drm/i915: Do not disable preemption for resets

Commit ade8a0f ("drm/i915: Make all GPU resets atomic") added a
preempt disable section over the hardware reset callback to prepare the
driver for being able to reset from atomic contexts.

In retrospect I can see that the work item at a time was about removing
the struct mutex from the reset path. Code base also briefly entertained
the idea of doing the reset under stop_machine in order to serialize
userspace mmap and temporary glitch in the fence registers (see
eb8d0f5 ("drm/i915: Remove GPU reset dependence on struct_mutex"),
but that never materialized and was soon removed in 2caffbf
("drm/i915: Revoke mmaps and prevent access to fence registers across
reset") and replaced with a SRCU based solution.

As such, as far as I can see, today we still have a requirement that
resets must not sleep (invoked from submission tasklets), but no need to
support invoking them from a truly atomic context.

Given that the preemption section is problematic on RT kernels, since the
uncore lock becomes a sleeping lock and so is invalid in such section,
lets try and remove it. Potential downside is that our short waits on GPU
to complete the reset may get extended if CPU scheduling interferes, but
in practice that probably isn't a deal breaker.

In terms of mechanics, since the preemption disabled block is being
removed we just need to replace a few of the wait_for_atomic macros into
busy looping versions which will work (and not complain) when called from
non-atomic sections.

v2:
 * Fix timeouts which are now in us. (Andi)
 * Update one comment as a drive by. (Andi)

Signed-off-by: Tvrtko Ursulin <[email protected]>
Cc: Chris Wilson <[email protected]>
Cc: Paul Gortmaker <[email protected]>
Cc: Sebastian Andrzej Siewior <[email protected]>
Cc: Andi Shyti <[email protected]>
Reviewed-by: Andi Shyti <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

Author: tursulin

drivers/gpu/drm/i915/gt/intel_reset.c

@@ -161,16 +161,16 @@ static int i915_do_reset(struct intel_gt *gt,
 	struct pci_dev *pdev = to_pci_dev(gt->i915->drm.dev);
 	int err;
 
-	/* Assert reset for at least 20 usec, and wait for acknowledgement. */
+	/* Assert reset for at least 50 usec, and wait for acknowledgement. */
 	pci_write_config_byte(pdev, I915_GDRST, GRDOM_RESET_ENABLE);
 	udelay(50);
-	err = wait_for_atomic(i915_in_reset(pdev), 50);
+	err = _wait_for_atomic(i915_in_reset(pdev), 50000, 0);
 
 	/* Clear the reset request. */
 	pci_write_config_byte(pdev, I915_GDRST, 0);
 	udelay(50);
 	if (!err)
-		err = wait_for_atomic(!i915_in_reset(pdev), 50);
+		err = _wait_for_atomic(!i915_in_reset(pdev), 50000, 0);
 
 	return err;
 }
@@ -190,7 +190,7 @@ static int g33_do_reset(struct intel_gt *gt,
 	struct pci_dev *pdev = to_pci_dev(gt->i915->drm.dev);
 
 	pci_write_config_byte(pdev, I915_GDRST, GRDOM_RESET_ENABLE);
-	return wait_for_atomic(g4x_reset_complete(pdev), 50);
+	return _wait_for_atomic(g4x_reset_complete(pdev), 50000, 0);
 }
 
 static int g4x_do_reset(struct intel_gt *gt,
@@ -207,15 +207,15 @@ static int g4x_do_reset(struct intel_gt *gt,
 
 	pci_write_config_byte(pdev, I915_GDRST,
 			      GRDOM_MEDIA | GRDOM_RESET_ENABLE);
-	ret =  wait_for_atomic(g4x_reset_complete(pdev), 50);
+	ret =  _wait_for_atomic(g4x_reset_complete(pdev), 50000, 0);
 	if (ret) {
 		GT_TRACE(gt, "Wait for media reset failed\n");
 		goto out;
 	}
 
 	pci_write_config_byte(pdev, I915_GDRST,
 			      GRDOM_RENDER | GRDOM_RESET_ENABLE);
-	ret =  wait_for_atomic(g4x_reset_complete(pdev), 50);
+	ret =  _wait_for_atomic(g4x_reset_complete(pdev), 50000, 0);
 	if (ret) {
 		GT_TRACE(gt, "Wait for render reset failed\n");
 		goto out;
@@ -785,9 +785,7 @@ int __intel_gt_reset(struct intel_gt *gt, intel_engine_mask_t engine_mask)
 		reset_mask = wa_14015076503_start(gt, engine_mask, !retry);
 
 		GT_TRACE(gt, "engine_mask=%x\n", reset_mask);
-		preempt_disable();
 		ret = reset(gt, reset_mask, retry);
-		preempt_enable();
 
 		wa_14015076503_end(gt, reset_mask);
 	}