[PATCH v3 0/9] gpu: nova-core: gsp: add continuation record support

[PATCH v3 0/9] gpu: nova-core: gsp: add continuation record support

[Eliot Courtney]

GSP commands over 16 pages need to be sent using "continuation records"
which essentially means splitting the payload over multiple commands.

This series adds a command type `ContinuationRecord` which just writes
its header and whatever payload it is given. It also adds types
`SplitState` and `SplitCommand` which support splitting a large RPC
into smaller ones while transparently letting regular sized RPCs be sent
without extra copies.

The send pathway uses `SplitState` to send all commands, but if
the command fits into 16 pages, it still writes directly into the
command queue. If it is larger than 16 pages and needs continuation
records, it writes into a staging buffer, so there is one copy.

Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
---
Changes in v3:
- Rebased onto drm-rust-next
- Added some doc comments
- Reworded command_size doc comment
- Link to v2: https://lore.kernel.org/r/20260219-cmdq-continuation-v2-0-2e8b7615536f@nvidia.com

Changes in v2:
- Added doccoments
- Renamed driver_bytes_available_to_write to driver_write_area_size
- allocate_command_with_timeout merged allocate_command with timeout parameter
- Replaced hardcoded GSP_PAGE_SIZE * 16 with bindings
- Changed oversized command error from EIO to EMSGSIZE
- Added EMSGSIZE to kernel/error.rs
- Split WrappingCommand functionality into SplitState + SplitCommand enum
- Made max_size a const (MAX_CMD_SIZE)
- Removed send_continuation_record + added comment for type inference
- send_single_command now consumes the command
- Extracted command_size + used in SplitState 
- Link to v1: https://lore.kernel.org/r/20260212-cmdq-continuation-v1-0-73079ded55e6@nvidia.com

---
Eliot Courtney (9):
      gpu: nova-core: gsp: sort MsgFunction variants alphabetically
      gpu: nova-core: gsp: add mechanism to wait for space on command queue
      rust: add EMSGSIZE error code
      gpu: nova-core: gsp: add checking oversized commands
      gpu: nova-core: gsp: clarify invariant on command queue
      gpu: nova-core: gsp: unconditionally call variable payload handling
      gpu: nova-core: gsp: add command_size helper
      gpu: nova-core: gsp: support large RPCs via continuation record
      gpu: nova-core: gsp: add tests for SplitState

 drivers/gpu/nova-core/gsp/cmdq.rs                 | 116 +++++++---
 drivers/gpu/nova-core/gsp/commands.rs             | 258 +++++++++++++++++++++-
 drivers/gpu/nova-core/gsp/fw.rs                   |  75 ++++---
 drivers/gpu/nova-core/gsp/fw/r570_144/bindings.rs |   1 +
 rust/kernel/error.rs                              |   1 +
 5 files changed, 393 insertions(+), 58 deletions(-)
---
base-commit: 4a49fe23e357b48845e31fe9c28a802c05458198
change-id: 20260203-cmdq-continuation-b99f3d5966c3

Best regards,
-- 
Eliot Courtney <ecourtney@nvidia.com>

[PATCH v3 1/9] gpu: nova-core: gsp: sort MsgFunction variants alphabetically

[Eliot Courtney]

There is no particular order required here and keeping them alphabetical
will help preventing future mistakes.

Tested-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
---
 drivers/gpu/nova-core/gsp/fw.rs | 67 +++++++++++++++++++++--------------------
 1 file changed, 35 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/nova-core/gsp/fw.rs b/drivers/gpu/nova-core/gsp/fw.rs
index f1797e1f0d9d..4b998485360b 100644
--- a/drivers/gpu/nova-core/gsp/fw.rs
+++ b/drivers/gpu/nova-core/gsp/fw.rs
@@ -191,34 +191,34 @@ pub(crate) fn new(gsp_firmware: &GspFirmware, fb_layout: &FbLayout) -> Self {
 #[repr(u32)]
 pub(crate) enum MsgFunction {
     // Common function codes
-    Nop = bindings::NV_VGPU_MSG_FUNCTION_NOP,
-    SetGuestSystemInfo = bindings::NV_VGPU_MSG_FUNCTION_SET_GUEST_SYSTEM_INFO,
-    AllocRoot = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_ROOT,
+    AllocChannelDma = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_CHANNEL_DMA,
+    AllocCtxDma = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_CTX_DMA,
     AllocDevice = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_DEVICE,
     AllocMemory = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_MEMORY,
-    AllocCtxDma = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_CTX_DMA,
-    AllocChannelDma = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_CHANNEL_DMA,
-    MapMemory = bindings::NV_VGPU_MSG_FUNCTION_MAP_MEMORY,
-    BindCtxDma = bindings::NV_VGPU_MSG_FUNCTION_BIND_CTX_DMA,
     AllocObject = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_OBJECT,
+    AllocRoot = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_ROOT,
+    BindCtxDma = bindings::NV_VGPU_MSG_FUNCTION_BIND_CTX_DMA,
     Free = bindings::NV_VGPU_MSG_FUNCTION_FREE,
-    Log = bindings::NV_VGPU_MSG_FUNCTION_LOG,
     GetGspStaticInfo = bindings::NV_VGPU_MSG_FUNCTION_GET_GSP_STATIC_INFO,
-    SetRegistry = bindings::NV_VGPU_MSG_FUNCTION_SET_REGISTRY,
-    GspSetSystemInfo = bindings::NV_VGPU_MSG_FUNCTION_GSP_SET_SYSTEM_INFO,
+    GetStaticInfo = bindings::NV_VGPU_MSG_FUNCTION_GET_STATIC_INFO,
     GspInitPostObjGpu = bindings::NV_VGPU_MSG_FUNCTION_GSP_INIT_POST_OBJGPU,
     GspRmControl = bindings::NV_VGPU_MSG_FUNCTION_GSP_RM_CONTROL,
-    GetStaticInfo = bindings::NV_VGPU_MSG_FUNCTION_GET_STATIC_INFO,
+    GspSetSystemInfo = bindings::NV_VGPU_MSG_FUNCTION_GSP_SET_SYSTEM_INFO,
+    Log = bindings::NV_VGPU_MSG_FUNCTION_LOG,
+    MapMemory = bindings::NV_VGPU_MSG_FUNCTION_MAP_MEMORY,
+    Nop = bindings::NV_VGPU_MSG_FUNCTION_NOP,
+    SetGuestSystemInfo = bindings::NV_VGPU_MSG_FUNCTION_SET_GUEST_SYSTEM_INFO,
+    SetRegistry = bindings::NV_VGPU_MSG_FUNCTION_SET_REGISTRY,
 
     // Event codes
     GspInitDone = bindings::NV_VGPU_MSG_EVENT_GSP_INIT_DONE,
+    GspLockdownNotice = bindings::NV_VGPU_MSG_EVENT_GSP_LOCKDOWN_NOTICE,
+    GspPostNoCat = bindings::NV_VGPU_MSG_EVENT_GSP_POST_NOCAT_RECORD,
     GspRunCpuSequencer = bindings::NV_VGPU_MSG_EVENT_GSP_RUN_CPU_SEQUENCER,
-    PostEvent = bindings::NV_VGPU_MSG_EVENT_POST_EVENT,
-    RcTriggered = bindings::NV_VGPU_MSG_EVENT_RC_TRIGGERED,
     MmuFaultQueued = bindings::NV_VGPU_MSG_EVENT_MMU_FAULT_QUEUED,
     OsErrorLog = bindings::NV_VGPU_MSG_EVENT_OS_ERROR_LOG,
-    GspPostNoCat = bindings::NV_VGPU_MSG_EVENT_GSP_POST_NOCAT_RECORD,
-    GspLockdownNotice = bindings::NV_VGPU_MSG_EVENT_GSP_LOCKDOWN_NOTICE,
+    PostEvent = bindings::NV_VGPU_MSG_EVENT_POST_EVENT,
+    RcTriggered = bindings::NV_VGPU_MSG_EVENT_RC_TRIGGERED,
     UcodeLibOsPrint = bindings::NV_VGPU_MSG_EVENT_UCODE_LIBOS_PRINT,
 }
 
@@ -227,38 +227,41 @@ impl TryFrom<u32> for MsgFunction {
 
     fn try_from(value: u32) -> Result<MsgFunction> {
         match value {
-            bindings::NV_VGPU_MSG_FUNCTION_NOP => Ok(MsgFunction::Nop),
-            bindings::NV_VGPU_MSG_FUNCTION_SET_GUEST_SYSTEM_INFO => {
-                Ok(MsgFunction::SetGuestSystemInfo)
-            }
-            bindings::NV_VGPU_MSG_FUNCTION_ALLOC_ROOT => Ok(MsgFunction::AllocRoot),
+            // Common function codes
+            bindings::NV_VGPU_MSG_FUNCTION_ALLOC_CHANNEL_DMA => Ok(MsgFunction::AllocChannelDma),
+            bindings::NV_VGPU_MSG_FUNCTION_ALLOC_CTX_DMA => Ok(MsgFunction::AllocCtxDma),
             bindings::NV_VGPU_MSG_FUNCTION_ALLOC_DEVICE => Ok(MsgFunction::AllocDevice),
             bindings::NV_VGPU_MSG_FUNCTION_ALLOC_MEMORY => Ok(MsgFunction::AllocMemory),
-            bindings::NV_VGPU_MSG_FUNCTION_ALLOC_CTX_DMA => Ok(MsgFunction::AllocCtxDma),
-            bindings::NV_VGPU_MSG_FUNCTION_ALLOC_CHANNEL_DMA => Ok(MsgFunction::AllocChannelDma),
-            bindings::NV_VGPU_MSG_FUNCTION_MAP_MEMORY => Ok(MsgFunction::MapMemory),
-            bindings::NV_VGPU_MSG_FUNCTION_BIND_CTX_DMA => Ok(MsgFunction::BindCtxDma),
             bindings::NV_VGPU_MSG_FUNCTION_ALLOC_OBJECT => Ok(MsgFunction::AllocObject),
+            bindings::NV_VGPU_MSG_FUNCTION_ALLOC_ROOT => Ok(MsgFunction::AllocRoot),
+            bindings::NV_VGPU_MSG_FUNCTION_BIND_CTX_DMA => Ok(MsgFunction::BindCtxDma),
             bindings::NV_VGPU_MSG_FUNCTION_FREE => Ok(MsgFunction::Free),
-            bindings::NV_VGPU_MSG_FUNCTION_LOG => Ok(MsgFunction::Log),
             bindings::NV_VGPU_MSG_FUNCTION_GET_GSP_STATIC_INFO => Ok(MsgFunction::GetGspStaticInfo),
-            bindings::NV_VGPU_MSG_FUNCTION_SET_REGISTRY => Ok(MsgFunction::SetRegistry),
-            bindings::NV_VGPU_MSG_FUNCTION_GSP_SET_SYSTEM_INFO => Ok(MsgFunction::GspSetSystemInfo),
+            bindings::NV_VGPU_MSG_FUNCTION_GET_STATIC_INFO => Ok(MsgFunction::GetStaticInfo),
             bindings::NV_VGPU_MSG_FUNCTION_GSP_INIT_POST_OBJGPU => {
                 Ok(MsgFunction::GspInitPostObjGpu)
             }
             bindings::NV_VGPU_MSG_FUNCTION_GSP_RM_CONTROL => Ok(MsgFunction::GspRmControl),
-            bindings::NV_VGPU_MSG_FUNCTION_GET_STATIC_INFO => Ok(MsgFunction::GetStaticInfo),
+            bindings::NV_VGPU_MSG_FUNCTION_GSP_SET_SYSTEM_INFO => Ok(MsgFunction::GspSetSystemInfo),
+            bindings::NV_VGPU_MSG_FUNCTION_LOG => Ok(MsgFunction::Log),
+            bindings::NV_VGPU_MSG_FUNCTION_MAP_MEMORY => Ok(MsgFunction::MapMemory),
+            bindings::NV_VGPU_MSG_FUNCTION_NOP => Ok(MsgFunction::Nop),
+            bindings::NV_VGPU_MSG_FUNCTION_SET_GUEST_SYSTEM_INFO => {
+                Ok(MsgFunction::SetGuestSystemInfo)
+            }
+            bindings::NV_VGPU_MSG_FUNCTION_SET_REGISTRY => Ok(MsgFunction::SetRegistry),
+
+            // Event codes
             bindings::NV_VGPU_MSG_EVENT_GSP_INIT_DONE => Ok(MsgFunction::GspInitDone),
+            bindings::NV_VGPU_MSG_EVENT_GSP_LOCKDOWN_NOTICE => Ok(MsgFunction::GspLockdownNotice),
+            bindings::NV_VGPU_MSG_EVENT_GSP_POST_NOCAT_RECORD => Ok(MsgFunction::GspPostNoCat),
             bindings::NV_VGPU_MSG_EVENT_GSP_RUN_CPU_SEQUENCER => {
                 Ok(MsgFunction::GspRunCpuSequencer)
             }
-            bindings::NV_VGPU_MSG_EVENT_POST_EVENT => Ok(MsgFunction::PostEvent),
-            bindings::NV_VGPU_MSG_EVENT_RC_TRIGGERED => Ok(MsgFunction::RcTriggered),
             bindings::NV_VGPU_MSG_EVENT_MMU_FAULT_QUEUED => Ok(MsgFunction::MmuFaultQueued),
             bindings::NV_VGPU_MSG_EVENT_OS_ERROR_LOG => Ok(MsgFunction::OsErrorLog),
-            bindings::NV_VGPU_MSG_EVENT_GSP_POST_NOCAT_RECORD => Ok(MsgFunction::GspPostNoCat),
-            bindings::NV_VGPU_MSG_EVENT_GSP_LOCKDOWN_NOTICE => Ok(MsgFunction::GspLockdownNotice),
+            bindings::NV_VGPU_MSG_EVENT_POST_EVENT => Ok(MsgFunction::PostEvent),
+            bindings::NV_VGPU_MSG_EVENT_RC_TRIGGERED => Ok(MsgFunction::RcTriggered),
             bindings::NV_VGPU_MSG_EVENT_UCODE_LIBOS_PRINT => Ok(MsgFunction::UcodeLibOsPrint),
             _ => Err(EINVAL),
         }

-- 
2.53.0

[PATCH v3 2/9] gpu: nova-core: gsp: add mechanism to wait for space on command queue

[Eliot Courtney]

Add a timeout to `allocate_command` which waits for space on the GSP
command queue. It uses a similar timeout to nouveau.

This lets `send_command` wait for space to free up in the command queue.
This is required to support continuation records which can fill up the
queue.

Tested-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
---
 drivers/gpu/nova-core/gsp/cmdq.rs | 39 +++++++++++++++++++++++++++------------
 1 file changed, 27 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/nova-core/gsp/cmdq.rs b/drivers/gpu/nova-core/gsp/cmdq.rs
index 87dbbd6d1be9..efbbc89f4d8a 100644
--- a/drivers/gpu/nova-core/gsp/cmdq.rs
+++ b/drivers/gpu/nova-core/gsp/cmdq.rs
@@ -250,6 +250,19 @@ fn new(dev: &device::Device<device::Bound>) -> Result<Self> {
         }
     }
 
+    /// Returns the size of the region of the CPU message queue that the driver is currently allowed
+    /// to write to, in bytes.
+    fn driver_write_area_size(&self) -> usize {
+        let tx = self.cpu_write_ptr();
+        let rx = self.gsp_read_ptr();
+
+        // `rx` and `tx` are both in `0..MSGQ_NUM_PAGES` per the invariants of `gsp_read_ptr` and
+        // `cpu_write_ptr`. The minimum value case is where `rx == 0` and `tx == MSGQ_NUM_PAGES -
+        // 1`, which gives `0 + MSGQ_NUM_PAGES - (MSGQ_NUM_PAGES - 1) - 1 == 0`.
+        let slots = (rx + MSGQ_NUM_PAGES - tx - 1) % MSGQ_NUM_PAGES;
+        num::u32_as_usize(slots) * GSP_PAGE_SIZE
+    }
+
     /// Returns the region of the GSP message queue that the driver is currently allowed to read
     /// from.
     ///
@@ -281,15 +294,22 @@ fn new(dev: &device::Device<device::Bound>) -> Result<Self> {
     }
 
     /// Allocates a region on the command queue that is large enough to send a command of `size`
-    /// bytes.
+    /// bytes, waiting for space to become available based on the provided timeout.
     ///
     /// This returns a [`GspCommand`] ready to be written to by the caller.
     ///
     /// # Errors
     ///
-    /// - `EAGAIN` if the driver area is too small to hold the requested command.
+    /// - `ETIMEDOUT` if space does not become available within the timeout.
     /// - `EIO` if the command header is not properly aligned.
-    fn allocate_command(&mut self, size: usize) -> Result<GspCommand<'_>> {
+    fn allocate_command(&mut self, size: usize, timeout: Delta) -> Result<GspCommand<'_>> {
+        read_poll_timeout(
+            || Ok(self.driver_write_area_size()),
+            |available_bytes| *available_bytes >= size_of::<GspMsgElement>() + size,
+            Delta::ZERO,
+            timeout,
+        )?;
+
         // Get the current writable area as an array of bytes.
         let (slice_1, slice_2) = {
             let (slice_1, slice_2) = self.driver_write_area();
@@ -298,13 +318,6 @@ fn allocate_command(&mut self, size: usize) -> Result<GspCommand<'_>> {
             (slice_1.as_flattened_mut(), slice_2.as_flattened_mut())
         };
 
-        // If the GSP is still processing previous messages the shared region
-        // may be full in which case we will have to retry once the GSP has
-        // processed the existing commands.
-        if size_of::<GspMsgElement>() + size > slice_1.len() + slice_2.len() {
-            return Err(EAGAIN);
-        }
-
         // Extract area for the `GspMsgElement`.
         let (header, slice_1) = GspMsgElement::from_bytes_mut_prefix(slice_1).ok_or(EIO)?;
 
@@ -497,7 +510,7 @@ fn notify_gsp(bar: &Bar0) {
     ///
     /// # Errors
     ///
-    /// - `EAGAIN` if there was not enough space in the command queue to send the command.
+    /// - `ETIMEDOUT` if space does not become available within the timeout.
     /// - `EIO` if the variable payload requested by the command has not been entirely
     ///   written to by its [`CommandToGsp::init_variable_payload`] method.
     ///
@@ -509,7 +522,9 @@ pub(crate) fn send_command<M>(&mut self, bar: &Bar0, command: M) -> Result
         Error: From<M::InitError>,
     {
         let command_size = size_of::<M::Command>() + command.variable_payload_len();
-        let dst = self.gsp_mem.allocate_command(command_size)?;
+        let dst = self
+            .gsp_mem
+            .allocate_command(command_size, Delta::from_secs(1))?;
 
         // Extract area for the command itself.
         let (cmd, payload_1) = M::Command::from_bytes_mut_prefix(dst.contents.0).ok_or(EIO)?;

-- 
2.53.0

[PATCH v3 3/9] rust: add EMSGSIZE error code

[Eliot Courtney]

Add the EMSGSIZE error code, which indicates that a message is too
long.

Tested-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
---
 rust/kernel/error.rs | 1 +
 1 file changed, 1 insertion(+)

diff --git a/rust/kernel/error.rs b/rust/kernel/error.rs
index 258b12afdcba..10fcf1f0404d 100644
--- a/rust/kernel/error.rs
+++ b/rust/kernel/error.rs
@@ -67,6 +67,7 @@ macro_rules! declare_err {
     declare_err!(EDOM, "Math argument out of domain of func.");
     declare_err!(ERANGE, "Math result not representable.");
     declare_err!(EOVERFLOW, "Value too large for defined data type.");
+    declare_err!(EMSGSIZE, "Message too long.");
     declare_err!(ETIMEDOUT, "Connection timed out.");
     declare_err!(ERESTARTSYS, "Restart the system call.");
     declare_err!(ERESTARTNOINTR, "System call was interrupted by a signal and will be restarted.");

-- 
2.53.0

[PATCH v3 4/9] gpu: nova-core: gsp: add checking oversized commands

[Eliot Courtney]

The limit is 16 pages for a single command sent to the GSP. Return an
error if `allocate_command` is called with a too large size.

Tested-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
---
 drivers/gpu/nova-core/gsp/cmdq.rs                 | 7 ++++++-
 drivers/gpu/nova-core/gsp/fw.rs                   | 4 ++++
 drivers/gpu/nova-core/gsp/fw/r570_144/bindings.rs | 1 +
 3 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/nova-core/gsp/cmdq.rs b/drivers/gpu/nova-core/gsp/cmdq.rs
index efbbc89f4d8a..3c2652be74bf 100644
--- a/drivers/gpu/nova-core/gsp/cmdq.rs
+++ b/drivers/gpu/nova-core/gsp/cmdq.rs
@@ -32,7 +32,8 @@
             GspMsgElement,
             MsgFunction,
             MsgqRxHeader,
-            MsgqTxHeader, //
+            MsgqTxHeader,
+            GSP_MSG_QUEUE_ELEMENT_SIZE_MAX, //
         },
         PteArray,
         GSP_PAGE_SHIFT,
@@ -300,9 +301,13 @@ fn driver_write_area_size(&self) -> usize {
     ///
     /// # Errors
     ///
+    /// - `EMSGSIZE` if the command is larger than [`GSP_MSG_QUEUE_ELEMENT_SIZE_MAX`].
     /// - `ETIMEDOUT` if space does not become available within the timeout.
     /// - `EIO` if the command header is not properly aligned.
     fn allocate_command(&mut self, size: usize, timeout: Delta) -> Result<GspCommand<'_>> {
+        if size_of::<GspMsgElement>() + size > GSP_MSG_QUEUE_ELEMENT_SIZE_MAX {
+            return Err(EMSGSIZE);
+        }
         read_poll_timeout(
             || Ok(self.driver_write_area_size()),
             |available_bytes| *available_bytes >= size_of::<GspMsgElement>() + size,
diff --git a/drivers/gpu/nova-core/gsp/fw.rs b/drivers/gpu/nova-core/gsp/fw.rs
index 4b998485360b..6005362450cb 100644
--- a/drivers/gpu/nova-core/gsp/fw.rs
+++ b/drivers/gpu/nova-core/gsp/fw.rs
@@ -39,6 +39,10 @@
     },
 };
 
+/// Maximum size of a single GSP message queue element in bytes.
+pub(crate) const GSP_MSG_QUEUE_ELEMENT_SIZE_MAX: usize =
+    num::u32_as_usize(bindings::GSP_MSG_QUEUE_ELEMENT_SIZE_MAX);
+
 /// Empty type to group methods related to heap parameters for running the GSP firmware.
 enum GspFwHeapParams {}
 
diff --git a/drivers/gpu/nova-core/gsp/fw/r570_144/bindings.rs b/drivers/gpu/nova-core/gsp/fw/r570_144/bindings.rs
index 6d25fe0bffa9..334e8be5fde8 100644
--- a/drivers/gpu/nova-core/gsp/fw/r570_144/bindings.rs
+++ b/drivers/gpu/nova-core/gsp/fw/r570_144/bindings.rs
@@ -43,6 +43,7 @@ fn fmt(&self, fmt: &mut ::core::fmt::Formatter<'_>) -> ::core::fmt::Result {
 pub const GSP_FW_WPR_META_REVISION: u32 = 1;
 pub const GSP_FW_WPR_META_MAGIC: i64 = -2577556379034558285;
 pub const REGISTRY_TABLE_ENTRY_TYPE_DWORD: u32 = 1;
+pub const GSP_MSG_QUEUE_ELEMENT_SIZE_MAX: u32 = 65536;
 pub type __u8 = ffi::c_uchar;
 pub type __u16 = ffi::c_ushort;
 pub type __u32 = ffi::c_uint;

-- 
2.53.0

[PATCH v3 5/9] gpu: nova-core: gsp: clarify invariant on command queue

[Eliot Courtney]

Clarify why using only the first returned slice from allocate_command
for the message headers is okay.

Tested-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
---
 drivers/gpu/nova-core/gsp/cmdq.rs | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/nova-core/gsp/cmdq.rs b/drivers/gpu/nova-core/gsp/cmdq.rs
index 3c2652be74bf..3653212efa7a 100644
--- a/drivers/gpu/nova-core/gsp/cmdq.rs
+++ b/drivers/gpu/nova-core/gsp/cmdq.rs
@@ -531,7 +531,9 @@ pub(crate) fn send_command<M>(&mut self, bar: &Bar0, command: M) -> Result
             .gsp_mem
             .allocate_command(command_size, Delta::from_secs(1))?;
 
-        // Extract area for the command itself.
+        // Extract area for the command itself. The GSP message header and the command header
+        // together are guaranteed to fit entirely into a single page, so it's ok to only look
+        // at `dst.contents.0` here.
         let (cmd, payload_1) = M::Command::from_bytes_mut_prefix(dst.contents.0).ok_or(EIO)?;
 
         // Fill the header and command in-place.

-- 
2.53.0

[PATCH v3 6/9] gpu: nova-core: gsp: unconditionally call variable payload handling

[Eliot Courtney]

Unconditionally call the variable length payload code, which is a no-op
if there is no such payload but could defensively catch some coding
errors by e.g. checking that the allocated size is completely filled.

Tested-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
---
 drivers/gpu/nova-core/gsp/cmdq.rs | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/nova-core/gsp/cmdq.rs b/drivers/gpu/nova-core/gsp/cmdq.rs
index 3653212efa7a..9f74f0898d90 100644
--- a/drivers/gpu/nova-core/gsp/cmdq.rs
+++ b/drivers/gpu/nova-core/gsp/cmdq.rs
@@ -545,16 +545,14 @@ pub(crate) fn send_command<M>(&mut self, bar: &Bar0, command: M) -> Result
             command.init().__init(core::ptr::from_mut(cmd))?;
         }
 
-        // Fill the variable-length payload.
-        if command_size > size_of::<M::Command>() {
-            let mut sbuffer =
-                SBufferIter::new_writer([&mut payload_1[..], &mut dst.contents.1[..]]);
-            command.init_variable_payload(&mut sbuffer)?;
+        // Fill the variable-length payload, which may be empty.
+        let mut sbuffer = SBufferIter::new_writer([&mut payload_1[..], &mut dst.contents.1[..]]);
+        command.init_variable_payload(&mut sbuffer)?;
 
-            if !sbuffer.is_empty() {
-                return Err(EIO);
-            }
+        if !sbuffer.is_empty() {
+            return Err(EIO);
         }
+        drop(sbuffer);
 
         // Compute checksum now that the whole message is ready.
         dst.header

-- 
2.53.0

[PATCH v3 7/9] gpu: nova-core: gsp: add command_size helper

[Eliot Courtney]

Add a helper function which computes the size of a command.

Tested-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
---
 drivers/gpu/nova-core/gsp/cmdq.rs | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/nova-core/gsp/cmdq.rs b/drivers/gpu/nova-core/gsp/cmdq.rs
index 9f74f0898d90..4a663a5b3437 100644
--- a/drivers/gpu/nova-core/gsp/cmdq.rs
+++ b/drivers/gpu/nova-core/gsp/cmdq.rs
@@ -450,6 +450,15 @@ struct GspMessage<'a> {
     contents: (&'a [u8], &'a [u8]),
 }
 
+/// Computes the total size of the command (including its variable-length payload) without the
+/// [`GspMsgElement`] header.
+fn command_size<M>(command: &M) -> usize
+where
+    M: CommandToGsp,
+{
+    size_of::<M::Command>() + command.variable_payload_len()
+}
+
 /// GSP command queue.
 ///
 /// Provides the ability to send commands and receive messages from the GSP using a shared memory
@@ -526,7 +535,7 @@ pub(crate) fn send_command<M>(&mut self, bar: &Bar0, command: M) -> Result
         // This allows all error types, including `Infallible`, to be used for `M::InitError`.
         Error: From<M::InitError>,
     {
-        let command_size = size_of::<M::Command>() + command.variable_payload_len();
+        let command_size = command_size(&command);
         let dst = self
             .gsp_mem
             .allocate_command(command_size, Delta::from_secs(1))?;

-- 
2.53.0

[PATCH v3 8/9] gpu: nova-core: gsp: support large RPCs via continuation record

[Eliot Courtney]

Splits large RPCs if necessary and sends the remaining parts using
continuation records. RPCs that do not need continuation records
continue to write directly into the command buffer. Ones that do write
into a staging buffer first, so there is one copy.

Continuation record for receive is not necessary to support at the
moment because those replies do not need to be read and are currently
drained by retrying `receive_msg` on ERANGE.

Tested-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
---
 drivers/gpu/nova-core/gsp/cmdq.rs     |  43 +++++++++-
 drivers/gpu/nova-core/gsp/commands.rs | 144 +++++++++++++++++++++++++++++++++-
 drivers/gpu/nova-core/gsp/fw.rs       |   4 +
 3 files changed, 187 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/nova-core/gsp/cmdq.rs b/drivers/gpu/nova-core/gsp/cmdq.rs
index 4a663a5b3437..d68b93ddf7cc 100644
--- a/drivers/gpu/nova-core/gsp/cmdq.rs
+++ b/drivers/gpu/nova-core/gsp/cmdq.rs
@@ -28,6 +28,10 @@
 use crate::{
     driver::Bar0,
     gsp::{
+        commands::{
+            ContinuationRecord,
+            SplitState, //
+        },
         fw::{
             GspMsgElement,
             MsgFunction,
@@ -452,7 +456,7 @@ struct GspMessage<'a> {
 
 /// Computes the total size of the command (including its variable-length payload) without the
 /// [`GspMsgElement`] header.
-fn command_size<M>(command: &M) -> usize
+pub(crate) fn command_size<M>(command: &M) -> usize
 where
     M: CommandToGsp,
 {
@@ -520,7 +524,7 @@ fn notify_gsp(bar: &Bar0) {
             .write(bar);
     }
 
-    /// Sends `command` to the GSP.
+    /// Sends `command` to the GSP, without splitting it.
     ///
     /// # Errors
     ///
@@ -529,7 +533,7 @@ fn notify_gsp(bar: &Bar0) {
     ///   written to by its [`CommandToGsp::init_variable_payload`] method.
     ///
     /// Error codes returned by the command initializers are propagated as-is.
-    pub(crate) fn send_command<M>(&mut self, bar: &Bar0, command: M) -> Result
+    fn send_single_command<M>(&mut self, bar: &Bar0, command: M) -> Result
     where
         M: CommandToGsp,
         // This allows all error types, including `Infallible`, to be used for `M::InitError`.
@@ -588,6 +592,39 @@ pub(crate) fn send_command<M>(&mut self, bar: &Bar0, command: M) -> Result
         Ok(())
     }
 
+    /// Sends `command` to the GSP.
+    ///
+    /// The command may be split into multiple messages if it is large.
+    ///
+    /// # Errors
+    ///
+    /// - `ETIMEDOUT` if space does not become available within the timeout.
+    /// - `EIO` if the variable payload requested by the command has not been entirely
+    ///   written to by its [`CommandToGsp::init_variable_payload`] method.
+    ///
+    /// Error codes returned by the command initializers are propagated as-is.
+    pub(crate) fn send_command<M>(&mut self, bar: &Bar0, command: M) -> Result
+    where
+        M: CommandToGsp,
+        Error: From<M::InitError>,
+    {
+        let mut state = SplitState::new(&command)?;
+
+        self.send_single_command(bar, state.command(command))?;
+
+        while let Some(continuation) = state.next_continuation_record() {
+            dev_dbg!(
+                &self.dev,
+                "GSP RPC: send continuation: size=0x{:x}\n",
+                command_size(&continuation),
+            );
+            // Turbofish needed because the compiler cannot infer M here.
+            self.send_single_command::<ContinuationRecord<'_>>(bar, continuation)?;
+        }
+
+        Ok(())
+    }
+
     /// Wait for a message to become available on the message queue.
     ///
     /// This works purely at the transport layer and does not interpret or validate the message
diff --git a/drivers/gpu/nova-core/gsp/commands.rs b/drivers/gpu/nova-core/gsp/commands.rs
index 8f270eca33be..6ffd0b9cbf05 100644
--- a/drivers/gpu/nova-core/gsp/commands.rs
+++ b/drivers/gpu/nova-core/gsp/commands.rs
@@ -4,6 +4,7 @@
     array,
     convert::Infallible,
     ffi::FromBytesUntilNulError,
+    marker::PhantomData,
     str::Utf8Error, //
 };
 
@@ -22,13 +23,16 @@
     driver::Bar0,
     gsp::{
         cmdq::{
+            command_size,
             Cmdq,
             CommandToGsp,
             MessageFromGsp, //
         },
         fw::{
             commands::*,
-            MsgFunction, //
+            GspMsgElement,
+            MsgFunction,
+            GSP_MSG_QUEUE_ELEMENT_SIZE_MAX, //
         },
     },
     sbuffer::SBufferIter,
@@ -242,3 +246,141 @@ pub(crate) fn get_gsp_info(cmdq: &mut Cmdq, bar: &Bar0) -> Result<GetGspStaticIn
         }
     }
 }
+
+/// The `ContinuationRecord` command.
+pub(crate) struct ContinuationRecord<'a> {
+    data: &'a [u8],
+}
+
+impl<'a> ContinuationRecord<'a> {
+    /// Creates a new `ContinuationRecord` command with the given data.
+    pub(crate) fn new(data: &'a [u8]) -> Self {
+        Self { data }
+    }
+}
+
+impl<'a> CommandToGsp for ContinuationRecord<'a> {
+    const FUNCTION: MsgFunction = MsgFunction::ContinuationRecord;
+    type Command = ();
+    type InitError = Infallible;
+
+    fn init(&self) -> impl Init<Self::Command, Self::InitError> {
+        <()>::init_zeroed()
+    }
+
+    fn variable_payload_len(&self) -> usize {
+        self.data.len()
+    }
+
+    fn init_variable_payload(
+        &self,
+        dst: &mut SBufferIter<core::array::IntoIter<&mut [u8], 2>>,
+    ) -> Result {
+        dst.write_all(self.data)
+    }
+}
+
+/// Wrapper that splits a command across continuation records if needed.
+pub(crate) struct SplitState<C: CommandToGsp> {
+    state: Option<(KVVec<u8>, usize)>,
+    _phantom: PhantomData<C>,
+}
+
+impl<C: CommandToGsp> SplitState<C> {
+    /// Maximum command size that fits in a single queue element.
+    const MAX_CMD_SIZE: usize = GSP_MSG_QUEUE_ELEMENT_SIZE_MAX - size_of::<GspMsgElement>();
+
+    /// Maximum size of the variable payload that can be sent in the main command.
+    const MAX_FIRST_PAYLOAD_SIZE: usize = Self::MAX_CMD_SIZE - size_of::<C::Command>();
+
+    /// Creates a new `SplitState` for the given command.
+    ///
+    /// If the command is too large, it will be split into a main command and some number of
+    /// continuation records.
+    pub(crate) fn new(inner: &C) -> Result<Self> {
+        if command_size(inner) > Self::MAX_CMD_SIZE {
+            let mut staging =
+                KVVec::<u8>::from_elem(0u8, inner.variable_payload_len(), GFP_KERNEL)?;
+            let mut sbuffer = SBufferIter::new_writer([staging.as_mut_slice(), &mut []]);
+            inner.init_variable_payload(&mut sbuffer)?;
+            if !sbuffer.is_empty() {
+                return Err(EIO);
+            }
+            drop(sbuffer);
+
+            Ok(Self {
+                state: Some((staging, Self::MAX_FIRST_PAYLOAD_SIZE)),
+                _phantom: PhantomData,
+            })
+        } else {
+            Ok(Self {
+                state: None,
+                _phantom: PhantomData,
+            })
+        }
+    }
+
+    /// Returns the main command.
+    pub(crate) fn command(&self, inner: C) -> SplitCommand<'_, C> {
+        if let Some((staging, _)) = &self.state {
+            SplitCommand::Split(inner, staging)
+        } else {
+            SplitCommand::Single(inner)
+        }
+    }
+
+    /// Returns the next continuation record, or `None` if there are no more.
+    pub(crate) fn next_continuation_record(&mut self) -> Option<ContinuationRecord<'_>> {
+        let (staging, offset) = self.state.as_mut()?;
+
+        let remaining = staging.len() - *offset;
+        if remaining > 0 {
+            let chunk_size = remaining.min(Self::MAX_CMD_SIZE);
+            let record = ContinuationRecord::new(&staging[*offset..(*offset + chunk_size)]);
+            *offset += chunk_size;
+            Some(record)
+        } else {
+            None
+        }
+    }
+}
+
+/// Wrapper enum that represents either a single command or a split command with its staging buffer.
+pub(crate) enum SplitCommand<'a, C: CommandToGsp> {
+    /// A command that fits in a single queue element.
+    Single(C),
+    /// A command split across continuation records, with its full payload in a staging buffer.
+    Split(C, &'a [u8]),
+}
+
+impl<'a, C: CommandToGsp> CommandToGsp for SplitCommand<'a, C> {
+    const FUNCTION: MsgFunction = C::FUNCTION;
+    type Command = C::Command;
+    type InitError = C::InitError;
+
+    fn init(&self) -> impl Init<Self::Command, Self::InitError> {
+        match self {
+            SplitCommand::Single(cmd) => cmd.init(),
+            SplitCommand::Split(cmd, _) => cmd.init(),
+        }
+    }
+
+    fn variable_payload_len(&self) -> usize {
+        match self {
+            SplitCommand::Single(cmd) => cmd.variable_payload_len(),
+            SplitCommand::Split(_, _) => SplitState::<C>::MAX_FIRST_PAYLOAD_SIZE,
+        }
+    }
+
+    fn init_variable_payload(
+        &self,
+        dst: &mut SBufferIter<core::array::IntoIter<&mut [u8], 2>>,
+    ) -> Result {
+        match self {
+            SplitCommand::Single(cmd) => cmd.init_variable_payload(dst),
+            SplitCommand::Split(_, staging) => {
+                dst.write_all(&staging[..self.variable_payload_len()])
+            }
+        }
+    }
+}
diff --git a/drivers/gpu/nova-core/gsp/fw.rs b/drivers/gpu/nova-core/gsp/fw.rs
index 6005362450cb..25fca1f6db2c 100644
--- a/drivers/gpu/nova-core/gsp/fw.rs
+++ b/drivers/gpu/nova-core/gsp/fw.rs
@@ -202,6 +202,7 @@ pub(crate) enum MsgFunction {
     AllocObject = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_OBJECT,
     AllocRoot = bindings::NV_VGPU_MSG_FUNCTION_ALLOC_ROOT,
     BindCtxDma = bindings::NV_VGPU_MSG_FUNCTION_BIND_CTX_DMA,
+    ContinuationRecord = bindings::NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD,
     Free = bindings::NV_VGPU_MSG_FUNCTION_FREE,
     GetGspStaticInfo = bindings::NV_VGPU_MSG_FUNCTION_GET_GSP_STATIC_INFO,
     GetStaticInfo = bindings::NV_VGPU_MSG_FUNCTION_GET_STATIC_INFO,
@@ -239,6 +240,9 @@ fn try_from(value: u32) -> Result<MsgFunction> {
             bindings::NV_VGPU_MSG_FUNCTION_ALLOC_OBJECT => Ok(MsgFunction::AllocObject),
             bindings::NV_VGPU_MSG_FUNCTION_ALLOC_ROOT => Ok(MsgFunction::AllocRoot),
             bindings::NV_VGPU_MSG_FUNCTION_BIND_CTX_DMA => Ok(MsgFunction::BindCtxDma),
+            bindings::NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD => {
+                Ok(MsgFunction::ContinuationRecord)
+            }
             bindings::NV_VGPU_MSG_FUNCTION_FREE => Ok(MsgFunction::Free),
             bindings::NV_VGPU_MSG_FUNCTION_GET_GSP_STATIC_INFO => Ok(MsgFunction::GetGspStaticInfo),
             bindings::NV_VGPU_MSG_FUNCTION_GET_STATIC_INFO => Ok(MsgFunction::GetStaticInfo),

-- 
2.53.0

[PATCH v3 9/9] gpu: nova-core: gsp: add tests for SplitState

[Eliot Courtney]

Add tests for SplitState. They cover boundary conditions at the split
points to make sure the right number of continuation records are made.
They also check that the data concatenated is correct.

Tested-by: Zhi Wang <zhiw@nvidia.com>
Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
---
 drivers/gpu/nova-core/gsp/commands.rs | 114 ++++++++++++++++++++++++++++++++++
 1 file changed, 114 insertions(+)

diff --git a/drivers/gpu/nova-core/gsp/commands.rs b/drivers/gpu/nova-core/gsp/commands.rs
index 6ffd0b9cbf05..74f875755e53 100644
--- a/drivers/gpu/nova-core/gsp/commands.rs
+++ b/drivers/gpu/nova-core/gsp/commands.rs
@@ -384,3 +384,117 @@ fn init_variable_payload(
         }
     }
 }
+
+#[kunit_tests(nova_core_gsp_commands)]
+mod tests {
+    use super::*;
+
+    struct TestPayload {
+        data: KVVec<u8>,
+    }
+
+    impl TestPayload {
+        fn generate_pattern(len: usize) -> Result<KVVec<u8>> {
+            let mut data = KVVec::with_capacity(len, GFP_KERNEL)?;
+            for i in 0..len {
+                data.push(i as u8, GFP_KERNEL)?;
+            }
+            Ok(data)
+        }
+
+        fn new(len: usize) -> Result<Self> {
+            Ok(Self {
+                data: Self::generate_pattern(len)?,
+            })
+        }
+    }
+
+    impl CommandToGsp for TestPayload {
+        const FUNCTION: MsgFunction = MsgFunction::Nop;
+        type Command = ();
+        type InitError = Infallible;
+
+        fn init(&self) -> impl Init<Self::Command, Self::InitError> {
+            <()>::init_zeroed()
+        }
+
+        fn variable_payload_len(&self) -> usize {
+            self.data.len()
+        }
+
+        fn init_variable_payload(
+            &self,
+            dst: &mut SBufferIter<core::array::IntoIter<&mut [u8], 2>>,
+        ) -> Result {
+            dst.write_all(self.data.as_slice())
+        }
+    }
+
+    const MAX_CMD_SIZE: usize = SplitState::<TestPayload>::MAX_CMD_SIZE;
+
+    fn read_payload(cmd: &impl CommandToGsp) -> Result<KVVec<u8>> {
+        let len = cmd.variable_payload_len();
+        let mut buf = KVVec::from_elem(0u8, len, GFP_KERNEL)?;
+        let mut sbuf = SBufferIter::new_writer([buf.as_mut_slice(), &mut []]);
+        cmd.init_variable_payload(&mut sbuf)?;
+        drop(sbuf);
+        Ok(buf)
+    }
+
+    struct SplitTest {
+        payload_size: usize,
+        num_continuations: usize,
+    }
+
+    fn check_split(t: SplitTest) -> Result {
+        let payload = TestPayload::new(t.payload_size)?;
+        let mut state = SplitState::new(&payload)?;
+
+        let mut buf = read_payload(&state.command(payload))?;
+        assert!(buf.len() <= MAX_CMD_SIZE);
+
+        let mut num_continuations = 0;
+        while let Some(cont) = state.next_continuation_record() {
+            let payload = read_payload(&cont)?;
+            assert!(payload.len() <= MAX_CMD_SIZE);
+            buf.extend_from_slice(&payload, GFP_KERNEL)?;
+            num_continuations += 1;
+        }
+
+        assert_eq!(num_continuations, t.num_continuations);
+        assert_eq!(
+            buf.as_slice(),
+            TestPayload::generate_pattern(t.payload_size)?.as_slice()
+        );
+        Ok(())
+    }
+
+    #[test]
+    fn split_command() -> Result {
+        check_split(SplitTest {
+            payload_size: 0,
+            num_continuations: 0,
+        })?;
+        check_split(SplitTest {
+            payload_size: MAX_CMD_SIZE,
+            num_continuations: 0,
+        })?;
+        check_split(SplitTest {
+            payload_size: MAX_CMD_SIZE + 1,
+            num_continuations: 1,
+        })?;
+        check_split(SplitTest {
+            payload_size: MAX_CMD_SIZE * 2,
+            num_continuations: 1,
+        })?;
+        check_split(SplitTest {
+            payload_size: MAX_CMD_SIZE * 2 + 1,
+            num_continuations: 2,
+        })?;
+        check_split(SplitTest {
+            payload_size: MAX_CMD_SIZE * 3 + MAX_CMD_SIZE / 2,
+            num_continuations: 3,
+        })?;
+        Ok(())
+    }
+}

-- 
2.53.0

Re: [PATCH v3 3/9] rust: add EMSGSIZE error code

[Miguel Ojeda]

On Thu, Feb 26, 2026 at 12:46 PM Eliot Courtney <ecourtney@nvidia.com> wrote:
>
> Add the EMSGSIZE error code, which indicates that a message is too
> long.
>
> Tested-by: Zhi Wang <zhiw@nvidia.com>
> Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>

Acked-by: Miguel Ojeda <ojeda@kernel.org>

I checked that it is placed in the right spot in the list :)

Thanks!

Cheers,
Miguel

Re: [PATCH v3 8/9] gpu: nova-core: gsp: support large RPCs via continuation record

[Alexandre Courbot]

Hi Eliot,

On Thu Feb 26, 2026 at 8:45 PM JST, Eliot Courtney wrote:
<snip>
> @@ -588,6 +592,39 @@ pub(crate) fn send_command<M>(&mut self, bar: &Bar0, command: M) -> Result
>          Ok(())
>      }
>  
> +    /// Sends `command` to the GSP.
> +    ///
> +    /// The command may be split into multiple messages if it is large.
> +    ///
> +    /// # Errors
> +    ///
> +    /// - `ETIMEDOUT` if space does not become available within the timeout.
> +    /// - `EIO` if the variable payload requested by the command has not been entirely
> +    ///   written to by its [`CommandToGsp::init_variable_payload`] method.
> +    ///
> +    /// Error codes returned by the command initializers are propagated as-is.
> +    pub(crate) fn send_command<M>(&mut self, bar: &Bar0, command: M) -> Result
> +    where
> +        M: CommandToGsp,
> +        Error: From<M::InitError>,
> +    {
> +        let mut state = SplitState::new(&command)?;
> +
> +        self.send_single_command(bar, state.command(command))?;

As we discussed offline these two lines were bothering me a bit, because
we pass `command` twice and there is a possibility (although purely
hypothetical) of API misuse. I think I found a way around it: we turn
`SplitState` into an enum with a `Single` variant, that contains the
original command, and a `Split` one, which contains the truncated
initial command and its continuation records.

This simple change cascades into more simplifications in the new types
introduced by this patch: `SplitCommand` is not an enum anymore, but the
original command with a shorter payload. And the continuation records
are their own self-contained type. This is achieved by allocating two
`KVVec`s when we split, and taking advantage of the `SBuffer` we
conveniently have at hand to fill them both.

I have pushed a branch with this idea implemented on top of yours in [1]
- please check it, but I think it brings a nice (if small) incremental
improvement. It removes the API misuse potential, makes every type
purely single-purpose, and doesn't need `PhantomData` or lifetimes,
making it simpler overall IMHO.

[1] https://github.com/Gnurou/linux/tree/review/continuations

> +
> +        while let Some(continuation) = state.next_continuation_record() {
> +            dev_dbg!(
> +                &self.dev,
> +                "GSP RPC: send continuation: size=0x{:x}\n",
> +                command_size(&continuation),
> +            );

`send_single_command` should already print the command, so I don't think
we need this additional debug print?

> +            // Turbofish needed because the compiler cannot infer M here.
> +            self.send_single_command::<ContinuationRecord<'_>>(bar, continuation)?;
> +        }
> +
> +        Ok(())
> +    }
> +
>      /// Wait for a message to become available on the message queue.
>      ///
>      /// This works purely at the transport layer and does not interpret or validate the message
> diff --git a/drivers/gpu/nova-core/gsp/commands.rs b/drivers/gpu/nova-core/gsp/commands.rs
> index 8f270eca33be..6ffd0b9cbf05 100644
> --- a/drivers/gpu/nova-core/gsp/commands.rs
> +++ b/drivers/gpu/nova-core/gsp/commands.rs
> @@ -4,6 +4,7 @@
>      array,
>      convert::Infallible,
>      ffi::FromBytesUntilNulError,
> +    marker::PhantomData,
>      str::Utf8Error, //
>  };
>  
> @@ -22,13 +23,16 @@
>      driver::Bar0,
>      gsp::{
>          cmdq::{
> +            command_size,
>              Cmdq,
>              CommandToGsp,
>              MessageFromGsp, //
>          },
>          fw::{
>              commands::*,
> -            MsgFunction, //
> +            GspMsgElement,
> +            MsgFunction,
> +            GSP_MSG_QUEUE_ELEMENT_SIZE_MAX, //
>          },
>      },
>      sbuffer::SBufferIter,
> @@ -242,3 +246,141 @@ pub(crate) fn get_gsp_info(cmdq: &mut Cmdq, bar: &Bar0) -> Result<GetGspStaticIn
>          }
>      }
>  }
> +
> +/// The `ContinuationRecord` command.
> +pub(crate) struct ContinuationRecord<'a> {

These new types are to be used by the command queue, and we don't want
to use them elsewhere, so `pub(super)` seems more appropriate to me.

And actually, since they are more command queue infrastructure than pure
"commands" that fulfill a given purpose, I'd suggest moving them into
their own sub-module of `gsp` (named `continuation` or `split` maybe?).
That way they don't get in the way of readers who just want to learn
about GSP commands.

Re: [PATCH v3 9/9] gpu: nova-core: gsp: add tests for SplitState

[Alexandre Courbot]

On Thu Feb 26, 2026 at 8:45 PM JST, Eliot Courtney wrote:
> Add tests for SplitState. They cover boundary conditions at the split
> points to make sure the right number of continuation records are made.
> They also check that the data concatenated is correct.
>
> Tested-by: Zhi Wang <zhiw@nvidia.com>
> Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>
> ---
>  drivers/gpu/nova-core/gsp/commands.rs | 114 ++++++++++++++++++++++++++++++++++
>  1 file changed, 114 insertions(+)
>
> diff --git a/drivers/gpu/nova-core/gsp/commands.rs b/drivers/gpu/nova-core/gsp/commands.rs
> index 6ffd0b9cbf05..74f875755e53 100644
> --- a/drivers/gpu/nova-core/gsp/commands.rs
> +++ b/drivers/gpu/nova-core/gsp/commands.rs
> @@ -384,3 +384,117 @@ fn init_variable_payload(
>          }
>      }
>  }
> +
> +#[kunit_tests(nova_core_gsp_commands)]
> +mod tests {
> +    use super::*;
> +
> +    struct TestPayload {
> +        data: KVVec<u8>,
> +    }
> +
> +    impl TestPayload {
> +        fn generate_pattern(len: usize) -> Result<KVVec<u8>> {
> +            let mut data = KVVec::with_capacity(len, GFP_KERNEL)?;
> +            for i in 0..len {
> +                data.push(i as u8, GFP_KERNEL)?;
> +            }
> +            Ok(data)
> +        }
> +
> +        fn new(len: usize) -> Result<Self> {
> +            Ok(Self {
> +                data: Self::generate_pattern(len)?,
> +            })
> +        }
> +    }
> +
> +    impl CommandToGsp for TestPayload {
> +        const FUNCTION: MsgFunction = MsgFunction::Nop;
> +        type Command = ();

Since we are testing the size of the data written on the command queue,
can you make the command itself larger than 0 bytes? Otherwise there is
a potential for missing errors.

I'm saying that because it actually happened to me, I used
`MAX_CMD_SIZE` instead of `MAX_CMD_SIZE - size_of::<C::Command>()`
somewhere, and despite it being broken the tests were passing.

Re: [PATCH v3 9/9] gpu: nova-core: gsp: add tests for SplitState

[Alexandre Courbot]

On Thu Feb 26, 2026 at 8:45 PM JST, Eliot Courtney wrote:
> Add tests for SplitState. They cover boundary conditions at the split
> points to make sure the right number of continuation records are made.
> They also check that the data concatenated is correct.
>
> Tested-by: Zhi Wang <zhiw@nvidia.com>
> Signed-off-by: Eliot Courtney <ecourtney@nvidia.com>

Also I'd call this patch "add tests for continuation records". That's
another reason to move the continuation record stuff to its own module,
that way these tests can be contained to it and named accordingly
instead of being mixed with potentially future command-related tests.

Re: [PATCH v3 9/9] gpu: nova-core: gsp: add tests for SplitState

[Eliot Courtney]

On Fri Feb 27, 2026 at 8:32 PM JST, Alexandre Courbot wrote:
>> +#[kunit_tests(nova_core_gsp_commands)]
>> +mod tests {
>> +    use super::*;
>> +
>> +    struct TestPayload {
>> +        data: KVVec<u8>,
>> +    }
>> +
>> +    impl TestPayload {
>> +        fn generate_pattern(len: usize) -> Result<KVVec<u8>> {
>> +            let mut data = KVVec::with_capacity(len, GFP_KERNEL)?;
>> +            for i in 0..len {
>> +                data.push(i as u8, GFP_KERNEL)?;
>> +            }
>> +            Ok(data)
>> +        }
>> +
>> +        fn new(len: usize) -> Result<Self> {
>> +            Ok(Self {
>> +                data: Self::generate_pattern(len)?,
>> +            })
>> +        }
>> +    }
>> +
>> +    impl CommandToGsp for TestPayload {
>> +        const FUNCTION: MsgFunction = MsgFunction::Nop;
>> +        type Command = ();
>
> Since we are testing the size of the data written on the command queue,
> can you make the command itself larger than 0 bytes? Otherwise there is
> a potential for missing errors.
>
> I'm saying that because it actually happened to me, I used
> `MAX_CMD_SIZE` instead of `MAX_CMD_SIZE - size_of::<C::Command>()`
> somewhere, and despite it being broken the tests were passing.

Yeah good idea, particularly since you ran into it too. Added.

Re: [PATCH v3 8/9] gpu: nova-core: gsp: support large RPCs via continuation record

[Eliot Courtney]

On Fri Feb 27, 2026 at 8:28 PM JST, Alexandre Courbot wrote:
>> +    /// Sends `command` to the GSP.
>> +    ///
>> +    /// The command may be split into multiple messages if it is large.
>> +    ///
>> +    /// # Errors
>> +    ///
>> +    /// - `ETIMEDOUT` if space does not become available within the timeout.
>> +    /// - `EIO` if the variable payload requested by the command has not been entirely
>> +    ///   written to by its [`CommandToGsp::init_variable_payload`] method.
>> +    ///
>> +    /// Error codes returned by the command initializers are propagated as-is.
>> +    pub(crate) fn send_command<M>(&mut self, bar: &Bar0, command: M) -> Result
>> +    where
>> +        M: CommandToGsp,
>> +        Error: From<M::InitError>,
>> +    {
>> +        let mut state = SplitState::new(&command)?;
>> +
>> +        self.send_single_command(bar, state.command(command))?;
>
> As we discussed offline these two lines were bothering me a bit, because
> we pass `command` twice and there is a possibility (although purely
> hypothetical) of API misuse. I think I found a way around it: we turn
> `SplitState` into an enum with a `Single` variant, that contains the
> original command, and a `Split` one, which contains the truncated
> initial command and its continuation records.
>
> This simple change cascades into more simplifications in the new types
> introduced by this patch: `SplitCommand` is not an enum anymore, but the
> original command with a shorter payload. And the continuation records
> are their own self-contained type. This is achieved by allocating two
> `KVVec`s when we split, and taking advantage of the `SBuffer` we
> conveniently have at hand to fill them both.
>
> I have pushed a branch with this idea implemented on top of yours in [1]
> - please check it, but I think it brings a nice (if small) incremental
> improvement. It removes the API misuse potential, makes every type
> purely single-purpose, and doesn't need `PhantomData` or lifetimes,
> making it simpler overall IMHO.
>
> [1] https://github.com/Gnurou/linux/tree/review/continuations

Thank you for looking into this and I agree this new design is better.
We pay an extra allocation but it's rare that that path will be hit, so
I agree it's worth it.

Added and thanks for your work on improving this API.

>> +
>> +/// The `ContinuationRecord` command.
>> +pub(crate) struct ContinuationRecord<'a> {
>
> These new types are to be used by the command queue, and we don't want
> to use them elsewhere, so `pub(super)` seems more appropriate to me.
>
> And actually, since they are more command queue infrastructure than pure
> "commands" that fulfill a given purpose, I'd suggest moving them into
> their own sub-module of `gsp` (named `continuation` or `split` maybe?).
> That way they don't get in the way of readers who just want to learn
> about GSP commands.

Done.