Skip to contentSkip to conversation

drm/hyperv: Fix double mouse pointers

PATCH Superseded Dexuan Cui 3 participants 7 messages Sep 13, 2021 – Sep 16, 2021

Subsystems (5)
DRM drivers DRM misc Hyper-V/Azure core and drivers Microsoft MANA RDMA driver +1 more subsystems

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore

Dexuan Cui

author maintainer
#0 lore raw

PATCH 1/1 drm/hyperv: Fix double mouse pointers

It looks like Hyper-V supports a hardware cursor feature. It is not used by Linux VM, but the Hyper-V host still draws a point as an extra mouse pointer, which is unwanted, especially when Xorg is running.

The hyperv_fb driver uses synthvid_send_ptr() to hide the unwanted pointer. When the hyperv_drm driver was developed, the function synthvid_send_ptr() was not copied from the hyperv_fb driver. Fix the issue by adding the function into hyperv_drm.

Fixes 76c56a5affeb ("drm/hyperv: Add DRM driver for hyperv synthetic video device")
Signed-off-by Dexuan Cui <[email protected]>
Cc Deepak Rawat <[email protected]>
Cc Haiyang Zhang <[email protected]>
---
 drivers/gpu/drm/hyperv/hyperv_drm.h         |  1 +
 drivers/gpu/drm/hyperv/hyperv_drm_modeset.c |  1 +
 drivers/gpu/drm/hyperv/hyperv_drm_proto.c   | 39 ++++++++++++++++++++-
 3 files changed, 40 insertions(+), 1 deletion(-)
drivers/gpu/drm/hyperv/hyperv_drm.h +1 -0
index 886add4f9cd0..27bfd27c05be 100644
@@ -46,6 +46,7 @@ int hyperv_mode_config_init(struct hyperv_drm_device *hv);
 int hyperv_update_vram_location(struct hv_device *hdev, phys_addr_t vram_pp);
 int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp,
 			    u32 w, u32 h, u32 pitch);
+int hyperv_send_ptr(struct hv_device *hdev);
 int hyperv_update_dirt(struct hv_device *hdev, struct drm_rect *rect);
 int hyperv_connect_vsp(struct hv_device *hdev);
drivers/gpu/drm/hyperv/hyperv_drm_modeset.c +1 -0
index 3aaee4730ec6..e21c82cf3326 100644
@@ -101,6 +101,7 @@ static void hyperv_pipe_enable(struct drm_simple_display_pipe *pipe,
 	struct hyperv_drm_device *hv = to_hv(pipe->crtc.dev);
 	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(plane_state);
 
+	hyperv_send_ptr(hv->hdev);
 	hyperv_update_situation(hv->hdev, 1,  hv->screen_depth,
 				crtc_state->mode.hdisplay,
 				crtc_state->mode.vdisplay,
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +38 -1
index 6d4bdccfbd1a..1ea7a0432320 100644
@@ -299,6 +299,40 @@ int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp,
 	return 0;
 }
 
+/* Send mouse pointer info to host */
+int hyperv_send_ptr(struct hv_device *hdev)
+{
+	struct synthvid_msg msg;
+
+	memset(&msg, 0, sizeof(struct synthvid_msg));
+	msg.vid_hdr.type = SYNTHVID_POINTER_POSITION;
+	msg.vid_hdr.size = sizeof(struct synthvid_msg_hdr) +
+		sizeof(struct synthvid_pointer_position);
+	msg.ptr_pos.is_visible = 1;
+	msg.ptr_pos.video_output = 0;
+	msg.ptr_pos.image_x = 0;
+	msg.ptr_pos.image_y = 0;
+	hyperv_sendpacket(hdev, &msg);
+
+	memset(&msg, 0, sizeof(struct synthvid_msg));
+	msg.vid_hdr.type = SYNTHVID_POINTER_SHAPE;
+	msg.vid_hdr.size = sizeof(struct synthvid_msg_hdr) +
+		sizeof(struct synthvid_pointer_shape);
+	msg.ptr_shape.part_idx = SYNTHVID_CURSOR_COMPLETE;
+	msg.ptr_shape.is_argb = 1;
+	msg.ptr_shape.width = 1;
+	msg.ptr_shape.height = 1;
+	msg.ptr_shape.hot_x = 0;
+	msg.ptr_shape.hot_y = 0;
+	msg.ptr_shape.data[0] = 0;
+	msg.ptr_shape.data[1] = 1;
+	msg.ptr_shape.data[2] = 1;
+	msg.ptr_shape.data[3] = 1;
+	hyperv_sendpacket(hdev, &msg);
+
+	return 0;
+}
+
 int hyperv_update_dirt(struct hv_device *hdev, struct drm_rect *rect)
 {
 	struct hyperv_drm_device *hv = hv_get_drvdata(hdev);

@@ -392,8 +426,11 @@ static void hyperv_receive_sub(struct hv_device *hdev)
 		return;
 	}
 
-	if (msg->vid_hdr.type == SYNTHVID_FEATURE_CHANGE)
+	if (msg->vid_hdr.type == SYNTHVID_FEATURE_CHANGE) {
 		hv->dirt_needed = msg->feature_chg.is_dirt_needed;
+		if (hv->dirt_needed)
+			hyperv_send_ptr(hv->hdev);
+	}
 }
 
 static void hyperv_receive(void *ctx)
Signature · 1 line
2.25.1
maintainer Reviewed-by on v1 1/1
#1 lore raw
Quoted 25 lines

-----Original Message-----
From: Dexuan Cui <[email protected]>
Sent: Monday, September 13, 2021 2:27 PM
To: [email protected]; Haiyang Zhang <[email protected]>; [email protected]; [email protected]; [email protected]; dri- [email protected]
Cc: [email protected]; [email protected]; Dexuan Cui <[email protected]>
Subject: [PATCH] drm/hyperv: Fix double mouse pointers

It looks like Hyper-V supports a hardware cursor feature. It is not used by Linux VM, but the Hyper-V host still draws a point as an extra mouse pointer, which is unwanted, especially when Xorg is running.

The hyperv_fb driver uses synthvid_send_ptr() to hide the unwanted pointer.
When the hyperv_drm driver was developed, the function synthvid_send_ptr()
was not copied from the hyperv_fb driver. Fix the issue by adding the function into hyperv_drm.

Fixes 76c56a5affeb ("drm/hyperv: Add DRM driver for hyperv synthetic

video device")

Quoted 8 lines
Signed-off-by Dexuan Cui <[email protected]>
Cc Deepak Rawat <[email protected]>
Cc Haiyang Zhang <[email protected]>
---
 drivers/gpu/drm/hyperv/hyperv_drm.h         |  1 +
 drivers/gpu/drm/hyperv/hyperv_drm_modeset.c |  1 +
 drivers/gpu/drm/hyperv/hyperv_drm_proto.c   | 39 ++++++++++++++++++++-
 3 files changed, 40 insertions(+), 1 deletion(-)

Thank you!

Reviewed-by Haiyang Zhang <[email protected]>
#2 lore raw
In reply to Dexuan Cui

Thanks Dexuan, for the patch. A minor comment below.

Quoted 52 lines · Dexuan Cui

It looks like Hyper-V supports a hardware cursor feature. It is not used by Linux VM, but the Hyper-V host still draws a point as an extra mouse pointer, which is unwanted, especially when Xorg is running.

The hyperv_fb driver uses synthvid_send_ptr() to hide the unwanted pointer. When the hyperv_drm driver was developed, the function synthvid_send_ptr() was not copied from the hyperv_fb driver. Fix the issue by adding the function into hyperv_drm.

Fixes 76c56a5affeb ("drm/hyperv: Add DRM driver for hyperv synthetic video device")
Signed-off-by Dexuan Cui <[email protected]>
Cc Deepak Rawat <[email protected]>
Cc Haiyang Zhang <[email protected]>
---
 drivers/gpu/drm/hyperv/hyperv_drm.h         |  1 +
 drivers/gpu/drm/hyperv/hyperv_drm_modeset.c |  1 +
 drivers/gpu/drm/hyperv/hyperv_drm_proto.c   | 39 ++++++++++++++++++++-
 3 files changed, 40 insertions(+), 1 deletion(-)
drivers/gpu/drm/hyperv/hyperv_drm.h +1 -0
index 886add4f9cd0..27bfd27c05be 100644
@@ -46,6 +46,7 @@ int hyperv_mode_config_init(struct hyperv_drm_device *hv);
 int hyperv_update_vram_location(struct hv_device *hdev, phys_addr_t vram_pp);
 int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp,
                            u32 w, u32 h, u32 pitch);
+int hyperv_send_ptr(struct hv_device *hdev);
 int hyperv_update_dirt(struct hv_device *hdev, struct drm_rect *rect);
 int hyperv_connect_vsp(struct hv_device *hdev);
drivers/gpu/drm/hyperv/hyperv_drm_modeset.c +1 -0
index 3aaee4730ec6..e21c82cf3326 100644
@@ -101,6 +101,7 @@ static void hyperv_pipe_enable(struct drm_simple_display_pipe *pipe,
        struct hyperv_drm_device *hv = to_hv(pipe->crtc.dev);
        struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(plane_state);
 
+       hyperv_send_ptr(hv->hdev);
        hyperv_update_situation(hv->hdev, 1,  hv->screen_depth,
                                crtc_state->mode.hdisplay,
                                crtc_state->mode.vdisplay,
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +0 -0
index 6d4bdccfbd1a..1ea7a0432320 100644
@@ -299,6 +299,40 @@ int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp,
        return 0;
 }

drivers/gpu/drm/hyperv/hyperv_drm_proto.c int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp, · line 311

Quoted 10 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +10 -0
+/* Send mouse pointer info to host */
+int hyperv_send_ptr(struct hv_device *hdev)
+{
+       struct synthvid_msg msg;
+
+       memset(&msg, 0, sizeof(struct synthvid_msg));
+       msg.vid_hdr.type = SYNTHVID_POINTER_POSITION;
+       msg.vid_hdr.size = sizeof(struct synthvid_msg_hdr) +
+               sizeof(struct synthvid_pointer_position);
+       msg.ptr_pos.is_visible = 1;

"is_visible" should be 0 since you want to hide the pointer. Maybe better, accept these from the caller.

Quoted 11 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +11 -0
+       msg.ptr_pos.video_output = 0;
+       msg.ptr_pos.image_x = 0;
+       msg.ptr_pos.image_y = 0;
+       hyperv_sendpacket(hdev, &msg);
+
+       memset(&msg, 0, sizeof(struct synthvid_msg));
+       msg.vid_hdr.type = SYNTHVID_POINTER_SHAPE;
+       msg.vid_hdr.size = sizeof(struct synthvid_msg_hdr) +
+               sizeof(struct synthvid_pointer_shape);
+       msg.ptr_shape.part_idx = SYNTHVID_CURSOR_COMPLETE;
+       msg.ptr_shape.is_argb = 1;

drivers/gpu/drm/hyperv/hyperv_drm_proto.c int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp, · line 332

Quoted 10 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +10 -0
+       msg.ptr_shape.width = 1;
+       msg.ptr_shape.height = 1;
+       msg.ptr_shape.hot_x = 0;
+       msg.ptr_shape.hot_y = 0;
+       msg.ptr_shape.data[0] = 0;
+       msg.ptr_shape.data[1] = 1;
+       msg.ptr_shape.data[2] = 1;
+       msg.ptr_shape.data[3] = 1;
+       hyperv_sendpacket(hdev, &msg);
+

Is it necessary to send SYNTHVID_POINTER_SHAPE here? Perhaps we should separate SYNTHVID_POINTER_POSITION and SYNTHVID_POINTER_SHAPE into different functions.

Quoted 22 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +7 -2
+       return 0;
+}
+
 int hyperv_update_dirt(struct hv_device *hdev, struct drm_rect *rect)
 {
        struct hyperv_drm_device *hv = hv_get_drvdata(hdev);

@@ -392,8 +426,11 @@ static void hyperv_receive_sub(struct hv_device *hdev)
                return;
        }
 
-       if (msg->vid_hdr.type == SYNTHVID_FEATURE_CHANGE)
+       if (msg->vid_hdr.type == SYNTHVID_FEATURE_CHANGE) {
                hv->dirt_needed = msg->feature_chg.is_dirt_needed;
+               if (hv->dirt_needed)
+                       hyperv_send_ptr(hv->hdev);
+       }
 }
 
 static void hyperv_receive(void *ctx)
--

2.25.1

maintainer
#3 lore raw
Quoted 82 lines

-----Original Message-----
From: Deepak Rawat <[email protected]>
Sent: Tuesday, September 14, 2021 11:59 AM
To: Dexuan Cui <[email protected]>
Cc: Haiyang Zhang <[email protected]>; David Airlie <[email protected]>; Daniel Vetter <[email protected]>; Thomas Zimmermann <[email protected]>; [email protected]; linux- [email protected]; [email protected]
Subject: Re: [PATCH] drm/hyperv: Fix double mouse pointers

Thanks Dexuan, for the patch. A minor comment below.

Quoted 2 lines · Dexuan Cui

It looks like Hyper-V supports a hardware cursor feature. It is not

used

Quoted 1 line

by Linux VM, but the Hyper-V host still draws a point as an extra

mouse

Quoted 3 lines

pointer, which is unwanted, especially when Xorg is running.

The hyperv_fb driver uses synthvid_send_ptr() to hide the unwanted

pointer.

Quoted 1 line

When the hyperv_drm driver was developed, the function

synthvid_send_ptr()

Quoted 4 lines

was not copied from the hyperv_fb driver. Fix the issue by adding the function into hyperv_drm.

Fixes 76c56a5affeb ("drm/hyperv: Add DRM driver for hyperv synthetic

video device")

Quoted 7 lines
Signed-off-by Dexuan Cui <[email protected]>
Cc Deepak Rawat <[email protected]>
Cc Haiyang Zhang <[email protected]>
---
 drivers/gpu/drm/hyperv/hyperv_drm.h         |  1 +
 drivers/gpu/drm/hyperv/hyperv_drm_modeset.c |  1 +
 drivers/gpu/drm/hyperv/hyperv_drm_proto.c   | 39

++++++++++++++++++++-

Review

Quoted 3 lines
 3 files changed, 40 insertions(+), 1 deletion(-)
(quoted diff) +0 -0

b/drivers/gpu/drm/hyperv/hyperv_drm.h

drivers/gpu/drm/hyperv/hyperv_drm.h int hyperv_mode_config_init(struct hyperv_drm_device

Quoted 4 lines

index 886add4f9cd0..27bfd27c05be 100644

drivers/gpu/drm/hyperv/hyperv_drm.h +0 -0
@@ -46,6 +46,7 @@ int hyperv_mode_config_init(struct hyperv_drm_device

*hv);

drivers/gpu/drm/hyperv/hyperv_drm.h int hyperv_mode_config_init(struct hyperv_drm_device · line 46

Quoted 1 line
drivers/gpu/drm/hyperv/hyperv_drm.h +0 -0
 int hyperv_update_vram_location(struct hv_device *hdev, phys_addr_t

vram_pp);

drivers/gpu/drm/hyperv/hyperv_drm.h int hyperv_mode_config_init(struct hyperv_drm_device · line 47

Quoted 1 line
drivers/gpu/drm/hyperv/hyperv_drm.h +0 -0
 int hyperv_update_situation(struct hv_device *hdev, u8 active, u32

bpp,

Review

Quoted 6 lines
drivers/gpu/drm/hyperv/hyperv_drm.h +1 -0
                            u32 w, u32 h, u32 pitch);
+int hyperv_send_ptr(struct hv_device *hdev);
 int hyperv_update_dirt(struct hv_device *hdev, struct drm_rect *rect);
 int hyperv_connect_vsp(struct hv_device *hdev);
(quoted diff) +0 -0

b/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c

drivers/gpu/drm/hyperv/hyperv_drm_modeset.c static void hyperv_pipe_enable(struct

Quoted 4 lines

index 3aaee4730ec6..e21c82cf3326 100644

drivers/gpu/drm/hyperv/hyperv_drm_modeset.c +0 -0
@@ -101,6 +101,7 @@ static void hyperv_pipe_enable(struct

drm_simple_display_pipe *pipe,

drivers/gpu/drm/hyperv/hyperv_drm_modeset.c static void hyperv_pipe_enable(struct · line 102

Quoted 2 lines
drivers/gpu/drm/hyperv/hyperv_drm_modeset.c +0 -0
        struct hyperv_drm_device *hv = to_hv(pipe->crtc.dev);
        struct drm_shadow_plane_state *shadow_plane_state =

to_drm_shadow_plane_state(plane_state);

Review

Quoted 6 lines
drivers/gpu/drm/hyperv/hyperv_drm_modeset.c +1 -0
+       hyperv_send_ptr(hv->hdev);
        hyperv_update_situation(hv->hdev, 1,  hv->screen_depth,
                                crtc_state->mode.hdisplay,
                                crtc_state->mode.vdisplay,
(quoted diff) +0 -0

b/drivers/gpu/drm/hyperv/hyperv_drm_proto.c

drivers/gpu/drm/hyperv/hyperv_drm_proto.c int hyperv_update_situation(struct hv_device

Quoted 4 lines

index 6d4bdccfbd1a..1ea7a0432320 100644

drivers/gpu/drm/hyperv/hyperv_drm_proto.c +0 -0
@@ -299,6 +299,40 @@ int hyperv_update_situation(struct hv_device

*hdev, u8 active, u32 bpp,

Quoted 3 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +0 -0
        return 0;
 }
Quoted 13 lines

drivers/gpu/drm/hyperv/hyperv_drm_proto.c int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp, · line 311

Quoted 10 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +10 -0
+/* Send mouse pointer info to host */
+int hyperv_send_ptr(struct hv_device *hdev)
+{
+       struct synthvid_msg msg;
+
+       memset(&msg, 0, sizeof(struct synthvid_msg));
+       msg.vid_hdr.type = SYNTHVID_POINTER_POSITION;
+       msg.vid_hdr.size = sizeof(struct synthvid_msg_hdr) +
+               sizeof(struct synthvid_pointer_position);
+       msg.ptr_pos.is_visible = 1;

"is_visible" should be 0 since you want to hide the pointer. Maybe better, accept these from the caller.

I believe I tried is_visible = 0 during my implementation of the hyperv_fb. It still have the host cursor visible... So I had to set the cursor color's alpha value = 0 to make it transparent. But you may try is_visible = 0 again.

Thanks,
- Haiyang

author maintainer
#4 lore raw
In reply to Deepak Rawat
Quoted 16 lines

From: Deepak Rawat <[email protected]>
Sent: Tuesday, September 14, 2021 8:59 AM
...

drivers/gpu/drm/hyperv/hyperv_drm_proto.c int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp, · line 311

Quoted 10 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +10 -0
+/* Send mouse pointer info to host */
+int hyperv_send_ptr(struct hv_device *hdev)
+{
+       struct synthvid_msg msg;
+
+       memset(&msg, 0, sizeof(struct synthvid_msg));
+       msg.vid_hdr.type = SYNTHVID_POINTER_POSITION;
+       msg.vid_hdr.size = sizeof(struct synthvid_msg_hdr) +
+               sizeof(struct synthvid_pointer_position);
+       msg.ptr_pos.is_visible = 1;

"is_visible" should be 0 since you want to hide the pointer. Maybe better, accept these from the caller.

According to my test, "is_visible = 0" doesn't work, i.e. can't hide the unwanted HW mouse poiner. It looks like the field is for some very old legacy Windows VMs like Windows Vista.

Haiyang also replied in another email, saying "is_visible = 0" doesn't work.

Quoted 15 lines
Quoted 15 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +15 -0
+       msg.ptr_pos.video_output = 0;
+       msg.ptr_pos.image_x = 0;
+       msg.ptr_pos.image_y = 0;
+       hyperv_sendpacket(hdev, &msg);
+
+       memset(&msg, 0, sizeof(struct synthvid_msg));
+       msg.vid_hdr.type = SYNTHVID_POINTER_SHAPE;
+       msg.vid_hdr.size = sizeof(struct synthvid_msg_hdr) +
+               sizeof(struct synthvid_pointer_shape);
+       msg.ptr_shape.part_idx = SYNTHVID_CURSOR_COMPLETE;
+       msg.ptr_shape.is_argb = 1;
+       msg.ptr_shape.width = 1;
+       msg.ptr_shape.height = 1;
+       msg.ptr_shape.hot_x = 0;
+       msg.ptr_shape.hot_y = 0;
Quoted 8 lines

drivers/gpu/drm/hyperv/hyperv_drm_proto.c int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp, · line 332

Quoted 6 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +6 -0
+       msg.ptr_shape.data[0] = 0;
+       msg.ptr_shape.data[1] = 1;
+       msg.ptr_shape.data[2] = 1;
+       msg.ptr_shape.data[3] = 1;
+       hyperv_sendpacket(hdev, &msg);
+

Is it necessary to send SYNTHVID_POINTER_SHAPE here? Perhaps we should

According to my test, yes. If I don't send a SYNTHVID_POINTER_SHAPE message, the unwanted mouse pointer can't be hidden. As we know, the protocol between the VSC and the VSP is not well documented to us. I can ask Hyper-V team for some clarification on this, but it's probably we can just use the current version of hiding the mouse pointer as-is -- this has been used for 10+ years in the hyperv_fb driver without any issue. :-)

Quoted 2 lines

separate SYNTHVID_POINTER_POSITION and SYNTHVID_POINTER_SHAPE into different functions.

Since the 2 messages are only used here, I suggest we keep it as-is.

Thanks,
-- Dexuan

Reviewed-by
#5 lore raw

HI Dexuan, thanks for confirming. Could you please add this as a comment to the function.

Reviewed-by Deepak Rawat <[email protected]>
Quoted 63 lines · Dexuan Cui
Quoted 4 lines

From: Deepak Rawat <[email protected]>
Sent: Tuesday, September 14, 2021 8:59 AM
...

Quoted 13 lines

drivers/gpu/drm/hyperv/hyperv_drm_proto.c int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp, · line 311

Quoted 10 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +10 -0
+/* Send mouse pointer info to host */
+int hyperv_send_ptr(struct hv_device *hdev)
+{
+       struct synthvid_msg msg;
+
+       memset(&msg, 0, sizeof(struct synthvid_msg));
+       msg.vid_hdr.type = SYNTHVID_POINTER_POSITION;
+       msg.vid_hdr.size = sizeof(struct synthvid_msg_hdr) +
+               sizeof(struct synthvid_pointer_position);
+       msg.ptr_pos.is_visible = 1;

"is_visible" should be 0 since you want to hide the pointer. Maybe better, accept these from the caller.

According to my test, "is_visible = 0" doesn't work, i.e. can't hide the unwanted HW mouse poiner. It looks like the field is for some very old legacy Windows VMs like Windows Vista.

Haiyang also replied in another email, saying "is_visible = 0" doesn't work.

Quoted 15 lines
Quoted 15 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +15 -0
+       msg.ptr_pos.video_output = 0;
+       msg.ptr_pos.image_x = 0;
+       msg.ptr_pos.image_y = 0;
+       hyperv_sendpacket(hdev, &msg);
+
+       memset(&msg, 0, sizeof(struct synthvid_msg));
+       msg.vid_hdr.type = SYNTHVID_POINTER_SHAPE;
+       msg.vid_hdr.size = sizeof(struct synthvid_msg_hdr) +
+               sizeof(struct synthvid_pointer_shape);
+       msg.ptr_shape.part_idx = SYNTHVID_CURSOR_COMPLETE;
+       msg.ptr_shape.is_argb = 1;
+       msg.ptr_shape.width = 1;
+       msg.ptr_shape.height = 1;
+       msg.ptr_shape.hot_x = 0;
+       msg.ptr_shape.hot_y = 0;
Quoted 8 lines

drivers/gpu/drm/hyperv/hyperv_drm_proto.c int hyperv_update_situation(struct hv_device *hdev, u8 active, u32 bpp, · line 332

Quoted 6 lines
drivers/gpu/drm/hyperv/hyperv_drm_proto.c +6 -0
+       msg.ptr_shape.data[0] = 0;
+       msg.ptr_shape.data[1] = 1;
+       msg.ptr_shape.data[2] = 1;
+       msg.ptr_shape.data[3] = 1;
+       hyperv_sendpacket(hdev, &msg);
+

Is it necessary to send SYNTHVID_POINTER_SHAPE here? Perhaps we should

According to my test, yes. If I don't send a SYNTHVID_POINTER_SHAPE message, the unwanted mouse pointer can't be hidden. As we know, the protocol between the VSC and the VSP is not well documented to us. I can ask Hyper-V team for some clarification on this, but it's probably we can just use the current version of hiding the mouse pointer as-is -- this has been used for 10+ years in the hyperv_fb driver without any issue. :-)

Quoted 2 lines

separate SYNTHVID_POINTER_POSITION and SYNTHVID_POINTER_SHAPE into different functions.

Since the 2 messages are only used here, I suggest we keep it as-is.

Thanks,
-- Dexuan

author maintainer
#6 lore raw
Quoted 9 lines

From: Deepak Rawat <[email protected]>
Sent: Thursday, September 16, 2021 9:17 AM
To: Dexuan Cui <[email protected]>
Cc: Haiyang Zhang <[email protected]>; David Airlie <[email protected]>; Daniel Vetter <[email protected]>; Thomas Zimmermann <[email protected]>; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH] drm/hyperv: Fix double mouse pointers

Quoted 4 lines

HI Dexuan, thanks for confirming. Could you please add this as a comment to the function.

Reviewed-by Deepak Rawat <[email protected]>

Thanks, Deepak!
Will do, and I'll rename the function to hyperv_hide_hw_ptr() to better express the intent of the function.