| Age | Commit message (Collapse) | Author |
|---|
|
This commit adds support for the GB20x GPUs found on GeForce RTX 50xx
series boards.
Beyond a few miscellaneous register moves and HW class ID plumbing,
this reuses most of the code added to support GH100/GB10x.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
This commit enables basic support for the GB100/GB102 Blackwell GPUs.
Beyond HW class ID plumbing there's very little change here vs GH100.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
This commit enables basic support for Hopper GPUs, and is intended
primarily as a base supporting Blackwell GPUs, which reuse most of
the code added here.
Advanced features such as Confidential Compute are not supported.
Beyond a few miscellaneous register moves and HW class ID plumbing,
the bulk of the changes implemented here are to support the GSP-RM
boot sequence used on Hopper/Blackwell GPUs, as well as a new page
table layout.
There should be no changes here that impact prior GPUs.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Co-developed-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
570.144 has incompatible changes to NV0000_ALLOC_PARAMETERS.
Factor out the common code so it can be shared.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
555.42.02 has incompatible changes to FBSR.
At the same time, move the calling of FBSR functions from the instmem
subdev's suspend/resume paths, to GSP's. This is needed to fix ordering
issues that arise from changes to FBSR in newer RM versions.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
545.23.06 increases the libos3 heap size requirements, and GH100/GBxxx
will need their own implementation entirely.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
With GSP-RM handling the majority of the HW programming, NVKM's usual
HALs are more elaborate than necessary, resulting in a fair amount of
duplicated boilerplate.
Adds 'nvkm_rm_gpu' which serves to provide GPU-specific constants and
functions in a more streamlined manner.
This is initially used in subsequent commits to store engine class IDs,
and replace the per-engine/engobj boilerplate with common code for all
GSP-RM supported engines - and is further extended when adding GH100,
GB10x and GB20x support.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
Split handling of NV01_DEVICE (and other related objects) out into its
own module.
Aside from moving the function pointers, no code change is intended.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
Split NV01_ROOT handling out into its own module.
Aside from moving the function pointers, no code change is intended.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
Split base RM_ALLOC handling out into its own module.
Aside from moving the function pointers, no code change is intended.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
Split base RM_CONTROL handling out into its own module.
Aside from moving the function pointers, no code change is intended.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
Later patches in the series add HALs around various RM APIs in order to
support a newer version of GSP-RM firmware. In order to do this, begin
by splitting the code up into "modules" that roughly represent RM's API
boundaries so they can be more easily managed.
Aside from moving the RPC function pointers, no code change is indended.
Signed-off-by: Ben Skeggs <bskeggs@nvidia.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Timur Tabi <ttabi@nvidia.com>
Tested-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
Some GSP RPC commands need a new reply policy: "caller don't care about
the message content but want to make sure a reply is received". To
support this case, a new reply policy is introduced.
NV_VGPU_MSG_FUNCTION_ALLOC_MEMORY is a large GSP RPC command. The actual
required policy is NVKM_GSP_RPC_REPLY_POLL. This can be observed from the
dump of the GSP message queue. After the large GSP RPC command is issued,
GSP will write only an empty RPC header in the queue as the reply.
Without this change, the policy "receiving the entire message" is used
for NV_VGPU_MSG_FUNCTION_ALLOC_MEMORY. This causes the timeout of receiving
the returned GSP message in the suspend/resume path.
Introduce the new reply policy NVKM_GSP_RPC_REPLY_POLL, which waits for
the returned GSP message but discards it for the caller. Use the new policy
NVKM_GSP_RPC_REPLY_POLL on the GSP RPC command
NV_VGPU_MSG_FUNCTION_ALLOC_MEMORY.
Fixes: 50f290053d79 ("drm/nouveau: support handling the return of large GSP message")
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Alexandre Courbot <acourbot@nvidia.com>
Tested-by: Ben Skeggs <bskeggs@nvidia.com>
Signed-off-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20250227013554.8269-3-zhiw@nvidia.com
|
|
There can be multiple cases of handling the GSP RPC messages, which are
the reply of GSP RPC commands according to the requirement of the
callers and the nature of the GSP RPC commands.
The current supported reply policies are "callers don't care" and "receive
the entire message" according to the requirement of the callers. To
introduce a new policy, factor out the current RPC command reply polices.
Also, centralize the handling of the reply in a single function.
Factor out NVKM_GSP_RPC_REPLY_NOWAIT as "callers don't care" and
NVKM_GSP_RPC_REPLY_RECV as "receive the entire message". Introduce a
kernel doc to document the policies. Factor out
r535_gsp_rpc_handle_reply().
No functional change is intended for small GSP RPC commands. For large GSP
commands, the caller decides the policy of how to handle the returned GSP
RPC message.
Cc: Ben Skeggs <bskeggs@nvidia.com>
Cc: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20250227013554.8269-2-zhiw@nvidia.com
|
|
The user of *rm_alloc_push() always pass 0 in repc.
Remove unused param repc since no user actually uses it.
No functional change is intended.
Signed-off-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-5-zhiw@nvidia.com
|
|
The LOGINIT, LOGINTR, LOGRM, and LOGPMU buffers are circular buffers
that have printf-like logs from GSP-RM and PMU encoded in them.
LOGINIT, LOGINTR, and LOGRM are allocated by Nouveau and their DMA
addresses are passed to GSP-RM during initialization. The buffers are
required for GSP-RM to initialize properly.
LOGPMU is also allocated by Nouveau, but its contents are updated
when Nouveau receives an NV_VGPU_MSG_EVENT_UCODE_LIBOS_PRINT RPC from
GSP-RM. Nouveau then copies the RPC to the buffer.
The messages are encoded as an array of variable-length structures that
contain the parameters to an NV_PRINTF call. The format string and
parameter count are stored in a special ELF image that contains only
logging strings. This image is not currently shipped with the Nvidia
driver.
There are two methods to extract the logs.
OpenRM tries to load the logging ELF, and if present, parses the log
buffers in real time and outputs the strings to the kernel console.
Alternatively, and this is the method used by this patch, the buffers
can be exposed to user space, and a user-space tool (along with the
logging ELF image) can parse the buffer and dump the logs.
This method has the advantage that it allows the buffers to be parsed
even when the logging ELF file is not available to the user. However,
it has the disadvantage the debugfs entries need to remain until the
driver is unloaded.
The buffers are exposed via debugfs. If GSP-RM fails to initialize, then
Nouveau immediately shuts down the GSP interface. This would normally
also deallocate the logging buffers, thereby preventing the user from
capturing the debug logs.
To avoid this, introduce the keep-gsp-logging command line parameter. If
specified, and if at least one logging buffer has content, then Nouveau
will migrate these buffers into new debugfs entries that are retained
until the driver unloads.
An end-user can capture the logs using the following commands:
cp /sys/kernel/debug/nouveau/<path>/loginit loginit
cp /sys/kernel/debug/nouveau/<path>/logrm logrm
cp /sys/kernel/debug/nouveau/<path>/logintr logintr
cp /sys/kernel/debug/nouveau/<path>/logpmu logpmu
where (for a PCI device) <path> is the PCI ID of the GPU (e.g.
0000:65:00.0).
Since LOGPMU is not needed for normal GSP-RM operation, it is only
created if debugfs is available. Otherwise, the
NV_VGPU_MSG_EVENT_UCODE_LIBOS_PRINT RPCs are ignored.
A simple way to test the buffer migration feature is to have
nvkm_gsp_init() return an error code.
Tested-by: Ben Skeggs <bskeggs@nvidia.com>
Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20241030202952.694055-2-ttabi@nvidia.com
|
|
Store the struct device pointer used to allocate the DMA buffer in
the nvkm_gsp_mem object. This allows nvkm_gsp_mem_dtor() to release
the buffer without needing the nvkm_gsp. This is needed so that
we can retain DMA buffers even after the nvkm_gsp object is deleted.
Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20241030202952.694055-1-ttabi@nvidia.com
|
|
The mutex field has two following semicolons, replace this with just
one semicolon.
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20240917120856.1877733-1-colin.i.king@gmail.com
|
|
Let's start the new release cycle.
Signed-off-by: Maxime Ripard <mripard@kernel.org>
|
|
Currently we allocate all 3 levels of radix3 page tables using
nvkm_gsp_mem_ctor(), which uses dma_alloc_coherent() for allocating all of
the relevant memory. This can end up failing in scenarios where the system
has very high memory fragmentation, and we can't find enough contiguous
memory to allocate level 2 of the page table.
Currently, this can result in runtime PM issues on systems where memory
fragmentation is high - as we'll fail to allocate the page table for our
suspend/resume buffer:
kworker/10:2: page allocation failure: order:7, mode:0xcc0(GFP_KERNEL),
nodemask=(null),cpuset=/,mems_allowed=0
CPU: 10 PID: 479809 Comm: kworker/10:2 Not tainted
6.8.6-201.ChopperV6.fc39.x86_64 #1
Hardware name: SLIMBOOK Executive/Executive, BIOS N.1.10GRU06 02/02/2024
Workqueue: pm pm_runtime_work
Call Trace:
<TASK>
dump_stack_lvl+0x64/0x80
warn_alloc+0x165/0x1e0
? __alloc_pages_direct_compact+0xb3/0x2b0
__alloc_pages_slowpath.constprop.0+0xd7d/0xde0
__alloc_pages+0x32d/0x350
__dma_direct_alloc_pages.isra.0+0x16a/0x2b0
dma_direct_alloc+0x70/0x270
nvkm_gsp_radix3_sg+0x5e/0x130 [nouveau]
r535_gsp_fini+0x1d4/0x350 [nouveau]
nvkm_subdev_fini+0x67/0x150 [nouveau]
nvkm_device_fini+0x95/0x1e0 [nouveau]
nvkm_udevice_fini+0x53/0x70 [nouveau]
nvkm_object_fini+0xb9/0x240 [nouveau]
nvkm_object_fini+0x75/0x240 [nouveau]
nouveau_do_suspend+0xf5/0x280 [nouveau]
nouveau_pmops_runtime_suspend+0x3e/0xb0 [nouveau]
pci_pm_runtime_suspend+0x67/0x1e0
? __pfx_pci_pm_runtime_suspend+0x10/0x10
__rpm_callback+0x41/0x170
? __pfx_pci_pm_runtime_suspend+0x10/0x10
rpm_callback+0x5d/0x70
? __pfx_pci_pm_runtime_suspend+0x10/0x10
rpm_suspend+0x120/0x6a0
pm_runtime_work+0x98/0xb0
process_one_work+0x171/0x340
worker_thread+0x27b/0x3a0
? __pfx_worker_thread+0x10/0x10
kthread+0xe5/0x120
? __pfx_kthread+0x10/0x10
ret_from_fork+0x31/0x50
? __pfx_kthread+0x10/0x10
ret_from_fork_asm+0x1b/0x30
Luckily, we don't actually need to allocate coherent memory for the page
table thanks to being able to pass the GPU a radix3 page table for
suspend/resume data. So, let's rewrite nvkm_gsp_radix3_sg() to use the sg
allocator for level 2. We continue using coherent allocations for lvl0 and
1, since they only take a single page.
V2:
* Don't forget to actually jump to the next scatterlist when we reach the
end of the scatterlist we're currently on when writing out the page table
for level 2
Signed-off-by: Lyude Paul <lyude@redhat.com>
Cc: stable@vger.kernel.org
Reviewed-by: Ben Skeggs <bskeggs@nvidia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240429182318.189668-2-lyude@redhat.com
|
|
Add the NVreg_RegistryDwords command line parameter, which allows
specifying additional registry keys to be sent to GSP-RM. This
allows additional configuration, debugging, and experimentation
with GSP-RM, which uses these keys to alter its behavior.
Note that these keys are passed as-is to GSP-RM, and Nouveau does
not parse them. This is in contrast to the Nvidia driver, which may
parse some of the keys to configure some functionality in concert with
GSP-RM. Therefore, any keys which also require action by the driver
may not function correctly when passed by Nouveau. Caveat emptor.
The name and format of NVreg_RegistryDwords is the same as used by
the Nvidia driver, to maintain compatibility.
Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Danilo Krummrich <dakr@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240417215317.3490856-1-ttabi@nvidia.com
|
|
Nouveau manages GSP-RM DMA buffers with nvkm_gsp_mem objects. Several of
these buffers are never dealloced. Some of them can be deallocated
right after GSP-RM is initialized, but the rest need to stay until the
driver unloads.
Also futher bullet-proof these objects by poisoning the buffer and
clearing the nvkm_gsp_mem object when it is deallocated. Poisoning
the buffer should trigger an error (or crash) from GSP-RM if it tries
to access the buffer after we've deallocated it, because we were wrong
about when it is safe to deallocate.
Finally, change the mem->size field to a size_t because that's the same
type that dma_alloc_coherent expects.
Cc: <stable@vger.kernel.org> # v6.7
Fixes: 176fdcbddfd2 ("drm/nouveau/gsp/r535: add support for booting GSP-RM")
Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Signed-off-by: Danilo Krummrich <dakr@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240202230608.1981026-1-ttabi@nvidia.com
|
|
It looks like for some messages the upper layers need to get access to the
results of the message so we can interpret it.
Rework the ctrl push interface to not free things and cleanup properly
whereever it errors out.
Requested-by: Lyude
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231222043308.3090089-9-airlied@gmail.com
|
|
Adds support for allocating GR classes from RM.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230918202149.4343-41-skeggsb@gmail.com
|
|
Adds support for modesetting on RM.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230918202149.4343-38-skeggsb@gmail.com
|
|
- Valid VRAM regions are read from GSP-RM, and used to construct our MM
- BAR1/BAR2 VMMs modified to be shared with RM
- Client VMMs have RM VASPACE objects created for them
- Adds FBSR to backup system objects in VRAM across suspend
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230918202149.4343-37-skeggsb@gmail.com
|
|
Fetches the interrupt table from RM, and hooks up the GSP interrupt
handler to message queue processing to catch async messages.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230918202149.4343-36-skeggsb@gmail.com
|
|
Adds the plumbing to be able to allocate and free RM objects, and
implements RM client/device/subdevice allocation with it.
These will be used by subsequent patches.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230918202149.4343-35-skeggsb@gmail.com
|
|
Adds the plumbing to start making RM control calls, and initialises
objects to represent internal RM objects provided to us during init.
These will be used by subsequent patches.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230918202149.4343-34-skeggsb@gmail.com
|
|
This commit adds the initial code needed to boot the GSP-RM firmware
provided by NVIDIA, bringing with it the beginnings of Ada support.
Until it's had more testing and time to bake, support is disabled by
default (except on Ada). GSP-RM usage can be enabled by passing the
"config=NvGspRm=1" module option.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230918202149.4343-33-skeggsb@gmail.com
|
|
- move TOP after GSP, so we can disable TOP if GSP is in use
- provide plumbing to support falcon-only and GSP-RM paths
- provide a method for subdevs to detect GSP-RM paths
- split tu102/tu116/ga100 paths from gv100, which can't support GSP-RM
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230918202149.4343-5-skeggsb@gmail.com
|
|
v2. fixup for ga103 early merge
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Gourav Samaiya <gsamaiya@nvidia.com>
|
|
Ampere.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
|
|
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
|
|
This will allow us to register the falcon with ACR, and further customise
its behaviour by providing the nvkm_falcon_func structure directly.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
|
|
We need this for Turing ACR, but it's present from Volta onwards.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
|
|
Exact meaning of the acronym is unknown, but we need this for Turing ACR.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
|
