.745
y low graphics
memory limits. The pinned console fb, mob cursors and graphical login
manager tend to run out of 16MB graphics memory that those guests use.
Fix it by making sure the coherent dumb buffers are only used on
configs with 3d enabled.
Signed-off-by: Zack Rusin
Fixes: d6667f0ddf46
d472d5 ("drm/vmwgfx: Add support for CursorMob and CursorBypass 4")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Cc: # v5.19+
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 13 +++--
drivers/gpu/drm/vmwgfx/vmwgfx_bo.h | 3 +++
2 f
to be used.
Fixes crashes in IGT's kms_prime with vgem. Regular desktop usage won't
trigger this due to the fact that virtual machines will not have
multiple GPUs but it enables better test coverage in IGT.
Signed-off-by: Zack Rusin
Fixes: b32233acceff ("drm/vmwgfx: Fix prime import/
This is the same series I've sent out earlier but with one extra patch,
that fixes the dumb buffer coherency on low mem systems.
The second patch has also been updated to not use math functions.
Zack Rusin (3):
drm/vmwgfx: Prevent unmapping active read buffers
drm/vmwgfx: Fix prime
On Thu, Aug 15, 2024 at 1:48 PM Christian Heusel wrote:
>
> Hello Zack,
>
> the user rdkehn (in CC) on the Arch Linux Forums reports that after
> updating to the 6.10.4 stable kernel inside of their VM Workstation the
> driver crashes with the error attached below. This error is also present
> on
to be used.
Fixes crashes in IGT's kms_prime with vgem. Regular desktop usage won't
trigger this due to the fact that virtual machines will not have
multiple GPUs but it enables better test coverage in IGT.
v2: Fix the diff rectangle computation
Signed-off-by: Zack Rusin
Fixes: b32233acce
Fixes: 485d98d472d5 ("drm/vmwgfx: Add support for CursorMob and CursorBypass 4")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Cc: # v5.19+
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 13 +++--
drivers/gpu/drm/vmwgf
to be used.
Fixes crashes in IGT's kms_prime with vgem. Regular desktop usage won't
trigger this due to the fact that virtual machines will not have
multiple GPUs but it enables better test coverage in IGT.
Signed-off-by: Zack Rusin
Fixes: b32233acceff ("drm/vmwgfx: Fix prime import/
Fixes: 485d98d472d5 ("drm/vmwgfx: Add support for CursorMob and CursorBypass 4")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Cc: # v5.19+
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 13 +++--
drivers/gpu/drm/vmwgf
On Mon, Aug 12, 2024 at 3:16 PM Maaz Mombasawala
wrote:
>
> This series introduces basic support for userspace managed surfaces. The
> lifetime and id's of these surfaces is managed by userspace submitted
> commands instead of relying on the kernel to manage them.
>
> Maaz Mombasawala (3):
> drm
On Tue, Aug 13, 2024 at 1:56 PM Ian Forbes wrote:
>
> In that case move `map_count` above `map` which should move it to a
> separate cache line and update the doc strings as needed.
Sorry, I'm not sure I understand. What are you trying to fix?
z
On Tue, Aug 13, 2024 at 1:29 PM Ian Forbes wrote:
>
> Remove `busy_places` now that it's unused. There's also probably a
> better place to put `map_count` in the struct layout to avoid false
> sharing with `cpu_writers`. I'd repack the whole struct if we're going
> to be adding and removing fields
gt; + if (IS_ERR(conn_state))
> + return PTR_ERR(conn_state);
> +
> du = vmw_connector_to_stdu(conn);
>
> if (!conn_state->crtc)
> --
> 2.34.1
>
Reviewed-by: Zack Rusin
z
_y2;
> };
>
> -#define VMWGFX_NUM_DISPLAY_UNITS 8
> -
> -
> #define vmw_framebuffer_to_vfb(x) \
> container_of(x, struct vmw_framebuffer, base)
> #define vmw_framebuffer_to_vfbs(x) \
> --
> 2.34.1
>
Looks good. Thanks.
Reviewed-by: Zack Rusin
z
to be used.
Fixes crashes in IGT's kms_prime with vgem. Regular desktop usage won't
trigger this due to the fact that virtual machines will not have
multiple GPUs but it enables better test coverage in IGT.
Signed-off-by: Zack Rusin
Fixes: b32233acceff ("drm/vmwgfx: Fix prime import/
d472d5 ("drm/vmwgfx: Add support for CursorMob and CursorBypass 4")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Cc: # v5.19+
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 13 +++--
drivers/gpu/drm/vmwgfx/vmwgfx_bo.h | 1 +
2 f
INT_MIN, INT_MAX);
> - if (IS_ERR(prop_y)) {
> + if (!prop_y) {
> drm_property_destroy(plane->dev, prop_x);
> - return PTR_ERR(prop_y);
> + return -ENOMEM;
> }
>
> drm_object_attach_property(&plane->base, prop_x, 0);
Thanks, that looks good to me.
Reviewed-by: Zack Rusin
z
On Fri, Jul 19, 2024 at 12:37 PM Ian Forbes wrote:
>
> This code was never updated to support Screen Targets.
> Fixes a bug where Xv playback displays a green screen instead of actual
> video contents when 3D acceleration is disabled in the guest.
>
> Fixes: c8261a961ece ("vmwgfx: Major KMS refact
Provide a way to query for the fixed support for dumb buffers with kms.
Lets mesa svga return a buffer id, instead of a surface id from
resource_to_handle which fixes a lot of userspace apps that assume
that those handles are gem buffers.
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx
by us, it's semantically completely broken.
Fix importing of external buffers by forwarding all memory access
requests to the importer.
Tested by the vmw_prime basic_vgem test.
Signed-off-by: Zack Rusin
Reviewed-by: Maaz Mombasawala
Reviewed-by: Martin Krastev
---
drivers/gpu/drm/vmwgfx/vm
d kms but internally switch
to vmw_surface's to properly track the dirty state of the objects between
the GPU and CPU.
Fixes new kwin and kde on wayland.
Signed-off-by: Zack Rusin
Fixes: b32233acceff ("drm/vmwgfx: Fix prime import/export")
Cc: Broadcom internal kernel review
s not always the code, in particular during
atomic disable, so to make sure the surface, when present, is not
being actively destroyed at the same time, hold a reference to it.
Signed-off-by: Zack Rusin
Fixes: 7b0062036c3b ("drm/vmwgfx: Implement virtual crc generation")
Cc: Zack Rusi
particular this fixes KDE stalls/deadlock.
Signed-off-by: Zack Rusin
Fixes: 2298e804e96e ("drm/vmwgfx: rework to new fence interface, v2")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Cc: # v6.2+
---
drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 17
or races due to buffer mapping not being reserved in
the third patch
v5: Use map.virtual to check whether an user buffer is mapped instead
of map.bo. There's no change in functionality but we do use map.virtual
to check whether a buffer has been mapped elsewhere so usage of it
is more consistent. S
d kms but internally switch
to vmw_surface's to properly track the dirty state of the objects between
the GPU and CPU.
Fixes new kwin and kde on wayland.
Signed-off-by: Zack Rusin
Fixes: b32233acceff ("drm/vmwgfx: Fix prime import/export")
Cc: Broadcom internal kernel review
by us, it's semantically completely broken.
Fix importing of external buffers by forwarding all memory access
requests to the importer.
Tested by the vmw_prime basic_vgem test.
Signed-off-by: Zack Rusin
Reviewed-by: Maaz Mombasawala
Reviewed-by: Martin Krastev
---
drivers/gpu/drm/vmwgfx/vm
s not always the code, in particular during
atomic disable, so to make sure the surface, when present, is not
being actively destroyed at the same time, hold a reference to it.
Signed-off-by: Zack Rusin
Fixes: 7b0062036c3b ("drm/vmwgfx: Implement virtual crc generation")
Cc: Zack Rusi
particular this fixes KDE stalls/deadlock.
Signed-off-by: Zack Rusin
Fixes: 2298e804e96e ("drm/vmwgfx: rework to new fence interface, v2")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Cc: # v6.2+
---
drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 17
or races due to buffer mapping not being reserved in
the third patch
Zack Rusin (4):
drm/vmwgfx: Fix a deadlock in dma buf fence polling
drm/vmwgfx: Make sure the screen surface is ref counted
drm/vmwgfx: Fix handling of dumb buffers
drm/vmwgfx: Add basic support for external buffers
drivers
1-65d02bb02...@suse.de/
> Tested-by: David Kaplan
> Reviewed-by: Daniel Vetter
> Cc: Thomas Zimmermann
> Cc: Dmitry Osipenko
> Cc: Christian König
> Cc: Zack Rusin
> Cc: Dave Airlie
> Cc: Gerd Hoffmann
> Cc: virtualizat...@lists.linux.dev
> Cc: spice-de...@
d kms but internally switch
to vmw_surface's to properly track the dirty state of the objects between
the GPU and CPU.
Fixes new kwin and kde on wayland.
Signed-off-by: Zack Rusin
Fixes: b32233acceff ("drm/vmwgfx: Fix prime import/export")
Cc: Broadcom internal kernel review
by us, it's semantically completely broken.
Fix importing of external buffers by forwarding all memory access
requests to the importer.
Tested by the vmw_prime basic_vgem test.
Signed-off-by: Zack Rusin
Reviewed-by: Maaz Mombasawala
Reviewed-by: Martin Krastev
---
drivers/gpu/drm/vmwgfx/vm
s not always the code, in particular during
atomic disable, so to make sure the surface, when present, is not
being actively destroyed at the same time, hold a reference to it.
Signed-off-by: Zack Rusin
Fixes: 7b0062036c3b ("drm/vmwgfx: Implement virtual crc generation")
Cc: Zack Rusi
particular this fixes KDE stalls/deadlock.
Signed-off-by: Zack Rusin
Fixes: 2298e804e96e ("drm/vmwgfx: rework to new fence interface, v2")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Cc: # v6.2+
---
drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 17
which always works and doesn't require any changes in userspace.
This fixes the known KDE (KWin's) buffer rendering issues.
v2: Fix compute_crc in the second patch, as spotted by Martin
v3: Simplify the first change which fixes the deadlock in the dma-buf
fence polling
Zack Rusin (4):
dr
particular this fixes KDE stalls/deadlock.
Signed-off-by: Zack Rusin
Fixes: 2298e804e96e ("drm/vmwgfx: rework to new fence interface, v2")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Cc: # v6.2+
---
drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 26 ++
s not always the code, in particular during
atomic disable, so to make sure the surface, when present, is not
being actively destroyed at the same time, hold a reference to it.
Signed-off-by: Zack Rusin
Fixes: 7b0062036c3b ("drm/vmwgfx: Implement virtual crc generation")
Cc: Zack Rusin
Cc:
d kms but internally switch
to vmw_surface's to properly track the dirty state of the objects between
the GPU and CPU.
Fixes new kwin and kde on wayland.
Signed-off-by: Zack Rusin
Fixes: b32233acceff ("drm/vmwgfx: Fix prime import/export")
Cc: Broadcom internal kernel review
by us, it's semantically completely broken.
Fix importing of external buffers by forwarding all memory access
requests to the importer.
Tested by the vmw_prime basic_vgem test.
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/vmwgfx_gem.c | 62 +++--
1 file c
which always works and doesn't require any changes in userspace.
This fixes the known KDE (KWin's) buffer rendering issues.
v2: Fix compute_crc in the second patch, as spotted by Martin
Zack Rusin (4):
drm/vmwgfx: Fix a deadlock in dma buf fence polling
drm/vmwgfx: Make sure the screen
On Thu, Jun 27, 2024 at 8:37 AM Martin Krastev
wrote:
>
> On Thu, Jun 27, 2024 at 8:34 AM Zack Rusin wrote:
> >
> > Fix races issues in virtual crc generation by making sure the surface
> > the code uses for crc computation is properly ref counted.
> >
> >
by us, it's semantically completely broken.
Fix importing of external buffers by forwarding all memory access
requests to the importer.
Tested by the vmw_prime basic_vgem test.
Signed-off-by: Zack Rusin
---
drivers/gpu/drm/vmwgfx/vmwgfx_gem.c | 62 +++--
1 file c
s not always the code, in particular during
atomic disable, so to make sure the surface, when present, is not
being actively destroyed at the same time, hold a reference to it.
Signed-off-by: Zack Rusin
Fixes: 7b0062036c3b ("drm/vmwgfx: Implement virtual crc generation")
Cc: Zack Rusin
Cc:
d kms but internally switch
to vmw_surface's to properly track the dirty state of the objects between
the GPU and CPU.
Fixes new kwin and kde on wayland.
Signed-off-by: Zack Rusin
Fixes: b32233acceff ("drm/vmwgfx: Fix prime import/export")
Cc: Broadcom internal kernel review
particular this fixes KDE stalls/deadlock.
Signed-off-by: Zack Rusin
Fixes: 2298e804e96e ("drm/vmwgfx: rework to new fence interface, v2")
Cc: Broadcom internal kernel review list
Cc: dri-devel@lists.freedesktop.org
Cc: # v6.2+
---
drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 26 ++
ffers both (raw buffers and surfaces referenced by the same
handle), which always works and doesn't require any changes in userspace.
This fixes the known KDE (KWin's) buffer rendering issues.
Zack Rusin (4):
dr
On Mon, Jun 17, 2024 at 6:02 AM Borislav Petkov wrote:
>
> On Mon, Jun 17, 2024 at 11:07:09AM +0200, Borislav Petkov wrote:
> > On Sat, Jun 15, 2024 at 06:25:10PM -0700, Alexey Makhalov wrote:
> > > VMWARE_HYPERCALL alternative will not work as intended without
> > > VMware guest code initializati
On Mon, Jun 3, 2024 at 1:24 PM Dr. David Alan Gilbert wrote:
>
> * li...@treblig.org (li...@treblig.org) wrote:
> > From: "Dr. David Alan Gilbert"
> >
> > 'vmw_stdu_dma' is unused since
> > commit 39985eea5a6d ("drm/vmwgfx: Abstract placement selection")
> > Remove it.
>
> Ping.
Thanks. I pushed
nged, 60 insertions(+), 37 deletions(-)
Looks great. For the series:
Reviewed-by: Zack Rusin
z
On Tue, May 14, 2024 at 3:00 AM Christian König
wrote:
>
> Am 14.05.24 um 06:15 schrieb Zack Rusin:
>
> On Mon, May 13, 2024 at 1:09 PM Christian König
> wrote:
>
> Am 10.05.24 um 18:34 schrieb Zack Rusin:
>
> Hey,
>
> so this is a bit of a silly problem but I&#x
On Mon, May 13, 2024 at 1:09 PM Christian König
wrote:
>
> Am 10.05.24 um 18:34 schrieb Zack Rusin:
> > Hey,
> >
> > so this is a bit of a silly problem but I'd still like to solve it
> > properly. The tldr is that virtualized drivers abuse
> > drm_dr
Hey,
so this is a bit of a silly problem but I'd still like to solve it
properly. The tldr is that virtualized drivers abuse
drm_driver::gem_prime_import_sg_table (at least vmwgfx and xen do,
virtgpu and xen punt on it) because there doesn't seem to be a
universally supported way of converting the
On Fri, May 3, 2024 at 6:29 PM Ian Forbes wrote:
>
> This function was removed in the referenced fixes commit and caused a
> regression. This is because the presence of this function, even though it
> is a noop, changes the behaviour of disable_outputs in
> drm_atomic_helper.c:1211.
>
> Fixes: 7b0
position(plane->dev->dev_private, false, 0, 0);
> + vmw_cursor_update_position(vmw_priv(plane->dev), false, 0, 0);
>
> for (i = 0; i < ARRAY_SIZE(vcp->cursor_mobs); i++)
> vmw_du_destroy_cursor_mob(&vcp->cursor_mobs[i]);
> --
> 2.34.1
>
Looks good.
Reviewed-by: Zack Rusin
z
1 - 100 of 630 matches
Mail list logo
