drm/panthor: Fix the fast-reset logic

If we do a GPU soft-reset, that's no longer fast reset. This also means
the slow reset fallback doesn't work because the MCU state is only reset
after a GPU soft-reset.

Let's move the retry logic to panthor_device_resume() to issue a
soft-reset between the fast and slow attempts, and patch
panthor_gpu_suspend() to only power-off the L2 when a fast reset is
requested.

v3:
- No changes

v2:
- Add R-b

Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20241211075419.2333731-6-boris.brezillon@collabora.com

diff --git a/drivers/gpu/drm/panthor/panthor_device.c b/drivers/gpu/drm/panthor/panthor_device.c
index 5f9602bddc539cad5012f757d0026c1e0f2e0cbc..0a37cfeeb181cbc48e72771ac7e4b3c5b55c96fc 100644 (file)
--- a/drivers/gpu/drm/panthor/panthor_device.c
+++ b/drivers/gpu/drm/panthor/panthor_device.c
@@ -435,6 +435,22 @@ int panthor_device_mmap_io(struct panthor_device *ptdev, struct vm_area_struct *
        return 0;
 }
 
+static int panthor_device_resume_hw_components(struct panthor_device *ptdev)
+{
+       int ret;
+
+       panthor_gpu_resume(ptdev);
+       panthor_mmu_resume(ptdev);
+
+       ret = panthor_fw_resume(ptdev);
+       if (!ret)
+               return 0;
+
+       panthor_mmu_suspend(ptdev);
+       panthor_gpu_suspend(ptdev);
+       return ret;
+}
+
 int panthor_device_resume(struct device *dev)
 {
        struct panthor_device *ptdev = dev_get_drvdata(dev);
@@ -461,16 +477,16 @@ int panthor_device_resume(struct device *dev)
 
        if (panthor_device_is_initialized(ptdev) &&
            drm_dev_enter(&ptdev->base, &cookie)) {
-               panthor_gpu_resume(ptdev);
-               panthor_mmu_resume(ptdev);
-               ret = panthor_fw_resume(ptdev);
-               if (!drm_WARN_ON(&ptdev->base, ret)) {
-                       panthor_sched_resume(ptdev);
-               } else {
-                       panthor_mmu_suspend(ptdev);
-                       panthor_gpu_suspend(ptdev);
+               ret = panthor_device_resume_hw_components(ptdev);
+               if (ret && ptdev->reset.fast) {
+                       drm_err(&ptdev->base, "Fast reset failed, trying a slow reset");
+                       ptdev->reset.fast = false;
+                       ret = panthor_device_resume_hw_components(ptdev);
                }
 
+               if (!ret)
+                       panthor_sched_resume(ptdev);
+
                drm_dev_exit(cookie);
 
                if (ret)

diff --git a/drivers/gpu/drm/panthor/panthor_device.h b/drivers/gpu/drm/panthor/panthor_device.h
index b6c4f25a5d6e575210bfdcd01df1c36a278f7c9e..da6574021664ba2b4350ab1a595c18ac7f5be8e6 100644 (file)
--- a/drivers/gpu/drm/panthor/panthor_device.h
+++ b/drivers/gpu/drm/panthor/panthor_device.h
@@ -157,6 +157,17 @@ struct panthor_device {
 
                /** @pending: Set to true if a reset is pending. */
                atomic_t pending;
+
+               /**
+                * @fast: True if the post_reset logic can proceed with a fast reset.
+                *
+                * A fast reset is just a reset where the driver doesn't reload the FW sections.
+                *
+                * Any time the firmware is properly suspended, a fast reset can take place.
+                * On the other hand, if the halt operation failed, the driver will reload
+                * all FW sections to make sure we start from a fresh state.
+                */
+               bool fast;
        } reset;
 
        /** @pm: Power management related data. */

diff --git a/drivers/gpu/drm/panthor/panthor_fw.c b/drivers/gpu/drm/panthor/panthor_fw.c
index 7cdf251281977b561c63130d503853063b03f4e2..68eb4fb4d3a8ae0ddf5998a3e93ebc55d7cfafc6 100644 (file)
--- a/drivers/gpu/drm/panthor/panthor_fw.c
+++ b/drivers/gpu/drm/panthor/panthor_fw.c
@@ -263,17 +263,6 @@ struct panthor_fw {
        /** @booted: True is the FW is booted */
        bool booted;
 
-       /**
-        * @fast_reset: True if the post_reset logic can proceed with a fast reset.
-        *
-        * A fast reset is just a reset where the driver doesn't reload the FW sections.
-        *
-        * Any time the firmware is properly suspended, a fast reset can take place.
-        * On the other hand, if the halt operation failed, the driver will reload
-        * all sections to make sure we start from a fresh state.
-        */
-       bool fast_reset;
-
        /** @irq: Job irq data. */
        struct panthor_irq irq;
 };
@@ -1090,7 +1079,7 @@ void panthor_fw_pre_reset(struct panthor_device *ptdev, bool on_hang)
        /* Make sure we won't be woken up by a ping. */
        cancel_delayed_work_sync(&ptdev->fw->watchdog.ping_work);
 
-       ptdev->fw->fast_reset = false;
+       ptdev->reset.fast = false;
 
        if (!on_hang) {
                struct panthor_fw_global_iface *glb_iface = panthor_fw_get_glb_iface(ptdev);
@@ -1100,7 +1089,7 @@ void panthor_fw_pre_reset(struct panthor_device *ptdev, bool on_hang)
                gpu_write(ptdev, CSF_DOORBELL(CSF_GLB_DOORBELL_ID), 1);
                if (!readl_poll_timeout(ptdev->iomem + MCU_STATUS, status,
                                        status == MCU_STATUS_HALT, 10, 100000)) {
-                       ptdev->fw->fast_reset = true;
+                       ptdev->reset.fast = true;
                } else {
                        drm_warn(&ptdev->base, "Failed to cleanly suspend MCU");
                }
@@ -1125,49 +1114,30 @@ int panthor_fw_post_reset(struct panthor_device *ptdev)
        if (ret)
                return ret;
 
-       /* If this is a fast reset, try to start the MCU without reloading
-        * the FW sections. If it fails, go for a full reset.
-        */
-       if (ptdev->fw->fast_reset) {
+       if (!ptdev->reset.fast) {
+               /* On a slow reset, reload all sections, including RO ones.
+                * We're not supposed to end up here anyway, let's just assume
+                * the overhead of reloading everything is acceptable.
+                */
+               panthor_reload_fw_sections(ptdev, true);
+       } else {
                /* The FW detects 0 -> 1 transitions. Make sure we reset
                 * the HALT bit before the FW is rebooted.
                 * This is not needed on a slow reset because FW sections are
                 * re-initialized.
                 */
                struct panthor_fw_global_iface *glb_iface = panthor_fw_get_glb_iface(ptdev);
-               panthor_fw_update_reqs(glb_iface, req, 0, GLB_HALT);
-
-               ret = panthor_fw_start(ptdev);
-               if (!ret)
-                       goto out;
 
-               /* Forcibly reset the MCU and force a slow reset, so we get a
-                * fresh boot on the next panthor_fw_start() call.
-                */
-               panthor_fw_stop(ptdev);
-               ptdev->fw->fast_reset = false;
-               drm_err(&ptdev->base, "FW fast reset failed, trying a slow reset");
-
-               ret = panthor_vm_flush_all(ptdev->fw->vm);
-               if (ret) {
-                       drm_err(&ptdev->base, "FW slow reset failed (couldn't flush FW's AS l2cache)");
-                       return ret;
-               }
+               panthor_fw_update_reqs(glb_iface, req, 0, GLB_HALT);
        }
 
-       /* Reload all sections, including RO ones. We're not supposed
-        * to end up here anyway, let's just assume the overhead of
-        * reloading everything is acceptable.
-        */
-       panthor_reload_fw_sections(ptdev, true);
-
        ret = panthor_fw_start(ptdev);
        if (ret) {
-               drm_err(&ptdev->base, "FW slow reset failed (couldn't start the FW )");
+               drm_err(&ptdev->base, "FW %s reset failed",
+                       ptdev->reset.fast ?  "fast" : "slow");
                return ret;
        }
 
-out:
        /* We must re-initialize the global interface even on fast-reset. */
        panthor_fw_init_global_iface(ptdev);
        return 0;

diff --git a/drivers/gpu/drm/panthor/panthor_gpu.c b/drivers/gpu/drm/panthor/panthor_gpu.c
index ee85a371bc3892138befa52810c4103d3e9e90ac..671049020afaa21fb5b5a76fddadf81fe3a704b5 100644 (file)
--- a/drivers/gpu/drm/panthor/panthor_gpu.c
+++ b/drivers/gpu/drm/panthor/panthor_gpu.c
@@ -470,11 +470,12 @@ int panthor_gpu_soft_reset(struct panthor_device *ptdev)
  */
 void panthor_gpu_suspend(struct panthor_device *ptdev)
 {
-       /*
-        * It may be preferable to simply power down the L2, but for now just
-        * soft-reset which will leave the L2 powered down.
-        */
-       panthor_gpu_soft_reset(ptdev);
+       /* On a fast reset, simply power down the L2. */
+       if (!ptdev->reset.fast)
+               panthor_gpu_soft_reset(ptdev);
+       else
+               panthor_gpu_power_off(ptdev, L2, 1, 20000);
+
        panthor_gpu_irq_suspend(&ptdev->gpu->irq);
 }