[PATCH] drm/dumb-buffers: Increase size limits to match current devices

[PATCH] drm/dumb-buffers: Increase size limits to match current devices

[Thomas Zimmermann]

Increase the limits for dumb-buffer arguments to sizes of 16384x16384
at 64 bpp. Such buffer sizes are still within 32-bit unsigned limits.
Fixes CI and high-end devices.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Fixes: 5ab62dd3687b ("drm: prevent integer overflows in dumb buffer creation helpers")
Reported-by: Jani Nikula <jani.nikula@linux.intel.com>
Closes: https://lore.kernel.org/dri-devel/ddf0233e50044059c85279f928661563ef6a55bf@intel.com/
Cc: Rajat Gupta <rajat.gupta@oss.qualcomm.com>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
---
 drivers/gpu/drm/drm_dumb_buffers.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/drm_dumb_buffers.c b/drivers/gpu/drm/drm_dumb_buffers.c
index e60130b2bb0c..9c4cfb3210aa 100644
--- a/drivers/gpu/drm/drm_dumb_buffers.c
+++ b/drivers/gpu/drm/drm_dumb_buffers.c
@@ -201,11 +201,12 @@ int drm_mode_create_dumb(struct drm_device *dev,
 	if (!args->width || !args->height || !args->bpp)
 		return -EINVAL;
 
-	/* Reject unreasonable inputs early.  Dumb buffers are for software
-	 * rendering; nothing legitimate needs more than 8192x8192 at 32bpp.
-	 * This prevents overflows in downstream alignment helpers.
+	/*
+	 * Reject unreasonable inputs early. Dumb buffers are for software
+	 * rendering; nothing legitimate needs more than 16384x16384 at
+	 * 64 bpp. This prevents overflows in downstream alignment helpers.
 	 */
-	if (args->width >= 8192 || args->height >= 8192 || args->bpp > 32)
+	if (args->width > 16384 || args->height > 16384 || args->bpp > 64)
 		return -EINVAL;
 
 	/* overflow checks for 32bit size calculations */

base-commit: a980196655477a8f5067112946401fe52e510664
-- 
2.54.0

RE: [PATCH] drm/dumb-buffers: Increase size limits to match current devices

[Saarinen, Jani]

Hi, 
> -----Original Message-----
> From: Thomas Zimmermann <tzimmermann@suse.de>
> Sent: Tuesday, 2 June 2026 10.30
> To: rajat.gupta@oss.qualcomm.com; jani.nikula@linux.intel.com; Saarinen,
> Jani <jani.saarinen@intel.com>; simona@ffwll.ch; airlied@gmail.com;
> mripard@kernel.org; maarten.lankhorst@linux.intel.com
> Cc: dri-devel@lists.freedesktop.org; intel-gfx@lists.freedesktop.org; intel-
> xe@lists.freedesktop.org; Thomas Zimmermann <tzimmermann@suse.de>
> Subject: [PATCH] drm/dumb-buffers: Increase size limits to match current
> devices
> 
> Increase the limits for dumb-buffer arguments to sizes of 16384x16384 at 64
> bpp. Such buffer sizes are still within 32-bit unsigned limits.
> Fixes CI and high-end devices.
> 
> Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
> Fixes: 5ab62dd3687b ("drm: prevent integer overflows in dumb buffer
> creation helpers")
> Reported-by: Jani Nikula <jani.nikula@linux.intel.com>
> Closes: https://lore.kernel.org/dri-
> devel/ddf0233e50044059c85279f928661563ef6a55bf@intel.com/

Seems not helping on i915 CI BAT issue https://lore.kernel.org/intel-gfx/178038951063.34290.3796875109342598028@6beec6c84f66/T/#u 
=> https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_167719v1/index.html?

> Cc: Rajat Gupta <rajat.gupta@oss.qualcomm.com>
> Cc: Thomas Zimmermann <tzimmermann@suse.de>
> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Cc: Maxime Ripard <mripard@kernel.org>

Br,
Jani

> ---
>  drivers/gpu/drm/drm_dumb_buffers.c | 9 +++++----
>  1 file changed, 5 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/gpu/drm/drm_dumb_buffers.c
> b/drivers/gpu/drm/drm_dumb_buffers.c
> index e60130b2bb0c..9c4cfb3210aa 100644
> --- a/drivers/gpu/drm/drm_dumb_buffers.c
> +++ b/drivers/gpu/drm/drm_dumb_buffers.c
> @@ -201,11 +201,12 @@ int drm_mode_create_dumb(struct drm_device
> *dev,
>  	if (!args->width || !args->height || !args->bpp)
>  		return -EINVAL;
> 
> -	/* Reject unreasonable inputs early.  Dumb buffers are for software
> -	 * rendering; nothing legitimate needs more than 8192x8192 at
> 32bpp.
> -	 * This prevents overflows in downstream alignment helpers.
> +	/*
> +	 * Reject unreasonable inputs early. Dumb buffers are for software
> +	 * rendering; nothing legitimate needs more than 16384x16384 at
> +	 * 64 bpp. This prevents overflows in downstream alignment helpers.
>  	 */
> -	if (args->width >= 8192 || args->height >= 8192 || args->bpp > 32)
> +	if (args->width > 16384 || args->height > 16384 || args->bpp > 64)
>  		return -EINVAL;
> 
>  	/* overflow checks for 32bit size calculations */
> 
> base-commit: a980196655477a8f5067112946401fe52e510664
> --
> 2.54.0

RE: [PATCH] drm/dumb-buffers: Increase size limits to match current devices

[Jani Nikula]

On Tue, 02 Jun 2026, "Saarinen, Jani" <jani.saarinen@intel.com> wrote:
> Hi, 
>> -----Original Message-----
>> From: Thomas Zimmermann <tzimmermann@suse.de>
>> Sent: Tuesday, 2 June 2026 10.30
>> To: rajat.gupta@oss.qualcomm.com; jani.nikula@linux.intel.com; Saarinen,
>> Jani <jani.saarinen@intel.com>; simona@ffwll.ch; airlied@gmail.com;
>> mripard@kernel.org; maarten.lankhorst@linux.intel.com
>> Cc: dri-devel@lists.freedesktop.org; intel-gfx@lists.freedesktop.org; intel-
>> xe@lists.freedesktop.org; Thomas Zimmermann <tzimmermann@suse.de>
>> Subject: [PATCH] drm/dumb-buffers: Increase size limits to match current
>> devices
>> 
>> Increase the limits for dumb-buffer arguments to sizes of 16384x16384 at 64
>> bpp. Such buffer sizes are still within 32-bit unsigned limits.
>> Fixes CI and high-end devices.
>> 
>> Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
>> Fixes: 5ab62dd3687b ("drm: prevent integer overflows in dumb buffer
>> creation helpers")
>> Reported-by: Jani Nikula <jani.nikula@linux.intel.com>
>> Closes: https://lore.kernel.org/dri-
>> devel/ddf0233e50044059c85279f928661563ef6a55bf@intel.com/
>
> Seems not helping on i915 CI BAT issue https://lore.kernel.org/intel-gfx/178038951063.34290.3796875109342598028@6beec6c84f66/T/#u 
> => https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_167719v1/index.html?

Right, so vgem_basic has this, among other things:

	bo.width = 1<<15;
	bo.height = 1<<15;
	bo.bpp = 16;
	vgem_create(fd, &bo);

i.e. 32768x32768.

Not so sure about this one.

Anyone else care to chime in?


BR,
Jani.

>
>> Cc: Rajat Gupta <rajat.gupta@oss.qualcomm.com>
>> Cc: Thomas Zimmermann <tzimmermann@suse.de>
>> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
>> Cc: Maxime Ripard <mripard@kernel.org>
>
> Br,
> Jani
>
>> ---
>>  drivers/gpu/drm/drm_dumb_buffers.c | 9 +++++----
>>  1 file changed, 5 insertions(+), 4 deletions(-)
>> 
>> diff --git a/drivers/gpu/drm/drm_dumb_buffers.c
>> b/drivers/gpu/drm/drm_dumb_buffers.c
>> index e60130b2bb0c..9c4cfb3210aa 100644
>> --- a/drivers/gpu/drm/drm_dumb_buffers.c
>> +++ b/drivers/gpu/drm/drm_dumb_buffers.c
>> @@ -201,11 +201,12 @@ int drm_mode_create_dumb(struct drm_device
>> *dev,
>>  	if (!args->width || !args->height || !args->bpp)
>>  		return -EINVAL;
>> 
>> -	/* Reject unreasonable inputs early.  Dumb buffers are for software
>> -	 * rendering; nothing legitimate needs more than 8192x8192 at
>> 32bpp.
>> -	 * This prevents overflows in downstream alignment helpers.
>> +	/*
>> +	 * Reject unreasonable inputs early. Dumb buffers are for software
>> +	 * rendering; nothing legitimate needs more than 16384x16384 at
>> +	 * 64 bpp. This prevents overflows in downstream alignment helpers.
>>  	 */
>> -	if (args->width >= 8192 || args->height >= 8192 || args->bpp > 32)
>> +	if (args->width > 16384 || args->height > 16384 || args->bpp > 64)
>>  		return -EINVAL;
>> 
>>  	/* overflow checks for 32bit size calculations */
>> 
>> base-commit: a980196655477a8f5067112946401fe52e510664
>> --
>> 2.54.0
>

-- 
Jani Nikula, Intel

Re: [PATCH] drm/dumb-buffers: Increase size limits to match current devices

[Thomas Zimmermann]

Hi

Am 02.06.26 um 12:53 schrieb Jani Nikula:
> On Tue, 02 Jun 2026, "Saarinen, Jani" <jani.saarinen@intel.com> wrote:
>> Hi,
>>> -----Original Message-----
>>> From: Thomas Zimmermann <tzimmermann@suse.de>
>>> Sent: Tuesday, 2 June 2026 10.30
>>> To: rajat.gupta@oss.qualcomm.com; jani.nikula@linux.intel.com; Saarinen,
>>> Jani <jani.saarinen@intel.com>; simona@ffwll.ch; airlied@gmail.com;
>>> mripard@kernel.org; maarten.lankhorst@linux.intel.com
>>> Cc: dri-devel@lists.freedesktop.org; intel-gfx@lists.freedesktop.org; intel-
>>> xe@lists.freedesktop.org; Thomas Zimmermann <tzimmermann@suse.de>
>>> Subject: [PATCH] drm/dumb-buffers: Increase size limits to match current
>>> devices
>>>
>>> Increase the limits for dumb-buffer arguments to sizes of 16384x16384 at 64
>>> bpp. Such buffer sizes are still within 32-bit unsigned limits.
>>> Fixes CI and high-end devices.
>>>
>>> Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
>>> Fixes: 5ab62dd3687b ("drm: prevent integer overflows in dumb buffer
>>> creation helpers")
>>> Reported-by: Jani Nikula <jani.nikula@linux.intel.com>
>>> Closes: https://lore.kernel.org/dri-
>>> devel/ddf0233e50044059c85279f928661563ef6a55bf@intel.com/
>> Seems not helping on i915 CI BAT issue https://lore.kernel.org/intel-gfx/178038951063.34290.3796875109342598028@6beec6c84f66/T/#u
>> => https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_167719v1/index.html?
> Right, so vgem_basic has this, among other things:
>
> 	bo.width = 1<<15;
> 	bo.height = 1<<15;
> 	bo.bpp = 16;
> 	vgem_create(fd, &bo);
>
> i.e. 32768x32768.
>
> Not so sure about this one.
>
> Anyone else care to chime in?

I give up. I'll send out a patch to remove the size limits in a bit. The 
rest of the overflow fixes in the broken patch should not be a problem, 
I think.

But TBH allocating a dumb buffer of that size makes no sense. Who uses a 
2-GiB buffer for software rendering?

And as I mentioned elsewhere, there are reports about devices with GiBs 
of VRAM but only a few hundered MiB of PCI-BAR ranges. As we don't have 
DMA here, there will be some size limits required to make mmap work 
correctly.

Best regards
Thomas

>
>
> BR,
> Jani.
>
>>> Cc: Rajat Gupta <rajat.gupta@oss.qualcomm.com>
>>> Cc: Thomas Zimmermann <tzimmermann@suse.de>
>>> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
>>> Cc: Maxime Ripard <mripard@kernel.org>
>> Br,
>> Jani
>>
>>> ---
>>>   drivers/gpu/drm/drm_dumb_buffers.c | 9 +++++----
>>>   1 file changed, 5 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/drm_dumb_buffers.c
>>> b/drivers/gpu/drm/drm_dumb_buffers.c
>>> index e60130b2bb0c..9c4cfb3210aa 100644
>>> --- a/drivers/gpu/drm/drm_dumb_buffers.c
>>> +++ b/drivers/gpu/drm/drm_dumb_buffers.c
>>> @@ -201,11 +201,12 @@ int drm_mode_create_dumb(struct drm_device
>>> *dev,
>>>   	if (!args->width || !args->height || !args->bpp)
>>>   		return -EINVAL;
>>>
>>> -	/* Reject unreasonable inputs early.  Dumb buffers are for software
>>> -	 * rendering; nothing legitimate needs more than 8192x8192 at
>>> 32bpp.
>>> -	 * This prevents overflows in downstream alignment helpers.
>>> +	/*
>>> +	 * Reject unreasonable inputs early. Dumb buffers are for software
>>> +	 * rendering; nothing legitimate needs more than 16384x16384 at
>>> +	 * 64 bpp. This prevents overflows in downstream alignment helpers.
>>>   	 */
>>> -	if (args->width >= 8192 || args->height >= 8192 || args->bpp > 32)
>>> +	if (args->width > 16384 || args->height > 16384 || args->bpp > 64)
>>>   		return -EINVAL;
>>>
>>>   	/* overflow checks for 32bit size calculations */
>>>
>>> base-commit: a980196655477a8f5067112946401fe52e510664
>>> --
>>> 2.54.0

-- 
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, Werner Knoblich, (HRB 36809, AG Nürnberg)

Claude review: drm/dumb-buffers: Increase size limits to match current devices

[Claude Code Review Bot]

Patch Review

**Correctness of the new limits:**

The worst-case arithmetic at the new limits (16384 x 16384 x 64bpp):
- `cpp = DIV_ROUND_UP(64, 8) = 8`
- `stride = 8 * 16384 = 131,072`
- `size = 16384 * 131,072 = 2,147,483,648` (0x80000000)

This fits within `u32` (max ~4.29 billion) and `PAGE_ALIGN(0x80000000)` is non-zero, so all downstream overflow checks in the old `drm_mode_create_dumb()` pass. The arithmetic is safe.

**The `>=` to `>` change is correct and intentional:**

```c
-	if (args->width >= 8192 || args->height >= 8192 || args->bpp > 32)
+	if (args->width > 16384 || args->height > 16384 || args->bpp > 64)
```

The old code rejected `width == 8192` (off-by-one — the comment said "8192x8192" but the `>=` operator excluded that size). The new code correctly allows `width == 16384` using `>`. This is a good fix.

**64bpp is a legitimate value:**

The current drm-next tree explicitly handles `case 64` in `drm_mode_size_dumb()` for `DRM_FORMAT_{XRGB,XBGR,ARGB,ABGR}16161616F` formats, confirming 64bpp is a real use case. The original 32bpp limit was too restrictive.

**Comment update is appropriate:**

```c
+	/*
+	 * Reject unreasonable inputs early. Dumb buffers are for software
+	 * rendering; nothing legitimate needs more than 16384x16384 at
+	 * 64 bpp. This prevents overflows in downstream alignment helpers.
 	 */
```

The comment accurately describes the new limits and the multi-line comment style is reformatted to kernel coding style (opening `/*` on its own line).

**Minor observations:**

1. The `Fixes:` tag placement is after `Signed-off-by:` — while not strictly wrong, the conventional ordering in the kernel is `Fixes:` before `Signed-off-by:`. This is a trivial style point.

2. The patch is well-scoped and minimal — it changes exactly what needs to change for the fix.

3. As noted above, this patch is **superseded on drm-next** where the hardcoded limits have been entirely replaced with overflow-safe arithmetic. If this is intended for stable backport to kernels that still have the original limiting code, it looks correct. If it's intended for drm-next, it will need to be dropped or reworked since the target code no longer exists.

**Verdict:** The patch is correct for the codebase version it targets. The arithmetic is safe within u32 bounds, the off-by-one fix (`>=` to `>`) is an improvement, and 64bpp / 16384 resolution are justified by real hardware and formats. The main concern is applicability — this should be applied to the branch containing `5ab62dd3687b`, not drm-next where the code has been refactored away.

---
Generated by Claude Code Patch Reviewer

Claude review: drm/dumb-buffers: Increase size limits to match current devices

[Claude Code Review Bot]

Overall Series Review

Subject: drm/dumb-buffers: Increase size limits to match current devices
Author: Thomas Zimmermann <tzimmermann@suse.de>
Patches: 4
Reviewed: 2026-06-04T13:08:23.071360

---

This is a single patch from Thomas Zimmermann that increases the hardcoded sanity limits for dumb buffer creation from 8192x8192 at 32bpp to 16384x16384 at 64bpp. It fixes commit `5ab62dd3687b` which introduced overly strict limits that broke CI and high-end devices (e.g., 64bpp formats like XRGB16161616F, and resolutions beyond 8K).

**The patch targets an older codebase.** On the current drm-next tree, the file has been significantly refactored — the hardcoded width/height/bpp limits this patch modifies have been entirely removed and replaced with proper overflow-safe arithmetic (`check_mul_overflow`, `DIV_ROUND_UP` with explicit `U32_MAX` bounds, and format-aware pitch calculation via `drm_mode_size_dumb()`). This explains why the patch does not apply cleanly.

This patch is likely intended for **stable backport** or an older branch where the refactoring hasn't landed yet. As a fix for the original limiting commit, the approach is reasonable.

---
Generated by Claude Code Patch Reviewer