[Linaro-mm-sig] Re: [PATCH 7/8] drm/panthor: Add support for entering and exiting protected mode

On Tue, 5 May 2026 16:05:13 +0200 Ketil Johnsen ketil.johnsen@arm.com wrote:

From: Florent Tomasin florent.tomasin@arm.com

This patch modifies the Panthor driver code to allow handling of the GPU HW protected mode enter and exit.

The logic added by this patch includes:

• the mechanisms needed for entering and exiting protected mode.
• the handling of protected mode IRQs and FW interactions.
• the scheduler changes needed to decide when to enter protected mode based on CSG scheduling.

Note that the submission of a protected mode jobs are done from the user space.

The following is a summary of how protected mode is entered and exited:

• When the GPU detects a protected mode job needs to be executed, an IRQ is sent to the CPU to notify the kernel driver that the job is blocked until the GPU has entered protected mode. The entering of protected mode is controlled by the kernel driver.
• The Mali Panthor CSF driver will schedule a tick and evaluate which CS in the CSG to schedule on slot needs protected mode. If the priority of the CSG is not sufficiently high, the protected mode job will not progress until the CSG is scheduled at top priority.
• The Panthor scheduler notifies the GPU that the blocked protected jobs will soon be able to progress.
• Once all CSG and CS slots are updated, the scheduler requests the GPU to enter protected mode and waits for it to be acknowledged.
• If successful, all protected mode jobs will resume execution while normal mode jobs block until the GPU exits protected mode, or the kernel driver rotates the CSGs and forces the GPU to exit protected mode.
• If unsuccessful, the scheduler will request a GPU reset.
• When a protected mode job is suspended as a result of the CSGs rotation, the GPU will send an IRQ to the CPU to notify that the protected mode job needs to resume.

This sequence will continue so long the user space is submitting protected mode jobs.

Signed-off-by: Florent Tomasin florent.tomasin@arm.com Co-developed-by: Paul Toadere paul.toadere@arm.com Signed-off-by: Paul Toadere paul.toadere@arm.com Co-developed-by: Samuel Percival samuel.percival@arm.com Signed-off-by: Samuel Percival samuel.percival@arm.com Co-developed-by: Ketil Johnsen ketil.johnsen@arm.com Signed-off-by: Ketil Johnsen ketil.johnsen@arm.com

---

drivers/gpu/drm/panthor/panthor_device.c | 1 + drivers/gpu/drm/panthor/panthor_device.h | 9 + drivers/gpu/drm/panthor/panthor_fw.c | 86 ++++++++- drivers/gpu/drm/panthor/panthor_fw.h | 5 + drivers/gpu/drm/panthor/panthor_gpu.c | 14 +- drivers/gpu/drm/panthor/panthor_gpu.h | 6 + drivers/gpu/drm/panthor/panthor_mmu.c | 10 + drivers/gpu/drm/panthor/panthor_sched.c | 224 +++++++++++++++++++++-- 8 files changed, 339 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/panthor/panthor_device.c b/drivers/gpu/drm/panthor/panthor_device.c index 3a5cdfa99e5fe..449f17b0f4c5c 100644 --- a/drivers/gpu/drm/panthor/panthor_device.c +++ b/drivers/gpu/drm/panthor/panthor_device.c @@ -207,6 +207,7 @@ int panthor_device_init(struct panthor_device *ptdev) ptdev->soc_data = of_device_get_match_data(ptdev->base.dev);

• init_rwsem(&ptdev->protm.lock); init_completion(&ptdev->unplug.done); ret = drmm_mutex_init(&ptdev->base, &ptdev->unplug.lock); if (ret)

diff --git a/drivers/gpu/drm/panthor/panthor_device.h b/drivers/gpu/drm/panthor/panthor_device.h index d51fec97fc5fa..ebeec45cf60a1 100644 --- a/drivers/gpu/drm/panthor/panthor_device.h +++ b/drivers/gpu/drm/panthor/panthor_device.h @@ -334,6 +334,15 @@ struct panthor_device { struct { /** @heap: Pointer to the protected heap */ struct dma_heap *heap;

• /**
  

•  * @lock: Lock to prevent VM operations during protected mode.
  

Here it says the lock prevents VM ops while the GPU is in protected mode, but...

•  *
  

•  * The MMU will not execute commands when the GPU is in
  

•  * protected mode, so we use this RW lock to sync access
  

•  * between VM_BIND and GPU protected mode.
  

•  */
  

• struct rw_semaphore lock;
  

  } protm;

}; diff --git a/drivers/gpu/drm/panthor/panthor_fw.c b/drivers/gpu/drm/panthor/panthor_fw.c index 1aba29b9779b6..281556530ddab 100644 --- a/drivers/gpu/drm/panthor/panthor_fw.c +++ b/drivers/gpu/drm/panthor/panthor_fw.c @@ -1057,7 +1057,9 @@ static void panthor_fw_init_global_iface(struct panthor_device *ptdev) GLB_CFG_PROGRESS_TIMER | GLB_CFG_POWEROFF_TIMER | GLB_IDLE_EN |

• 			 GLB_IDLE;
  

• 			 GLB_IDLE |
  

• 			 GLB_PROTM_ENTER |
  

• 			 GLB_PROTM_EXIT;
  

if (panthor_fw_has_glb_state(ptdev)) glb_iface->input->ack_irq_mask |= GLB_STATE_MASK; @@ -1456,6 +1458,88 @@ static void panthor_fw_ping_work(struct work_struct *work) } } +int panthor_fw_protm_enter(struct panthor_device *ptdev) +{

• struct panthor_fw_global_iface *glb_iface;
• u32 acked;
• u32 status;
• int ret;
• down_write(&ptdev->protm.lock);
• glb_iface = panthor_fw_get_glb_iface(ptdev);
• panthor_fw_toggle_reqs(glb_iface, req, ack, GLB_PROTM_ENTER);
• gpu_write(ptdev, CSF_DOORBELL(CSF_GLB_DOORBELL_ID), 1);
• ret = panthor_fw_glb_wait_acks(ptdev, GLB_PROTM_ENTER, &acked, 4000);
• if (ret) {

• drm_err(&ptdev->base, "Wait for FW protected mode acknowledge timed out");
  

• up_write(&ptdev->protm.lock);
  

• return ret;
  

• }
• /* Wait for the GPU to actually enter protected mode.

• * There would be some time gap between FW sending the
  

• * ACK for GLB_PROTM_ENTER and GPU entering protected mode.
  

• */
  

• if (gpu_read_poll_timeout(ptdev, GPU_STATUS, status,

• 		  (status & GPU_STATUS_PROTM_ACTIVE) ||
  

• 			  ((glb_iface->input->req ^ glb_iface->output->ack) &
  

• 			   GLB_PROTM_EXIT),
  

• 		  10, 500000)) {
  

• drm_err(&ptdev->base, "Wait for GPU protected mode enter timed out");
  

• ret = -ETIMEDOUT;
  

• }
• up_write(&ptdev->protm.lock);

... here I see the lock being released right after we've entered protected mode. Meaning the MMU layer can proceed with any pending VM ops even though the GPU only exists PROTM when panthor_fw_protm_exit() is called. If this is expected, the protm::lock doc should be updated to reflect that.

Also, I don't think a rw_semaphore alone is enough to cover the kind of critical section you're trying to declare, because it requires that the lock is taken/released from the same thread, and panthor_fw_protm_enter()/panthor_fw_protm_exit() will be called from different work items. This probably explains why the doc no longer matches the implementation.

I guess this could be reworked to use a combination of rwlock+completion, where the VM logic does something like:

down_read(protm.lock); while (!try_wait_for_completion(protm.complete)) { up_read(protm.lock); if (!wait_for_completion_timeout(protm.complete, timeout)) { schedule_reset(); return -ETIMEDOUT; } down_read(protm.lock); }

// proceed with the VM op

up_read(protm.lock);

in panthor_fw_protm_enter(), you'd take the lock in write mode, reinit the completion object, release the lock, and proceed with the PROTM operation. If it fails, you call complete_all() immediately, if it works, you defer the complete_all() to the panthor_fw_protm_exit() path.

• return ret;

+}

+void panthor_fw_protm_exit(struct panthor_device *ptdev) +{

• struct panthor_fw_global_iface *glb_iface = panthor_fw_get_glb_iface(ptdev);
• /* Acknowledge the protm exit. */
• panthor_fw_update_reqs(glb_iface, req, glb_iface->output->ack, GLB_PROTM_EXIT);

+}

+int panthor_fw_protm_exit_wait_event_timeout(struct panthor_device *ptdev) +{

• struct panthor_fw_global_iface *glb_iface = panthor_fw_get_glb_iface(ptdev);
• int ret = 0;
• /* Send PING request to force an exit */
• panthor_fw_toggle_reqs(glb_iface, req, ack, GLB_PING);

Uh, if a PING triggers a PROTM exit, we should probably pause the PING (or reschedule it) right before entering PROTM, otherwise timings might make it so PROTM is exited almost immediately after enter.

• gpu_write(ptdev, CSF_DOORBELL(CSF_GLB_DOORBELL_ID), 1);
• ret = wait_event_timeout(ptdev->fw->req_waitqueue,

• 		 !(gpu_read(ptdev, GPU_STATUS) & GPU_STATUS_PROTM_ACTIVE),
  

• 		 msecs_to_jiffies(500));
  

• if (!ret) {

• drm_err(&ptdev->base, "Wait for forced protected mode exit timed out");
  

• panthor_device_schedule_reset(ptdev);
  

• return -ETIMEDOUT;
  

• }
• return 0;

+}

+void panthor_fw_protm_exit_sync(struct panthor_device *ptdev) +{

• u32 status;
• /* Busy-wait (5ms) for FW to exit protected mode on its own */
• if (!gpu_read_poll_timeout(ptdev, GPU_STATUS, status,

• 		   !(status & GPU_STATUS_PROTM_ACTIVE), 10,
  

• 		   5000))
  

• return;
  

• panthor_fw_protm_exit_wait_event_timeout(ptdev);

+}

I'll stop there for now.

Regards,

Boris