summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/i915/gt
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/i915/gt')
-rw-r--r--drivers/gpu/drm/i915/gt/intel_gt.c8
-rw-r--r--drivers/gpu/drm/i915/gt/intel_gt_regs.h14
-rw-r--r--drivers/gpu/drm/i915/gt/intel_reset.c34
-rw-r--r--drivers/gpu/drm/i915/gt/intel_workarounds.c14
-rw-r--r--drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c42
5 files changed, 83 insertions, 29 deletions
diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c
index 767e329e1cc5..9c18b5f2e789 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt.c
@@ -1109,9 +1109,15 @@ static void mmio_invalidate_full(struct intel_gt *gt)
continue;
if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 50)) {
+ u32 val = BIT(engine->instance);
+
+ if (engine->class == VIDEO_DECODE_CLASS ||
+ engine->class == VIDEO_ENHANCEMENT_CLASS ||
+ engine->class == COMPUTE_CLASS)
+ val = _MASKED_BIT_ENABLE(val);
intel_gt_mcr_multicast_write_fw(gt,
xehp_regs[engine->class],
- BIT(engine->instance));
+ val);
} else {
rb = get_reg_and_bit(engine, regs == gen8_regs, regs, num);
if (!i915_mmio_reg_offset(rb.reg))
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
index c3cd92691795..a5454af2a9cf 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h
+++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
@@ -406,10 +406,10 @@
#define GEN9_WM_CHICKEN3 _MMIO(0x5588)
#define GEN9_FACTOR_IN_CLR_VAL_HIZ (1 << 9)
-#define CHICKEN_RASTER_1 _MMIO(0x6204)
+#define CHICKEN_RASTER_1 MCR_REG(0x6204)
#define DIS_SF_ROUND_NEAREST_EVEN REG_BIT(8)
-#define CHICKEN_RASTER_2 _MMIO(0x6208)
+#define CHICKEN_RASTER_2 MCR_REG(0x6208)
#define TBIMR_FAST_CLIP REG_BIT(5)
#define VFLSKPD MCR_REG(0x62a8)
@@ -429,9 +429,10 @@
#define RC_OP_FLUSH_ENABLE (1 << 0)
#define HIZ_RAW_STALL_OPT_DISABLE (1 << 2)
#define CACHE_MODE_1 _MMIO(0x7004) /* IVB+ */
-#define PIXEL_SUBSPAN_COLLECT_OPT_DISABLE (1 << 6)
-#define GEN8_4x4_STC_OPTIMIZATION_DISABLE (1 << 6)
-#define GEN9_PARTIAL_RESOLVE_IN_VC_DISABLE (1 << 1)
+#define MSAA_OPTIMIZATION_REDUC_DISABLE REG_BIT(11)
+#define PIXEL_SUBSPAN_COLLECT_OPT_DISABLE REG_BIT(6)
+#define GEN8_4x4_STC_OPTIMIZATION_DISABLE REG_BIT(6)
+#define GEN9_PARTIAL_RESOLVE_IN_VC_DISABLE REG_BIT(1)
#define GEN7_GT_MODE _MMIO(0x7008)
#define GEN9_IZ_HASHING_MASK(slice) (0x3 << ((slice) * 2))
@@ -457,6 +458,9 @@
#define GEN8_L3CNTLREG _MMIO(0x7034)
#define GEN8_ERRDETBCTRL (1 << 9)
+#define PSS_MODE2 _MMIO(0x703c)
+#define SCOREBOARD_STALL_FLUSH_CONTROL REG_BIT(5)
+
#define GEN7_SC_INSTDONE _MMIO(0x7100)
#define GEN12_SC_INSTDONE_EXTRA _MMIO(0x7104)
#define GEN12_SC_INSTDONE_EXTRA2 _MMIO(0x7108)
diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c
index 24736ebee17c..78dc5e493c62 100644
--- a/drivers/gpu/drm/i915/gt/intel_reset.c
+++ b/drivers/gpu/drm/i915/gt/intel_reset.c
@@ -278,6 +278,7 @@ out:
static int gen6_hw_domain_reset(struct intel_gt *gt, u32 hw_domain_mask)
{
struct intel_uncore *uncore = gt->uncore;
+ int loops = 2;
int err;
/*
@@ -285,18 +286,39 @@ static int gen6_hw_domain_reset(struct intel_gt *gt, u32 hw_domain_mask)
* for fifo space for the write or forcewake the chip for
* the read
*/
- intel_uncore_write_fw(uncore, GEN6_GDRST, hw_domain_mask);
+ do {
+ intel_uncore_write_fw(uncore, GEN6_GDRST, hw_domain_mask);
- /* Wait for the device to ack the reset requests */
- err = __intel_wait_for_register_fw(uncore,
- GEN6_GDRST, hw_domain_mask, 0,
- 500, 0,
- NULL);
+ /*
+ * Wait for the device to ack the reset requests.
+ *
+ * On some platforms, e.g. Jasperlake, we see that the
+ * engine register state is not cleared until shortly after
+ * GDRST reports completion, causing a failure as we try
+ * to immediately resume while the internal state is still
+ * in flux. If we immediately repeat the reset, the second
+ * reset appears to serialise with the first, and since
+ * it is a no-op, the registers should retain their reset
+ * value. However, there is still a concern that upon
+ * leaving the second reset, the internal engine state
+ * is still in flux and not ready for resuming.
+ */
+ err = __intel_wait_for_register_fw(uncore, GEN6_GDRST,
+ hw_domain_mask, 0,
+ 2000, 0,
+ NULL);
+ } while (err == 0 && --loops);
if (err)
GT_TRACE(gt,
"Wait for 0x%08x engines reset failed\n",
hw_domain_mask);
+ /*
+ * As we have observed that the engine state is still volatile
+ * after GDRST is acked, impose a small delay to let everything settle.
+ */
+ udelay(50);
+
return err;
}
diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index 2afb4f80a954..949c19339015 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -645,7 +645,7 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine,
static void dg2_ctx_gt_tuning_init(struct intel_engine_cs *engine,
struct i915_wa_list *wal)
{
- wa_masked_en(wal, CHICKEN_RASTER_2, TBIMR_FAST_CLIP);
+ wa_mcr_masked_en(wal, CHICKEN_RASTER_2, TBIMR_FAST_CLIP);
wa_mcr_write_clr_set(wal, XEHP_L3SQCREG5, L3_PWM_TIMER_INIT_VAL_MASK,
REG_FIELD_PREP(L3_PWM_TIMER_INIT_VAL_MASK, 0x7f));
wa_mcr_add(wal,
@@ -771,11 +771,19 @@ static void dg2_ctx_workarounds_init(struct intel_engine_cs *engine,
/* Wa_14014947963:dg2 */
if (IS_DG2_GRAPHICS_STEP(engine->i915, G10, STEP_B0, STEP_FOREVER) ||
- IS_DG2_G11(engine->i915) || IS_DG2_G12(engine->i915))
+ IS_DG2_G11(engine->i915) || IS_DG2_G12(engine->i915))
wa_masked_field_set(wal, VF_PREEMPTION, PREEMPTION_VERTEX_COUNT, 0x4000);
+ /* Wa_18018764978:dg2 */
+ if (IS_DG2_GRAPHICS_STEP(engine->i915, G10, STEP_C0, STEP_FOREVER) ||
+ IS_DG2_G11(engine->i915) || IS_DG2_G12(engine->i915))
+ wa_masked_en(wal, PSS_MODE2, SCOREBOARD_STALL_FLUSH_CONTROL);
+
/* Wa_15010599737:dg2 */
- wa_masked_en(wal, CHICKEN_RASTER_1, DIS_SF_ROUND_NEAREST_EVEN);
+ wa_mcr_masked_en(wal, CHICKEN_RASTER_1, DIS_SF_ROUND_NEAREST_EVEN);
+
+ /* Wa_18019271663:dg2 */
+ wa_masked_en(wal, CACHE_MODE_1, MSAA_OPTIMIZATION_REDUC_DISABLE);
}
static void fakewa_disable_nestedbb_mode(struct intel_engine_cs *engine,
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index 0c80ba51a4bd..2bcdd192f814 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -545,6 +545,32 @@ static int check_ccs_header(struct intel_gt *gt,
return 0;
}
+static int try_firmware_load(struct intel_uc_fw *uc_fw, const struct firmware **fw)
+{
+ struct intel_gt *gt = __uc_fw_to_gt(uc_fw);
+ struct device *dev = gt->i915->drm.dev;
+ int err;
+
+ err = firmware_request_nowarn(fw, uc_fw->file_selected.path, dev);
+
+ if (err)
+ return err;
+
+ if ((*fw)->size > INTEL_UC_RSVD_GGTT_PER_FW) {
+ drm_err(&gt->i915->drm,
+ "%s firmware %s: size (%zuKB) exceeds max supported size (%uKB)\n",
+ intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path,
+ (*fw)->size / SZ_1K, INTEL_UC_RSVD_GGTT_PER_FW / SZ_1K);
+
+ /* try to find another blob to load */
+ release_firmware(*fw);
+ *fw = NULL;
+ return -ENOENT;
+ }
+
+ return 0;
+}
+
/**
* intel_uc_fw_fetch - fetch uC firmware
* @uc_fw: uC firmware
@@ -558,7 +584,6 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw)
struct intel_gt *gt = __uc_fw_to_gt(uc_fw);
struct drm_i915_private *i915 = gt->i915;
struct intel_uc_fw_file file_ideal;
- struct device *dev = i915->drm.dev;
struct drm_i915_gem_object *obj;
const struct firmware *fw = NULL;
bool old_ver = false;
@@ -574,20 +599,9 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw)
__force_fw_fetch_failures(uc_fw, -EINVAL);
__force_fw_fetch_failures(uc_fw, -ESTALE);
- err = firmware_request_nowarn(&fw, uc_fw->file_selected.path, dev);
+ err = try_firmware_load(uc_fw, &fw);
memcpy(&file_ideal, &uc_fw->file_wanted, sizeof(file_ideal));
- if (!err && fw->size > INTEL_UC_RSVD_GGTT_PER_FW) {
- drm_err(&i915->drm,
- "%s firmware %s: size (%zuKB) exceeds max supported size (%uKB)\n",
- intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path,
- fw->size / SZ_1K, INTEL_UC_RSVD_GGTT_PER_FW / SZ_1K);
-
- /* try to find another blob to load */
- release_firmware(fw);
- err = -ENOENT;
- }
-
/* Any error is terminal if overriding. Don't bother searching for older versions */
if (err && intel_uc_fw_is_overridden(uc_fw))
goto fail;
@@ -608,7 +622,7 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw)
break;
}
- err = firmware_request_nowarn(&fw, uc_fw->file_selected.path, dev);
+ err = try_firmware_load(uc_fw, &fw);
}
if (err)
generated by cgit v1.2.3 (git 2.39.1) at 2024-09-27 17:42:02 +0000