[PATCH] drm/nouveau: always set RMDevidCheckIgnore for GSP-RM

[PATCH] drm/nouveau: always set RMDevidCheckIgnore for GSP-RM

[Timur Tabi]

Always set the RMDevidCheckIgnore registry key for GSP-RM so that it
will continue support newer variants of already supported GPUs.

GSP-RM maintains an internal list of PCI IDs of GPUs that it supports,
and checks if the current GPU is on this list.  While the actual GPU
architecture (as specified in the BOOT_0/BOOT_42 registers) determines
how to enable the GPU, the PCI ID is used for the product name, e.g.
"NVIDIA GeForce RTX 5090".

Unfortunately, if there is no match, GSP-RM will refuse to initialize,
even if the device is fully supported.  Nouveau will get an error
return code, but by then it's too late.  This behavior may be corrected
in a future version of GSP-RM, but that does not help Nouveau today.

Fortunately, GSP-RM supports an undocumented registry key that tells it
to ignore the mismatch.  In such cases, the product name returned will
be a blank string, but otherwise GSP-RM will continue.

Unlike Nvidia's proprietary driver, Nouveau cannot update to newer
firmware versions to keep up with every new hardware release.  Instead,
we can permanently set this registry key, and GSP-RM will continue
to function the same with known hardware.

Signed-off-by: Timur Tabi <ttabi@nvidia.com>
---
 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c
index baf42339f93e..259d83ba595e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c
@@ -582,10 +582,13 @@ struct nv_gsp_registry_entries {
  * RMSecBusResetEnable - enables PCI secondary bus reset
  * RMForcePcieConfigSave - forces GSP-RM to preserve PCI configuration
  *   registers on any PCI reset.
+ * RMDevidCheckIgnore - allows GSP-RM to boot even if the PCI dev ID
+ *   is not found in the internal product name database.
  */
 static const struct nv_gsp_registry_entries r535_registry_entries[] = {
 	{ "RMSecBusResetEnable", 1 },
 	{ "RMForcePcieConfigSave", 1 },
+	{ "RMDevidCheckIgnore", 1 },
 };
 #define NV_GSP_REG_NUM_ENTRIES ARRAY_SIZE(r535_registry_entries)
 
-- 
2.43.0

Re: [PATCH] drm/nouveau: always set RMDevidCheckIgnore for GSP-RM

[Danilo Krummrich]

On 8/8/25 9:13 PM, Timur Tabi wrote:
> Always set the RMDevidCheckIgnore registry key for GSP-RM so that it
> will continue support newer variants of already supported GPUs.
> 
> GSP-RM maintains an internal list of PCI IDs of GPUs that it supports,
> and checks if the current GPU is on this list.  While the actual GPU
> architecture (as specified in the BOOT_0/BOOT_42 registers) determines
> how to enable the GPU, the PCI ID is used for the product name, e.g.
> "NVIDIA GeForce RTX 5090".
> 
> Unfortunately, if there is no match, GSP-RM will refuse to initialize,
> even if the device is fully supported.  Nouveau will get an error
> return code, but by then it's too late.  This behavior may be corrected
> in a future version of GSP-RM, but that does not help Nouveau today.

Are there any GPUs in the field today which technically are supported, but yet
refused by GSP due to this?

> Fortunately, GSP-RM supports an undocumented registry key that tells it
> to ignore the mismatch.  In such cases, the product name returned will
> be a blank string, but otherwise GSP-RM will continue.

I assume you refer to internal documentation when you say "undocumented"?

> Unlike Nvidia's proprietary driver, Nouveau cannot update to newer
> firmware versions to keep up with every new hardware release.  Instead,
> we can permanently set this registry key, and GSP-RM will continue
> to function the same with known hardware.

Re: [PATCH] drm/nouveau: always set RMDevidCheckIgnore for GSP-RM

[Timur Tabi]

On Sat, 2025-08-09 at 13:12 +0200, Danilo Krummrich wrote:
> Are there any GPUs in the field today which technically are supported, but yet
> refused by GSP due to this?

I don't think so.  This issue was discovered through a combination of an internal board with a new
PCI ID as well as code inspection.  The consensus is that GSP-RM should not be rejecting boards by
default.  That is, RMDevidCheckIgnore=1 should be the default behavior of GSP-RM.

> > Fortunately, GSP-RM supports an undocumented registry key that tells it
> > to ignore the mismatch.  In such cases, the product name returned will
> > be a blank string, but otherwise GSP-RM will continue.
> 
> I assume you refer to internal documentation when you say "undocumented"?

No, I'm referring to public documentation.  

Re: [PATCH] drm/nouveau: always set RMDevidCheckIgnore for GSP-RM

[Danilo Krummrich]

On 8/8/25 9:13 PM, Timur Tabi wrote:
> Always set the RMDevidCheckIgnore registry key for GSP-RM so that it
> will continue support newer variants of already supported GPUs.

Applied to drm-misc-next, thanks!