Support Linux DRM backend #69

Bennctu · 2024-11-03T14:58:55Z

Currently, using the DRM legacy interface is suitable for Mado's backend because we only need basic window display without advanced features.

backend/drm.c

jserv · 2024-11-03T15:13:53Z

Currently, using the DRM legacy interface is suitable for Mado's backend because we only need basic window display without advanced features.

What compatibility risks exist when using DRM legacy interfaces with newer Linux kernels? Could you verify the compatibility status of libdrm across different kernel versions?

What prevents the DRM backend from rendering content across the full screen resolution?

jserv · 2024-11-03T15:28:17Z

@a1091150, You might check DRM backend for your environment as well.

Makefile

huaxinliao · 2024-11-03T18:37:44Z

backend/drm.c

+
+    return true;
+
+bail_crtc:


I'm confused about this. It should be as below.

/* Retrieve resources */ if (!get_resources(tx->drm_dri_fd, &RES(tx))) goto bail_res; .... bail_fb: drmModeRmFB(tx->drm_dri_fd, tx->fb_id); munmap(tx->fb_base, tx->fb_len); bail_crtc: drmModeFreeCrtc(CRTC(tx)); bail_conn: drmModeFreeConnector(CONN(tx)); bail_res: drmModeFreeResources(RES(tx));

jserv

Rebase the latest main branch and resolve conflicts.

README.md

backend/linux_vt.c

Bennctu · 2024-11-06T16:40:59Z

Currently, using the DRM legacy interface is suitable for Mado's backend because we only need basic window display without advanced features.

What compatibility risks exist when using DRM legacy interfaces with newer Linux kernels? Could you verify the compatibility status of libdrm across different kernel versions?

What prevents the DRM backend from rendering content across the full screen resolution?

I need more time to research and adjust.

backend/drm.c

configs/Kconfig

backend/drm.c

jserv · 2024-11-11T02:48:58Z

Check the DRM implementation for reference: https://github.com/zlgopen/awtk-linux-fb/blob/master/awtk-port/lcd_linux/lcd_linux_drm.c

jserv · 2024-11-23T07:23:11Z

backend/drm.c

+
+/* Register the Linux DRM backend */
+
+const twin_backend_t g_twin_backend = {


You should add poll operation as it is required in recent main branch.

As the original implementation, we creates another thread to handle the events in function twin_linux_evdev_thread. So do you suggest refactoring linux_input.c#L255~278?

Or we just assign NULL to .poll?

/* Register the Linux DRM backend */ const twin_backend_t g_twin_backend = { .init = twin_drm_init, .poll = NULL, /*without implementation*/ .configure = twin_drm_configure, .exit = twin_drm_exit, };

jserv · 2024-11-23T09:38:18Z

README.md

+$ sudo ./demo-drm
+```
+
+The DRM device can be assigened via the environment variable `DRI_CARD`.


DRI_CARD is not common for understanding. How about DRMDEVICE?

Implemented DRM-based framebuffer setup, including: - Opening DRM device - Creating dumb buffers and mapping them to framebuffers - Setting mode and handling CRTC for connected display output Currently, using the DRM legacy interface is suitable for Mado's backend if we only need basic window display without advanced features. The DRM legacy is simpler to implement. Close sysprog21#60

jserv

Refine git commit messages:

Use Close #60 instead of Close sysprog21#60
Always use third-person perspective: Avoid personal pronouns like "I", "you", and "we", focusing on an objective description of changes.
More explanation for using legacy DRM interface. Provide context for why the alternatives are not suitable.

huaxinliao · 2025-01-26T13:12:47Z

backend/drm.c

+    if (drmModeAddFB(tx->drm_dri_fd, tx->width, tx->height, 24, 32,
+                     create_req.pitch, create_req.handle, &tx->fb_id) != 0) {
+        log_error("Cannot create framebubber");
+        return false;


I consider that this line should be changed to goto bail_dumb.

bail_dumb: struct drm_mode_destroy_dumb destroy = { .handle = create_req.handle }; ioctl(tx->drm_dri_fd, DRM_IOCTL_MODE_DESTROY_DUMB, &destroy); return false;

huaxinliao · 2025-01-26T13:15:57Z

backend/drm.c

+    /* Create framebuffer object for the dumb-buffer */
+    if (drmModeAddFB(tx->drm_dri_fd, tx->width, tx->height, 24, 32,
+                     create_req.pitch, create_req.handle, &tx->fb_id) != 0) {
+        log_error("Cannot create framebubber");


Correct the typo in framebubber.

huaxinliao · 2025-01-26T16:32:31Z

backend/drm.c

+    twin_screen_resize(SCREEN(ctx), width, height);
+}
+
+static void twin_drm_exit(twin_context_t *ctx)


I consider that twin_drm_exit should handle ioctl(tx->drm_dri_fd, DRM_IOCTL_MODE_DESTROY_DUMB, &destroy) to ensure allocated resources are properly released.

huaxinliao · 2025-01-26T17:34:36Z

I ran the DRM backend on a VM with Ubuntu 24.04. However, I noticed that the screen froze.

drm.mov

Bennctu · 2025-02-03T13:41:57Z

I ran the DRM backend on a VM with Ubuntu 24.04. However, I noticed that the screen froze.
drm.mov

@huaxinliao Thank you for reporting this issue.

cubic-dev-ai

1 issue found across 4 files

Prompt for AI agents (all 1 issues)


Understand the root cause of the following 1 issues and fix them.


<file name="README.md">

<violation number="1" location="README.md:140">
Documentation points users to set `DRI_CARD`, but the DRM backend actually reads `$DRMDEVICE`, so following the new instructions will not select the intended card and the line also contains a typo.</violation>
</file>

Since this is your first cubic review, here's how it works:

cubic automatically reviews your code and comments on bugs and improvements
Teach cubic by replying to its comments. cubic learns from your replies and gets better over time
Ask questions if you need clarification on any suggestion

_{React with 👍 or 👎 to teach cubic. Mention @cubic-dev-ai to give feedback, ask questions, or re-run the review.}

cubic-dev-ai · 2025-10-16T16:22:14Z

README.md

+$ sudo ./demo-drm
+```
+
+The DRM device can be assigened via the environment variable `DRI_CARD`.


Documentation points users to set DRI_CARD, but the DRM backend actually reads $DRMDEVICE, so following the new instructions will not select the intended card and the line also contains a typo.

Prompt for AI agents

Address the following comment on README.md at line 140: <comment>Documentation points users to set `DRI_CARD`, but the DRM backend actually reads `$DRMDEVICE`, so following the new instructions will not select the intended card and the line also contains a typo.</comment> <file context> @@ -128,6 +131,14 @@ $ sudo usermod -a -G video $USERNAME +$ sudo ./demo-drm +``` + +The DRM device can be assigened via the environment variable `DRI_CARD`. + To run demo program with the VNC backend: </file context>

Suggested change

The DRM device can be assigened via the environment variable `DRI_CARD`.

The DRM device can be assigned via the environment variable `DRMDEVICE`.

cubic-dev-ai

2 issues found across 4 files

Prompt for AI agents (all 2 issues)


Understand the root cause of the following 2 issues and fix them.


<file name="README.md">

<violation number="1" location="README.md:140">
Typo: &quot;assigened&quot; should be spelled &quot;assigned&quot; to keep the documentation clear.</violation>
</file>

<file name="backend/drm.c">

<violation number="1" location="backend/drm.c:64">
Frame buffer writes should advance by the mapped buffer pitch, but this code uses the visible width, so scanlines misalign whenever the dumb buffer pitch exceeds the mode width. Use the stride returned by create_fb() instead of tx-&gt;width to compute offsets.</violation>
</file>

Since this is your first cubic review, here's how it works:

cubic automatically reviews your code and comments on bugs and improvements
Teach cubic by replying to its comments. cubic learns from your replies and gets better over time
Ask questions if you need clarification on any suggestion

_{React with 👍 or 👎 to teach cubic. Mention @cubic-dev-ai to give feedback, ask questions, or re-run the review.}

cubic-dev-ai · 2025-10-16T16:37:06Z

README.md

+$ sudo ./demo-drm
+```
+
+The DRM device can be assigened via the environment variable `DRI_CARD`.


Typo: "assigened" should be spelled "assigned" to keep the documentation clear.

Prompt for AI agents

Address the following comment on README.md at line 140: <comment>Typo: "assigened" should be spelled "assigned" to keep the documentation clear.</comment> <file context> @@ -128,6 +131,14 @@ $ sudo usermod -a -G video $USERNAME +$ sudo ./demo-drm +``` + +The DRM device can be assigened via the environment variable `DRI_CARD`. + To run demo program with the VNC backend: </file context>

Suggested change

The DRM device can be assigened via the environment variable `DRI_CARD`.

The DRM device can be assigned via the environment variable `DRI_CARD`.

cubic-dev-ai · 2025-10-16T16:37:07Z

backend/drm.c

+        return;
+
+    twin_coord_t width = right - left;
+    off_t off = tx->width * top + left;


Frame buffer writes should advance by the mapped buffer pitch, but this code uses the visible width, so scanlines misalign whenever the dumb buffer pitch exceeds the mode width. Use the stride returned by create_fb() instead of tx->width to compute offsets.

Prompt for AI agents

Address the following comment on backend/drm.c at line 64: <comment>Frame buffer writes should advance by the mapped buffer pitch, but this code uses the visible width, so scanlines misalign whenever the dumb buffer pitch exceeds the mode width. Use the stride returned by create_fb() instead of tx->width to compute offsets.</comment> <file context> @@ -0,0 +1,302 @@ + return; + + twin_coord_t width = right - left; + off_t off = tx->width * top + left; + uint32_t *dest = + (uint32_t *) ((uintptr_t) tx->fb_base + (off * sizeof(*dest))); </file context>

jserv reviewed Nov 3, 2024

View reviewed changes

backend/drm.c Outdated Show resolved Hide resolved

jserv requested a review from shengwen-tw November 3, 2024 15:05

jserv reviewed Nov 3, 2024

View reviewed changes

backend/drm.c Outdated Show resolved Hide resolved

jserv reviewed Nov 3, 2024

View reviewed changes

Makefile Show resolved Hide resolved

huaxinliao reviewed Nov 3, 2024

View reviewed changes

jserv requested changes Nov 4, 2024

View reviewed changes

Bennctu force-pushed the drm_dev branch from f8729f4 to 8cd4c32 Compare November 6, 2024 16:30

jserv reviewed Nov 6, 2024

View reviewed changes

README.md Outdated Show resolved Hide resolved

jserv reviewed Nov 6, 2024

View reviewed changes

backend/linux_vt.c Outdated Show resolved Hide resolved

jserv reviewed Nov 6, 2024

View reviewed changes

backend/drm.c Outdated Show resolved Hide resolved

jserv reviewed Nov 6, 2024

View reviewed changes

backend/drm.c Outdated Show resolved Hide resolved

a1091150 mentioned this pull request Nov 7, 2024

Fix memory region in framebuffer #65

Merged

shengwen-tw reviewed Nov 9, 2024

View reviewed changes

configs/Kconfig Outdated Show resolved Hide resolved

backend/drm.c Outdated Show resolved Hide resolved

backend/drm.c Outdated Show resolved Hide resolved

backend/drm.c Outdated Show resolved Hide resolved

backend/drm.c Show resolved Hide resolved

This comment was marked as resolved.

Sign in to view

Bennctu force-pushed the drm_dev branch 2 times, most recently from 32e2217 to 3cba956 Compare November 23, 2024 07:19

jserv reviewed Nov 23, 2024

View reviewed changes

jserv requested a review from alanjian85 November 28, 2024 16:23

This comment was marked as resolved.

Sign in to view

Mention sample program execution with Linux DRM

20d4707

Bennctu force-pushed the drm_dev branch from 3cba956 to f711c33 Compare January 22, 2025 12:46

Bennctu force-pushed the drm_dev branch from f711c33 to e275994 Compare January 22, 2025 13:51

jserv requested a review from huaxinliao January 22, 2025 14:21

jserv requested changes Jan 24, 2025

View reviewed changes

huaxinliao reviewed Jan 26, 2025

View reviewed changes

sysprog21 deleted a comment from cubic-dev-ai bot Oct 16, 2025

cubic-dev-ai bot reviewed Oct 16, 2025

View reviewed changes

sysprog21 deleted a comment from cubic-dev-ai bot Oct 16, 2025

cubic-dev-ai bot reviewed Oct 16, 2025

View reviewed changes


		/* Register the Linux DRM backend */

		const twin_backend_t g_twin_backend = {

	The DRM device can be assigened via the environment variable `DRI_CARD`.
	The DRM device can be assigned via the environment variable `DRMDEVICE`.

Support Linux DRM backend #69

Are you sure you want to change the base?

Support Linux DRM backend #69

Uh oh!

Conversation

Bennctu commented Nov 3, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

jserv commented Nov 3, 2024

Uh oh!

jserv commented Nov 3, 2024

Uh oh!

Uh oh!

huaxinliao Nov 3, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jserv left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Bennctu commented Nov 6, 2024

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

jserv commented Nov 11, 2024

Uh oh!

This comment was marked as resolved.

Uh oh!

jserv Nov 23, 2024

Choose a reason for hiding this comment

Uh oh!

Bennctu Nov 28, 2024

Choose a reason for hiding this comment

Uh oh!

jserv Nov 23, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

This comment was marked as resolved.

Uh oh!

jserv left a comment

Choose a reason for hiding this comment

Uh oh!

huaxinliao Jan 26, 2025

Choose a reason for hiding this comment

Uh oh!

huaxinliao Jan 26, 2025

Choose a reason for hiding this comment

Uh oh!

huaxinliao Jan 26, 2025

Choose a reason for hiding this comment

Uh oh!

huaxinliao commented Jan 26, 2025

Uh oh!

Bennctu commented Feb 3, 2025

Uh oh!

cubic-dev-ai bot left a comment

Choose a reason for hiding this comment

Uh oh!

cubic-dev-ai bot Oct 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cubic-dev-ai bot left a comment

Choose a reason for hiding this comment

Uh oh!

cubic-dev-ai bot Oct 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cubic-dev-ai bot Oct 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Bennctu commented Nov 3, 2024 •

edited

Loading

huaxinliao Nov 3, 2024 •

edited

Loading

jserv Nov 23, 2024 •

edited

Loading

cubic-dev-ai bot Oct 16, 2025 •

edited

Loading

cubic-dev-ai bot Oct 16, 2025 •

edited

Loading

cubic-dev-ai bot Oct 16, 2025 •

edited

Loading