[PATCH 5.9 013/133] drm/i915/gt: Delay execlist processing for tgl

[Greg Kroah-Hartman]

From: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>

commit 9b99e5ba3e5d68039bd6b657e4bbe520a3521f4c upstream.

When running gem_exec_nop, it floods the system with many requests (with
the goal of userspace submitting faster than the HW can process a single
empty batch). This causes the driver to continually resubmit new
requests onto the end of an active context, a flood of lite-restore
preemptions. If we time this just right, Tigerlake hangs.

Inserting a small delay between the processing of CS events and
submitting the next context, prevents the hang. Naturally it does not
occur with debugging enabled. The suspicion then is that this is related
to the issues with the CS event buffer, and inserting an mmio read of
the CS pointer status appears to be very successful in preventing the
hang. Other registers, or uncached reads, or plain mb, do not prevent
the hang, suggesting that register is key -- but that the hang can be
prevented by a simple udelay, suggests it is just a timing issue like
that encountered by commit 233c1ae3c83f ("drm/i915/gt: Wait for CSB
entries on Tigerlake"). Also note that the hang is not prevented by
applying CTX_DESC_FORCE_RESTORE, or by inserting a delay on the GPU
between requests.

Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Cc: Mika Kuoppala <mika.kuoppala@xxxxxxxxxxxxxxx>
Cc: Bruce Chang <yu.bruce.chang@xxxxxxxxx>
Cc: Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
Acked-by: Mika Kuoppala <mika.kuoppala@xxxxxxxxxxxxxxx>
Link: https://patchwork.freedesktop.org/patch/msgid/20201015195023.32346-1-chris@xxxxxxxxxxxxxxxxxx
(cherry picked from commit 6ca7217dffaf1abba91558e67a2efb655ac91405)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@xxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
 drivers/gpu/drm/i915/gt/intel_lrc.c |    3 +++
 1 file changed, 3 insertions(+)

--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -2661,6 +2661,9 @@ static void process_csb(struct intel_eng
 			smp_wmb(); /* complete the seqlock */
 			WRITE_ONCE(execlists->active, execlists->inflight);
 
+			/* XXX Magic delay for tgl */
+			ENGINE_POSTING_READ(engine, RING_CONTEXT_STATUS_PTR);
+
 			WRITE_ONCE(execlists->pending[0], NULL);
 		} else {
 			if (GEM_WARN_ON(!*execlists->active)) {