From: Ramalingam C <ramalingam.c@intel.com>
To: intel-gfx <intel-gfx@lists.freedesktop.org>,
dri-devel <dri-devel@lists.freedesktop.org>
Cc: Hellstrom Thomas <thomas.hellstrom@intel.com>,
Matthew Auld <matthew.auld@intel.com>,
Chris Wilson <chris.p.wilson@intel.com>
Subject: [PATCH 1/5] drm/i915/dg2: Add Wa_22011450934
Date: Sat, 29 Jan 2022 00:22:05 +0530 [thread overview]
Message-ID: <20220128185209.18077-2-ramalingam.c@intel.com> (raw)
In-Reply-To: <20220128185209.18077-1-ramalingam.c@intel.com>
An indirect ctx wabb is implemented as per Wa_22011450934 to avoid rcs
restore hang during context restore of a preempted context in GPGPU mode
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
cc: Chris Wilson <chris.p.wilson@intel.com>
---
drivers/gpu/drm/i915/gt/intel_lrc.c | 28 ++++++++++++++++++++++++++++
drivers/gpu/drm/i915/i915_reg.h | 4 ++++
2 files changed, 32 insertions(+)
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index 89a95a125fc8..8440f4b0f613 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -1165,6 +1165,29 @@ gen12_emit_cmd_buf_wa(const struct intel_context *ce, u32 *cs)
return cs;
}
+/*
+ * On DG2 during context restore of a preempted context in GPGPU mode,
+ * RCS restore hang is detected. This is extremely timing dependent.
+ * To address this below sw wabb is implemented for DG2 A steppings.
+ */
+static u32 *
+dg2_emit_rcs_hang_wabb(const struct intel_context *ce, u32 *cs)
+{
+ *cs++ = MI_LOAD_REGISTER_IMM(1);
+ *cs++ = i915_mmio_reg_offset(GEN12_STATE_ACK_DEBUG);
+ *cs++ = 0x21;
+
+ *cs++ = MI_LOAD_REGISTER_REG;
+ *cs++ = i915_mmio_reg_offset(RING_NOPID(ce->engine->mmio_base));
+ *cs++ = i915_mmio_reg_offset(GEN12_CULLBIT1);
+
+ *cs++ = MI_LOAD_REGISTER_REG;
+ *cs++ = i915_mmio_reg_offset(RING_NOPID(ce->engine->mmio_base));
+ *cs++ = i915_mmio_reg_offset(GEN12_CULLBIT2);
+
+ return cs;
+}
+
static u32 *
gen12_emit_indirect_ctx_rcs(const struct intel_context *ce, u32 *cs)
{
@@ -1172,6 +1195,11 @@ gen12_emit_indirect_ctx_rcs(const struct intel_context *ce, u32 *cs)
cs = gen12_emit_cmd_buf_wa(ce, cs);
cs = gen12_emit_restore_scratch(ce, cs);
+ /* Wa_22011450934:dg2 */
+ if (IS_DG2_GRAPHICS_STEP(ce->engine->i915, G10, STEP_A0, STEP_B0) ||
+ IS_DG2_GRAPHICS_STEP(ce->engine->i915, G11, STEP_A0, STEP_B0))
+ cs = dg2_emit_rcs_hang_wabb(ce, cs);
+
/* Wa_16013000631:dg2 */
if (IS_DG2_GRAPHICS_STEP(ce->engine->i915, G10, STEP_B0, STEP_C0) ||
IS_DG2_G11(ce->engine->i915))
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index ec48406eb37a..3f94f4b5ef6c 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -12026,4 +12026,8 @@ enum skl_power_gate {
#define SLICE_COMMON_ECO_CHICKEN1 _MMIO(0x731C)
#define MSC_MSAA_REODER_BUF_BYPASS_DISABLE REG_BIT(14)
+#define GEN12_CULLBIT1 _MMIO(0x6100)
+#define GEN12_CULLBIT2 _MMIO(0x7030)
+#define GEN12_STATE_ACK_DEBUG _MMIO(0x20BC)
+
#endif /* _I915_REG_H_ */
--
2.20.1
next prev parent reply other threads:[~2022-01-28 18:52 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-01-28 18:52 [PATCH 0/5] Misc DG2 enabling patches Ramalingam C
2022-01-28 18:52 ` Ramalingam C [this message]
2022-02-07 11:48 ` [Intel-gfx] [PATCH 1/5] drm/i915/dg2: Add Wa_22011450934 Matthew Auld
2022-02-07 11:52 ` Matthew Auld
2022-02-11 13:34 ` Ramalingam C
2022-01-28 18:52 ` [PATCH 2/5] drm/i915: align the plane_vma to min_page_size of stolen mem Ramalingam C
2022-01-31 10:18 ` Matthew Auld
2022-01-31 10:26 ` [Intel-gfx] " Matthew Auld
2022-01-28 18:52 ` [PATCH 3/5] drm/i915: More gt idling time with guc submission Ramalingam C
2022-01-31 10:19 ` Matthew Auld
2022-01-28 18:52 ` [PATCH 4/5] drm/i915/dg2: Add Wa_22011100796 Ramalingam C
2022-01-31 11:06 ` [Intel-gfx] " Matthew Auld
2022-01-28 18:52 ` [PATCH 5/5] drm/i915/guc: Allow user to override driver load failure without GuC Ramalingam C
2022-02-07 16:55 ` [Intel-gfx] " Daniele Ceraolo Spurio
2022-02-11 13:32 ` Ramalingam C
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20220128185209.18077-2-ramalingam.c@intel.com \
--to=ramalingam.c@intel.com \
--cc=chris.p.wilson@intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=matthew.auld@intel.com \
--cc=thomas.hellstrom@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).