Skip to contentSkip to conversation

Refactor to expand subslice mask

PATCH In review Stuart Summers 6 participants 34 messages May 1, 2019 – May 3, 2019

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

cover Refactor to expand subslice mask

This patch series contains a few code clean-up patches, followed by a patch which changes the storage of the subslice mask to better match the userspace access through the I915_QUERY_TOPOLOGY_INFO ioctl. The index into the subslice_mask array is then calculated:
slice * subslice stride + subslice index / 8

v2: fix i915_pm_sseu test failure
v3: no changes to patches in the series, just resending to pick up
in CI correctly
v4: rebase
v5: fix header test
v6: address review comments from Jari
address minor checkpatch warning in existing code use eu_stride for EU div-by-8
v7: another rebase

Stuart Summers (6):
drm/i915: Use local variable for SSEU info in GETPARAM ioctl drm/i915: Add macro for SSEU stride calculation drm/i915: Move calculation of subslices per slice to new function drm/i915: Move sseu helper functions to intel_sseu.h drm/i915: Remove inline from sseu helper functions drm/i915: Expand subslice mask

 drivers/gpu/drm/i915/gt/intel_engine_cs.c    |   6 +-
 drivers/gpu/drm/i915/gt/intel_engine_types.h |  32 +--
 drivers/gpu/drm/i915/gt/intel_hangcheck.c    |   3 +-
 drivers/gpu/drm/i915/gt/intel_sseu.c         |  85 ++++++++
 drivers/gpu/drm/i915/gt/intel_sseu.h         |  30 ++-
 drivers/gpu/drm/i915/gt/intel_workarounds.c  |   2 +-
 drivers/gpu/drm/i915/i915_debugfs.c          |  50 +++--
 drivers/gpu/drm/i915/i915_drv.c              |  15 +-
 drivers/gpu/drm/i915/i915_gpu_error.c        |   5 +-
 drivers/gpu/drm/i915/i915_query.c            |  15 +-
 drivers/gpu/drm/i915/intel_device_info.c     | 209 +++++++++++--------
 drivers/gpu/drm/i915/intel_device_info.h     |  47 -----
 12 files changed, 302 insertions(+), 197 deletions(-)
Signature · 6 lines
2.21.0.5.gaeb582a983

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#1 lore raw

PATCH 1/6 drm/i915: Use local variable for SSEU info in GETPARAM ioctl

In the GETPARAM ioctl handler, use a local variable to consolidate usage of SSEU runtime info.

v2: add const to sseu_dev_info variable

Cc Daniele Ceraolo Spurio <[email protected]>
---
 drivers/gpu/drm/i915/i915_drv.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/gpu/drm/i915/i915_drv.c +6 -5
index 21dac5a09fbe..c376244c19c4 100644
@@ -324,6 +324,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct pci_dev *pdev = dev_priv->drm.pdev;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
 	drm_i915_getparam_t *param = data;
 	int value;
 
@@ -377,12 +378,12 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
 		value = i915_cmd_parser_get_version(dev_priv);
 		break;
 	case I915_PARAM_SUBSLICE_TOTAL:
-		value = sseu_subslice_total(&RUNTIME_INFO(dev_priv)->sseu);
+		value = sseu_subslice_total(sseu);
 		if (!value)
 			return -ENODEV;
 		break;
 	case I915_PARAM_EU_TOTAL:
-		value = RUNTIME_INFO(dev_priv)->sseu.eu_total;
+		value = sseu->eu_total;
 		if (!value)
 			return -ENODEV;
 		break;
@@ -399,7 +400,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
 		value = HAS_POOLED_EU(dev_priv);
 		break;
 	case I915_PARAM_MIN_EU_IN_POOL:
-		value = RUNTIME_INFO(dev_priv)->sseu.min_eu_in_pool;
+		value = sseu->min_eu_in_pool;
 		break;
 	case I915_PARAM_HUC_STATUS:
 		value = intel_huc_check_status(&dev_priv->huc);
@@ -449,12 +450,12 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
 		value = intel_engines_has_context_isolation(dev_priv);
 		break;
 	case I915_PARAM_SLICE_MASK:
-		value = RUNTIME_INFO(dev_priv)->sseu.slice_mask;
+		value = sseu->slice_mask;
 		if (!value)
 			return -ENODEV;
 		break;
 	case I915_PARAM_SUBSLICE_MASK:
-		value = RUNTIME_INFO(dev_priv)->sseu.subslice_mask[0];
+		value = sseu->subslice_mask[0];
 		if (!value)
 			return -ENODEV;
 		break;
Signature · 6 lines
2.21.0.5.gaeb582a983

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#4 lore raw

PATCH 2/6 drm/i915: Add macro for SSEU stride calculation

In reply to Stuart Summers

Subslice stride and EU stride are calculated multiple times in i915_query. Move this calculation to a macro to reduce code duplication.

v2: update headers in intel_sseu.h

Cc Daniele Ceraolo Spurio <[email protected]>
---
 drivers/gpu/drm/i915/gt/intel_sseu.h |  2 ++
 drivers/gpu/drm/i915/i915_query.c    | 17 ++++++++---------
 2 files changed, 10 insertions(+), 9 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.h +2 -0
index 73bc824094e8..c0b16b248d4c 100644
@@ -8,11 +8,13 @@
 #define __INTEL_SSEU_H__
 
 #include <linux/types.h>
+#include <linux/kernel.h>
 
 struct drm_i915_private;
 
 #define GEN_MAX_SLICES		(6) /* CNL upper bound */
 #define GEN_MAX_SUBSLICES	(8) /* ICL upper bound */
+#define GEN_SSEU_STRIDE(bits) DIV_ROUND_UP(bits, BITS_PER_BYTE)
 
 struct sseu_dev_info {
 	u8 slice_mask;
drivers/gpu/drm/i915/i915_query.c +8 -9
index 782183b78f49..7c1708c22811 100644
@@ -37,6 +37,8 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
 	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
 	struct drm_i915_query_topology_info topo;
 	u32 slice_length, subslice_length, eu_length, total_length;
+	u8 subslice_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
+	u8 eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
 	int ret;
 
 	if (query_item->flags != 0)
@@ -48,12 +50,10 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
 	BUILD_BUG_ON(sizeof(u8) != sizeof(sseu->slice_mask));
 
 	slice_length = sizeof(sseu->slice_mask);
-	subslice_length = sseu->max_slices *
-		DIV_ROUND_UP(sseu->max_subslices, BITS_PER_BYTE);
-	eu_length = sseu->max_slices * sseu->max_subslices *
-		DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE);
-
-	total_length = sizeof(topo) + slice_length + subslice_length + eu_length;
+	subslice_length = sseu->max_slices * subslice_stride;
+	eu_length = sseu->max_slices * sseu->max_subslices * eu_stride;
+	total_length = sizeof(topo) + slice_length + subslice_length +
+		       eu_length;
 
 	ret = copy_query_item(&topo, sizeof(topo), total_length,
 			      query_item);
@@ -69,10 +69,9 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
 	topo.max_eus_per_subslice = sseu->max_eus_per_subslice;
 
 	topo.subslice_offset = slice_length;
-	topo.subslice_stride = DIV_ROUND_UP(sseu->max_subslices, BITS_PER_BYTE);
+	topo.subslice_stride = subslice_stride;
 	topo.eu_offset = slice_length + subslice_length;
-	topo.eu_stride =
-		DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE);
+	topo.eu_stride = eu_stride;
 
 	if (__copy_to_user(u64_to_user_ptr(query_item->data_ptr),
 			   &topo, sizeof(topo)))
Signature · 6 lines
2.21.0.5.gaeb582a983

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#7 lore raw

PATCH 3/6 drm/i915: Move calculation of subslices per slice to new function

In reply to Stuart Summers

Add a new function to return the number of subslices per slice to consolidate code usage.

v2: rebase on changes to move sseu struct to intel_sseu.h

Cc Daniele Ceraolo Spurio <[email protected]>
---
 drivers/gpu/drm/i915/gt/intel_sseu.h     | 6 ++++++
 drivers/gpu/drm/i915/i915_debugfs.c      | 2 +-
 drivers/gpu/drm/i915/intel_device_info.c | 4 ++--
 3 files changed, 9 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.h +6 -0
index c0b16b248d4c..f5ff6b7a756a 100644
@@ -63,6 +63,12 @@ intel_sseu_from_device_info(const struct sseu_dev_info *sseu)
 	return value;
 }
 
+static inline unsigned int
+sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice)
+{
+	return hweight8(sseu->subslice_mask[slice]);
+}
+
 u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
 			 const struct intel_sseu *req_sseu);
 
drivers/gpu/drm/i915/i915_debugfs.c +1 -1
index 0e4dffcd4da4..fe854c629a32 100644
@@ -4185,7 +4185,7 @@ static void i915_print_sseu_info(struct seq_file *m, bool is_available_info,
 		   sseu_subslice_total(sseu));
 	for (s = 0; s < fls(sseu->slice_mask); s++) {
 		seq_printf(m, "  %s Slice%i subslices: %u\n", type,
-			   s, hweight8(sseu->subslice_mask[s]));
+			   s, sseu_subslices_per_slice(sseu, s));
 	}
 	seq_printf(m, "  %s EU Total: %u\n", type,
 		   sseu->eu_total);
drivers/gpu/drm/i915/intel_device_info.c +2 -2
index 6af480b95bc6..559cf0d0628e 100644
@@ -93,7 +93,7 @@ static void sseu_dump(const struct sseu_dev_info *sseu, struct drm_printer *p)
 	drm_printf(p, "subslice total: %u\n", sseu_subslice_total(sseu));
 	for (s = 0; s < sseu->max_slices; s++) {
 		drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
-			   s, hweight8(sseu->subslice_mask[s]),
+			   s, sseu_subslices_per_slice(sseu, s),
 			   sseu->subslice_mask[s]);
 	}
 	drm_printf(p, "EU total: %u\n", sseu->eu_total);
@@ -126,7 +126,7 @@ void intel_device_info_dump_topology(const struct sseu_dev_info *sseu,
 
 	for (s = 0; s < sseu->max_slices; s++) {
 		drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
-			   s, hweight8(sseu->subslice_mask[s]),
+			   s, sseu_subslices_per_slice(sseu, s),
 			   sseu->subslice_mask[s]);
 
 		for (ss = 0; ss < sseu->max_subslices; ss++) {
Signature · 6 lines
2.21.0.5.gaeb582a983

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#10 lore raw

PATCH 4/6 drm/i915: Move sseu helper functions to intel_sseu.h

In reply to Stuart Summers

v2: fix spacing from checkpatch warning

---
 drivers/gpu/drm/i915/gt/intel_sseu.h     | 47 ++++++++++++++++++++++++
 drivers/gpu/drm/i915/intel_device_info.h | 47 ------------------------
 2 files changed, 47 insertions(+), 47 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.h +47 -0
index f5ff6b7a756a..029e71d8f140 100644
@@ -63,12 +63,59 @@ intel_sseu_from_device_info(const struct sseu_dev_info *sseu)
 	return value;
 }
 
+static inline unsigned int sseu_subslice_total(const struct sseu_dev_info *sseu)
+{
+	unsigned int i, total = 0;
+
+	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
+		total += hweight8(sseu->subslice_mask[i]);
+
+	return total;
+}
+
 static inline unsigned int
 sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice)
 {
 	return hweight8(sseu->subslice_mask[slice]);
 }
 
+static inline int sseu_eu_idx(const struct sseu_dev_info *sseu,
+			      int slice, int subslice)
+{
+	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
+					   BITS_PER_BYTE);
+	int slice_stride = sseu->max_subslices * subslice_stride;
+
+	return slice * slice_stride + subslice * subslice_stride;
+}
+
+static inline u16 sseu_get_eus(const struct sseu_dev_info *sseu,
+			       int slice, int subslice)
+{
+	int i, offset = sseu_eu_idx(sseu, slice, subslice);
+	u16 eu_mask = 0;
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
+			(i * BITS_PER_BYTE);
+	}
+
+	return eu_mask;
+}
+
+static inline void sseu_set_eus(struct sseu_dev_info *sseu,
+				int slice, int subslice, u16 eu_mask)
+{
+	int i, offset = sseu_eu_idx(sseu, slice, subslice);
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+		sseu->eu_mask[offset + i] =
+			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
+	}
+}
+
 u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
 			 const struct intel_sseu *req_sseu);
 
drivers/gpu/drm/i915/intel_device_info.h +0 -47

53 lines not on this page · raw message

Signature · 6 lines
2.21.0.5.gaeb582a983

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#13 lore raw

PATCH 5/6 drm/i915: Remove inline from sseu helper functions

In reply to Stuart Summers

Additionally, ensure these are all prefixed with intel_sseu_* to match the convention of other functions in i915.

---
 drivers/gpu/drm/i915/gt/intel_sseu.c     | 54 +++++++++++++++++++
 drivers/gpu/drm/i915/gt/intel_sseu.h     | 57 +++-----------------
 drivers/gpu/drm/i915/i915_debugfs.c      |  6 +--
 drivers/gpu/drm/i915/i915_drv.c          |  2 +-
 drivers/gpu/drm/i915/intel_device_info.c | 69 ++++++++++++------------
 5 files changed, 102 insertions(+), 86 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.c +54 -0
index 7f448f3bea0b..4a0b82fc108c 100644
@@ -8,6 +8,60 @@
 #include "intel_lrc_reg.h"
 #include "intel_sseu.h"
 
+unsigned int
+intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
+{
+	unsigned int i, total = 0;
+
+	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
+		total += hweight8(sseu->subslice_mask[i]);
+
+	return total;
+}
+
+unsigned int
+intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice)
+{
+	return hweight8(sseu->subslice_mask[slice]);
+}
+
+static int intel_sseu_eu_idx(const struct sseu_dev_info *sseu, int slice,
+			     int subslice)
+{
+	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
+					   BITS_PER_BYTE);
+	int slice_stride = sseu->max_subslices * subslice_stride;
+
+	return slice * slice_stride + subslice * subslice_stride;
+}
+
+u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int slice,
+		       int subslice)
+{
+	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
+	u16 eu_mask = 0;
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
+			(i * BITS_PER_BYTE);
+	}
+
+	return eu_mask;
+}
+
+void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice, int subslice,
+			u16 eu_mask)
+{
+	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+		sseu->eu_mask[offset + i] =
+			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
+	}
+}
+
 u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
 			 const struct intel_sseu *req_sseu)
 {
drivers/gpu/drm/i915/gt/intel_sseu.h +8 -49

66 lines not on this page · raw message

drivers/gpu/drm/i915/i915_debugfs.c +3 -3
index fe854c629a32..3f3ee83ac315 100644
@@ -4158,7 +4158,7 @@ static void broadwell_sseu_device_status(struct drm_i915_private *dev_priv,
 				RUNTIME_INFO(dev_priv)->sseu.subslice_mask[s];
 		}
 		sseu->eu_total = sseu->eu_per_subslice *
-				 sseu_subslice_total(sseu);
+				 intel_sseu_subslice_total(sseu);
 
 		/* subtract fused off EU(s) from enabled slice(s) */
 		for (s = 0; s < fls(sseu->slice_mask); s++) {
@@ -4182,10 +4182,10 @@ static void i915_print_sseu_info(struct seq_file *m, bool is_available_info,
 	seq_printf(m, "  %s Slice Total: %u\n", type,
 		   hweight8(sseu->slice_mask));
 	seq_printf(m, "  %s Subslice Total: %u\n", type,
-		   sseu_subslice_total(sseu));
+		   intel_sseu_subslice_total(sseu));
 	for (s = 0; s < fls(sseu->slice_mask); s++) {
 		seq_printf(m, "  %s Slice%i subslices: %u\n", type,
-			   s, sseu_subslices_per_slice(sseu, s));
+			   s, intel_sseu_subslices_per_slice(sseu, s));
 	}
 	seq_printf(m, "  %s EU Total: %u\n", type,
 		   sseu->eu_total);
drivers/gpu/drm/i915/i915_drv.c +1 -1
index c376244c19c4..130c5140db0d 100644
@@ -378,7 +378,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
 		value = i915_cmd_parser_get_version(dev_priv);
 		break;
 	case I915_PARAM_SUBSLICE_TOTAL:
-		value = sseu_subslice_total(sseu);
+		value = intel_sseu_subslice_total(sseu);
 		if (!value)
 			return -ENODEV;
 		break;
drivers/gpu/drm/i915/intel_device_info.c +36 -33

181 lines not on this page · raw message

Signature · 6 lines
2.21.0.5.gaeb582a983

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#22 lore raw

PATCH 6/6 drm/i915: Expand subslice mask

In reply to Stuart Summers

Currently, the subslice_mask runtime parameter is stored as an array of subslices per slice. Expand the subslice mask array to better match what is presented to userspace through the I915_QUERY_TOPOLOGY_INFO ioctl. The index into this array is then calculated:
slice * subslice stride + subslice index / 8

v2: fix spacing in set_sseu_info args
use set_sseu_info to initialize sseu data when building device status in debugfs
rename variables in intel_engine_types.h to avoid checkpatch warnings
v3: update headers in intel_sseu.h
v4: add const to some sseu_dev_info variables

    use sseu->eu_stride for EU stride calculations
Cc Daniele Ceraolo Spurio <[email protected]>
---
 drivers/gpu/drm/i915/gt/intel_engine_cs.c    |   6 +-
 drivers/gpu/drm/i915/gt/intel_engine_types.h |  32 +++--
 drivers/gpu/drm/i915/gt/intel_hangcheck.c    |   3 +-
 drivers/gpu/drm/i915/gt/intel_sseu.c         |  49 +++++--
 drivers/gpu/drm/i915/gt/intel_sseu.h         |  16 ++-
 drivers/gpu/drm/i915/gt/intel_workarounds.c  |   2 +-
 drivers/gpu/drm/i915/i915_debugfs.c          |  44 +++---
 drivers/gpu/drm/i915/i915_drv.c              |   6 +-
 drivers/gpu/drm/i915/i915_gpu_error.c        |   5 +-
 drivers/gpu/drm/i915/i915_query.c            |  10 +-
 drivers/gpu/drm/i915/intel_device_info.c     | 142 +++++++++++--------
 11 files changed, 198 insertions(+), 117 deletions(-)
drivers/gpu/drm/i915/gt/intel_engine_cs.c +4 -2
index 6e40f8ea9a6a..8f7967cc9a50 100644
@@ -914,7 +914,7 @@ u32 intel_calculate_mcr_s_ss_select(struct drm_i915_private *dev_priv)
 	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
 	u32 mcr_s_ss_select;
 	u32 slice = fls(sseu->slice_mask);
-	u32 subslice = fls(sseu->subslice_mask[slice]);
+	u32 subslice = fls(sseu->subslice_mask[slice * sseu->ss_stride]);
 
 	if (IS_GEN(dev_priv, 10))
 		mcr_s_ss_select = GEN8_MCR_SLICE(slice) |
@@ -990,6 +990,7 @@ void intel_engine_get_instdone(struct intel_engine_cs *engine,
 			       struct intel_instdone *instdone)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
 	struct intel_uncore *uncore = engine->uncore;
 	u32 mmio_base = engine->mmio_base;
 	int slice;
@@ -1007,7 +1008,8 @@ void intel_engine_get_instdone(struct intel_engine_cs *engine,
 
 		instdone->slice_common =
 			intel_uncore_read(uncore, GEN7_SC_INSTDONE);
-		for_each_instdone_slice_subslice(dev_priv, slice, subslice) {
+		for_each_instdone_slice_subslice(dev_priv, sseu, slice,
+						 subslice) {
 			instdone->sampler[slice][subslice] =
 				read_subslice_reg(dev_priv, slice, subslice,
 						  GEN7_SAMPLER_INSTDONE);
drivers/gpu/drm/i915/gt/intel_engine_types.h +17 -15
index 9d64e33f8427..1710546a2446 100644
@@ -534,20 +534,22 @@ intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine)
 	return engine->flags & I915_ENGINE_NEEDS_BREADCRUMB_TASKLET;
 }
 
-#define instdone_slice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.slice_mask)
-
-#define instdone_subslice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.subslice_mask[0])
-
-#define for_each_instdone_slice_subslice(dev_priv__, slice__, subslice__) \
-	for ((slice__) = 0, (subslice__) = 0; \
-	     (slice__) < I915_MAX_SLICES; \
-	     (subslice__) = ((subslice__) + 1) < I915_MAX_SUBSLICES ? (subslice__) + 1 : 0, \
-	       (slice__) += ((subslice__) == 0)) \
-		for_each_if((BIT(slice__) & instdone_slice_mask(dev_priv__)) && \
-			    (BIT(subslice__) & instdone_subslice_mask(dev_priv__)))
+#define instdone_has_slice(dev_priv___, sseu___, slice___) \
+	((IS_GEN(dev_priv___, 7) ? \
+	  1 : (sseu___)->slice_mask) & \
+	BIT(slice___)) \
+
+#define instdone_has_subslice(dev_priv__, sseu__, slice__, subslice__) \
+	((IS_GEN(dev_priv__, 7) ? \
+	  1 : (sseu__)->subslice_mask[slice__ * (sseu__)->ss_stride + \
+				      subslice__ / BITS_PER_BYTE]) & \
+	 BIT(subslice__ % BITS_PER_BYTE)) \
+
+#define for_each_instdone_slice_subslice(dev_priv_, sseu_, slice_, subslice_) \
+	for ((slice_) = 0, (subslice_) = 0; (slice_) < I915_MAX_SLICES; \
+	     (subslice_) = ((subslice_) + 1) < I915_MAX_SUBSLICES ? (subslice_) + 1 : 0, \
+	       (slice_) += ((subslice_) == 0)) \
+		for_each_if(instdone_has_slice(dev_priv_, sseu_, slice) && \
+			    instdone_has_subslice(dev_priv_, sseu_, slice_, subslice_)) \
 
 #endif /* __INTEL_ENGINE_TYPES_H__ */
drivers/gpu/drm/i915/gt/intel_hangcheck.c +2 -1
index e5eaa06fe74d..53c1c98161e1 100644
@@ -50,6 +50,7 @@ static bool instdone_unchanged(u32 current_instdone, u32 *old_instdone)
 static bool subunits_stuck(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *dev_priv = engine->i915;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
 	struct intel_instdone instdone;
 	struct intel_instdone *accu_instdone = &engine->hangcheck.instdone;
 	bool stuck;
@@ -71,7 +72,7 @@ static bool subunits_stuck(struct intel_engine_cs *engine)
 	stuck &= instdone_unchanged(instdone.slice_common,
 				    &accu_instdone->slice_common);
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice) {
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice, subslice) {
 		stuck &= instdone_unchanged(instdone.sampler[slice][subslice],
 					    &accu_instdone->sampler[slice][subslice]);
 		stuck &= instdone_unchanged(instdone.row[slice][subslice],
drivers/gpu/drm/i915/gt/intel_sseu.c +40 -9

79 lines not on this page · raw message

drivers/gpu/drm/i915/gt/intel_sseu.h +15 -1
index 56e3721ae83f..bf01f338a8cc 100644
@@ -9,16 +9,18 @@
 
 #include <linux/types.h>
 #include <linux/kernel.h>
+#include <linux/string.h>
 
 struct drm_i915_private;
 
 #define GEN_MAX_SLICES		(6) /* CNL upper bound */
 #define GEN_MAX_SUBSLICES	(8) /* ICL upper bound */
 #define GEN_SSEU_STRIDE(bits) DIV_ROUND_UP(bits, BITS_PER_BYTE)
+#define GEN_MAX_SUBSLICE_STRIDE GEN_SSEU_STRIDE(GEN_MAX_SUBSLICES)
 
 struct sseu_dev_info {
 	u8 slice_mask;
-	u8 subslice_mask[GEN_MAX_SLICES];
+	u8 subslice_mask[GEN_MAX_SLICES * GEN_MAX_SUBSLICE_STRIDE];
 	u16 eu_total;
 	u8 eu_per_subslice;
 	u8 min_eu_in_pool;
@@ -33,6 +35,9 @@ struct sseu_dev_info {
 	u8 max_subslices;
 	u8 max_eus_per_subslice;
 
+	u8 ss_stride;
+	u8 eu_stride;
+
 	/* We don't have more than 8 eus per subslice at the moment and as we
 	 * store eus enabled using bits, no need to multiply by eus per
 	 * subslice.
@@ -63,12 +68,21 @@ intel_sseu_from_device_info(const struct sseu_dev_info *sseu)
 	return value;
 }
 
+void intel_sseu_set_info(struct sseu_dev_info *sseu, u8 max_slices,
+			 u8 max_subslices, u8 max_eus_per_subslice);
+
 unsigned int
 intel_sseu_subslice_total(const struct sseu_dev_info *sseu);
 
 unsigned int
 intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice);
 
+void intel_sseu_copy_subslices(const struct sseu_dev_info *sseu, int slice,
+			       u8 *to_mask, const u8 *from_mask);
+
+void intel_sseu_set_subslices(struct sseu_dev_info *sseu, int slice,
+			      u32 ss_mask);
+
 u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int slice,
 		       int subslice);
 
drivers/gpu/drm/i915/gt/intel_workarounds.c +1 -1
index 43e290306551..7c7e9556c1c5 100644
@@ -767,7 +767,7 @@ wa_init_mcr(struct drm_i915_private *i915, struct i915_wa_list *wal)
 		u32 slice = fls(sseu->slice_mask);
 		u32 fuse3 =
 			intel_uncore_read(&i915->uncore, GEN10_MIRROR_FUSE3);
-		u8 ss_mask = sseu->subslice_mask[slice];
+		u8 ss_mask = sseu->subslice_mask[slice * sseu->ss_stride];
 
 		u8 enabled_mask = (ss_mask | ss_mask >>
 				   GEN10_L3BANK_PAIR_COUNT) & GEN10_L3BANK_MASK;
drivers/gpu/drm/i915/i915_debugfs.c +25 -19

112 lines not on this page · raw message

drivers/gpu/drm/i915/i915_drv.c +4 -2
index 130c5140db0d..6afe4e3afea4 100644
@@ -326,7 +326,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
 	struct pci_dev *pdev = dev_priv->drm.pdev;
 	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
 	drm_i915_getparam_t *param = data;
-	int value;
+	int value = 0;
 
 	switch (param->param) {
 	case I915_PARAM_IRQ_ACTIVE:
@@ -455,7 +455,9 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
 			return -ENODEV;
 		break;
 	case I915_PARAM_SUBSLICE_MASK:
-		value = sseu->subslice_mask[0];
+		/* Only copy bits from the first subslice */
+		memcpy(&value, sseu->subslice_mask,
+		       min(sseu->ss_stride, (u8)sizeof(value)));
 		if (!value)
 			return -ENODEV;
 		break;
drivers/gpu/drm/i915/i915_gpu_error.c +3 -2
index e1b858bd1d32..140918dd9b7d 100644
@@ -407,6 +407,7 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
 static void error_print_instdone(struct drm_i915_error_state_buf *m,
 				 const struct drm_i915_error_engine *ee)
 {
+	struct sseu_dev_info *sseu = &RUNTIME_INFO(m->i915)->sseu;
 	int slice;
 	int subslice;
 
@@ -422,12 +423,12 @@ static void error_print_instdone(struct drm_i915_error_state_buf *m,
 	if (INTEL_GEN(m->i915) <= 6)
 		return;
 
-	for_each_instdone_slice_subslice(m->i915, slice, subslice)
+	for_each_instdone_slice_subslice(m->i915, sseu, slice, subslice)
 		err_printf(m, "  SAMPLER_INSTDONE[%d][%d]: 0x%08x\n",
 			   slice, subslice,
 			   ee->instdone.sampler[slice][subslice]);
 
-	for_each_instdone_slice_subslice(m->i915, slice, subslice)
+	for_each_instdone_slice_subslice(m->i915, sseu, slice, subslice)
 		err_printf(m, "  ROW_INSTDONE[%d][%d]: 0x%08x\n",
 			   slice, subslice,
 			   ee->instdone.row[slice][subslice]);
drivers/gpu/drm/i915/i915_query.c +4 -6
index 7c1708c22811..000dcb145ce0 100644
@@ -37,8 +37,6 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
 	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
 	struct drm_i915_query_topology_info topo;
 	u32 slice_length, subslice_length, eu_length, total_length;
-	u8 subslice_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
-	u8 eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
 	int ret;
 
 	if (query_item->flags != 0)
@@ -50,8 +48,8 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
 	BUILD_BUG_ON(sizeof(u8) != sizeof(sseu->slice_mask));
 
 	slice_length = sizeof(sseu->slice_mask);
-	subslice_length = sseu->max_slices * subslice_stride;
-	eu_length = sseu->max_slices * sseu->max_subslices * eu_stride;
+	subslice_length = sseu->max_slices * sseu->ss_stride;
+	eu_length = sseu->max_slices * sseu->max_subslices * sseu->eu_stride;
 	total_length = sizeof(topo) + slice_length + subslice_length +
 		       eu_length;
 
@@ -69,9 +67,9 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
 	topo.max_eus_per_subslice = sseu->max_eus_per_subslice;
 
 	topo.subslice_offset = slice_length;
-	topo.subslice_stride = subslice_stride;
+	topo.subslice_stride = sseu->ss_stride;
 	topo.eu_offset = slice_length + subslice_length;
-	topo.eu_stride = eu_stride;
+	topo.eu_stride = sseu->eu_stride;
 
 	if (__copy_to_user(u64_to_user_ptr(query_item->data_ptr),
 			   &topo, sizeof(topo)))
drivers/gpu/drm/i915/intel_device_info.c +83 -59

300 lines not on this page · raw message

Signature · 6 lines
2.21.0.5.gaeb582a983

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#30 lore raw

✗ Fi.CI.CHECKPATCH: warning for Refactor to expand subslice mask (rev7)

In reply to Stuart Summers
== Series Details ==

== Series Details ==

Series: Refactor to expand subslice mask (rev7)
URL : https://patchwork.freedesktop.org/series/59742/
State : warning

== Summary ==

$ dim checkpatch origin/drm-tip
47812364a20c drm/i915: Use local variable for SSEU info in GETPARAM ioctl
b0f425609a7b drm/i915: Add macro for SSEU stride calculation
be21e44a771d drm/i915: Move calculation of subslices per slice to new function
0dbec2982376 drm/i915: Move sseu helper functions to intel_sseu.h
bf909e717bcf drm/i915: Remove inline from sseu helper functions
3416f0a72f5d drm/i915: Expand subslice mask
-:84: CHECK:MACRO_ARG_REUSE: Macro argument reuse 'sseu__' - possible side-effects?
#84: FILE: drivers/gpu/drm/i915/gt/intel_engine_types.h:542:
+#define instdone_has_subslice(dev_priv__, sseu__, slice__, subslice__) \
+	((IS_GEN(dev_priv__, 7) ? \
+	  1 : (sseu__)->subslice_mask[slice__ * (sseu__)->ss_stride + \
+				      subslice__ / BITS_PER_BYTE]) & \
+	 BIT(subslice__ % BITS_PER_BYTE)) \
+

-:84: CHECK:MACRO_ARG_PRECEDENCE: Macro argument 'slice__' may be better as '(slice__)' to avoid precedence issues #84: FILE: drivers/gpu/drm/i915/gt/intel_engine_types.h:542:
+#define instdone_has_subslice(dev_priv__, sseu__, slice__, subslice__) \
+ ((IS_GEN(dev_priv__, 7) ? \
+ 1 : (sseu__)->subslice_mask[slice__ * (sseu__)->ss_stride + \
+ subslice__ / BITS_PER_BYTE]) & \
+ BIT(subslice__ % BITS_PER_BYTE)) \
+

-:84: CHECK:MACRO_ARG_REUSE: Macro argument reuse 'subslice__' - possible side-effects? #84: FILE: drivers/gpu/drm/i915/gt/intel_engine_types.h:542:
+#define instdone_has_subslice(dev_priv__, sseu__, slice__, subslice__) \
+ ((IS_GEN(dev_priv__, 7) ? \
+ 1 : (sseu__)->subslice_mask[slice__ * (sseu__)->ss_stride + \
+ subslice__ / BITS_PER_BYTE]) & \
+ BIT(subslice__ % BITS_PER_BYTE)) \
+

-:84: CHECK:MACRO_ARG_PRECEDENCE: Macro argument 'subslice__' may be better as '(subslice__)' to avoid precedence issues #84: FILE: drivers/gpu/drm/i915/gt/intel_engine_types.h:542:
+#define instdone_has_subslice(dev_priv__, sseu__, slice__, subslice__) \
+ ((IS_GEN(dev_priv__, 7) ? \
+ 1 : (sseu__)->subslice_mask[slice__ * (sseu__)->ss_stride + \
+ subslice__ / BITS_PER_BYTE]) & \
+ BIT(subslice__ % BITS_PER_BYTE)) \
+

-:90: CHECK:MACRO_ARG_REUSE: Macro argument reuse 'dev_priv_' - possible side-effects? #90: FILE: drivers/gpu/drm/i915/gt/intel_engine_types.h:548:
+#define for_each_instdone_slice_subslice(dev_priv_, sseu_, slice_, subslice_) \
+ for ((slice_) = 0, (subslice_) = 0; (slice_) < I915_MAX_SLICES; \
+ (subslice_) = ((subslice_) + 1) < I915_MAX_SUBSLICES ? (subslice_) + 1 : 0, \
+ (slice_) += ((subslice_) == 0)) \
+ for_each_if(instdone_has_slice(dev_priv_, sseu_, slice) && \
+ instdone_has_subslice(dev_priv_, sseu_, slice_, subslice_)) \

-:90: CHECK:MACRO_ARG_REUSE: Macro argument reuse 'sseu_' - possible side-effects? #90: FILE: drivers/gpu/drm/i915/gt/intel_engine_types.h:548:
+#define for_each_instdone_slice_subslice(dev_priv_, sseu_, slice_, subslice_) \
+ for ((slice_) = 0, (subslice_) = 0; (slice_) < I915_MAX_SLICES; \
+ (subslice_) = ((subslice_) + 1) < I915_MAX_SUBSLICES ? (subslice_) + 1 : 0, \
+ (slice_) += ((subslice_) == 0)) \
+ for_each_if(instdone_has_slice(dev_priv_, sseu_, slice) && \
+ instdone_has_subslice(dev_priv_, sseu_, slice_, subslice_)) \

-:90: CHECK:MACRO_ARG_REUSE: Macro argument reuse 'slice_' - possible side-effects? #90: FILE: drivers/gpu/drm/i915/gt/intel_engine_types.h:548:
+#define for_each_instdone_slice_subslice(dev_priv_, sseu_, slice_, subslice_) \
+ for ((slice_) = 0, (subslice_) = 0; (slice_) < I915_MAX_SLICES; \
+ (subslice_) = ((subslice_) + 1) < I915_MAX_SUBSLICES ? (subslice_) + 1 : 0, \
+ (slice_) += ((subslice_) == 0)) \
+ for_each_if(instdone_has_slice(dev_priv_, sseu_, slice) && \
+ instdone_has_subslice(dev_priv_, sseu_, slice_, subslice_)) \

-:90: CHECK:MACRO_ARG_REUSE: Macro argument reuse 'subslice_' - possible side-effects? #90: FILE: drivers/gpu/drm/i915/gt/intel_engine_types.h:548:
+#define for_each_instdone_slice_subslice(dev_priv_, sseu_, slice_, subslice_) \
+ for ((slice_) = 0, (subslice_) = 0; (slice_) < I915_MAX_SLICES; \
+ (subslice_) = ((subslice_) + 1) < I915_MAX_SUBSLICES ? (subslice_) + 1 : 0, \
+ (slice_) += ((subslice_) == 0)) \
+ for_each_if(instdone_has_slice(dev_priv_, sseu_, slice) && \
+ instdone_has_subslice(dev_priv_, sseu_, slice_, subslice_)) \

total: 0 errors, 0 warnings, 8 checks, 692 lines checked

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#31 lore raw

✗ Fi.CI.SPARSE: warning for Refactor to expand subslice mask (rev7)

In reply to Stuart Summers
== Series Details ==

== Series Details ==

Series: Refactor to expand subslice mask (rev7)
URL : https://patchwork.freedesktop.org/series/59742/
State : warning

== Summary ==

$ dim sparse origin/drm-tip
Sparse version: v0.5.2
Commit: drm/i915: Use local variable for SSEU info in GETPARAM ioctl
Okay!

Commit: drm/i915: Add macro for SSEU stride calculation Okay!

Commit: drm/i915: Move calculation of subslices per slice to new function Okay!

Commit: drm/i915: Move sseu helper functions to intel_sseu.h Okay!

Commit: drm/i915: Remove inline from sseu helper functions Okay!

Commit: drm/i915: Expand subslice mask +drivers/gpu/drm/i915/i915_drv.c:460:24: warning: expression using sizeof(void)

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#32 lore raw

✓ Fi.CI.BAT: success for Refactor to expand subslice mask (rev7)

In reply to Stuart Summers
== Series Details ==

== Series Details ==

Series: Refactor to expand subslice mask (rev7)
URL : https://patchwork.freedesktop.org/series/59742/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_6021 -> Patchwork_12926
====================================================

Summary
-------

**SUCCESS**

No regressions found.

External URL: https://patchwork.freedesktop.org/api/1.0/series/59742/revisions/7/mbox/

Known issues
------------

Here are the changes found in Patchwork_12926 that come from known issues:

### IGT changes ###

#### Issues hit ####

* igt@gem_ctx_create@basic-files:
- fi-icl-y: [PASS][1] -> [INCOMPLETE][2] ([fdo#107713] / [fdo#109100])
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/fi-icl-y/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/fi-icl-y/igt@[email protected]

* igt@i915_selftest@live_contexts:
- fi-bdw-gvtdvm: [PASS][3] -> [DMESG-FAIL][4] ([fdo#110235])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/fi-bdw-gvtdvm/igt@i915_selftest@live_contexts.html
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/fi-bdw-gvtdvm/igt@i915_selftest@live_contexts.html

* igt@kms_cursor_legacy@basic-flip-after-cursor-varying-size:
- fi-glk-dsi: [PASS][5] -> [INCOMPLETE][6] ([fdo#103359] / [k.org#198133])
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/fi-glk-dsi/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/fi-glk-dsi/igt@[email protected]

#### Possible fixes ####

* igt@kms_pipe_crc_basic@suspend-read-crc-pipe-a:
- fi-blb-e6850: [INCOMPLETE][7] ([fdo#107718]) -> [PASS][8]
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/fi-blb-e6850/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/fi-blb-e6850/igt@[email protected]

#### Warnings ####

* igt@i915_pm_rpm@basic-pci-d3-state:
- fi-kbl-guc: [SKIP][9] ([fdo#109271]) -> [INCOMPLETE][10] ([fdo#107807])
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/fi-kbl-guc/igt@[email protected]
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/fi-kbl-guc/igt@[email protected]

[fdo#103359]: https://bugs.freedesktop.org/show_bug.cgi?id=103359
[fdo#107713]: https://bugs.freedesktop.org/show_bug.cgi?id=107713
[fdo#107718]: https://bugs.freedesktop.org/show_bug.cgi?id=107718
[fdo#107807]: https://bugs.freedesktop.org/show_bug.cgi?id=107807
[fdo#109100]: https://bugs.freedesktop.org/show_bug.cgi?id=109100
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#110235]: https://bugs.freedesktop.org/show_bug.cgi?id=110235
[k.org#198133]: https://bugzilla.kernel.org/show_bug.cgi?id=198133

Participating hosts (52 -> 44)
------------------------------

Additional (1): fi-pnv-d510
Missing (9): fi-kbl-soraka fi-ilk-m540 fi-hsw-4200u fi-bsw-cyan fi-ctg-p8600 fi-cfl-8109u fi-icl-u3 fi-byt-clapper fi-bdw-samus

Build changes
-------------

* Linux: CI_DRM_6021 -> Patchwork_12926

CI_DRM_6021: 850aa4220e8bf7609b03bf89bce146305704bec6 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4971: fc5e0467eb6913d21ad932aa8a31c77fdb5a9c77 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_12926: 3416f0a72f5df37142ab06ae57f55bff003ecf5d @ git://anongit.freedesktop.org/gfx-ci/linux

== Linux commits ==

3416f0a72f5d drm/i915: Expand subslice mask
bf909e717bcf drm/i915: Remove inline from sseu helper functions 0dbec2982376 drm/i915: Move sseu helper functions to intel_sseu.h be21e44a771d drm/i915: Move calculation of subslices per slice to new function b0f425609a7b drm/i915: Add macro for SSEU stride calculation 47812364a20c drm/i915: Use local variable for SSEU info in GETPARAM ioctl

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Reviewed-by on v1 1/6
#2 lore raw

Re: [PATCH 1/6] drm/i915: Use local variable for SSEU info in GETPARAM ioctl

In reply to Stuart Summers
Quoted 7 lines · Stuart Summers

In the GETPARAM ioctl handler, use a local variable to consolidate usage of SSEU runtime info.

v2: add const to sseu_dev_info variable

Cc Daniele Ceraolo Spurio <[email protected]>
Quoted 46 lines
---
  drivers/gpu/drm/i915/i915_drv.c | 11 ++++++-----
  1 file changed, 6 insertions(+), 5 deletions(-)
drivers/gpu/drm/i915/i915_drv.c +4 -4
index 21dac5a09fbe..c376244c19c4 100644
@@ -324,6 +324,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
  {
  	struct drm_i915_private *dev_priv = to_i915(dev);
  	struct pci_dev *pdev = dev_priv->drm.pdev;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	drm_i915_getparam_t *param = data;
  	int value;
 
@@ -377,12 +378,12 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
  		value = i915_cmd_parser_get_version(dev_priv);
  		break;
  	case I915_PARAM_SUBSLICE_TOTAL:
-		value = sseu_subslice_total(&RUNTIME_INFO(dev_priv)->sseu);
+		value = sseu_subslice_total(sseu);
  		if (!value)
  			return -ENODEV;
  		break;
  	case I915_PARAM_EU_TOTAL:
-		value = RUNTIME_INFO(dev_priv)->sseu.eu_total;
+		value = sseu->eu_total;
  		if (!value)
  			return -ENODEV;
  		break;
@@ -399,7 +400,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
  		value = HAS_POOLED_EU(dev_priv);
  		break;
  	case I915_PARAM_MIN_EU_IN_POOL:
-		value = RUNTIME_INFO(dev_priv)->sseu.min_eu_in_pool;
+		value = sseu->min_eu_in_pool;
  		break;
  	case I915_PARAM_HUC_STATUS:
  		value = intel_huc_check_status(&dev_priv->huc);
@@ -449,12 +450,12 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
  		value = intel_engines_has_context_isolation(dev_priv);
  		break;
  	case I915_PARAM_SLICE_MASK:
-		value = RUNTIME_INFO(dev_priv)->sseu.slice_mask;

drivers/gpu/drm/i915/i915_drv.c static int i915_getparam_ioctl(struct drm_device *dev, void *data, · line 461

Quoted 10 lines
drivers/gpu/drm/i915/i915_drv.c +2 -1
+		value = sseu->slice_mask;
  		if (!value)
  			return -ENODEV;
  		break;
  	case I915_PARAM_SUBSLICE_MASK:
-		value = RUNTIME_INFO(dev_priv)->sseu.subslice_mask[0];
+		value = sseu->subslice_mask[0];
  		if (!value)
  			return -ENODEV;
  		break;

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Reviewed-by on v1 2/6
#5 lore raw

Re: [PATCH 2/6] drm/i915: Add macro for SSEU stride calculation

In reply to Stuart Summers
Quoted 17 lines · Stuart Summers

Subslice stride and EU stride are calculated multiple times in i915_query. Move this calculation to a macro to reduce code duplication.

v2: update headers in intel_sseu.h

Cc Daniele Ceraolo Spurio <[email protected]>
---
  drivers/gpu/drm/i915/gt/intel_sseu.h |  2 ++
  drivers/gpu/drm/i915/i915_query.c    | 17 ++++++++---------
  2 files changed, 10 insertions(+), 9 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.h +0 -0
index 73bc824094e8..c0b16b248d4c 100644

drivers/gpu/drm/i915/gt/intel_sseu.h line 17

Quoted 10 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +2 -0
@@ -8,11 +8,13 @@
  #define __INTEL_SSEU_H__
 
  #include <linux/types.h>
+#include <linux/kernel.h>
 
  struct drm_i915_private;
 
  #define GEN_MAX_SLICES		(6) /* CNL upper bound */
  #define GEN_MAX_SUBSLICES	(8) /* ICL upper bound */
+#define GEN_SSEU_STRIDE(bits) DIV_ROUND_UP(bits, BITS_PER_BYTE)

What we pass to this macro isn't really a bits count but the maximum amount of s/ss/eus. s/bits/max_entry/, or something like that? with that:

Daniele

Quoted 37 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +0 -0
  struct sseu_dev_info {
  	u8 slice_mask;
drivers/gpu/drm/i915/i915_query.c +6 -6
index 782183b78f49..7c1708c22811 100644
@@ -37,6 +37,8 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	struct drm_i915_query_topology_info topo;
  	u32 slice_length, subslice_length, eu_length, total_length;
+	u8 subslice_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
+	u8 eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
  	int ret;
 
  	if (query_item->flags != 0)
@@ -48,12 +50,10 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
  	BUILD_BUG_ON(sizeof(u8) != sizeof(sseu->slice_mask));
 
  	slice_length = sizeof(sseu->slice_mask);
-	subslice_length = sseu->max_slices *
-		DIV_ROUND_UP(sseu->max_subslices, BITS_PER_BYTE);
-	eu_length = sseu->max_slices * sseu->max_subslices *
-		DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE);
-
-	total_length = sizeof(topo) + slice_length + subslice_length + eu_length;
+	subslice_length = sseu->max_slices * subslice_stride;
+	eu_length = sseu->max_slices * sseu->max_subslices * eu_stride;
+	total_length = sizeof(topo) + slice_length + subslice_length +
+		       eu_length;
 
  	ret = copy_query_item(&topo, sizeof(topo), total_length,
  			      query_item);
@@ -69,10 +69,9 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
  	topo.max_eus_per_subslice = sseu->max_eus_per_subslice;
 

drivers/gpu/drm/i915/i915_query.c static int query_topology_info(struct drm_i915_private *dev_priv, · line 77

Quoted 10 lines
drivers/gpu/drm/i915/i915_query.c +2 -3
  	topo.subslice_offset = slice_length;
-	topo.subslice_stride = DIV_ROUND_UP(sseu->max_subslices, BITS_PER_BYTE);
+	topo.subslice_stride = subslice_stride;
  	topo.eu_offset = slice_length + subslice_length;
-	topo.eu_stride =
-		DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE);
+	topo.eu_stride = eu_stride;
 
  	if (__copy_to_user(u64_to_user_ptr(query_item->data_ptr),
  			   &topo, sizeof(topo)))

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Reviewed-by on v1 3/6
#8 lore raw

Re: [PATCH 3/6] drm/i915: Move calculation of subslices per slice to new function

In reply to Stuart Summers
Quoted 18 lines · Stuart Summers

Add a new function to return the number of subslices per slice to consolidate code usage.

v2: rebase on changes to move sseu struct to intel_sseu.h

Cc Daniele Ceraolo Spurio <[email protected]>
---
  drivers/gpu/drm/i915/gt/intel_sseu.h     | 6 ++++++
  drivers/gpu/drm/i915/i915_debugfs.c      | 2 +-
  drivers/gpu/drm/i915/intel_device_info.c | 4 ++--
  3 files changed, 9 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.h +0 -0
index c0b16b248d4c..f5ff6b7a756a 100644

drivers/gpu/drm/i915/gt/intel_sseu.h intel_sseu_from_device_info(const struct sseu_dev_info *sseu) · line 67

Quoted 5 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +2 -0
@@ -63,6 +63,12 @@ intel_sseu_from_device_info(const struct sseu_dev_info *sseu)
  	return value;
  }
 
+static inline unsigned int
+sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice)

This is exposed, so needs an intel_* prefix. with that:

Daniele

Quoted 35 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +4 -0
+{
+	return hweight8(sseu->subslice_mask[slice]);
+}
+
  u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
  			 const struct intel_sseu *req_sseu);
 
drivers/gpu/drm/i915/i915_debugfs.c +1 -1
index 0e4dffcd4da4..fe854c629a32 100644
@@ -4185,7 +4185,7 @@ static void i915_print_sseu_info(struct seq_file *m, bool is_available_info,
  		   sseu_subslice_total(sseu));
  	for (s = 0; s < fls(sseu->slice_mask); s++) {
  		seq_printf(m, "  %s Slice%i subslices: %u\n", type,
-			   s, hweight8(sseu->subslice_mask[s]));
+			   s, sseu_subslices_per_slice(sseu, s));
  	}
  	seq_printf(m, "  %s EU Total: %u\n", type,
  		   sseu->eu_total);
drivers/gpu/drm/i915/intel_device_info.c +1 -1
index 6af480b95bc6..559cf0d0628e 100644
@@ -93,7 +93,7 @@ static void sseu_dump(const struct sseu_dev_info *sseu, struct drm_printer *p)
  	drm_printf(p, "subslice total: %u\n", sseu_subslice_total(sseu));
  	for (s = 0; s < sseu->max_slices; s++) {
  		drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
-			   s, hweight8(sseu->subslice_mask[s]),
+			   s, sseu_subslices_per_slice(sseu, s),
  			   sseu->subslice_mask[s]);
  	}
  	drm_printf(p, "EU total: %u\n", sseu->eu_total);

drivers/gpu/drm/i915/intel_device_info.c void intel_device_info_dump_topology(const struct sseu_dev_info *sseu, · line 132

Quoted 8 lines
drivers/gpu/drm/i915/intel_device_info.c +1 -1
@@ -126,7 +126,7 @@ void intel_device_info_dump_topology(const struct sseu_dev_info *sseu,
 
  	for (s = 0; s < sseu->max_slices; s++) {
  		drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
-			   s, hweight8(sseu->subslice_mask[s]),
+			   s, sseu_subslices_per_slice(sseu, s),
  			   sseu->subslice_mask[s]);
 
  		for (ss = 0; ss < sseu->max_subslices; ss++) {

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#23 lore raw

Re: [PATCH 6/6] drm/i915: Expand subslice mask

In reply to Stuart Summers

Just one drive by below...

Quoted 501 lines · Stuart Summers

Currently, the subslice_mask runtime parameter is stored as an array of subslices per slice. Expand the subslice mask array to better match what is presented to userspace through the I915_QUERY_TOPOLOGY_INFO ioctl. The index into this array is then calculated:
slice * subslice stride + subslice index / 8

v2: fix spacing in set_sseu_info args
use set_sseu_info to initialize sseu data when building device status in debugfs
rename variables in intel_engine_types.h to avoid checkpatch warnings
v3: update headers in intel_sseu.h
v4: add const to some sseu_dev_info variables

     use sseu->eu_stride for EU stride calculations
Cc Daniele Ceraolo Spurio <[email protected]>
---
  drivers/gpu/drm/i915/gt/intel_engine_cs.c    |   6 +-
  drivers/gpu/drm/i915/gt/intel_engine_types.h |  32 +++--
  drivers/gpu/drm/i915/gt/intel_hangcheck.c    |   3 +-
  drivers/gpu/drm/i915/gt/intel_sseu.c         |  49 +++++--
  drivers/gpu/drm/i915/gt/intel_sseu.h         |  16 ++-
  drivers/gpu/drm/i915/gt/intel_workarounds.c  |   2 +-
  drivers/gpu/drm/i915/i915_debugfs.c          |  44 +++---
  drivers/gpu/drm/i915/i915_drv.c              |   6 +-
  drivers/gpu/drm/i915/i915_gpu_error.c        |   5 +-
  drivers/gpu/drm/i915/i915_query.c            |  10 +-
  drivers/gpu/drm/i915/intel_device_info.c     | 142 +++++++++++--------
  11 files changed, 198 insertions(+), 117 deletions(-)
drivers/gpu/drm/i915/gt/intel_engine_cs.c +4 -2
index 6e40f8ea9a6a..8f7967cc9a50 100644
@@ -914,7 +914,7 @@ u32 intel_calculate_mcr_s_ss_select(struct drm_i915_private *dev_priv)
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	u32 mcr_s_ss_select;
  	u32 slice = fls(sseu->slice_mask);
-	u32 subslice = fls(sseu->subslice_mask[slice]);
+	u32 subslice = fls(sseu->subslice_mask[slice * sseu->ss_stride]);
 
  	if (IS_GEN(dev_priv, 10))
  		mcr_s_ss_select = GEN8_MCR_SLICE(slice) |
@@ -990,6 +990,7 @@ void intel_engine_get_instdone(struct intel_engine_cs *engine,
  			       struct intel_instdone *instdone)
  {
  	struct drm_i915_private *dev_priv = engine->i915;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	struct intel_uncore *uncore = engine->uncore;
  	u32 mmio_base = engine->mmio_base;
  	int slice;
@@ -1007,7 +1008,8 @@ void intel_engine_get_instdone(struct intel_engine_cs *engine,
 
  		instdone->slice_common =
  			intel_uncore_read(uncore, GEN7_SC_INSTDONE);
-		for_each_instdone_slice_subslice(dev_priv, slice, subslice) {
+		for_each_instdone_slice_subslice(dev_priv, sseu, slice,
+						 subslice) {
  			instdone->sampler[slice][subslice] =
  				read_subslice_reg(dev_priv, slice, subslice,
  						  GEN7_SAMPLER_INSTDONE);
drivers/gpu/drm/i915/gt/intel_engine_types.h +17 -15
index 9d64e33f8427..1710546a2446 100644
@@ -534,20 +534,22 @@ intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine)
  	return engine->flags & I915_ENGINE_NEEDS_BREADCRUMB_TASKLET;
  }
 
-#define instdone_slice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.slice_mask)
-
-#define instdone_subslice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.subslice_mask[0])
-
-#define for_each_instdone_slice_subslice(dev_priv__, slice__, subslice__) \
-	for ((slice__) = 0, (subslice__) = 0; \
-	     (slice__) < I915_MAX_SLICES; \
-	     (subslice__) = ((subslice__) + 1) < I915_MAX_SUBSLICES ? (subslice__) + 1 : 0, \
-	       (slice__) += ((subslice__) == 0)) \
-		for_each_if((BIT(slice__) & instdone_slice_mask(dev_priv__)) && \
-			    (BIT(subslice__) & instdone_subslice_mask(dev_priv__)))
+#define instdone_has_slice(dev_priv___, sseu___, slice___) \
+	((IS_GEN(dev_priv___, 7) ? \
+	  1 : (sseu___)->slice_mask) & \
+	BIT(slice___)) \
+
+#define instdone_has_subslice(dev_priv__, sseu__, slice__, subslice__) \
+	((IS_GEN(dev_priv__, 7) ? \
+	  1 : (sseu__)->subslice_mask[slice__ * (sseu__)->ss_stride + \
+				      subslice__ / BITS_PER_BYTE]) & \
+	 BIT(subslice__ % BITS_PER_BYTE)) \
+
+#define for_each_instdone_slice_subslice(dev_priv_, sseu_, slice_, subslice_) \
+	for ((slice_) = 0, (subslice_) = 0; (slice_) < I915_MAX_SLICES; \
+	     (subslice_) = ((subslice_) + 1) < I915_MAX_SUBSLICES ? (subslice_) + 1 : 0, \
+	       (slice_) += ((subslice_) == 0)) \
+		for_each_if(instdone_has_slice(dev_priv_, sseu_, slice) && \
+			    instdone_has_subslice(dev_priv_, sseu_, slice_, subslice_)) \
 
  #endif /* __INTEL_ENGINE_TYPES_H__ */
drivers/gpu/drm/i915/gt/intel_hangcheck.c +2 -1
index e5eaa06fe74d..53c1c98161e1 100644
@@ -50,6 +50,7 @@ static bool instdone_unchanged(u32 current_instdone, u32 *old_instdone)
  static bool subunits_stuck(struct intel_engine_cs *engine)
  {
  	struct drm_i915_private *dev_priv = engine->i915;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	struct intel_instdone instdone;
  	struct intel_instdone *accu_instdone = &engine->hangcheck.instdone;
  	bool stuck;
@@ -71,7 +72,7 @@ static bool subunits_stuck(struct intel_engine_cs *engine)
  	stuck &= instdone_unchanged(instdone.slice_common,
  				    &accu_instdone->slice_common);
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice) {
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice, subslice) {
  		stuck &= instdone_unchanged(instdone.sampler[slice][subslice],
  					    &accu_instdone->sampler[slice][subslice]);
  		stuck &= instdone_unchanged(instdone.row[slice][subslice],
drivers/gpu/drm/i915/gt/intel_sseu.c +40 -9
index 4a0b82fc108c..49316b7ef074 100644
@@ -8,6 +8,17 @@
  #include "intel_lrc_reg.h"
  #include "intel_sseu.h"
 
+void intel_sseu_set_info(struct sseu_dev_info *sseu, u8 max_slices,
+			 u8 max_subslices, u8 max_eus_per_subslice)
+{
+	sseu->max_slices = max_slices;
+	sseu->max_subslices = max_subslices;
+	sseu->max_eus_per_subslice = max_eus_per_subslice;
+
+	sseu->ss_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
+	sseu->eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
+}
+
  unsigned int
  intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
  {
@@ -22,17 +33,39 @@ intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
  unsigned int
  intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice)
  {
-	return hweight8(sseu->subslice_mask[slice]);
+	unsigned int i, total = 0;
+
+	for (i = 0; i < sseu->ss_stride; i++)
+		total += hweight8(sseu->subslice_mask[slice * sseu->ss_stride +
+						      i]);
+
+	return total;
+}
+
+void intel_sseu_copy_subslices(const struct sseu_dev_info *sseu, int slice,
+			       u8 *to_mask, const u8 *from_mask)
+{
+	int offset = slice * sseu->ss_stride;
+
+	memcpy(&to_mask[offset], &from_mask[offset], sseu->ss_stride);
+}
+
+void intel_sseu_set_subslices(struct sseu_dev_info *sseu, int slice,
+			      u32 ss_mask)
+{
+	int i, offset = slice * sseu->ss_stride;
+
+	for (i = 0; i < sseu->ss_stride; i++)
+		sseu->subslice_mask[offset + i] =
+			(ss_mask >> (BITS_PER_BYTE * i)) & 0xff;
  }
 
  static int intel_sseu_eu_idx(const struct sseu_dev_info *sseu, int slice,
  			     int subslice)
  {
-	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
-					   BITS_PER_BYTE);
-	int slice_stride = sseu->max_subslices * subslice_stride;
+	int slice_stride = sseu->max_subslices * sseu->eu_stride;
 
-	return slice * slice_stride + subslice * subslice_stride;
+	return slice * slice_stride + subslice * sseu->eu_stride;
  }
 
  u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int slice,
@@ -41,8 +74,7 @@ u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int slice,
  	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
  	u16 eu_mask = 0;
 
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+	for (i = 0; i < sseu->eu_stride; i++) {
  		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
  			(i * BITS_PER_BYTE);
  	}
@@ -55,8 +87,7 @@ void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice, int subslice,
  {
  	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
 
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+	for (i = 0; i < sseu->eu_stride; i++) {
  		sseu->eu_mask[offset + i] =
  			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
  	}
drivers/gpu/drm/i915/gt/intel_sseu.h +15 -1
index 56e3721ae83f..bf01f338a8cc 100644
@@ -9,16 +9,18 @@
 
  #include <linux/types.h>
  #include <linux/kernel.h>
+#include <linux/string.h>
 
  struct drm_i915_private;
 
  #define GEN_MAX_SLICES		(6) /* CNL upper bound */
  #define GEN_MAX_SUBSLICES	(8) /* ICL upper bound */
  #define GEN_SSEU_STRIDE(bits) DIV_ROUND_UP(bits, BITS_PER_BYTE)
+#define GEN_MAX_SUBSLICE_STRIDE GEN_SSEU_STRIDE(GEN_MAX_SUBSLICES)
 
  struct sseu_dev_info {
  	u8 slice_mask;
-	u8 subslice_mask[GEN_MAX_SLICES];
+	u8 subslice_mask[GEN_MAX_SLICES * GEN_MAX_SUBSLICE_STRIDE];
  	u16 eu_total;
  	u8 eu_per_subslice;
  	u8 min_eu_in_pool;
@@ -33,6 +35,9 @@ struct sseu_dev_info {
  	u8 max_subslices;
  	u8 max_eus_per_subslice;
 
+	u8 ss_stride;
+	u8 eu_stride;
+
  	/* We don't have more than 8 eus per subslice at the moment and as we
  	 * store eus enabled using bits, no need to multiply by eus per
  	 * subslice.
@@ -63,12 +68,21 @@ intel_sseu_from_device_info(const struct sseu_dev_info *sseu)
  	return value;
  }
 
+void intel_sseu_set_info(struct sseu_dev_info *sseu, u8 max_slices,
+			 u8 max_subslices, u8 max_eus_per_subslice);
+
  unsigned int
  intel_sseu_subslice_total(const struct sseu_dev_info *sseu);
 
  unsigned int
  intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice);
 
+void intel_sseu_copy_subslices(const struct sseu_dev_info *sseu, int slice,
+			       u8 *to_mask, const u8 *from_mask);
+
+void intel_sseu_set_subslices(struct sseu_dev_info *sseu, int slice,
+			      u32 ss_mask);
+
  u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int slice,
  		       int subslice);
 
drivers/gpu/drm/i915/gt/intel_workarounds.c +1 -1
index 43e290306551..7c7e9556c1c5 100644
@@ -767,7 +767,7 @@ wa_init_mcr(struct drm_i915_private *i915, struct i915_wa_list *wal)
  		u32 slice = fls(sseu->slice_mask);
  		u32 fuse3 =
  			intel_uncore_read(&i915->uncore, GEN10_MIRROR_FUSE3);
-		u8 ss_mask = sseu->subslice_mask[slice];
+		u8 ss_mask = sseu->subslice_mask[slice * sseu->ss_stride];
 
  		u8 enabled_mask = (ss_mask | ss_mask >>
  				   GEN10_L3BANK_PAIR_COUNT) & GEN10_L3BANK_MASK;
drivers/gpu/drm/i915/i915_debugfs.c +25 -19
index 3f3ee83ac315..08089c24db25 100644
@@ -1257,6 +1257,7 @@ static void i915_instdone_info(struct drm_i915_private *dev_priv,
  			       struct seq_file *m,
  			       struct intel_instdone *instdone)
  {
+	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	int slice;
  	int subslice;
 
@@ -1272,11 +1273,11 @@ static void i915_instdone_info(struct drm_i915_private *dev_priv,
  	if (INTEL_GEN(dev_priv) <= 6)
  		return;
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice)
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice, subslice)
  		seq_printf(m, "\t\tSAMPLER_INSTDONE[%d][%d]: 0x%08x\n",
  			   slice, subslice, instdone->sampler[slice][subslice]);
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice)
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice, subslice)
  		seq_printf(m, "\t\tROW_INSTDONE[%d][%d]: 0x%08x\n",
  			   slice, subslice, instdone->row[slice][subslice]);
  }
@@ -4066,7 +4067,9 @@ static void gen10_sseu_device_status(struct drm_i915_private *dev_priv,
  			continue;
 
  		sseu->slice_mask |= BIT(s);
-		sseu->subslice_mask[s] = info->sseu.subslice_mask[s];
+		intel_sseu_copy_subslices(&info->sseu, s,
+					  sseu->subslice_mask,
+					  info->sseu.subslice_mask);
 
  		for (ss = 0; ss < info->sseu.max_subslices; ss++) {
  			unsigned int eu_cnt;
@@ -4117,18 +4120,22 @@ static void gen9_sseu_device_status(struct drm_i915_private *dev_priv,
  		sseu->slice_mask |= BIT(s);
 
  		if (IS_GEN9_BC(dev_priv))
-			sseu->subslice_mask[s] =
-				RUNTIME_INFO(dev_priv)->sseu.subslice_mask[s];
+			intel_sseu_copy_subslices(&info->sseu, s,
+						  sseu->subslice_mask,
+						  info->sseu.subslice_mask);
 
  		for (ss = 0; ss < info->sseu.max_subslices; ss++) {
  			unsigned int eu_cnt;
+			u8 ss_idx = s * info->sseu.ss_stride +
+				    ss / BITS_PER_BYTE;
 
  			if (IS_GEN9_LP(dev_priv)) {
  				if (!(s_reg[s] & (GEN9_PGCTL_SS_ACK(ss))))
  					/* skip disabled subslice */
  					continue;
 
-				sseu->subslice_mask[s] |= BIT(ss);
+				sseu->subslice_mask[ss_idx] |=
+					BIT(ss % BITS_PER_BYTE);
  			}
 
  			eu_cnt = 2 * hweight32(eu_reg[2*s + ss/2] &
@@ -4145,25 +4152,24 @@ static void gen9_sseu_device_status(struct drm_i915_private *dev_priv,
  static void broadwell_sseu_device_status(struct drm_i915_private *dev_priv,
  					 struct sseu_dev_info *sseu)
  {
+	struct intel_runtime_info *info = RUNTIME_INFO(dev_priv);
  	u32 slice_info = I915_READ(GEN8_GT_SLICE_INFO);
  	int s;
 
  	sseu->slice_mask = slice_info & GEN8_LSLICESTAT_MASK;
 
  	if (sseu->slice_mask) {
-		sseu->eu_per_subslice =
-			RUNTIME_INFO(dev_priv)->sseu.eu_per_subslice;
-		for (s = 0; s < fls(sseu->slice_mask); s++) {
-			sseu->subslice_mask[s] =
-				RUNTIME_INFO(dev_priv)->sseu.subslice_mask[s];
-		}
+		sseu->eu_per_subslice = info->sseu.eu_per_subslice;
+		for (s = 0; s < fls(sseu->slice_mask); s++)
+			intel_sseu_copy_subslices(&info->sseu, s,
+						  sseu->subslice_mask,
+						  info->sseu.subslice_mask);
  		sseu->eu_total = sseu->eu_per_subslice *
  				 intel_sseu_subslice_total(sseu);
 
  		/* subtract fused off EU(s) from enabled slice(s) */
  		for (s = 0; s < fls(sseu->slice_mask); s++) {
-			u8 subslice_7eu =
-				RUNTIME_INFO(dev_priv)->sseu.subslice_7eu[s];
+			u8 subslice_7eu = info->sseu.subslice_7eu[s];
 
  			sseu->eu_total -= hweight8(subslice_7eu);
  		}
@@ -4210,6 +4216,7 @@ static void i915_print_sseu_info(struct seq_file *m, bool is_available_info,
  static int i915_sseu_status(struct seq_file *m, void *unused)
  {
  	struct drm_i915_private *dev_priv = node_to_i915(m->private);
+	const struct intel_runtime_info *info = RUNTIME_INFO(dev_priv);
  	struct sseu_dev_info sseu;
  	intel_wakeref_t wakeref;
 
@@ -4217,14 +4224,13 @@ static int i915_sseu_status(struct seq_file *m, void *unused)
  		return -ENODEV;
 
  	seq_puts(m, "SSEU Device Info\n");
-	i915_print_sseu_info(m, true, &RUNTIME_INFO(dev_priv)->sseu);
+	i915_print_sseu_info(m, true, &info->sseu);
 
  	seq_puts(m, "SSEU Device Status\n");
  	memset(&sseu, 0, sizeof(sseu));
-	sseu.max_slices = RUNTIME_INFO(dev_priv)->sseu.max_slices;
-	sseu.max_subslices = RUNTIME_INFO(dev_priv)->sseu.max_subslices;
-	sseu.max_eus_per_subslice =
-		RUNTIME_INFO(dev_priv)->sseu.max_eus_per_subslice;
+	intel_sseu_set_info(&sseu, info->sseu.max_slices,
+			    info->sseu.max_subslices,
+			    info->sseu.max_eus_per_subslice);
 
  	with_intel_runtime_pm(dev_priv, wakeref) {
  		if (IS_CHERRYVIEW(dev_priv))
drivers/gpu/drm/i915/i915_drv.c +4 -2
index 130c5140db0d..6afe4e3afea4 100644
@@ -326,7 +326,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
  	struct pci_dev *pdev = dev_priv->drm.pdev;
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	drm_i915_getparam_t *param = data;
-	int value;
+	int value = 0;
 
  	switch (param->param) {
  	case I915_PARAM_IRQ_ACTIVE:
@@ -455,7 +455,9 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
  			return -ENODEV;
  		break;
  	case I915_PARAM_SUBSLICE_MASK:
-		value = sseu->subslice_mask[0];
+		/* Only copy bits from the first subslice */
+		memcpy(&value, sseu->subslice_mask,
+		       min(sseu->ss_stride, (u8)sizeof(value)));
  		if (!value)
  			return -ENODEV;
  		break;
drivers/gpu/drm/i915/i915_gpu_error.c +3 -2
index e1b858bd1d32..140918dd9b7d 100644
@@ -407,6 +407,7 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
  static void error_print_instdone(struct drm_i915_error_state_buf *m,
  				 const struct drm_i915_error_engine *ee)
  {
+	struct sseu_dev_info *sseu = &RUNTIME_INFO(m->i915)->sseu;
  	int slice;
  	int subslice;
 
@@ -422,12 +423,12 @@ static void error_print_instdone(struct drm_i915_error_state_buf *m,
  	if (INTEL_GEN(m->i915) <= 6)
  		return;
 
-	for_each_instdone_slice_subslice(m->i915, slice, subslice)
+	for_each_instdone_slice_subslice(m->i915, sseu, slice, subslice)
  		err_printf(m, "  SAMPLER_INSTDONE[%d][%d]: 0x%08x\n",
  			   slice, subslice,
  			   ee->instdone.sampler[slice][subslice]);
 
-	for_each_instdone_slice_subslice(m->i915, slice, subslice)
+	for_each_instdone_slice_subslice(m->i915, sseu, slice, subslice)
  		err_printf(m, "  ROW_INSTDONE[%d][%d]: 0x%08x\n",
  			   slice, subslice,
  			   ee->instdone.row[slice][subslice]);
drivers/gpu/drm/i915/i915_query.c +4 -6
index 7c1708c22811..000dcb145ce0 100644
@@ -37,8 +37,6 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	struct drm_i915_query_topology_info topo;
  	u32 slice_length, subslice_length, eu_length, total_length;
-	u8 subslice_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
-	u8 eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
  	int ret;
 
  	if (query_item->flags != 0)
@@ -50,8 +48,8 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
  	BUILD_BUG_ON(sizeof(u8) != sizeof(sseu->slice_mask));
 
  	slice_length = sizeof(sseu->slice_mask);
-	subslice_length = sseu->max_slices * subslice_stride;
-	eu_length = sseu->max_slices * sseu->max_subslices * eu_stride;
+	subslice_length = sseu->max_slices * sseu->ss_stride;
+	eu_length = sseu->max_slices * sseu->max_subslices * sseu->eu_stride;
  	total_length = sizeof(topo) + slice_length + subslice_length +
  		       eu_length;
 
@@ -69,9 +67,9 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
  	topo.max_eus_per_subslice = sseu->max_eus_per_subslice;
 
  	topo.subslice_offset = slice_length;
-	topo.subslice_stride = subslice_stride;
+	topo.subslice_stride = sseu->ss_stride;
  	topo.eu_offset = slice_length + subslice_length;
-	topo.eu_stride = eu_stride;
+	topo.eu_stride = sseu->eu_stride;
 
  	if (__copy_to_user(u64_to_user_ptr(query_item->data_ptr),
  			   &topo, sizeof(topo)))
drivers/gpu/drm/i915/intel_device_info.c +1 -0
index e1dbccf04cd9..bbbc0a8c2183 100644
@@ -84,17 +84,42 @@ void intel_device_info_dump_flags(const struct intel_device_info *info,
  #undef PRINT_FLAG
  }
 
+#define SS_STR_MAX_SIZE (GEN_MAX_SUBSLICE_STRIDE * 2)

drivers/gpu/drm/i915/intel_device_info.c void intel_device_info_dump_flags(const struct intel_device_info *info, · line 97

Quoted 10 lines
drivers/gpu/drm/i915/intel_device_info.c +10 -0
+
+static u8 *
+subslice_per_slice_str(u8 *buf, const struct sseu_dev_info *sseu, u8 slice)
+{
+	int i;
+	u8 ss_offset = slice * sseu->ss_stride;
+
+	GEM_BUG_ON(slice >= sseu->max_slices);
+
+	memset(buf, 0, SS_STR_MAX_SIZE);

I suggest a more hardened approach of caller passing in the buffer size, since it is their buffer.

Regards,

Tvrtko

Quoted 294 lines
drivers/gpu/drm/i915/intel_device_info.c +67 -59
+
+	/*
+	 * Print subslice information in reverse order to match
+	 * userspace expectations.
+	 */
+	for (i = 0; i < sseu->ss_stride; i++)
+		sprintf(&buf[i * 2], "%02x",
+			sseu->subslice_mask[ss_offset + sseu->ss_stride -
+					    (i + 1)]);
+
+	return buf;
+}
+
  static void sseu_dump(const struct sseu_dev_info *sseu, struct drm_printer *p)
  {
  	int s;
+	u8 buf[SS_STR_MAX_SIZE];
 
  	drm_printf(p, "slice total: %u, mask=%04x\n",
  		   hweight8(sseu->slice_mask), sseu->slice_mask);
  	drm_printf(p, "subslice total: %u\n", intel_sseu_subslice_total(sseu));
  	for (s = 0; s < sseu->max_slices; s++) {
-		drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
+		drm_printf(p, "slice%d: %u subslices, mask=%s\n",
  			   s, intel_sseu_subslices_per_slice(sseu, s),
-			   sseu->subslice_mask[s]);
+			   subslice_per_slice_str(buf, sseu, s));
  	}
  	drm_printf(p, "EU total: %u\n", sseu->eu_total);
  	drm_printf(p, "EU per subslice: %u\n", sseu->eu_per_subslice);
@@ -118,6 +143,7 @@ void intel_device_info_dump_topology(const struct sseu_dev_info *sseu,
  				     struct drm_printer *p)
  {
  	int s, ss;
+	u8 buf[SS_STR_MAX_SIZE];
 
  	if (sseu->max_slices == 0) {
  		drm_printf(p, "Unavailable\n");
@@ -125,9 +151,9 @@ void intel_device_info_dump_topology(const struct sseu_dev_info *sseu,
  	}
 
  	for (s = 0; s < sseu->max_slices; s++) {
-		drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
+		drm_printf(p, "slice%d: %u subslice(s) (0x%s):\n",
  			   s, intel_sseu_subslices_per_slice(sseu, s),
-			   sseu->subslice_mask[s]);
+			   subslice_per_slice_str(buf, sseu, s));
 
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			u16 enabled_eus = intel_sseu_get_eus(sseu, s, ss);
@@ -156,15 +182,10 @@ static void gen11_sseu_info_init(struct drm_i915_private *dev_priv)
  	u8 eu_en;
  	int s;
 
-	if (IS_ELKHARTLAKE(dev_priv)) {
-		sseu->max_slices = 1;
-		sseu->max_subslices = 4;
-		sseu->max_eus_per_subslice = 8;
-	} else {
-		sseu->max_slices = 1;
-		sseu->max_subslices = 8;
-		sseu->max_eus_per_subslice = 8;
-	}
+	if (IS_ELKHARTLAKE(dev_priv))
+		intel_sseu_set_info(sseu, 1, 4, 8);
+	else
+		intel_sseu_set_info(sseu, 1, 8, 8);
 
  	s_en = I915_READ(GEN11_GT_SLICE_ENABLE) & GEN11_GT_S_ENA_MASK;
  	ss_en = ~I915_READ(GEN11_GT_SUBSLICE_DISABLE);
@@ -177,9 +198,11 @@ static void gen11_sseu_info_init(struct drm_i915_private *dev_priv)
  			int ss;
 
  			sseu->slice_mask |= BIT(s);
-			sseu->subslice_mask[s] = (ss_en >> ss_idx) & ss_en_mask;
+			sseu->subslice_mask[s * sseu->ss_stride] =
+				(ss_en >> ss_idx) & ss_en_mask;
  			for (ss = 0; ss < sseu->max_subslices; ss++) {
-				if (sseu->subslice_mask[s] & BIT(ss))
+				if (sseu->subslice_mask[s * sseu->ss_stride] &
+				    BIT(ss))
  					intel_sseu_set_eus(sseu, s, ss, eu_en);
  			}
  		}
@@ -201,23 +224,10 @@ static void gen10_sseu_info_init(struct drm_i915_private *dev_priv)
  	const int eu_mask = 0xff;
  	u32 subslice_mask, eu_en;
 
+	intel_sseu_set_info(sseu, 6, 4, 8);
+
  	sseu->slice_mask = (fuse2 & GEN10_F2_S_ENA_MASK) >>
  			    GEN10_F2_S_ENA_SHIFT;
-	sseu->max_slices = 6;
-	sseu->max_subslices = 4;
-	sseu->max_eus_per_subslice = 8;
-
-	subslice_mask = (1 << 4) - 1;
-	subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
-			   GEN10_F2_SS_DIS_SHIFT);
-
-	/*
-	 * Slice0 can have up to 3 subslices, but there are only 2 in
-	 * slice1/2.
-	 */
-	sseu->subslice_mask[0] = subslice_mask;
-	for (s = 1; s < sseu->max_slices; s++)
-		sseu->subslice_mask[s] = subslice_mask & 0x3;
 
  	/* Slice0 */
  	eu_en = ~I915_READ(GEN8_EU_DISABLE0);
@@ -242,14 +252,22 @@ static void gen10_sseu_info_init(struct drm_i915_private *dev_priv)
  	eu_en = ~I915_READ(GEN10_EU_DISABLE3);
  	intel_sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
 
-	/* Do a second pass where we mark the subslices disabled if all their
-	 * eus are off.
-	 */
+	subslice_mask = (1 << 4) - 1;
+	subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
+			   GEN10_F2_SS_DIS_SHIFT);
+
  	for (s = 0; s < sseu->max_slices; s++) {
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			if (intel_sseu_get_eus(sseu, s, ss) == 0)
-				sseu->subslice_mask[s] &= ~BIT(ss);
+				subslice_mask &= ~BIT(ss);
  		}
+
+		/*
+		 * Slice0 can have up to 3 subslices, but there are only 2 in
+		 * slice1/2.
+		 */
+		intel_sseu_set_subslices(sseu, s, s == 0 ? subslice_mask :
+							   subslice_mask & 0x3);
  	}
 
  	sseu->eu_total = compute_eu_total(sseu);
@@ -275,13 +293,12 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  {
  	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	u32 fuse;
+	u8 subslice_mask;
 
  	fuse = I915_READ(CHV_FUSE_GT);
 
  	sseu->slice_mask = BIT(0);
-	sseu->max_slices = 1;
-	sseu->max_subslices = 2;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, 1, 2, 8);
 
  	if (!(fuse & CHV_FGT_DISABLE_SS0)) {
  		u8 disabled_mask =
@@ -290,7 +307,7 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  			(((fuse & CHV_FGT_EU_DIS_SS0_R1_MASK) >>
  			  CHV_FGT_EU_DIS_SS0_R1_SHIFT) << 4);
 
-		sseu->subslice_mask[0] |= BIT(0);
+		subslice_mask |= BIT(0);
  		intel_sseu_set_eus(sseu, 0, 0, ~disabled_mask);
  	}
 
@@ -301,10 +318,12 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  			(((fuse & CHV_FGT_EU_DIS_SS1_R1_MASK) >>
  			  CHV_FGT_EU_DIS_SS1_R1_SHIFT) << 4);
 
-		sseu->subslice_mask[0] |= BIT(1);
+		subslice_mask |= BIT(1);
  		intel_sseu_set_eus(sseu, 0, 1, ~disabled_mask);
  	}
 
+	intel_sseu_set_subslices(sseu, 0, subslice_mask);
+
  	sseu->eu_total = compute_eu_total(sseu);
 
  	/*
@@ -312,7 +331,8 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  	 * across subslices.
  	*/
  	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
-				sseu->eu_total / intel_sseu_subslice_total(sseu) :
+				sseu->eu_total /
+					intel_sseu_subslice_total(sseu) :
  				0;
  	/*
  	 * CHV supports subslice power gating on devices with more than
@@ -336,9 +356,8 @@ static void gen9_sseu_info_init(struct drm_i915_private *dev_priv)
  	sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >> GEN8_F2_S_ENA_SHIFT;
 
  	/* BXT has a single slice and at most 3 subslices. */
-	sseu->max_slices = IS_GEN9_LP(dev_priv) ? 1 : 3;
-	sseu->max_subslices = IS_GEN9_LP(dev_priv) ? 3 : 4;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, IS_GEN9_LP(dev_priv) ? 1 : 3,
+			    IS_GEN9_LP(dev_priv) ? 3 : 4, 8);
 
  	/*
  	 * The subslice disable field is global, i.e. it applies
@@ -357,14 +376,16 @@ static void gen9_sseu_info_init(struct drm_i915_private *dev_priv)
  			/* skip disabled slice */
  			continue;
 
-		sseu->subslice_mask[s] = subslice_mask;
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
 
  		eu_disable = I915_READ(GEN9_EU_DISABLE(s));
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			int eu_per_ss;
  			u8 eu_disabled_mask;
+			u8 ss_idx = s * sseu->ss_stride + ss / BITS_PER_BYTE;
 
-			if (!(sseu->subslice_mask[s] & BIT(ss)))
+			if (!(sseu->subslice_mask[ss_idx] &
+			      BIT(ss % BITS_PER_BYTE)))
  				/* skip disabled subslice */
  				continue;
 
@@ -437,9 +458,7 @@ static void broadwell_sseu_info_init(struct drm_i915_private *dev_priv)
 
  	fuse2 = I915_READ(GEN8_FUSE2);
  	sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >> GEN8_F2_S_ENA_SHIFT;
-	sseu->max_slices = 3;
-	sseu->max_subslices = 3;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, 3, 3, 8);
 
  	/*
  	 * The subslice disable field is global, i.e. it applies
@@ -466,18 +485,21 @@ static void broadwell_sseu_info_init(struct drm_i915_private *dev_priv)
  			/* skip disabled slice */
  			continue;
 
-		sseu->subslice_mask[s] = subslice_mask;
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
 
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			u8 eu_disabled_mask;
+			u8 ss_idx = s * sseu->ss_stride + ss / BITS_PER_BYTE;
  			u32 n_disabled;
 
-			if (!(sseu->subslice_mask[s] & BIT(ss)))
+			if (!(sseu->subslice_mask[ss_idx] &
+			      BIT(ss % BITS_PER_BYTE)))
  				/* skip disabled subslice */
  				continue;
 
  			eu_disabled_mask =
-				eu_disable[s] >> (ss * sseu->max_eus_per_subslice);
+				eu_disable[s] >>
+					(ss * sseu->max_eus_per_subslice);
 
  			intel_sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
 
@@ -517,6 +539,7 @@ static void haswell_sseu_info_init(struct drm_i915_private *dev_priv)
  	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	u32 fuse1;
  	int s, ss;
+	u32 subslice_mask;
 
  	/*
  	 * There isn't a register to tell us how many slices/subslices. We
@@ -528,22 +551,18 @@ static void haswell_sseu_info_init(struct drm_i915_private *dev_priv)
  		/* fall through */
  	case 1:
  		sseu->slice_mask = BIT(0);
-		sseu->subslice_mask[0] = BIT(0);
+		subslice_mask = BIT(0);
  		break;
  	case 2:
  		sseu->slice_mask = BIT(0);
-		sseu->subslice_mask[0] = BIT(0) | BIT(1);
+		subslice_mask = BIT(0) | BIT(1);
  		break;
  	case 3:
  		sseu->slice_mask = BIT(0) | BIT(1);
-		sseu->subslice_mask[0] = BIT(0) | BIT(1);
-		sseu->subslice_mask[1] = BIT(0) | BIT(1);
+		subslice_mask = BIT(0) | BIT(1);
  		break;
  	}
 
-	sseu->max_slices = hweight8(sseu->slice_mask);
-	sseu->max_subslices = hweight8(sseu->subslice_mask[0]);
-
  	fuse1 = I915_READ(HSW_PAVP_FUSE1);
  	switch ((fuse1 & HSW_F1_EU_DIS_MASK) >> HSW_F1_EU_DIS_SHIFT) {
  	default:
@@ -560,9 +579,14 @@ static void haswell_sseu_info_init(struct drm_i915_private *dev_priv)
  		sseu->eu_per_subslice = 6;
  		break;
  	}
-	sseu->max_eus_per_subslice = sseu->eu_per_subslice;
+

drivers/gpu/drm/i915/intel_device_info.c static void haswell_sseu_info_init(struct drm_i915_private *dev_priv) · line 592

Quoted 10 lines
drivers/gpu/drm/i915/intel_device_info.c +5 -0
+	intel_sseu_set_info(sseu, hweight8(sseu->slice_mask),
+			    hweight8(subslice_mask),
+			    sseu->eu_per_subslice);
 
  	for (s = 0; s < sseu->max_slices; s++) {
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
+
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			intel_sseu_set_eus(sseu, s, ss,
  					   (1UL << sseu->eu_per_subslice) - 1);

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#24 lore raw

Re: [PATCH 6/6] drm/i915: Expand subslice mask

On 01/05/2019 19:22, Tvrtko Ursulin wrote:

[snip]

Quoted 6 lines
Quoted 6 lines
drivers/gpu/drm/i915/intel_device_info.c +4 -0
+#define SS_STR_MAX_SIZE (GEN_MAX_SUBSLICE_STRIDE * 2)
+
+static u8 *
+subslice_per_slice_str(u8 *buf, const struct sseu_dev_info *sseu, u8 

slice)
+{

Quoted 8 lines

drivers/gpu/drm/i915/intel_device_info.c void intel_device_info_dump_flags(const struct intel_device_info *info, · line 97

Quoted 6 lines

+    int i;
+    u8 ss_offset = slice * sseu->ss_stride;

drivers/gpu/drm/i915/intel_device_info.c +4 -0
+
+    GEM_BUG_ON(slice >= sseu->max_slices);
+
+    memset(buf, 0, SS_STR_MAX_SIZE);

I suggest a more hardened approach of caller passing in the buffer size, since it is their buffer.

Having said this..

Quoted 9 lines
Quoted 9 lines

+
+    /*
+     * Print subslice information in reverse order to match +     * userspace expectations.
+     */
+    for (i = 0; i < sseu->ss_stride; i++)
+        sprintf(&buf[i * 2], "%02x",
+            sseu->subslice_mask[ss_offset + sseu->ss_stride - +                        (i + 1)]);

...sprintf also needs to check against overflowing the buffer. (Relationship between loop boundary (ss_stride) and buffer size is a bit decoupled.)

And buffer should probably be char *.

Regards,

Tvrtko

Quoted 269 lines
Quoted 269 lines
drivers/gpu/drm/i915/intel_device_info.c +4 -0
+
+    return buf;
+}
+

  static void sseu_dump(const struct sseu_dev_info *sseu, struct drm_printer *p) {
      int s;
+    u8 buf[SS_STR_MAX_SIZE];
      drm_printf(p, "slice total: %u, mask=%04x\n", hweight8(sseu->slice_mask), sseu->slice_mask); drm_printf(p, "subslice total: %u\n", intel_sseu_subslice_total(sseu)); for (s = 0; s < sseu->max_slices; s++) {
-        drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
+        drm_printf(p, "slice%d: %u subslices, mask=%s\n",
                 s, intel_sseu_subslices_per_slice(sseu, s),
-               sseu->subslice_mask[s]);
+               subslice_per_slice_str(buf, sseu, s));
      }
      drm_printf(p, "EU total: %u\n", sseu->eu_total); drm_printf(p, "EU per subslice: %u\n", sseu->eu_per_subslice);

drivers/gpu/drm/i915/intel_device_info.c +13 -13
@@ -118,6 +143,7 @@ void intel_device_info_dump_topology(const struct 
 sseu_dev_info *sseu,
                        struct drm_printer *p)
   {
       int s, ss;
+    u8 buf[SS_STR_MAX_SIZE];
       if (sseu->max_slices == 0) {
           drm_printf(p, "Unavailable\n");
@@ -125,9 +151,9 @@ void intel_device_info_dump_topology(const struct 
 sseu_dev_info *sseu,
       }
       for (s = 0; s < sseu->max_slices; s++) {
-        drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
+        drm_printf(p, "slice%d: %u subslice(s) (0x%s):\n",
                  s, intel_sseu_subslices_per_slice(sseu, s),
-               sseu->subslice_mask[s]);
+               subslice_per_slice_str(buf, sseu, s));
           for (ss = 0; ss < sseu->max_subslices; ss++) {
               u16 enabled_eus = intel_sseu_get_eus(sseu, s, ss);
@@ -156,15 +182,10 @@ static void gen11_sseu_info_init(struct 
 drm_i915_private *dev_priv)
       u8 eu_en;
       int s;
-    if (IS_ELKHARTLAKE(dev_priv)) {
-        sseu->max_slices = 1;
-        sseu->max_subslices = 4;
-        sseu->max_eus_per_subslice = 8;
-    } else {
-        sseu->max_slices = 1;
-        sseu->max_subslices = 8;
-        sseu->max_eus_per_subslice = 8;
-    }
+    if (IS_ELKHARTLAKE(dev_priv))
+        intel_sseu_set_info(sseu, 1, 4, 8);
+    else
+        intel_sseu_set_info(sseu, 1, 8, 8);
       s_en = I915_READ(GEN11_GT_SLICE_ENABLE) & GEN11_GT_S_ENA_MASK;
       ss_en = ~I915_READ(GEN11_GT_SUBSLICE_DISABLE);
@@ -177,9 +198,11 @@ static void gen11_sseu_info_init(struct 
 drm_i915_private *dev_priv)
               int ss;
               sseu->slice_mask |= BIT(s);
-            sseu->subslice_mask[s] = (ss_en >> ss_idx) & ss_en_mask;
+            sseu->subslice_mask[s * sseu->ss_stride] =
+                (ss_en >> ss_idx) & ss_en_mask;
               for (ss = 0; ss < sseu->max_subslices; ss++) {
-                if (sseu->subslice_mask[s] & BIT(ss))
+                if (sseu->subslice_mask[s * sseu->ss_stride] &
+                    BIT(ss))
                       intel_sseu_set_eus(sseu, s, ss, eu_en);
               }
           }
@@ -201,23 +224,10 @@ static void gen10_sseu_info_init(struct 
 drm_i915_private *dev_priv)
       const int eu_mask = 0xff;
       u32 subslice_mask, eu_en;
+    intel_sseu_set_info(sseu, 6, 4, 8);
+

      sseu->slice_mask = (fuse2 & GEN10_F2_S_ENA_MASK) >> GEN10_F2_S_ENA_SHIFT;
-    sseu->max_slices = 6;
-    sseu->max_subslices = 4;
-    sseu->max_eus_per_subslice = 8;

drivers/gpu/drm/i915/intel_device_info.c +0 -12
-
-    subslice_mask = (1 << 4) - 1;
-    subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
-               GEN10_F2_SS_DIS_SHIFT);
-
-    /*
-     * Slice0 can have up to 3 subslices, but there are only 2 in
-     * slice1/2.
-     */
-    sseu->subslice_mask[0] = subslice_mask;
-    for (s = 1; s < sseu->max_slices; s++)
-        sseu->subslice_mask[s] = subslice_mask & 0x3;

      /* Slice0 */
      eu_en = ~I915_READ(GEN8_EU_DISABLE0);

drivers/gpu/drm/i915/intel_device_info.c +0 -1
@@ -242,14 +252,22 @@ static void gen10_sseu_info_init(struct 
 drm_i915_private *dev_priv)
       eu_en = ~I915_READ(GEN10_EU_DISABLE3);
       intel_sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
-    /* Do a second pass where we mark the subslices disabled if all 

their
-     * eus are off.
-     */
+    subslice_mask = (1 << 4) - 1;
+    subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
+               GEN10_F2_SS_DIS_SHIFT);
+
      for (s = 0; s < sseu->max_slices; s++) {
          for (ss = 0; ss < sseu->max_subslices; ss++) {
              if (intel_sseu_get_eus(sseu, s, ss) == 0)
-                sseu->subslice_mask[s] &= ~BIT(ss);
+                subslice_mask &= ~BIT(ss);
          }
+
+        /*
+         * Slice0 can have up to 3 subslices, but there are only 2 in +         * slice1/2.
+         */
+        intel_sseu_set_subslices(sseu, s, s == 0 ? subslice_mask :
+                               subslice_mask & 0x3);
      }
      sseu->eu_total = compute_eu_total(sseu);

drivers/gpu/drm/i915/intel_device_info.c +2 -3
@@ -275,13 +293,12 @@ static void cherryview_sseu_info_init(struct 
 drm_i915_private *dev_priv)
   {
       struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
       u32 fuse;
+    u8 subslice_mask;
       fuse = I915_READ(CHV_FUSE_GT);
       sseu->slice_mask = BIT(0);
-    sseu->max_slices = 1;
-    sseu->max_subslices = 2;
-    sseu->max_eus_per_subslice = 8;
+    intel_sseu_set_info(sseu, 1, 2, 8);
       if (!(fuse & CHV_FGT_DISABLE_SS0)) {

          u8 disabled_mask =

(quoted diff) +0 -0
@@ -290,7 +307,7 @@ static void cherryview_sseu_info_init(struct 
 drm_i915_private *dev_priv)

              (((fuse & CHV_FGT_EU_DIS_SS0_R1_MASK) >> CHV_FGT_EU_DIS_SS0_R1_SHIFT) << 4); -        sseu->subslice_mask[0] |= BIT(0);
+        subslice_mask |= BIT(0); intel_sseu_set_eus(sseu, 0, 0, ~disabled_mask); }

(quoted diff) +0 -0
@@ -301,10 +318,12 @@ static void cherryview_sseu_info_init(struct 
 drm_i915_private *dev_priv)

              (((fuse & CHV_FGT_EU_DIS_SS1_R1_MASK) >> CHV_FGT_EU_DIS_SS1_R1_SHIFT) << 4); -        sseu->subslice_mask[0] |= BIT(1);
+        subslice_mask |= BIT(1); intel_sseu_set_eus(sseu, 0, 1, ~disabled_mask); }
+    intel_sseu_set_subslices(sseu, 0, subslice_mask); +
      sseu->eu_total = compute_eu_total(sseu); /*

drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -312,7 +331,8 @@ static void cherryview_sseu_info_init(struct 
 drm_i915_private *dev_priv)
        * across subslices.
       */

      sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ? -                sseu->eu_total / intel_sseu_subslice_total(sseu) :
+                sseu->eu_total /
+                    intel_sseu_subslice_total(sseu) :
                  0;
      /*
       * CHV supports subslice power gating on devices with more than

drivers/gpu/drm/i915/intel_device_info.c +11 -10
@@ -336,9 +356,8 @@ static void gen9_sseu_info_init(struct 
 drm_i915_private *dev_priv)
       sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >> GEN8_F2_S_ENA_SHIFT;
       /* BXT has a single slice and at most 3 subslices. */
-    sseu->max_slices = IS_GEN9_LP(dev_priv) ? 1 : 3;
-    sseu->max_subslices = IS_GEN9_LP(dev_priv) ? 3 : 4;
-    sseu->max_eus_per_subslice = 8;
+    intel_sseu_set_info(sseu, IS_GEN9_LP(dev_priv) ? 1 : 3,
+                IS_GEN9_LP(dev_priv) ? 3 : 4, 8);
       /*
        * The subslice disable field is global, i.e. it applies
@@ -357,14 +376,16 @@ static void gen9_sseu_info_init(struct 
 drm_i915_private *dev_priv)
               /* skip disabled slice */
               continue;
-        sseu->subslice_mask[s] = subslice_mask;
+        intel_sseu_set_subslices(sseu, s, subslice_mask);
           eu_disable = I915_READ(GEN9_EU_DISABLE(s));
           for (ss = 0; ss < sseu->max_subslices; ss++) {
               int eu_per_ss;
               u8 eu_disabled_mask;
+            u8 ss_idx = s * sseu->ss_stride + ss / BITS_PER_BYTE;
-            if (!(sseu->subslice_mask[s] & BIT(ss)))
+            if (!(sseu->subslice_mask[ss_idx] &
+                  BIT(ss % BITS_PER_BYTE)))
                   /* skip disabled subslice */
                   continue;
@@ -437,9 +458,7 @@ static void broadwell_sseu_info_init(struct 
 drm_i915_private *dev_priv)
       fuse2 = I915_READ(GEN8_FUSE2);
       sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >> GEN8_F2_S_ENA_SHIFT;
-    sseu->max_slices = 3;
-    sseu->max_subslices = 3;
-    sseu->max_eus_per_subslice = 8;
+    intel_sseu_set_info(sseu, 3, 3, 8);
       /*
        * The subslice disable field is global, i.e. it applies
@@ -466,18 +485,21 @@ static void broadwell_sseu_info_init(struct 
 drm_i915_private *dev_priv)
               /* skip disabled slice */
               continue;
-        sseu->subslice_mask[s] = subslice_mask;
+        intel_sseu_set_subslices(sseu, s, subslice_mask);
           for (ss = 0; ss < sseu->max_subslices; ss++) {
               u8 eu_disabled_mask;
+            u8 ss_idx = s * sseu->ss_stride + ss / BITS_PER_BYTE;
               u32 n_disabled;
-            if (!(sseu->subslice_mask[s] & BIT(ss)))
+            if (!(sseu->subslice_mask[ss_idx] &
+                  BIT(ss % BITS_PER_BYTE)))
                   /* skip disabled subslice */
                   continue;

              eu_disabled_mask =
-                eu_disable[s] >> (ss * sseu->max_eus_per_subslice); +                eu_disable[s] >>
+                    (ss * sseu->max_eus_per_subslice); intel_sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);

drivers/gpu/drm/i915/intel_device_info.c +1 -0
@@ -517,6 +539,7 @@ static void haswell_sseu_info_init(struct 
 drm_i915_private *dev_priv)
       struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
       u32 fuse1;
       int s, ss;
+    u32 subslice_mask;
       /*
        * There isn't a register to tell us how many slices/subslices. We
@@ -528,22 +551,18 @@ static void haswell_sseu_info_init(struct 
 drm_i915_private *dev_priv)
           /* fall through */

      case 1:
          sseu->slice_mask = BIT(0);
-        sseu->subslice_mask[0] = BIT(0);
+        subslice_mask = BIT(0);
          break;
      case 2:
          sseu->slice_mask = BIT(0);
-        sseu->subslice_mask[0] = BIT(0) | BIT(1);
+        subslice_mask = BIT(0) | BIT(1);
          break;
      case 3:
          sseu->slice_mask = BIT(0) | BIT(1);
-        sseu->subslice_mask[0] = BIT(0) | BIT(1);
-        sseu->subslice_mask[1] = BIT(0) | BIT(1);
+        subslice_mask = BIT(0) | BIT(1);
          break;
      }
-    sseu->max_slices = hweight8(sseu->slice_mask);
-    sseu->max_subslices = hweight8(sseu->subslice_mask[0]); -
      fuse1 = I915_READ(HSW_PAVP_FUSE1);
      switch ((fuse1 & HSW_F1_EU_DIS_MASK) >> HSW_F1_EU_DIS_SHIFT) { default:

drivers/gpu/drm/i915/intel_device_info.c +3 -1
@@ -560,9 +579,14 @@ static void haswell_sseu_info_init(struct 
 drm_i915_private *dev_priv)
           sseu->eu_per_subslice = 6;
           break;
       }
-    sseu->max_eus_per_subslice = sseu->eu_per_subslice;
+
+    intel_sseu_set_info(sseu, hweight8(sseu->slice_mask),
+                hweight8(subslice_mask),
Quoted 8 lines

drivers/gpu/drm/i915/intel_device_info.c static void haswell_sseu_info_init(struct drm_i915_private *dev_priv) · line 592

Quoted 8 lines
drivers/gpu/drm/i915/intel_device_info.c +3 -0
+                sseu->eu_per_subslice);
       for (s = 0; s < sseu->max_slices; s++) {
+        intel_sseu_set_subslices(sseu, s, subslice_mask);
+
           for (ss = 0; ss < sseu->max_subslices; ss++) {
               intel_sseu_set_eus(sseu, s, ss,
                          (1UL << sseu->eu_per_subslice) - 1);

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#11 lore raw

Re: [PATCH 4/6] drm/i915: Move sseu helper functions to intel_sseu.h

In reply to Stuart Summers
Quoted 59 lines · Stuart Summers

v2: fix spacing from checkpatch warning

---
  drivers/gpu/drm/i915/gt/intel_sseu.h     | 47 ++++++++++++++++++++++++
  drivers/gpu/drm/i915/intel_device_info.h | 47 ------------------------
  2 files changed, 47 insertions(+), 47 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.h +37 -0
index f5ff6b7a756a..029e71d8f140 100644
@@ -63,12 +63,59 @@ intel_sseu_from_device_info(const struct sseu_dev_info *sseu)
  	return value;
  }
 
+static inline unsigned int sseu_subslice_total(const struct sseu_dev_info *sseu)
+{
+	unsigned int i, total = 0;
+
+	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
+		total += hweight8(sseu->subslice_mask[i]);
+
+	return total;
+}
+
  static inline unsigned int
  sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice)
  {
  	return hweight8(sseu->subslice_mask[slice]);
  }
 
+static inline int sseu_eu_idx(const struct sseu_dev_info *sseu,
+			      int slice, int subslice)
+{
+	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
+					   BITS_PER_BYTE);
+	int slice_stride = sseu->max_subslices * subslice_stride;
+
+	return slice * slice_stride + subslice * subslice_stride;
+}
+
+static inline u16 sseu_get_eus(const struct sseu_dev_info *sseu,
+			       int slice, int subslice)
+{
+	int i, offset = sseu_eu_idx(sseu, slice, subslice);
+	u16 eu_mask = 0;
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
+			(i * BITS_PER_BYTE);
+	}
+
+	return eu_mask;
+}
+
+static inline void sseu_set_eus(struct sseu_dev_info *sseu,
+				int slice, int subslice, u16 eu_mask)

drivers/gpu/drm/i915/gt/intel_sseu.h intel_sseu_from_device_info(const struct sseu_dev_info *sseu) · line 118

Quoted 10 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +10 -0
+{
+	int i, offset = sseu_eu_idx(sseu, slice, subslice);
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+		sseu->eu_mask[offset + i] =
+			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
+	}
+}
+

AFAICS sseu_get_eus() and sseu_set_eus() are only used by sseu-related functions in device_info.c and sseu_eu_idx() is only used by those 2, so we can make all 3 of them static in that file. We should also migrate all of the sseu code from device_info.c to intel_sseu.c for consistency; I'm ok with that being done as a follow up if you prefer to avoid it in this series.

I was also about to mention adding the intel_* prefix to the remaining functions but then I realized you add it in the next patch. My personal preference would be to do that in this patch, but I'm not going to block on it.

Daniele

Quoted 52 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +0 -0
  u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
  			 const struct intel_sseu *req_sseu);
 
drivers/gpu/drm/i915/intel_device_info.h +0 -40
index 5a2e17d6146b..6412a9c72898 100644
@@ -218,53 +218,6 @@ struct intel_driver_caps {
  	bool has_logical_contexts:1;
  };
 
-static inline unsigned int sseu_subslice_total(const struct sseu_dev_info *sseu)
-{
-	unsigned int i, total = 0;
-
-	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
-		total += hweight8(sseu->subslice_mask[i]);
-
-	return total;
-}
-
-static inline int sseu_eu_idx(const struct sseu_dev_info *sseu,
-			      int slice, int subslice)
-{
-	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
-					   BITS_PER_BYTE);
-	int slice_stride = sseu->max_subslices * subslice_stride;
-
-	return slice * slice_stride + subslice * subslice_stride;
-}
-
-static inline u16 sseu_get_eus(const struct sseu_dev_info *sseu,
-			       int slice, int subslice)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-	u16 eu_mask = 0;
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
-		eu_mask |= ((u16) sseu->eu_mask[offset + i]) <<
-			(i * BITS_PER_BYTE);
-	}
-
-	return eu_mask;
-}
-
-static inline void sseu_set_eus(struct sseu_dev_info *sseu,
-				int slice, int subslice, u16 eu_mask)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-

drivers/gpu/drm/i915/intel_device_info.h struct intel_driver_caps { · line 223

Quoted 10 lines
drivers/gpu/drm/i915/intel_device_info.h +0 -7
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
-		sseu->eu_mask[offset + i] =
-			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
-	}
-}
-
  const char *intel_platform_name(enum intel_platform platform);
 
  void intel_device_info_subplatform_init(struct drm_i915_private *dev_priv);

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#12 lore raw

Re: [PATCH 4/6] drm/i915: Move sseu helper functions to intel_sseu.h

Quoted 92 lines · Daniele Ceraolo Spurio
Quoted 72 lines · Stuart Summers

v2: fix spacing from checkpatch warning

---
  drivers/gpu/drm/i915/gt/intel_sseu.h     | 47
++++++++++++++++++++++++
  drivers/gpu/drm/i915/intel_device_info.h | 47 -----------------
-------
  2 files changed, 47 insertions(+), 47 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.h +28 -0
index f5ff6b7a756a..029e71d8f140 100644
@@ -63,12 +63,59 @@ intel_sseu_from_device_info(const struct
 sseu_dev_info *sseu)
  	return value;
  }
 
+static inline unsigned int sseu_subslice_total(const struct
 sseu_dev_info *sseu)
+{
+	unsigned int i, total = 0;
+
+	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
+		total += hweight8(sseu->subslice_mask[i]);
+
+	return total;
+}
+
  static inline unsigned int
  sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8
 slice)
  {
  	return hweight8(sseu->subslice_mask[slice]);
  }
 
+static inline int sseu_eu_idx(const struct sseu_dev_info *sseu,
+			      int slice, int subslice)
+{
+	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
+					   BITS_PER_BYTE);
+	int slice_stride = sseu->max_subslices * subslice_stride;
+
+	return slice * slice_stride + subslice * subslice_stride;
+}
+
+static inline u16 sseu_get_eus(const struct sseu_dev_info *sseu,
+			       int slice, int subslice)
+{
+	int i, offset = sseu_eu_idx(sseu, slice, subslice);
+	u16 eu_mask = 0;
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.h +14 -0
+		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
+			(i * BITS_PER_BYTE);
+	}
+
+	return eu_mask;
+}
+
+static inline void sseu_set_eus(struct sseu_dev_info *sseu,
+				int slice, int subslice, u16 eu_mask)
+{
+	int i, offset = sseu_eu_idx(sseu, slice, subslice);
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.h intel_sseu_from_device_info(const struct sseu_dev_info *sseu) · line 118

Quoted 5 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +5 -0
+		sseu->eu_mask[offset + i] =
+			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
+	}
+}
+

AFAICS sseu_get_eus() and sseu_set_eus() are only used by sseu- related
functions in device_info.c and sseu_eu_idx() is only used by those 2, so
we can make all 3 of them static in that file. We should also migrate
all of the sseu code from device_info.c to intel_sseu.c for consistency;
I'm ok with that being done as a follow up if you prefer to avoid it in
this series.

Quoted 7 lines

I was also about to mention adding the intel_* prefix to the remaining
functions but then I realized you add it in the next patch. My personal
preference would be to do that in this patch, but I'm not going to block
on it.

Not sure why I missed this one, but good catch! I'll post a fix in the next series update.

-Stuart

Quoted 67 lines

Daniele

Quoted 64 lines

u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
const struct intel_sseu *req_sseu);

drivers/gpu/drm/i915/intel_device_info.h +0 -47
index 5a2e17d6146b..6412a9c72898 100644
@@ -218,53 +218,6 @@ struct intel_driver_caps {
  	bool has_logical_contexts:1;
  };
 
-static inline unsigned int sseu_subslice_total(const struct
 sseu_dev_info *sseu)
-{
-	unsigned int i, total = 0;
-
-	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
-		total += hweight8(sseu->subslice_mask[i]);
-
-	return total;
-}
-
-static inline int sseu_eu_idx(const struct sseu_dev_info *sseu,
-			      int slice, int subslice)
-{
-	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
-					   BITS_PER_BYTE);
-	int slice_stride = sseu->max_subslices * subslice_stride;
-
-	return slice * slice_stride + subslice * subslice_stride;
-}
-
-static inline u16 sseu_get_eus(const struct sseu_dev_info *sseu,
-			       int slice, int subslice)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-	u16 eu_mask = 0;
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,
 BITS_PER_BYTE); i++) {
-		eu_mask |= ((u16) sseu->eu_mask[offset + i]) <<
-			(i * BITS_PER_BYTE);
-	}
-
-	return eu_mask;
-}
-
-static inline void sseu_set_eus(struct sseu_dev_info *sseu,
-				int slice, int subslice, u16 eu_mask)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,
 BITS_PER_BYTE); i++) {
-		sseu->eu_mask[offset + i] =
-			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
-	}
-}
-
  const char *intel_platform_name(enum intel_platform platform);
Quoted 3 lines
Quoted 3 lines

void intel_device_info_subplatform_init(struct drm_i915_private
*dev_priv);

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#9 lore raw

Re: [PATCH 3/6] drm/i915: Move calculation of subslices per slice to new function

In reply to Daniele Ceraolo Spurio
Quoted 25 lines · Daniele Ceraolo Spurio
Quoted 23 lines · Stuart Summers

Add a new function to return the number of subslices per slice to consolidate code usage.

v2: rebase on changes to move sseu struct to intel_sseu.h

Cc Daniele Ceraolo Spurio <[email protected]>
---
  drivers/gpu/drm/i915/gt/intel_sseu.h     | 6 ++++++
  drivers/gpu/drm/i915/i915_debugfs.c      | 2 +-
  drivers/gpu/drm/i915/intel_device_info.c | 4 ++--
  3 files changed, 9 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.h +0 -0
index c0b16b248d4c..f5ff6b7a756a 100644
@@ -63,6 +63,12 @@ intel_sseu_from_device_info(const struct
 sseu_dev_info *sseu)
  	return value;
  }
 
Quoted 5 lines

drivers/gpu/drm/i915/gt/intel_sseu.h intel_sseu_from_device_info(const struct · line 69

Quoted 3 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +2 -0
+static inline unsigned int
+sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8
 slice)

This is exposed, so needs an intel_* prefix. with that:

Will change in the next series update. Thanks for the review!

-Stuart

Quoted 46 lines

Daniele

Quoted 41 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +4 -0
+{
+	return hweight8(sseu->subslice_mask[slice]);
+}
+
  u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,

const struct intel_sseu *req_sseu);

drivers/gpu/drm/i915/i915_debugfs.c +1 -1
index 0e4dffcd4da4..fe854c629a32 100644
@@ -4185,7 +4185,7 @@ static void i915_print_sseu_info(struct
 seq_file *m, bool is_available_info,
  		   sseu_subslice_total(sseu));
  	for (s = 0; s < fls(sseu->slice_mask); s++) {
  		seq_printf(m, "  %s Slice%i subslices: %u\n", type,
-			   s, hweight8(sseu->subslice_mask[s]));
+			   s, sseu_subslices_per_slice(sseu, s));
  	}
  	seq_printf(m, "  %s EU Total: %u\n", type,

sseu->eu_total);

drivers/gpu/drm/i915/intel_device_info.c +1 -1
index 6af480b95bc6..559cf0d0628e 100644
@@ -93,7 +93,7 @@ static void sseu_dump(const struct sseu_dev_info
 *sseu, struct drm_printer *p)
  	drm_printf(p, "subslice total: %u\n",
 sseu_subslice_total(sseu));
  	for (s = 0; s < sseu->max_slices; s++) {
  		drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
-			   s, hweight8(sseu->subslice_mask[s]),
+			   s, sseu_subslices_per_slice(sseu, s),
  			   sseu->subslice_mask[s]);
  	}
  	drm_printf(p, "EU total: %u\n", sseu->eu_total);
drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -126,7 +126,7 @@ void intel_device_info_dump_topology(const
 struct sseu_dev_info *sseu,
 
Quoted 8 lines
Quoted 8 lines
drivers/gpu/drm/i915/intel_device_info.c +1 -1
  	for (s = 0; s < sseu->max_slices; s++) {
  		drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
-			   s, hweight8(sseu->subslice_mask[s]),
+			   s, sseu_subslices_per_slice(sseu, s),
  			   sseu->subslice_mask[s]);
 
  		for (ss = 0; ss < sseu->max_subslices; ss++) {

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#6 lore raw

Re: [PATCH 2/6] drm/i915: Add macro for SSEU stride calculation

In reply to Daniele Ceraolo Spurio
Quoted 28 lines · Daniele Ceraolo Spurio
Quoted 26 lines · Stuart Summers

Subslice stride and EU stride are calculated multiple times in i915_query. Move this calculation to a macro to reduce code duplication.

v2: update headers in intel_sseu.h

Cc Daniele Ceraolo Spurio <[email protected]>
---
  drivers/gpu/drm/i915/gt/intel_sseu.h |  2 ++
  drivers/gpu/drm/i915/i915_query.c    | 17 ++++++++---------
  2 files changed, 10 insertions(+), 9 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.h +1 -0
index 73bc824094e8..c0b16b248d4c 100644
@@ -8,11 +8,13 @@
  #define __INTEL_SSEU_H__
 
  #include <linux/types.h>
+#include <linux/kernel.h>
 
  struct drm_i915_private;
 
Quoted 7 lines

drivers/gpu/drm/i915/gt/intel_sseu.h line 17

Quoted 3 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +1 -0
  #define GEN_MAX_SLICES		(6) /* CNL upper bound */
  #define GEN_MAX_SUBSLICES	(8) /* ICL upper bound */
+#define GEN_SSEU_STRIDE(bits) DIV_ROUND_UP(bits, BITS_PER_BYTE)

What we pass to this macro isn't really a bits count but the maximum amount of s/ss/eus. s/bits/max_entry/, or something like that? with that:

Makes sense, I'll make the change in the next series post. Thanks for the review!

-Stuart

Quoted 58 lines

Daniele

Quoted 53 lines
drivers/gpu/drm/i915/gt/intel_sseu.h +0 -0
  struct sseu_dev_info {
  	u8 slice_mask;
drivers/gpu/drm/i915/i915_query.c +2 -0
index 782183b78f49..7c1708c22811 100644
@@ -37,6 +37,8 @@ static int query_topology_info(struct
 drm_i915_private *dev_priv,
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 >sseu;
  	struct drm_i915_query_topology_info topo;
  	u32 slice_length, subslice_length, eu_length, total_length;
+	u8 subslice_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
+	u8 eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
  	int ret;

if (query_item->flags != 0)

drivers/gpu/drm/i915/i915_query.c +4 -6
@@ -48,12 +50,10 @@ static int query_topology_info(struct
 drm_i915_private *dev_priv,
  	BUILD_BUG_ON(sizeof(u8) != sizeof(sseu->slice_mask));
 
  	slice_length = sizeof(sseu->slice_mask);
-	subslice_length = sseu->max_slices *
-		DIV_ROUND_UP(sseu->max_subslices, BITS_PER_BYTE);
-	eu_length = sseu->max_slices * sseu->max_subslices *
-		DIV_ROUND_UP(sseu->max_eus_per_subslice,
 BITS_PER_BYTE);
-
-	total_length = sizeof(topo) + slice_length + subslice_length +
 eu_length;
+	subslice_length = sseu->max_slices * subslice_stride;
+	eu_length = sseu->max_slices * sseu->max_subslices * eu_stride;
+	total_length = sizeof(topo) + slice_length + subslice_length +
+		       eu_length;

ret = copy_query_item(&topo, sizeof(topo), total_length,
query_item);

drivers/gpu/drm/i915/i915_query.c +2 -3
@@ -69,10 +69,9 @@ static int query_topology_info(struct
 drm_i915_private *dev_priv,
  	topo.max_eus_per_subslice = sseu->max_eus_per_subslice;
 
  	topo.subslice_offset = slice_length;
-	topo.subslice_stride = DIV_ROUND_UP(sseu->max_subslices,
 BITS_PER_BYTE);
+	topo.subslice_stride = subslice_stride;
  	topo.eu_offset = slice_length + subslice_length;
-	topo.eu_stride =
-		DIV_ROUND_UP(sseu->max_eus_per_subslice,
 BITS_PER_BYTE);
+	topo.eu_stride = eu_stride;
Quoted 3 lines
Quoted 3 lines

if (__copy_to_user(u64_to_user_ptr(query_item->data_ptr),
&topo, sizeof(topo)))

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#3 lore raw

Re: [PATCH 1/6] drm/i915: Use local variable for SSEU info in GETPARAM ioctl

In reply to Daniele Ceraolo Spurio
Quoted 10 lines · Daniele Ceraolo Spurio
Quoted 7 lines · Stuart Summers

In the GETPARAM ioctl handler, use a local variable to consolidate usage of SSEU runtime info.

v2: add const to sseu_dev_info variable

Cc Daniele Ceraolo Spurio <[email protected]>

Thanks for the review!

-Stuart

Quoted 57 lines
Quoted 55 lines
---
  drivers/gpu/drm/i915/i915_drv.c | 11 ++++++-----
  1 file changed, 6 insertions(+), 5 deletions(-)
drivers/gpu/drm/i915/i915_drv.c +1 -0
index 21dac5a09fbe..c376244c19c4 100644
@@ -324,6 +324,7 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
  {
  	struct drm_i915_private *dev_priv = to_i915(dev);
  	struct pci_dev *pdev = dev_priv->drm.pdev;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 >sseu;
  	drm_i915_getparam_t *param = data;

int value;

drivers/gpu/drm/i915/i915_drv.c +2 -2
@@ -377,12 +378,12 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
  		value = i915_cmd_parser_get_version(dev_priv);
  		break;
  	case I915_PARAM_SUBSLICE_TOTAL:
-		value = sseu_subslice_total(&RUNTIME_INFO(dev_priv)-
 >sseu);
+		value = sseu_subslice_total(sseu);
  		if (!value)
  			return -ENODEV;
  		break;
  	case I915_PARAM_EU_TOTAL:
-		value = RUNTIME_INFO(dev_priv)->sseu.eu_total;
+		value = sseu->eu_total;
  		if (!value)

return -ENODEV;
break;

drivers/gpu/drm/i915/i915_drv.c +1 -1
@@ -399,7 +400,7 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
  		value = HAS_POOLED_EU(dev_priv);
  		break;
  	case I915_PARAM_MIN_EU_IN_POOL:
-		value = RUNTIME_INFO(dev_priv)->sseu.min_eu_in_pool;
+		value = sseu->min_eu_in_pool;
  		break;
  	case I915_PARAM_HUC_STATUS:
  		value = intel_huc_check_status(&dev_priv->huc);
drivers/gpu/drm/i915/i915_drv.c +1 -1
@@ -449,12 +450,12 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
  		value = intel_engines_has_context_isolation(dev_priv);
  		break;
  	case I915_PARAM_SLICE_MASK:
-		value = RUNTIME_INFO(dev_priv)->sseu.slice_mask;
+		value = sseu->slice_mask;
  		if (!value)
  			return -ENODEV;
Quoted 8 lines
Quoted 8 lines
drivers/gpu/drm/i915/i915_drv.c +1 -1
  		break;
  	case I915_PARAM_SUBSLICE_MASK:
-		value = RUNTIME_INFO(dev_priv)->sseu.subslice_mask[0];
+		value = sseu->subslice_mask[0];
  		if (!value)
  			return -ENODEV;

break;

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#25 lore raw

Re: [PATCH 6/6] drm/i915: Expand subslice mask

In reply to Tvrtko Ursulin
Quoted 21 lines · Tvrtko Ursulin

On 01/05/2019 19:22, Tvrtko Ursulin wrote:

[snip]

Quoted 17 lines

drivers/gpu/drm/i915/intel_device_info.c void intel_device_info_dump_flags(const struct intel_device_info *info, · line 97

Quoted 13 lines
drivers/gpu/drm/i915/intel_device_info.c +4 -0
+#define SS_STR_MAX_SIZE (GEN_MAX_SUBSLICE_STRIDE * 2)
+
+static u8 *
+subslice_per_slice_str(u8 *buf, const struct sseu_dev_info

*sseu, u8
slice)

drivers/gpu/drm/i915/intel_device_info.c +7 -0
+{
+    int i;
+    u8 ss_offset = slice * sseu->ss_stride;
+
+    GEM_BUG_ON(slice >= sseu->max_slices);
+
+    memset(buf, 0, SS_STR_MAX_SIZE);

I suggest a more hardened approach of caller passing in the buffer size,
since it is their buffer.

Not a bad idea. I had the define to make this explicit and handle the future cases, but probably right it's better to isolate this. I'll make the change in the next series update.

Quoted 13 lines

Having said this..

Quoted 9 lines

drivers/gpu/drm/i915/intel_device_info.c void intel_device_info_dump_flags(const struct intel_device_info *info, · line 106

Quoted 9 lines
drivers/gpu/drm/i915/intel_device_info.c +9 -0
+
+    /*
+     * Print subslice information in reverse order to match
+     * userspace expectations.
+     */
+    for (i = 0; i < sseu->ss_stride; i++)
+        sprintf(&buf[i * 2], "%02x",
+            sseu->subslice_mask[ss_offset + sseu->ss_stride -
+                        (i + 1)]);
Quoted 4 lines

...sprintf also needs to check against overflowing the buffer. (Relationship between loop boundary (ss_stride) and buffer size is a bit
decoupled.)

I'll add the check, makes sense.

Quoted 2 lines

And buffer should probably be char *.

No problem. I'll make this change. Thanks for the feedback!

- Stuart

Quoted 297 lines

Regards,

Tvrtko

Quoted 292 lines
Quoted 292 lines
drivers/gpu/drm/i915/intel_device_info.c +4 -0
+
+    return buf;
+}
+
  static void sseu_dump(const struct sseu_dev_info *sseu, struct 

drm_printer *p)

  {
      int s;
+    u8 buf[SS_STR_MAX_SIZE];
      drm_printf(p, "slice total: %u, mask=%04x\n",
             hweight8(sseu->slice_mask), sseu->slice_mask);
      drm_printf(p, "subslice total: %u\n", 
intel_sseu_subslice_total(sseu));
      for (s = 0; s < sseu->max_slices; s++) {
drivers/gpu/drm/i915/intel_device_info.c +2 -2
-        drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
+        drm_printf(p, "slice%d: %u subslices, mask=%s\n",
                 s, intel_sseu_subslices_per_slice(sseu, s),
-               sseu->subslice_mask[s]);
+               subslice_per_slice_str(buf, sseu, s));
      }
      drm_printf(p, "EU total: %u\n", sseu->eu_total);
      drm_printf(p, "EU per subslice: %u\n", sseu-
Quoted 1 line

eu_per_subslice);

(quoted diff) +0 -0
@@ -118,6 +143,7 @@ void intel_device_info_dump_topology(const

struct
sseu_dev_info *sseu,
struct drm_printer *p)

  {
      int s, ss;
drivers/gpu/drm/i915/intel_device_info.c +1 -0
+    u8 buf[SS_STR_MAX_SIZE];
      if (sseu->max_slices == 0) {
          drm_printf(p, "Unavailable\n");
@@ -125,9 +151,9 @@ void intel_device_info_dump_topology(const

struct
sseu_dev_info *sseu,

      }
      for (s = 0; s < sseu->max_slices; s++) {
drivers/gpu/drm/i915/intel_device_info.c +10 -13
-        drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
+        drm_printf(p, "slice%d: %u subslice(s) (0x%s):\n",
                 s, intel_sseu_subslices_per_slice(sseu, s),
-               sseu->subslice_mask[s]);
+               subslice_per_slice_str(buf, sseu, s));
          for (ss = 0; ss < sseu->max_subslices; ss++) {
              u16 enabled_eus = intel_sseu_get_eus(sseu, s, ss);
@@ -156,15 +182,10 @@ static void gen11_sseu_info_init(struct 
 drm_i915_private *dev_priv)
      u8 eu_en;
      int s;
-    if (IS_ELKHARTLAKE(dev_priv)) {
-        sseu->max_slices = 1;
-        sseu->max_subslices = 4;
-        sseu->max_eus_per_subslice = 8;
-    } else {
-        sseu->max_slices = 1;
-        sseu->max_subslices = 8;
-        sseu->max_eus_per_subslice = 8;
-    }
+    if (IS_ELKHARTLAKE(dev_priv))
+        intel_sseu_set_info(sseu, 1, 4, 8);
+    else
+        intel_sseu_set_info(sseu, 1, 8, 8);
      s_en = I915_READ(GEN11_GT_SLICE_ENABLE) &
 GEN11_GT_S_ENA_MASK;
      ss_en = ~I915_READ(GEN11_GT_SUBSLICE_DISABLE);
@@ -177,9 +198,11 @@ static void gen11_sseu_info_init(struct 
 drm_i915_private *dev_priv)
              int ss;
              sseu->slice_mask |= BIT(s);
-            sseu->subslice_mask[s] = (ss_en >> ss_idx) &
 ss_en_mask;
+            sseu->subslice_mask[s * sseu->ss_stride] =
+                (ss_en >> ss_idx) & ss_en_mask;
              for (ss = 0; ss < sseu->max_subslices; ss++) {
-                if (sseu->subslice_mask[s] & BIT(ss))
+                if (sseu->subslice_mask[s * sseu->ss_stride] &
+                    BIT(ss))
                      intel_sseu_set_eus(sseu, s, ss, eu_en);
              }
          }
drivers/gpu/drm/i915/intel_device_info.c +2 -10
@@ -201,23 +224,10 @@ static void gen10_sseu_info_init(struct 
 drm_i915_private *dev_priv)
      const int eu_mask = 0xff;
      u32 subslice_mask, eu_en;
+    intel_sseu_set_info(sseu, 6, 4, 8);
+
      sseu->slice_mask = (fuse2 & GEN10_F2_S_ENA_MASK) >>
                  GEN10_F2_S_ENA_SHIFT;
-    sseu->max_slices = 6;
-    sseu->max_subslices = 4;
-    sseu->max_eus_per_subslice = 8;
-
-    subslice_mask = (1 << 4) - 1;
-    subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
-               GEN10_F2_SS_DIS_SHIFT);
-
-    /*
-     * Slice0 can have up to 3 subslices, but there are only 2

in

drivers/gpu/drm/i915/intel_device_info.c +0 -6
-     * slice1/2.
-     */
-    sseu->subslice_mask[0] = subslice_mask;
-    for (s = 1; s < sseu->max_slices; s++)
-        sseu->subslice_mask[s] = subslice_mask & 0x3;
      /* Slice0 */
      eu_en = ~I915_READ(GEN8_EU_DISABLE0);
@@ -242,14 +252,22 @@ static void gen10_sseu_info_init(struct 
 drm_i915_private *dev_priv)
      eu_en = ~I915_READ(GEN10_EU_DISABLE3);
      intel_sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
-    /* Do a second pass where we mark the subslices disabled if

all
their

drivers/gpu/drm/i915/intel_device_info.c +8 -3
-     * eus are off.
-     */
+    subslice_mask = (1 << 4) - 1;
+    subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
+               GEN10_F2_SS_DIS_SHIFT);
+
      for (s = 0; s < sseu->max_slices; s++) {
          for (ss = 0; ss < sseu->max_subslices; ss++) {
              if (intel_sseu_get_eus(sseu, s, ss) == 0)
-                sseu->subslice_mask[s] &= ~BIT(ss);
+                subslice_mask &= ~BIT(ss);
          }
+
+        /*
+         * Slice0 can have up to 3 subslices, but there are only

2 in

drivers/gpu/drm/i915/intel_device_info.c +3 -0
+         * slice1/2.
+         */
+        intel_sseu_set_subslices(sseu, s, s == 0 ? subslice_mask

:

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+                               subslice_mask & 0x3);
      }
      sseu->eu_total = compute_eu_total(sseu);
@@ -275,13 +293,12 @@ static void

cherryview_sseu_info_init(struct
drm_i915_private *dev_priv)

  {
      struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
      u32 fuse;
drivers/gpu/drm/i915/intel_device_info.c +3 -4
+    u8 subslice_mask;
      fuse = I915_READ(CHV_FUSE_GT);
      sseu->slice_mask = BIT(0);
-    sseu->max_slices = 1;
-    sseu->max_subslices = 2;
-    sseu->max_eus_per_subslice = 8;
+    intel_sseu_set_info(sseu, 1, 2, 8);
      if (!(fuse & CHV_FGT_DISABLE_SS0)) {
          u8 disabled_mask =
@@ -290,7 +307,7 @@ static void cherryview_sseu_info_init(struct 
 drm_i915_private *dev_priv)
              (((fuse & CHV_FGT_EU_DIS_SS0_R1_MASK) >>
                CHV_FGT_EU_DIS_SS0_R1_SHIFT) << 4);
-        sseu->subslice_mask[0] |= BIT(0);
+        subslice_mask |= BIT(0);
          intel_sseu_set_eus(sseu, 0, 0, ~disabled_mask);
      }
@@ -301,10 +318,12 @@ static void

cherryview_sseu_info_init(struct
drm_i915_private *dev_priv)

              (((fuse & CHV_FGT_EU_DIS_SS1_R1_MASK) >>
                CHV_FGT_EU_DIS_SS1_R1_SHIFT) << 4);
drivers/gpu/drm/i915/intel_device_info.c +3 -2
-        sseu->subslice_mask[0] |= BIT(1);
+        subslice_mask |= BIT(1);
          intel_sseu_set_eus(sseu, 0, 1, ~disabled_mask);
      }
+    intel_sseu_set_subslices(sseu, 0, subslice_mask);
+
      sseu->eu_total = compute_eu_total(sseu);
      /*
@@ -312,7 +331,8 @@ static void cherryview_sseu_info_init(struct 
 drm_i915_private *dev_priv)
       * across subslices.
      */
      sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
-                sseu->eu_total / intel_sseu_subslice_total(sseu)

:

drivers/gpu/drm/i915/intel_device_info.c +2 -0
+                sseu->eu_total /
+                    intel_sseu_subslice_total(sseu) :
                  0;
      /*
       * CHV supports subslice power gating on devices with more

than

drivers/gpu/drm/i915/intel_device_info.c +7 -8
@@ -336,9 +356,8 @@ static void gen9_sseu_info_init(struct 
 drm_i915_private *dev_priv)
      sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >> 
 GEN8_F2_S_ENA_SHIFT;
      /* BXT has a single slice and at most 3 subslices. */
-    sseu->max_slices = IS_GEN9_LP(dev_priv) ? 1 : 3;
-    sseu->max_subslices = IS_GEN9_LP(dev_priv) ? 3 : 4;
-    sseu->max_eus_per_subslice = 8;
+    intel_sseu_set_info(sseu, IS_GEN9_LP(dev_priv) ? 1 : 3,
+                IS_GEN9_LP(dev_priv) ? 3 : 4, 8);
      /*
       * The subslice disable field is global, i.e. it applies
@@ -357,14 +376,16 @@ static void gen9_sseu_info_init(struct 
 drm_i915_private *dev_priv)
              /* skip disabled slice */
              continue;
-        sseu->subslice_mask[s] = subslice_mask;
+        intel_sseu_set_subslices(sseu, s, subslice_mask);
          eu_disable = I915_READ(GEN9_EU_DISABLE(s));
          for (ss = 0; ss < sseu->max_subslices; ss++) {
              int eu_per_ss;
              u8 eu_disabled_mask;
+            u8 ss_idx = s * sseu->ss_stride + ss /
 BITS_PER_BYTE;
-            if (!(sseu->subslice_mask[s] & BIT(ss)))
+            if (!(sseu->subslice_mask[ss_idx] &
+                  BIT(ss % BITS_PER_BYTE)))
                  /* skip disabled subslice */
                  continue;
@@ -437,9 +458,7 @@ static void broadwell_sseu_info_init(struct 
 drm_i915_private *dev_priv)
      fuse2 = I915_READ(GEN8_FUSE2);
      sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >> 
 GEN8_F2_S_ENA_SHIFT;
-    sseu->max_slices = 3;
-    sseu->max_subslices = 3;
-    sseu->max_eus_per_subslice = 8;
+    intel_sseu_set_info(sseu, 3, 3, 8);
      /*
       * The subslice disable field is global, i.e. it applies
@@ -466,18 +485,21 @@ static void

broadwell_sseu_info_init(struct
drm_i915_private *dev_priv)

              /* skip disabled slice */
              continue;
drivers/gpu/drm/i915/intel_device_info.c +2 -1
-        sseu->subslice_mask[s] = subslice_mask;
+        intel_sseu_set_subslices(sseu, s, subslice_mask);
          for (ss = 0; ss < sseu->max_subslices; ss++) {
              u8 eu_disabled_mask;
+            u8 ss_idx = s * sseu->ss_stride + ss /
BITS_PER_BYTE;
              u32 n_disabled;
drivers/gpu/drm/i915/intel_device_info.c +2 -2
-            if (!(sseu->subslice_mask[s] & BIT(ss)))
+            if (!(sseu->subslice_mask[ss_idx] &
+                  BIT(ss % BITS_PER_BYTE)))
                  /* skip disabled subslice */
                  continue;
              eu_disabled_mask =
-                eu_disable[s] >> (ss * sseu-
Quoted 1 line

max_eus_per_subslice);

drivers/gpu/drm/i915/intel_device_info.c +3 -0
+                eu_disable[s] >>
+                    (ss * sseu->max_eus_per_subslice);
              intel_sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
@@ -517,6 +539,7 @@ static void haswell_sseu_info_init(struct 
 drm_i915_private *dev_priv)
      struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
      u32 fuse1;
      int s, ss;
+    u32 subslice_mask;
      /*
       * There isn't a register to tell us how many

slices/subslices. We

drivers/gpu/drm/i915/intel_device_info.c +3 -7
@@ -528,22 +551,18 @@ static void haswell_sseu_info_init(struct 
 drm_i915_private *dev_priv)
          /* fall through */
      case 1:
          sseu->slice_mask = BIT(0);
-        sseu->subslice_mask[0] = BIT(0);
+        subslice_mask = BIT(0);
          break;
      case 2:
          sseu->slice_mask = BIT(0);
-        sseu->subslice_mask[0] = BIT(0) | BIT(1);
+        subslice_mask = BIT(0) | BIT(1);
          break;
      case 3:
          sseu->slice_mask = BIT(0) | BIT(1);
-        sseu->subslice_mask[0] = BIT(0) | BIT(1);
-        sseu->subslice_mask[1] = BIT(0) | BIT(1);
+        subslice_mask = BIT(0) | BIT(1);
          break;
      }
-    sseu->max_slices = hweight8(sseu->slice_mask);
-    sseu->max_subslices = hweight8(sseu->subslice_mask[0]);
-
      fuse1 = I915_READ(HSW_PAVP_FUSE1);
      switch ((fuse1 & HSW_F1_EU_DIS_MASK) >>
 HSW_F1_EU_DIS_SHIFT) {

default:

drivers/gpu/drm/i915/intel_device_info.c +3 -1
@@ -560,9 +579,14 @@ static void haswell_sseu_info_init(struct 
 drm_i915_private *dev_priv)
          sseu->eu_per_subslice = 6;
          break;
      }
-    sseu->max_eus_per_subslice = sseu->eu_per_subslice;
+
+    intel_sseu_set_info(sseu, hweight8(sseu->slice_mask),
+                hweight8(subslice_mask),
Quoted 8 lines
Quoted 8 lines

drivers/gpu/drm/i915/intel_device_info.c static void haswell_sseu_info_init(struct drm_i915_private *dev_priv) · line 592

Quoted 8 lines
drivers/gpu/drm/i915/intel_device_info.c +3 -0
+                sseu->eu_per_subslice);
      for (s = 0; s < sseu->max_slices; s++) {
+        intel_sseu_set_subslices(sseu, s, subslice_mask);
+
          for (ss = 0; ss < sseu->max_subslices; ss++) {
              intel_sseu_set_eus(sseu, s, ss,
                         (1UL << sseu->eu_per_subslice) - 1);

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#14 lore raw

Re: [PATCH 5/6] drm/i915: Remove inline from sseu helper functions

In reply to Stuart Summers

Can you elaborate a bit more on what's the rationale for this? do you just want to avoid having too many inlines since the paths they're used in are not critical, or do you have some more functional reason? This is not a critic to the patch, I just want to understand where you're coming from ;)

BTW, looking at this patch I realized there are a few more DIV_ROUND_UP(..., BITS_PER_BYTE) that could be converted to GEN_SSEU_STRIDE() in patch 2. I noticed you update them to a new variable in the next patch, but for consistency it might still be worth updating them all in patch 2 or at least mention in the commit message of patch 2 that the remaining cases are updated by a follow-up patch in the series. Patch 2 is quite small, so you could also just squash it into patch 6 to avoid the split.

Daniele

Quoted 378 lines · Stuart Summers

Additionally, ensure these are all prefixed with intel_sseu_* to match the convention of other functions in i915.

---
  drivers/gpu/drm/i915/gt/intel_sseu.c     | 54 +++++++++++++++++++
  drivers/gpu/drm/i915/gt/intel_sseu.h     | 57 +++-----------------
  drivers/gpu/drm/i915/i915_debugfs.c      |  6 +--
  drivers/gpu/drm/i915/i915_drv.c          |  2 +-
  drivers/gpu/drm/i915/intel_device_info.c | 69 ++++++++++++------------
  5 files changed, 102 insertions(+), 86 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.c +54 -0
index 7f448f3bea0b..4a0b82fc108c 100644
@@ -8,6 +8,60 @@
  #include "intel_lrc_reg.h"
  #include "intel_sseu.h"
 
+unsigned int
+intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
+{
+	unsigned int i, total = 0;
+
+	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
+		total += hweight8(sseu->subslice_mask[i]);
+
+	return total;
+}
+
+unsigned int
+intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice)
+{
+	return hweight8(sseu->subslice_mask[slice]);
+}
+
+static int intel_sseu_eu_idx(const struct sseu_dev_info *sseu, int slice,
+			     int subslice)
+{
+	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
+					   BITS_PER_BYTE);
+	int slice_stride = sseu->max_subslices * subslice_stride;
+
+	return slice * slice_stride + subslice * subslice_stride;
+}
+
+u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int slice,
+		       int subslice)
+{
+	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
+	u16 eu_mask = 0;
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
+			(i * BITS_PER_BYTE);
+	}
+
+	return eu_mask;
+}
+
+void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice, int subslice,
+			u16 eu_mask)
+{
+	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+		sseu->eu_mask[offset + i] =
+			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
+	}
+}
+
  u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
  			 const struct intel_sseu *req_sseu)
  {
drivers/gpu/drm/i915/gt/intel_sseu.h +8 -49
index 029e71d8f140..56e3721ae83f 100644
@@ -63,58 +63,17 @@ intel_sseu_from_device_info(const struct sseu_dev_info *sseu)
  	return value;
  }
 
-static inline unsigned int sseu_subslice_total(const struct sseu_dev_info *sseu)
-{
-	unsigned int i, total = 0;
-
-	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
-		total += hweight8(sseu->subslice_mask[i]);
+unsigned int
+intel_sseu_subslice_total(const struct sseu_dev_info *sseu);
 
-	return total;
-}
+unsigned int
+intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice);
 
-static inline unsigned int
-sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice)
-{
-	return hweight8(sseu->subslice_mask[slice]);
-}
-
-static inline int sseu_eu_idx(const struct sseu_dev_info *sseu,
-			      int slice, int subslice)
-{
-	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
-					   BITS_PER_BYTE);
-	int slice_stride = sseu->max_subslices * subslice_stride;
-
-	return slice * slice_stride + subslice * subslice_stride;
-}
+u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int slice,
+		       int subslice);
 
-static inline u16 sseu_get_eus(const struct sseu_dev_info *sseu,
-			       int slice, int subslice)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-	u16 eu_mask = 0;
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
-		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
-			(i * BITS_PER_BYTE);
-	}
-
-	return eu_mask;
-}
-
-static inline void sseu_set_eus(struct sseu_dev_info *sseu,
-				int slice, int subslice, u16 eu_mask)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
-		sseu->eu_mask[offset + i] =
-			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
-	}
-}
+void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice, int subslice,
+			u16 eu_mask);
 
  u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
  			 const struct intel_sseu *req_sseu);
drivers/gpu/drm/i915/i915_debugfs.c +3 -3
index fe854c629a32..3f3ee83ac315 100644
@@ -4158,7 +4158,7 @@ static void broadwell_sseu_device_status(struct drm_i915_private *dev_priv,
  				RUNTIME_INFO(dev_priv)->sseu.subslice_mask[s];
  		}
  		sseu->eu_total = sseu->eu_per_subslice *
-				 sseu_subslice_total(sseu);
+				 intel_sseu_subslice_total(sseu);
 
  		/* subtract fused off EU(s) from enabled slice(s) */
  		for (s = 0; s < fls(sseu->slice_mask); s++) {
@@ -4182,10 +4182,10 @@ static void i915_print_sseu_info(struct seq_file *m, bool is_available_info,
  	seq_printf(m, "  %s Slice Total: %u\n", type,
  		   hweight8(sseu->slice_mask));
  	seq_printf(m, "  %s Subslice Total: %u\n", type,
-		   sseu_subslice_total(sseu));
+		   intel_sseu_subslice_total(sseu));
  	for (s = 0; s < fls(sseu->slice_mask); s++) {
  		seq_printf(m, "  %s Slice%i subslices: %u\n", type,
-			   s, sseu_subslices_per_slice(sseu, s));
+			   s, intel_sseu_subslices_per_slice(sseu, s));
  	}
  	seq_printf(m, "  %s EU Total: %u\n", type,
  		   sseu->eu_total);
drivers/gpu/drm/i915/i915_drv.c +1 -1
index c376244c19c4..130c5140db0d 100644
@@ -378,7 +378,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
  		value = i915_cmd_parser_get_version(dev_priv);
  		break;
  	case I915_PARAM_SUBSLICE_TOTAL:
-		value = sseu_subslice_total(sseu);
+		value = intel_sseu_subslice_total(sseu);
  		if (!value)
  			return -ENODEV;
  		break;
drivers/gpu/drm/i915/intel_device_info.c +34 -31
index 559cf0d0628e..e1dbccf04cd9 100644
@@ -90,10 +90,10 @@ static void sseu_dump(const struct sseu_dev_info *sseu, struct drm_printer *p)
 
  	drm_printf(p, "slice total: %u, mask=%04x\n",
  		   hweight8(sseu->slice_mask), sseu->slice_mask);
-	drm_printf(p, "subslice total: %u\n", sseu_subslice_total(sseu));
+	drm_printf(p, "subslice total: %u\n", intel_sseu_subslice_total(sseu));
  	for (s = 0; s < sseu->max_slices; s++) {
  		drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
-			   s, sseu_subslices_per_slice(sseu, s),
+			   s, intel_sseu_subslices_per_slice(sseu, s),
  			   sseu->subslice_mask[s]);
  	}
  	drm_printf(p, "EU total: %u\n", sseu->eu_total);
@@ -126,11 +126,11 @@ void intel_device_info_dump_topology(const struct sseu_dev_info *sseu,
 
  	for (s = 0; s < sseu->max_slices; s++) {
  		drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
-			   s, sseu_subslices_per_slice(sseu, s),
+			   s, intel_sseu_subslices_per_slice(sseu, s),
  			   sseu->subslice_mask[s]);
 
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
-			u16 enabled_eus = sseu_get_eus(sseu, s, ss);
+			u16 enabled_eus = intel_sseu_get_eus(sseu, s, ss);
 
  			drm_printf(p, "\tsubslice%d: %u EUs (0x%hx)\n",
  				   ss, hweight16(enabled_eus), enabled_eus);
@@ -180,7 +180,7 @@ static void gen11_sseu_info_init(struct drm_i915_private *dev_priv)
  			sseu->subslice_mask[s] = (ss_en >> ss_idx) & ss_en_mask;
  			for (ss = 0; ss < sseu->max_subslices; ss++) {
  				if (sseu->subslice_mask[s] & BIT(ss))
-					sseu_set_eus(sseu, s, ss, eu_en);
+					intel_sseu_set_eus(sseu, s, ss, eu_en);
  			}
  		}
  	}
@@ -222,32 +222,32 @@ static void gen10_sseu_info_init(struct drm_i915_private *dev_priv)
  	/* Slice0 */
  	eu_en = ~I915_READ(GEN8_EU_DISABLE0);
  	for (ss = 0; ss < sseu->max_subslices; ss++)
-		sseu_set_eus(sseu, 0, ss, (eu_en >> (8 * ss)) & eu_mask);
+		intel_sseu_set_eus(sseu, 0, ss, (eu_en >> (8 * ss)) & eu_mask);
  	/* Slice1 */
-	sseu_set_eus(sseu, 1, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 1, 0, (eu_en >> 24) & eu_mask);
  	eu_en = ~I915_READ(GEN8_EU_DISABLE1);
-	sseu_set_eus(sseu, 1, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 1, 1, eu_en & eu_mask);
  	/* Slice2 */
-	sseu_set_eus(sseu, 2, 0, (eu_en >> 8) & eu_mask);
-	sseu_set_eus(sseu, 2, 1, (eu_en >> 16) & eu_mask);
+	intel_sseu_set_eus(sseu, 2, 0, (eu_en >> 8) & eu_mask);
+	intel_sseu_set_eus(sseu, 2, 1, (eu_en >> 16) & eu_mask);
  	/* Slice3 */
-	sseu_set_eus(sseu, 3, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 3, 0, (eu_en >> 24) & eu_mask);
  	eu_en = ~I915_READ(GEN8_EU_DISABLE2);
-	sseu_set_eus(sseu, 3, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 3, 1, eu_en & eu_mask);
  	/* Slice4 */
-	sseu_set_eus(sseu, 4, 0, (eu_en >> 8) & eu_mask);
-	sseu_set_eus(sseu, 4, 1, (eu_en >> 16) & eu_mask);
+	intel_sseu_set_eus(sseu, 4, 0, (eu_en >> 8) & eu_mask);
+	intel_sseu_set_eus(sseu, 4, 1, (eu_en >> 16) & eu_mask);
  	/* Slice5 */
-	sseu_set_eus(sseu, 5, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 5, 0, (eu_en >> 24) & eu_mask);
  	eu_en = ~I915_READ(GEN10_EU_DISABLE3);
-	sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
 
  	/* Do a second pass where we mark the subslices disabled if all their
  	 * eus are off.
  	 */
  	for (s = 0; s < sseu->max_slices; s++) {
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
-			if (sseu_get_eus(sseu, s, ss) == 0)
+			if (intel_sseu_get_eus(sseu, s, ss) == 0)
  				sseu->subslice_mask[s] &= ~BIT(ss);
  		}
  	}
@@ -260,9 +260,10 @@ static void gen10_sseu_info_init(struct drm_i915_private *dev_priv)
  	 * EU in any one subslice may be fused off for die
  	 * recovery.
  	 */
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
  				DIV_ROUND_UP(sseu->eu_total,
-					     sseu_subslice_total(sseu)) : 0;
+					     intel_sseu_subslice_total(sseu)) :
+				0;
 
  	/* No restrictions on Power Gating */
  	sseu->has_slice_pg = 1;
@@ -290,7 +291,7 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  			  CHV_FGT_EU_DIS_SS0_R1_SHIFT) << 4);
 
  		sseu->subslice_mask[0] |= BIT(0);
-		sseu_set_eus(sseu, 0, 0, ~disabled_mask);
+		intel_sseu_set_eus(sseu, 0, 0, ~disabled_mask);
  	}
 
  	if (!(fuse & CHV_FGT_DISABLE_SS1)) {
@@ -301,7 +302,7 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  			  CHV_FGT_EU_DIS_SS1_R1_SHIFT) << 4);
 
  		sseu->subslice_mask[0] |= BIT(1);
-		sseu_set_eus(sseu, 0, 1, ~disabled_mask);
+		intel_sseu_set_eus(sseu, 0, 1, ~disabled_mask);
  	}
 
  	sseu->eu_total = compute_eu_total(sseu);
@@ -310,8 +311,8 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  	 * CHV expected to always have a uniform distribution of EU
  	 * across subslices.
  	*/
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
-				sseu->eu_total / sseu_subslice_total(sseu) :
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
+				sseu->eu_total / intel_sseu_subslice_total(sseu) :
  				0;
  	/*
  	 * CHV supports subslice power gating on devices with more than
@@ -319,7 +320,7 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  	 * more than one EU pair per subslice.
  	*/
  	sseu->has_slice_pg = 0;
-	sseu->has_subslice_pg = sseu_subslice_total(sseu) > 1;
+	sseu->has_subslice_pg = intel_sseu_subslice_total(sseu) > 1;
  	sseu->has_eu_pg = (sseu->eu_per_subslice > 2);
  }
 
@@ -369,7 +370,7 @@ static void gen9_sseu_info_init(struct drm_i915_private *dev_priv)
 
  			eu_disabled_mask = (eu_disable >> (ss * 8)) & eu_mask;
 
-			sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
+			intel_sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
 
  			eu_per_ss = sseu->max_eus_per_subslice -
  				hweight8(eu_disabled_mask);
@@ -393,9 +394,10 @@ static void gen9_sseu_info_init(struct drm_i915_private *dev_priv)
  	 * recovery. BXT is expected to be perfectly uniform in EU
  	 * distribution.
  	*/
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
  				DIV_ROUND_UP(sseu->eu_total,
-					     sseu_subslice_total(sseu)) : 0;
+					     intel_sseu_subslice_total(sseu)) :
+				0;
  	/*
  	 * SKL+ supports slice power gating on devices with more than
  	 * one slice, and supports EU power gating on devices with
@@ -407,7 +409,7 @@ static void gen9_sseu_info_init(struct drm_i915_private *dev_priv)
  	sseu->has_slice_pg =
  		!IS_GEN9_LP(dev_priv) && hweight8(sseu->slice_mask) > 1;
  	sseu->has_subslice_pg =
-		IS_GEN9_LP(dev_priv) && sseu_subslice_total(sseu) > 1;
+		IS_GEN9_LP(dev_priv) && intel_sseu_subslice_total(sseu) > 1;
  	sseu->has_eu_pg = sseu->eu_per_subslice > 2;
 
  	if (IS_GEN9_LP(dev_priv)) {
@@ -477,7 +479,7 @@ static void broadwell_sseu_info_init(struct drm_i915_private *dev_priv)
  			eu_disabled_mask =
  				eu_disable[s] >> (ss * sseu->max_eus_per_subslice);
 
-			sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
+			intel_sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
 
  			n_disabled = hweight8(eu_disabled_mask);
 
@@ -496,9 +498,10 @@ static void broadwell_sseu_info_init(struct drm_i915_private *dev_priv)
  	 * subslices with the exception that any one EU in any one subslice may
  	 * be fused off for die recovery.
  	 */
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
  				DIV_ROUND_UP(sseu->eu_total,
-					     sseu_subslice_total(sseu)) : 0;
+					     intel_sseu_subslice_total(sseu)) :
+				0;
 
  	/*
  	 * BDW supports slice power gating on devices with more than

drivers/gpu/drm/i915/intel_device_info.c static void haswell_sseu_info_init(struct drm_i915_private *dev_priv) · line 571

Quoted 10 lines
drivers/gpu/drm/i915/intel_device_info.c +2 -2
@@ -561,8 +564,8 @@ static void haswell_sseu_info_init(struct drm_i915_private *dev_priv)
 
  	for (s = 0; s < sseu->max_slices; s++) {
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
-			sseu_set_eus(sseu, s, ss,
-				     (1UL << sseu->eu_per_subslice) - 1);
+			intel_sseu_set_eus(sseu, s, ss,
+					   (1UL << sseu->eu_per_subslice) - 1);
  		}
  	}
 

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#15 lore raw

Re: [PATCH 5/6] drm/i915: Remove inline from sseu helper functions

Quoted 9 lines · Daniele Ceraolo Spurio

Can you elaborate a bit more on what's the rationale for this? do you
just want to avoid having too many inlines since the paths they're used
in are not critical, or do you have some more functional reason? This is
not a critic to the patch, I just want to understand where you're coming
from ;)

This was a request from Jani Nikula in a previous series update. I don't have a strong preference either way personally. If you don't have any major concerns, I'd prefer to keep the series as-is to prevent too much thrash here, but let me know.

Quoted 12 lines

BTW, looking at this patch I realized there are a few more DIV_ROUND_UP(..., BITS_PER_BYTE) that could be converted to GEN_SSEU_STRIDE() in patch 2. I noticed you update them to a new variable in the next patch, but for consistency it might still be worth
updating them all in patch 2 or at least mention in the commit message
of patch 2 that the remaining cases are updated by a follow-up patch in
the series. Patch 2 is quite small, so you could also just squash it into patch 6 to avoid the split.

I'm happy to squash them. I did try to isolate this a bit, but you're right that I ended up pushing some of these DIV_ROUND_UP... stride calculations to the last patch in the series. If you don't have any objection, to keep the finaly patch a bit simpler, I'd rather pull those changes into the earlier patch. I realize you already have a RB on that patch. Any issues doing this?

Thanks,
Stuart

Quoted 459 lines

Daniele

Quoted 455 lines · Stuart Summers

Additionally, ensure these are all prefixed with intel_sseu_* to match the convention of other functions in i915.

---
  drivers/gpu/drm/i915/gt/intel_sseu.c     | 54 +++++++++++++++++++
  drivers/gpu/drm/i915/gt/intel_sseu.h     | 57 +++--------------
---
  drivers/gpu/drm/i915/i915_debugfs.c      |  6 +--
  drivers/gpu/drm/i915/i915_drv.c          |  2 +-
  drivers/gpu/drm/i915/intel_device_info.c | 69 ++++++++++++-------
-----
  5 files changed, 102 insertions(+), 86 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.c +35 -0
index 7f448f3bea0b..4a0b82fc108c 100644
@@ -8,6 +8,60 @@
  #include "intel_lrc_reg.h"
  #include "intel_sseu.h"
 
+unsigned int
+intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
+{
+	unsigned int i, total = 0;
+
+	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
+		total += hweight8(sseu->subslice_mask[i]);
+
+	return total;
+}
+
+unsigned int
+intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu,
 u8 slice)
+{
+	return hweight8(sseu->subslice_mask[slice]);
+}
+
+static int intel_sseu_eu_idx(const struct sseu_dev_info *sseu, int
 slice,
+			     int subslice)
+{
+	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
+					   BITS_PER_BYTE);
+	int slice_stride = sseu->max_subslices * subslice_stride;
+
+	return slice * slice_stride + subslice * subslice_stride;
+}
+
+u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int
 slice,
+		       int subslice)
+{
+	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
+	u16 eu_mask = 0;
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.c +8 -0
+		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
+			(i * BITS_PER_BYTE);
+	}
+
+	return eu_mask;
+}
+
+void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice, int

subslice,

drivers/gpu/drm/i915/gt/intel_sseu.c +6 -0
+			u16 eu_mask)
+{
+	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.c +5 -0
+		sseu->eu_mask[offset + i] =
+			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
+	}
+}
+
  u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
  			 const struct intel_sseu *req_sseu)
  {
drivers/gpu/drm/i915/gt/intel_sseu.h +6 -31
index 029e71d8f140..56e3721ae83f 100644
@@ -63,58 +63,17 @@ intel_sseu_from_device_info(const struct
 sseu_dev_info *sseu)
  	return value;
  }
 
-static inline unsigned int sseu_subslice_total(const struct
 sseu_dev_info *sseu)
-{
-	unsigned int i, total = 0;
-
-	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
-		total += hweight8(sseu->subslice_mask[i]);
+unsigned int
+intel_sseu_subslice_total(const struct sseu_dev_info *sseu);
 
-	return total;
-}
+unsigned int
+intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu,
 u8 slice);
 
-static inline unsigned int
-sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8
 slice)
-{
-	return hweight8(sseu->subslice_mask[slice]);
-}
-
-static inline int sseu_eu_idx(const struct sseu_dev_info *sseu,
-			      int slice, int subslice)
-{
-	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
-					   BITS_PER_BYTE);
-	int slice_stride = sseu->max_subslices * subslice_stride;
-
-	return slice * slice_stride + subslice * subslice_stride;
-}
+u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int
 slice,
+		       int subslice);
 
-static inline u16 sseu_get_eus(const struct sseu_dev_info *sseu,
-			       int slice, int subslice)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-	u16 eu_mask = 0;
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.h +0 -14
-		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
-			(i * BITS_PER_BYTE);
-	}
-
-	return eu_mask;
-}
-
-static inline void sseu_set_eus(struct sseu_dev_info *sseu,
-				int slice, int subslice, u16 eu_mask)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.h +1 -4
-		sseu->eu_mask[offset + i] =
-			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
-	}
-}
+void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice, int

subslice,

drivers/gpu/drm/i915/gt/intel_sseu.h +1 -0
+			u16 eu_mask);
 
  u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
  			 const struct intel_sseu *req_sseu);
drivers/gpu/drm/i915/i915_debugfs.c +1 -1
index fe854c629a32..3f3ee83ac315 100644
@@ -4158,7 +4158,7 @@ static void
 broadwell_sseu_device_status(struct drm_i915_private *dev_priv,
  				RUNTIME_INFO(dev_priv)-
 >sseu.subslice_mask[s];
  		}
  		sseu->eu_total = sseu->eu_per_subslice *
-				 sseu_subslice_total(sseu);
+				 intel_sseu_subslice_total(sseu);
 
  		/* subtract fused off EU(s) from enabled slice(s) */
  		for (s = 0; s < fls(sseu->slice_mask); s++) {
drivers/gpu/drm/i915/i915_debugfs.c +2 -2
@@ -4182,10 +4182,10 @@ static void i915_print_sseu_info(struct
 seq_file *m, bool is_available_info,
  	seq_printf(m, "  %s Slice Total: %u\n", type,
  		   hweight8(sseu->slice_mask));
  	seq_printf(m, "  %s Subslice Total: %u\n", type,
-		   sseu_subslice_total(sseu));
+		   intel_sseu_subslice_total(sseu));
  	for (s = 0; s < fls(sseu->slice_mask); s++) {
  		seq_printf(m, "  %s Slice%i subslices: %u\n", type,
-			   s, sseu_subslices_per_slice(sseu, s));
+			   s, intel_sseu_subslices_per_slice(sseu, s));
  	}
  	seq_printf(m, "  %s EU Total: %u\n", type,

sseu->eu_total);

drivers/gpu/drm/i915/i915_drv.c +1 -1
index c376244c19c4..130c5140db0d 100644
@@ -378,7 +378,7 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
  		value = i915_cmd_parser_get_version(dev_priv);
  		break;
  	case I915_PARAM_SUBSLICE_TOTAL:
-		value = sseu_subslice_total(sseu);
+		value = intel_sseu_subslice_total(sseu);
  		if (!value)
  			return -ENODEV;

break;

drivers/gpu/drm/i915/intel_device_info.c +2 -2
index 559cf0d0628e..e1dbccf04cd9 100644
@@ -90,10 +90,10 @@ static void sseu_dump(const struct
 sseu_dev_info *sseu, struct drm_printer *p)
 
  	drm_printf(p, "slice total: %u, mask=%04x\n",
  		   hweight8(sseu->slice_mask), sseu->slice_mask);
-	drm_printf(p, "subslice total: %u\n",
 sseu_subslice_total(sseu));
+	drm_printf(p, "subslice total: %u\n",
 intel_sseu_subslice_total(sseu));
  	for (s = 0; s < sseu->max_slices; s++) {
  		drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
-			   s, sseu_subslices_per_slice(sseu, s),
+			   s, intel_sseu_subslices_per_slice(sseu, s),
  			   sseu->subslice_mask[s]);
  	}
  	drm_printf(p, "EU total: %u\n", sseu->eu_total);
drivers/gpu/drm/i915/intel_device_info.c +2 -2
@@ -126,11 +126,11 @@ void intel_device_info_dump_topology(const
 struct sseu_dev_info *sseu,
 
  	for (s = 0; s < sseu->max_slices; s++) {
  		drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
-			   s, sseu_subslices_per_slice(sseu, s),
+			   s, intel_sseu_subslices_per_slice(sseu, s),
  			   sseu->subslice_mask[s]);
 
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
-			u16 enabled_eus = sseu_get_eus(sseu, s, ss);
+			u16 enabled_eus = intel_sseu_get_eus(sseu, s,
 ss);
 

drm_printf(p, "\tsubslice%d: %u EUs (0x%hx)\n",
ss, hweight16(enabled_eus),
enabled_eus);

drivers/gpu/drm/i915/intel_device_info.c +1 -1
@@ -180,7 +180,7 @@ static void gen11_sseu_info_init(struct
 drm_i915_private *dev_priv)
  			sseu->subslice_mask[s] = (ss_en >> ss_idx) &
 ss_en_mask;
  			for (ss = 0; ss < sseu->max_subslices; ss++) {
  				if (sseu->subslice_mask[s] & BIT(ss))
-					sseu_set_eus(sseu, s, ss,
 eu_en);
+					intel_sseu_set_eus(sseu, s, ss,

eu_en);
}
}
}

drivers/gpu/drm/i915/intel_device_info.c +11 -11
@@ -222,32 +222,32 @@ static void gen10_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	/* Slice0 */
  	eu_en = ~I915_READ(GEN8_EU_DISABLE0);
  	for (ss = 0; ss < sseu->max_subslices; ss++)
-		sseu_set_eus(sseu, 0, ss, (eu_en >> (8 * ss)) &
 eu_mask);
+		intel_sseu_set_eus(sseu, 0, ss, (eu_en >> (8 * ss)) &
 eu_mask);
  	/* Slice1 */
-	sseu_set_eus(sseu, 1, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 1, 0, (eu_en >> 24) & eu_mask);
  	eu_en = ~I915_READ(GEN8_EU_DISABLE1);
-	sseu_set_eus(sseu, 1, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 1, 1, eu_en & eu_mask);
  	/* Slice2 */
-	sseu_set_eus(sseu, 2, 0, (eu_en >> 8) & eu_mask);
-	sseu_set_eus(sseu, 2, 1, (eu_en >> 16) & eu_mask);
+	intel_sseu_set_eus(sseu, 2, 0, (eu_en >> 8) & eu_mask);
+	intel_sseu_set_eus(sseu, 2, 1, (eu_en >> 16) & eu_mask);
  	/* Slice3 */
-	sseu_set_eus(sseu, 3, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 3, 0, (eu_en >> 24) & eu_mask);
  	eu_en = ~I915_READ(GEN8_EU_DISABLE2);
-	sseu_set_eus(sseu, 3, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 3, 1, eu_en & eu_mask);
  	/* Slice4 */
-	sseu_set_eus(sseu, 4, 0, (eu_en >> 8) & eu_mask);
-	sseu_set_eus(sseu, 4, 1, (eu_en >> 16) & eu_mask);
+	intel_sseu_set_eus(sseu, 4, 0, (eu_en >> 8) & eu_mask);
+	intel_sseu_set_eus(sseu, 4, 1, (eu_en >> 16) & eu_mask);
  	/* Slice5 */
-	sseu_set_eus(sseu, 5, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 5, 0, (eu_en >> 24) & eu_mask);
  	eu_en = ~I915_READ(GEN10_EU_DISABLE3);
-	sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
 
  	/* Do a second pass where we mark the subslices disabled if all

their
* eus are off.
*/
for (s = 0; s < sseu->max_slices; s++) {
for (ss = 0; ss < sseu->max_subslices; ss++) {

drivers/gpu/drm/i915/intel_device_info.c +3 -3
-			if (sseu_get_eus(sseu, s, ss) == 0)
+			if (intel_sseu_get_eus(sseu, s, ss) == 0)
  				sseu->subslice_mask[s] &= ~BIT(ss);
  		}
  	}
@@ -260,9 +260,10 @@ static void gen10_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	 * EU in any one subslice may be fused off for die
  	 * recovery.
  	 */
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
  				DIV_ROUND_UP(sseu->eu_total,
-					     sseu_subslice_total(sseu))
 : 0;
+					     intel_sseu_subslice_total(

sseu)) :

drivers/gpu/drm/i915/intel_device_info.c +2 -1
+				0;
 
  	/* No restrictions on Power Gating */
  	sseu->has_slice_pg = 1;
@@ -290,7 +291,7 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
  			  CHV_FGT_EU_DIS_SS0_R1_SHIFT) << 4);
 
  		sseu->subslice_mask[0] |= BIT(0);
-		sseu_set_eus(sseu, 0, 0, ~disabled_mask);
+		intel_sseu_set_eus(sseu, 0, 0, ~disabled_mask);
  	}
 
  	if (!(fuse & CHV_FGT_DISABLE_SS1)) {
drivers/gpu/drm/i915/intel_device_info.c +1 -1
@@ -301,7 +302,7 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
  			  CHV_FGT_EU_DIS_SS1_R1_SHIFT) << 4);
 
  		sseu->subslice_mask[0] |= BIT(1);
-		sseu_set_eus(sseu, 0, 1, ~disabled_mask);
+		intel_sseu_set_eus(sseu, 0, 1, ~disabled_mask);
  	}
 
  	sseu->eu_total = compute_eu_total(sseu);
drivers/gpu/drm/i915/intel_device_info.c +0 -2
@@ -310,8 +311,8 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	 * CHV expected to always have a uniform distribution of EU
  	 * across subslices.
  	*/
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
-				sseu->eu_total /

sseu_subslice_total(sseu) :

drivers/gpu/drm/i915/intel_device_info.c +2 -0
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
+				sseu->eu_total /

intel_sseu_subslice_total(sseu) :
0;
/*
* CHV supports subslice power gating on devices with more than

drivers/gpu/drm/i915/intel_device_info.c +2 -2
@@ -319,7 +320,7 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	 * more than one EU pair per subslice.
  	*/
  	sseu->has_slice_pg = 0;
-	sseu->has_subslice_pg = sseu_subslice_total(sseu) > 1;
+	sseu->has_subslice_pg = intel_sseu_subslice_total(sseu) > 1;
  	sseu->has_eu_pg = (sseu->eu_per_subslice > 2);
  }
@@ -369,7 +370,7 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
 
  			eu_disabled_mask = (eu_disable >> (ss * 8)) &
 eu_mask;
 
-			sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
+			intel_sseu_set_eus(sseu, s, ss,
 ~eu_disabled_mask);
  
  			eu_per_ss = sseu->max_eus_per_subslice -
  				hweight8(eu_disabled_mask);
drivers/gpu/drm/i915/intel_device_info.c +2 -2
@@ -393,9 +394,10 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	 * recovery. BXT is expected to be perfectly uniform in EU
  	 * distribution.
  	*/
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
  				DIV_ROUND_UP(sseu->eu_total,
-					     sseu_subslice_total(sseu))
 : 0;
+					     intel_sseu_subslice_total(

sseu)) :

drivers/gpu/drm/i915/intel_device_info.c +2 -1
+				0;
  	/*
  	 * SKL+ supports slice power gating on devices with more than
  	 * one slice, and supports EU power gating on devices with
@@ -407,7 +409,7 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	sseu->has_slice_pg =
  		!IS_GEN9_LP(dev_priv) && hweight8(sseu->slice_mask) >
 1;
  	sseu->has_subslice_pg =
-		IS_GEN9_LP(dev_priv) && sseu_subslice_total(sseu) > 1;
+		IS_GEN9_LP(dev_priv) && intel_sseu_subslice_total(sseu)
 > 1;
  	sseu->has_eu_pg = sseu->eu_per_subslice > 2;
  
  	if (IS_GEN9_LP(dev_priv)) {
drivers/gpu/drm/i915/intel_device_info.c +1 -1
@@ -477,7 +479,7 @@ static void broadwell_sseu_info_init(struct
 drm_i915_private *dev_priv)
  			eu_disabled_mask =
  				eu_disable[s] >> (ss * sseu-
 >max_eus_per_subslice);
 
-			sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
+			intel_sseu_set_eus(sseu, s, ss,
 ~eu_disabled_mask);
  
  			n_disabled = hweight8(eu_disabled_mask);
drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -496,9 +498,10 @@ static void broadwell_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	 * subslices with the exception that any one EU in any one

subslice may
* be fused off for die recovery.
*/

drivers/gpu/drm/i915/intel_device_info.c +1 -2
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
  				DIV_ROUND_UP(sseu->eu_total,
-					     sseu_subslice_total(sseu))

: 0;

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+					     intel_sseu_subslice_total(

sseu)) :

drivers/gpu/drm/i915/intel_device_info.c +3 -2
+				0;
 
  	/*
  	 * BDW supports slice power gating on devices with more than
@@ -561,8 +564,8 @@ static void haswell_sseu_info_init(struct
 drm_i915_private *dev_priv)
 
  	for (s = 0; s < sseu->max_slices; s++) {
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
-			sseu_set_eus(sseu, s, ss,
-				     (1UL << sseu->eu_per_subslice) -
 1);
+			intel_sseu_set_eus(sseu, s, ss,
+					   (1UL << sseu-
 >eu_per_subslice) - 1);
  		}
  	}

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#16 lore raw

Re: [PATCH 5/6] drm/i915: Remove inline from sseu helper functions

Quoted 11 lines · Stuart
Quoted 9 lines · Daniele Ceraolo Spurio

Can you elaborate a bit more on what's the rationale for this? do you
just want to avoid having too many inlines since the paths they're used
in are not critical, or do you have some more functional reason? This is
not a critic to the patch, I just want to understand where you're coming
from ;)

Quoted 5 lines

This was a request from Jani Nikula in a previous series update. I don't have a strong preference either way personally. If you don't have any major concerns, I'd prefer to keep the series as-is to prevent too much thrash here, but let me know.

No concerns, just please update the commit message to explain that we're moving them because there is no need for them to be inline since they're not on a critical path where we need preformance.

Quoted 13 lines
Quoted 12 lines

BTW, looking at this patch I realized there are a few more DIV_ROUND_UP(..., BITS_PER_BYTE) that could be converted to GEN_SSEU_STRIDE() in patch 2. I noticed you update them to a new variable in the next patch, but for consistency it might still be worth
updating them all in patch 2 or at least mention in the commit message
of patch 2 that the remaining cases are updated by a follow-up patch in
the series. Patch 2 is quite small, so you could also just squash it into patch 6 to avoid the split.

Quoted 7 lines

I'm happy to squash them. I did try to isolate this a bit, but you're right that I ended up pushing some of these DIV_ROUND_UP... stride calculations to the last patch in the series. If you don't have any objection, to keep the finaly patch a bit simpler, I'd rather pull those changes into the earlier patch. I realize you already have a RB on that patch. Any issues doing this?

If you're changing all of them from DIV_ROUND_UP to GEN_SSEU_STRIDE in patch 2 I'm ok for you to keep the r-b. If you want to port the other logic for saving sseu->ss_stride to that patch then I'll have another quick look at it after you re-send as that is a more complex change.

Daniele

Quoted 462 lines

Thanks,
Stuart

Quoted 459 lines

Daniele

Quoted 455 lines · Stuart Summers

Additionally, ensure these are all prefixed with intel_sseu_* to match the convention of other functions in i915.

---
   drivers/gpu/drm/i915/gt/intel_sseu.c     | 54 +++++++++++++++++++
   drivers/gpu/drm/i915/gt/intel_sseu.h     | 57 +++--------------
---
   drivers/gpu/drm/i915/i915_debugfs.c      |  6 +--
   drivers/gpu/drm/i915/i915_drv.c          |  2 +-
   drivers/gpu/drm/i915/intel_device_info.c | 69 ++++++++++++-------
-----
   5 files changed, 102 insertions(+), 86 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.c +35 -0
index 7f448f3bea0b..4a0b82fc108c 100644
@@ -8,6 +8,60 @@
   #include "intel_lrc_reg.h"
   #include "intel_sseu.h"
 
+unsigned int
+intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
+{
+	unsigned int i, total = 0;
+
+	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
+		total += hweight8(sseu->subslice_mask[i]);
+
+	return total;
+}
+
+unsigned int
+intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu,
 u8 slice)
+{
+	return hweight8(sseu->subslice_mask[slice]);
+}
+
+static int intel_sseu_eu_idx(const struct sseu_dev_info *sseu, int
 slice,
+			     int subslice)
+{
+	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
+					   BITS_PER_BYTE);
+	int slice_stride = sseu->max_subslices * subslice_stride;
+
+	return slice * slice_stride + subslice * subslice_stride;
+}
+
+u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int
 slice,
+		       int subslice)
+{
+	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
+	u16 eu_mask = 0;
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.c +8 -0
+		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
+			(i * BITS_PER_BYTE);
+	}
+
+	return eu_mask;
+}
+
+void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice, int

subslice,

drivers/gpu/drm/i915/gt/intel_sseu.c +6 -0
+			u16 eu_mask)
+{
+	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.c +5 -0
+		sseu->eu_mask[offset + i] =
+			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
+	}
+}
+
   u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
   			 const struct intel_sseu *req_sseu)
   {
drivers/gpu/drm/i915/gt/intel_sseu.h +6 -31
index 029e71d8f140..56e3721ae83f 100644
@@ -63,58 +63,17 @@ intel_sseu_from_device_info(const struct
 sseu_dev_info *sseu)
   	return value;
   }
 
-static inline unsigned int sseu_subslice_total(const struct
 sseu_dev_info *sseu)
-{
-	unsigned int i, total = 0;
-
-	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
-		total += hweight8(sseu->subslice_mask[i]);
+unsigned int
+intel_sseu_subslice_total(const struct sseu_dev_info *sseu);
 
-	return total;
-}
+unsigned int
+intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu,
 u8 slice);
 
-static inline unsigned int
-sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8
 slice)
-{
-	return hweight8(sseu->subslice_mask[slice]);
-}
-
-static inline int sseu_eu_idx(const struct sseu_dev_info *sseu,
-			      int slice, int subslice)
-{
-	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
-					   BITS_PER_BYTE);
-	int slice_stride = sseu->max_subslices * subslice_stride;
-
-	return slice * slice_stride + subslice * subslice_stride;
-}
+u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int
 slice,
+		       int subslice);
 
-static inline u16 sseu_get_eus(const struct sseu_dev_info *sseu,
-			       int slice, int subslice)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-	u16 eu_mask = 0;
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.h +0 -14
-		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
-			(i * BITS_PER_BYTE);
-	}
-
-	return eu_mask;
-}
-
-static inline void sseu_set_eus(struct sseu_dev_info *sseu,
-				int slice, int subslice, u16 eu_mask)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.h +1 -4
-		sseu->eu_mask[offset + i] =
-			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
-	}
-}
+void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice, int

subslice,

drivers/gpu/drm/i915/gt/intel_sseu.h +1 -0
+			u16 eu_mask);
 
   u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
   			 const struct intel_sseu *req_sseu);
drivers/gpu/drm/i915/i915_debugfs.c +1 -1
index fe854c629a32..3f3ee83ac315 100644
@@ -4158,7 +4158,7 @@ static void
 broadwell_sseu_device_status(struct drm_i915_private *dev_priv,
   				RUNTIME_INFO(dev_priv)-
 > sseu.subslice_mask[s];
   		}
   		sseu->eu_total = sseu->eu_per_subslice *
-				 sseu_subslice_total(sseu);
+				 intel_sseu_subslice_total(sseu);
 
   		/* subtract fused off EU(s) from enabled slice(s) */
   		for (s = 0; s < fls(sseu->slice_mask); s++) {
drivers/gpu/drm/i915/i915_debugfs.c +2 -2
@@ -4182,10 +4182,10 @@ static void i915_print_sseu_info(struct
 seq_file *m, bool is_available_info,
   	seq_printf(m, "  %s Slice Total: %u\n", type,
   		   hweight8(sseu->slice_mask));
   	seq_printf(m, "  %s Subslice Total: %u\n", type,
-		   sseu_subslice_total(sseu));
+		   intel_sseu_subslice_total(sseu));
   	for (s = 0; s < fls(sseu->slice_mask); s++) {
   		seq_printf(m, "  %s Slice%i subslices: %u\n", type,
-			   s, sseu_subslices_per_slice(sseu, s));
+			   s, intel_sseu_subslices_per_slice(sseu, s));
   	}
   	seq_printf(m, "  %s EU Total: %u\n", type,
   		   sseu->eu_total);
drivers/gpu/drm/i915/i915_drv.c +1 -1
index c376244c19c4..130c5140db0d 100644
@@ -378,7 +378,7 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
   		value = i915_cmd_parser_get_version(dev_priv);
   		break;
   	case I915_PARAM_SUBSLICE_TOTAL:
-		value = sseu_subslice_total(sseu);
+		value = intel_sseu_subslice_total(sseu);
   		if (!value)
   			return -ENODEV;
   		break;
drivers/gpu/drm/i915/intel_device_info.c +2 -2
index 559cf0d0628e..e1dbccf04cd9 100644
@@ -90,10 +90,10 @@ static void sseu_dump(const struct
 sseu_dev_info *sseu, struct drm_printer *p)
 
   	drm_printf(p, "slice total: %u, mask=%04x\n",
   		   hweight8(sseu->slice_mask), sseu->slice_mask);
-	drm_printf(p, "subslice total: %u\n",
 sseu_subslice_total(sseu));
+	drm_printf(p, "subslice total: %u\n",
 intel_sseu_subslice_total(sseu));
   	for (s = 0; s < sseu->max_slices; s++) {
   		drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
-			   s, sseu_subslices_per_slice(sseu, s),
+			   s, intel_sseu_subslices_per_slice(sseu, s),
   			   sseu->subslice_mask[s]);
   	}
   	drm_printf(p, "EU total: %u\n", sseu->eu_total);
drivers/gpu/drm/i915/intel_device_info.c +2 -2
@@ -126,11 +126,11 @@ void intel_device_info_dump_topology(const
 struct sseu_dev_info *sseu,
 
   	for (s = 0; s < sseu->max_slices; s++) {
   		drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
-			   s, sseu_subslices_per_slice(sseu, s),
+			   s, intel_sseu_subslices_per_slice(sseu, s),
   			   sseu->subslice_mask[s]);
 
   		for (ss = 0; ss < sseu->max_subslices; ss++) {
-			u16 enabled_eus = sseu_get_eus(sseu, s, ss);
+			u16 enabled_eus = intel_sseu_get_eus(sseu, s,
 ss);
 
   			drm_printf(p, "\tsubslice%d: %u EUs (0x%hx)\n",
   				   ss, hweight16(enabled_eus),

enabled_eus);

drivers/gpu/drm/i915/intel_device_info.c +1 -1
@@ -180,7 +180,7 @@ static void gen11_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			sseu->subslice_mask[s] = (ss_en >> ss_idx) &
 ss_en_mask;
   			for (ss = 0; ss < sseu->max_subslices; ss++) {
   				if (sseu->subslice_mask[s] & BIT(ss))
-					sseu_set_eus(sseu, s, ss,
 eu_en);
+					intel_sseu_set_eus(sseu, s, ss,
eu_en);
   			}
   		}
   	}
drivers/gpu/drm/i915/intel_device_info.c +11 -11
@@ -222,32 +222,32 @@ static void gen10_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	/* Slice0 */
   	eu_en = ~I915_READ(GEN8_EU_DISABLE0);
   	for (ss = 0; ss < sseu->max_subslices; ss++)
-		sseu_set_eus(sseu, 0, ss, (eu_en >> (8 * ss)) &
 eu_mask);
+		intel_sseu_set_eus(sseu, 0, ss, (eu_en >> (8 * ss)) &
 eu_mask);
   	/* Slice1 */
-	sseu_set_eus(sseu, 1, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 1, 0, (eu_en >> 24) & eu_mask);
   	eu_en = ~I915_READ(GEN8_EU_DISABLE1);
-	sseu_set_eus(sseu, 1, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 1, 1, eu_en & eu_mask);
   	/* Slice2 */
-	sseu_set_eus(sseu, 2, 0, (eu_en >> 8) & eu_mask);
-	sseu_set_eus(sseu, 2, 1, (eu_en >> 16) & eu_mask);
+	intel_sseu_set_eus(sseu, 2, 0, (eu_en >> 8) & eu_mask);
+	intel_sseu_set_eus(sseu, 2, 1, (eu_en >> 16) & eu_mask);
   	/* Slice3 */
-	sseu_set_eus(sseu, 3, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 3, 0, (eu_en >> 24) & eu_mask);
   	eu_en = ~I915_READ(GEN8_EU_DISABLE2);
-	sseu_set_eus(sseu, 3, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 3, 1, eu_en & eu_mask);
   	/* Slice4 */
-	sseu_set_eus(sseu, 4, 0, (eu_en >> 8) & eu_mask);
-	sseu_set_eus(sseu, 4, 1, (eu_en >> 16) & eu_mask);
+	intel_sseu_set_eus(sseu, 4, 0, (eu_en >> 8) & eu_mask);
+	intel_sseu_set_eus(sseu, 4, 1, (eu_en >> 16) & eu_mask);
   	/* Slice5 */
-	sseu_set_eus(sseu, 5, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 5, 0, (eu_en >> 24) & eu_mask);
   	eu_en = ~I915_READ(GEN10_EU_DISABLE3);
-	sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
 
   	/* Do a second pass where we mark the subslices disabled if all

their
* eus are off.
*/

   	for (s = 0; s < sseu->max_slices; s++) {
   		for (ss = 0; ss < sseu->max_subslices; ss++) {
drivers/gpu/drm/i915/intel_device_info.c +3 -3
-			if (sseu_get_eus(sseu, s, ss) == 0)
+			if (intel_sseu_get_eus(sseu, s, ss) == 0)
   				sseu->subslice_mask[s] &= ~BIT(ss);
   		}
   	}
@@ -260,9 +260,10 @@ static void gen10_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	 * EU in any one subslice may be fused off for die
   	 * recovery.
   	 */
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
   				DIV_ROUND_UP(sseu->eu_total,
-					     sseu_subslice_total(sseu))
 : 0;
+					     intel_sseu_subslice_total(

sseu)) :

drivers/gpu/drm/i915/intel_device_info.c +2 -1
+				0;
 
   	/* No restrictions on Power Gating */
   	sseu->has_slice_pg = 1;
@@ -290,7 +291,7 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			  CHV_FGT_EU_DIS_SS0_R1_SHIFT) << 4);
 
   		sseu->subslice_mask[0] |= BIT(0);
-		sseu_set_eus(sseu, 0, 0, ~disabled_mask);
+		intel_sseu_set_eus(sseu, 0, 0, ~disabled_mask);
   	}
 
   	if (!(fuse & CHV_FGT_DISABLE_SS1)) {
drivers/gpu/drm/i915/intel_device_info.c +1 -1
@@ -301,7 +302,7 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			  CHV_FGT_EU_DIS_SS1_R1_SHIFT) << 4);
 
   		sseu->subslice_mask[0] |= BIT(1);
-		sseu_set_eus(sseu, 0, 1, ~disabled_mask);
+		intel_sseu_set_eus(sseu, 0, 1, ~disabled_mask);
   	}
 
   	sseu->eu_total = compute_eu_total(sseu);
drivers/gpu/drm/i915/intel_device_info.c +0 -2
@@ -310,8 +311,8 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	 * CHV expected to always have a uniform distribution of EU
   	 * across subslices.
   	*/
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
-				sseu->eu_total /

sseu_subslice_total(sseu) :

drivers/gpu/drm/i915/intel_device_info.c +2 -0
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
+				sseu->eu_total /

intel_sseu_subslice_total(sseu) :

   				0;
   	/*
   	 * CHV supports subslice power gating on devices with more than
drivers/gpu/drm/i915/intel_device_info.c +2 -2
@@ -319,7 +320,7 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	 * more than one EU pair per subslice.
   	*/
   	sseu->has_slice_pg = 0;
-	sseu->has_subslice_pg = sseu_subslice_total(sseu) > 1;
+	sseu->has_subslice_pg = intel_sseu_subslice_total(sseu) > 1;
   	sseu->has_eu_pg = (sseu->eu_per_subslice > 2);
   }
@@ -369,7 +370,7 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
 
   			eu_disabled_mask = (eu_disable >> (ss * 8)) &
 eu_mask;
 
-			sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
+			intel_sseu_set_eus(sseu, s, ss,
 ~eu_disabled_mask);
   
   			eu_per_ss = sseu->max_eus_per_subslice -
   				hweight8(eu_disabled_mask);
drivers/gpu/drm/i915/intel_device_info.c +2 -2
@@ -393,9 +394,10 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	 * recovery. BXT is expected to be perfectly uniform in EU
   	 * distribution.
   	*/
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
   				DIV_ROUND_UP(sseu->eu_total,
-					     sseu_subslice_total(sseu))
 : 0;
+					     intel_sseu_subslice_total(

sseu)) :

drivers/gpu/drm/i915/intel_device_info.c +2 -1
+				0;
   	/*
   	 * SKL+ supports slice power gating on devices with more than
   	 * one slice, and supports EU power gating on devices with
@@ -407,7 +409,7 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	sseu->has_slice_pg =
   		!IS_GEN9_LP(dev_priv) && hweight8(sseu->slice_mask) >
 1;
   	sseu->has_subslice_pg =
-		IS_GEN9_LP(dev_priv) && sseu_subslice_total(sseu) > 1;
+		IS_GEN9_LP(dev_priv) && intel_sseu_subslice_total(sseu)
 > 1;
   	sseu->has_eu_pg = sseu->eu_per_subslice > 2;
   
   	if (IS_GEN9_LP(dev_priv)) {
drivers/gpu/drm/i915/intel_device_info.c +1 -1
@@ -477,7 +479,7 @@ static void broadwell_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			eu_disabled_mask =
   				eu_disable[s] >> (ss * sseu-
 > max_eus_per_subslice);
 
-			sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
+			intel_sseu_set_eus(sseu, s, ss,
 ~eu_disabled_mask);
   
   			n_disabled = hweight8(eu_disabled_mask);
drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -496,9 +498,10 @@ static void broadwell_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	 * subslices with the exception that any one EU in any one

subslice may
* be fused off for die recovery.
*/

drivers/gpu/drm/i915/intel_device_info.c +1 -2
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
   				DIV_ROUND_UP(sseu->eu_total,
-					     sseu_subslice_total(sseu))

: 0;

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+					     intel_sseu_subslice_total(

sseu)) :

drivers/gpu/drm/i915/intel_device_info.c +3 -2
+				0;
 
   	/*
   	 * BDW supports slice power gating on devices with more than
@@ -561,8 +564,8 @@ static void haswell_sseu_info_init(struct
 drm_i915_private *dev_priv)
 
   	for (s = 0; s < sseu->max_slices; s++) {
   		for (ss = 0; ss < sseu->max_subslices; ss++) {
-			sseu_set_eus(sseu, s, ss,
-				     (1UL << sseu->eu_per_subslice) -
 1);
+			intel_sseu_set_eus(sseu, s, ss,
+					   (1UL << sseu-
 > eu_per_subslice) - 1);
   		}
   	}

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#17 lore raw

Re: [PATCH 5/6] drm/i915: Remove inline from sseu helper functions

Quoted 23 lines · Daniele Ceraolo Spurio
Quoted 13 lines · Stuart
Quoted 11 lines · Daniele Ceraolo Spurio

Can you elaborate a bit more on what's the rationale for this? do you
just want to avoid having too many inlines since the paths they're
used
in are not critical, or do you have some more functional reason? This
is
not a critic to the patch, I just want to understand where you're coming
from ;)

Quoted 7 lines

This was a request from Jani Nikula in a previous series update. I don't have a strong preference either way personally. If you don't have
any major concerns, I'd prefer to keep the series as-is to prevent too
much thrash here, but let me know.

Quoted 5 lines

No concerns, just please update the commit message to explain that we're
moving them because there is no need for them to be inline since they're
not on a critical path where we need preformance.

Sounds great.

Quoted 26 lines
Quoted 16 lines
Quoted 15 lines

BTW, looking at this patch I realized there are a few more DIV_ROUND_UP(..., BITS_PER_BYTE) that could be converted to GEN_SSEU_STRIDE() in patch 2. I noticed you update them to a new variable in the next patch, but for consistency it might still be worth
updating them all in patch 2 or at least mention in the commit message
of patch 2 that the remaining cases are updated by a follow-up patch
in
the series. Patch 2 is quite small, so you could also just squash it
into patch 6 to avoid the split.

Quoted 9 lines

I'm happy to squash them. I did try to isolate this a bit, but you're
right that I ended up pushing some of these DIV_ROUND_UP... stride calculations to the last patch in the series. If you don't have any objection, to keep the finaly patch a bit simpler, I'd rather pull those changes into the earlier patch. I realize you already have a RB
on that patch. Any issues doing this?

Quoted 7 lines

If you're changing all of them from DIV_ROUND_UP to GEN_SSEU_STRIDE in
patch 2 I'm ok for you to keep the r-b. If you want to port the other
logic for saving sseu->ss_stride to that patch then I'll have another
quick look at it after you re-send as that is a more complex change.

I'll do the former, then convert those to the new structure layout in the subsequent patches.

Thanks,
Stuart

Quoted 552 lines

Daniele

Quoted 549 lines

Thanks,
Stuart

Quoted 546 lines

Daniele

Quoted 542 lines · Stuart Summers

Additionally, ensure these are all prefixed with intel_sseu_* to match the convention of other functions in i915.

---
   drivers/gpu/drm/i915/gt/intel_sseu.c     | 54
+++++++++++++++++++
   drivers/gpu/drm/i915/gt/intel_sseu.h     | 57 +++-----------
---
---
   drivers/gpu/drm/i915/i915_debugfs.c      |  6 +--
   drivers/gpu/drm/i915/i915_drv.c          |  2 +-
   drivers/gpu/drm/i915/intel_device_info.c | 69 ++++++++++++
-------
-----
   5 files changed, 102 insertions(+), 86 deletions(-)
drivers/gpu/drm/i915/gt/intel_sseu.c +18 -0
index 7f448f3bea0b..4a0b82fc108c 100644
@@ -8,6 +8,60 @@
   #include "intel_lrc_reg.h"
   #include "intel_sseu.h"
 
+unsigned int
+intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
+{
+	unsigned int i, total = 0;
+
+	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
+		total += hweight8(sseu->subslice_mask[i]);
+
+	return total;
+}
+
+unsigned int
+intel_sseu_subslices_per_slice(const struct sseu_dev_info
 *sseu,
 u8 slice)
+{
+	return hweight8(sseu->subslice_mask[slice]);
+}
+
+static int intel_sseu_eu_idx(const struct sseu_dev_info *sseu,

int
slice,

drivers/gpu/drm/i915/gt/intel_sseu.c +3 -0
+			     int subslice)
+{
+	int subslice_stride = DIV_ROUND_UP(sseu-
Quoted 1 line

max_eus_per_subslice,

drivers/gpu/drm/i915/gt/intel_sseu.c +2 -0
+					   BITS_PER_BYTE);
+	int slice_stride = sseu->max_subslices *

subslice_stride;

drivers/gpu/drm/i915/gt/intel_sseu.c +2 -0
+
+	return slice * slice_stride + subslice *

subslice_stride;

drivers/gpu/drm/i915/gt/intel_sseu.c +3 -0
+}
+
+u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int

slice,

drivers/gpu/drm/i915/gt/intel_sseu.c +3 -0
+		       int subslice)
+{
+	int i, offset = intel_sseu_eu_idx(sseu, slice,

subslice);

drivers/gpu/drm/i915/gt/intel_sseu.c +4 -0
+	u16 eu_mask = 0;
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.c +8 -0
+		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
+			(i * BITS_PER_BYTE);
+	}
+
+	return eu_mask;
+}
+
+void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice,

int
subslice,

drivers/gpu/drm/i915/gt/intel_sseu.c +3 -0
+			u16 eu_mask)
+{
+	int i, offset = intel_sseu_eu_idx(sseu, slice,

subslice);

drivers/gpu/drm/i915/gt/intel_sseu.c +3 -0
+
+	for (i = 0;
+	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.c +2 -0
+		sseu->eu_mask[offset + i] =
+			(eu_mask >> (BITS_PER_BYTE * i)) &

0xff;

drivers/gpu/drm/i915/gt/intel_sseu.c +3 -0
+	}
+}
+
   u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
   			 const struct intel_sseu *req_sseu)
   {
drivers/gpu/drm/i915/gt/intel_sseu.h +4 -22
index 029e71d8f140..56e3721ae83f 100644
@@ -63,58 +63,17 @@ intel_sseu_from_device_info(const struct
 sseu_dev_info *sseu)
   	return value;
   }
 
-static inline unsigned int sseu_subslice_total(const struct
 sseu_dev_info *sseu)
-{
-	unsigned int i, total = 0;
-
-	for (i = 0; i < ARRAY_SIZE(sseu->subslice_mask); i++)
-		total += hweight8(sseu->subslice_mask[i]);
+unsigned int
+intel_sseu_subslice_total(const struct sseu_dev_info *sseu);
 
-	return total;
-}
+unsigned int
+intel_sseu_subslices_per_slice(const struct sseu_dev_info
 *sseu,
 u8 slice);
 
-static inline unsigned int
-sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8
 slice)
-{
-	return hweight8(sseu->subslice_mask[slice]);
-}
-
-static inline int sseu_eu_idx(const struct sseu_dev_info
 *sseu,
-			      int slice, int subslice)
-{
-	int subslice_stride = DIV_ROUND_UP(sseu-
 >max_eus_per_subslice,
-					   BITS_PER_BYTE);
-	int slice_stride = sseu->max_subslices *
 subslice_stride;
-
-	return slice * slice_stride + subslice *

subslice_stride;

drivers/gpu/drm/i915/gt/intel_sseu.h +1 -1
-}
+u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int

slice,

drivers/gpu/drm/i915/gt/intel_sseu.h +1 -1
+		       int subslice);
 
-static inline u16 sseu_get_eus(const struct sseu_dev_info

*sseu,

drivers/gpu/drm/i915/gt/intel_sseu.h +0 -7
-			       int slice, int subslice)
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-	u16 eu_mask = 0;
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.h +0 -9
-		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
-			(i * BITS_PER_BYTE);
-	}
-
-	return eu_mask;
-}
-
-static inline void sseu_set_eus(struct sseu_dev_info *sseu,
-				int slice, int subslice, u16

eu_mask)

drivers/gpu/drm/i915/gt/intel_sseu.h +0 -5
-{
-	int i, offset = sseu_eu_idx(sseu, slice, subslice);
-
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,

BITS_PER_BYTE); i++) {

drivers/gpu/drm/i915/gt/intel_sseu.h +0 -2
-		sseu->eu_mask[offset + i] =
-			(eu_mask >> (BITS_PER_BYTE * i)) &

0xff;

drivers/gpu/drm/i915/gt/intel_sseu.h +1 -2
-	}
-}
+void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice,

int
subslice,

drivers/gpu/drm/i915/gt/intel_sseu.h +1 -0
+			u16 eu_mask);
 
   u32 intel_sseu_make_rpcs(struct drm_i915_private *i915,
   			 const struct intel_sseu *req_sseu);
drivers/gpu/drm/i915/i915_debugfs.c +1 -1
index fe854c629a32..3f3ee83ac315 100644
@@ -4158,7 +4158,7 @@ static void
 broadwell_sseu_device_status(struct drm_i915_private *dev_priv,
   				RUNTIME_INFO(dev_priv)-
 > sseu.subslice_mask[s];
 
   		}
   		sseu->eu_total = sseu->eu_per_subslice *
-				 sseu_subslice_total(sseu);
+				 intel_sseu_subslice_total(sseu
);
   
   		/* subtract fused off EU(s) from enabled

slice(s) */
for (s = 0; s < fls(sseu->slice_mask); s++) {

drivers/gpu/drm/i915/i915_debugfs.c +2 -2
@@ -4182,10 +4182,10 @@ static void i915_print_sseu_info(struct
 seq_file *m, bool is_available_info,
   	seq_printf(m, "  %s Slice Total: %u\n", type,
   		   hweight8(sseu->slice_mask));
   	seq_printf(m, "  %s Subslice Total: %u\n", type,
-		   sseu_subslice_total(sseu));
+		   intel_sseu_subslice_total(sseu));
   	for (s = 0; s < fls(sseu->slice_mask); s++) {
   		seq_printf(m, "  %s Slice%i subslices: %u\n",
 type,
-			   s, sseu_subslices_per_slice(sseu,
 s));
+			   s,
intel_sseu_subslices_per_slice(sseu, s));
   	}
   	seq_printf(m, "  %s EU Total: %u\n", type,
   		   sseu->eu_total);
drivers/gpu/drm/i915/i915_drv.c +1 -1
index c376244c19c4..130c5140db0d 100644
@@ -378,7 +378,7 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
   		value = i915_cmd_parser_get_version(dev_priv);
   		break;
   	case I915_PARAM_SUBSLICE_TOTAL:
-		value = sseu_subslice_total(sseu);
+		value = intel_sseu_subslice_total(sseu);
   		if (!value)
   			return -ENODEV;
   		break;
drivers/gpu/drm/i915/intel_device_info.c +1 -1
index 559cf0d0628e..e1dbccf04cd9 100644
@@ -90,10 +90,10 @@ static void sseu_dump(const struct
 sseu_dev_info *sseu, struct drm_printer *p)
 
   	drm_printf(p, "slice total: %u, mask=%04x\n",
   		   hweight8(sseu->slice_mask), sseu-
 >slice_mask);
-	drm_printf(p, "subslice total: %u\n",
 sseu_subslice_total(sseu));
+	drm_printf(p, "subslice total: %u\n",
 intel_sseu_subslice_total(sseu));
   	for (s = 0; s < sseu->max_slices; s++) {
   		drm_printf(p, "slice%d: %u subslices,

mask=%04x\n",

drivers/gpu/drm/i915/intel_device_info.c +0 -1
-			   s, sseu_subslices_per_slice(sseu,

s),

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+			   s,

intel_sseu_subslices_per_slice(sseu, s),

   			   sseu->subslice_mask[s]);
   	}
   	drm_printf(p, "EU total: %u\n", sseu->eu_total);
drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -126,11 +126,11 @@ void

intel_device_info_dump_topology(const
struct sseu_dev_info *sseu,

for (s = 0; s < sseu->max_slices; s++) {
drm_printf(p, "slice%d: %u subslice(s)
(0x%hhx):\n",

drivers/gpu/drm/i915/intel_device_info.c +0 -1
-			   s, sseu_subslices_per_slice(sseu,

s),

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+			   s,

intel_sseu_subslices_per_slice(sseu, s),

   			   sseu->subslice_mask[s]);
   
   		for (ss = 0; ss < sseu->max_subslices; ss++) {
drivers/gpu/drm/i915/intel_device_info.c +0 -1
-			u16 enabled_eus = sseu_get_eus(sseu, s,

ss);

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+			u16 enabled_eus =

intel_sseu_get_eus(sseu, s,
ss);

drm_printf(p, "\tsubslice%d: %u EUs
(0x%hx)\n",
ss, hweight16(enabled_eus),
enabled_eus);

drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -180,7 +180,7 @@ static void gen11_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			sseu->subslice_mask[s] = (ss_en >>

ss_idx) &

ss_en_mask;
   			for (ss = 0; ss < sseu->max_subslices;
ss++) {
   				if (sseu->subslice_mask[s] &

BIT(ss))

drivers/gpu/drm/i915/intel_device_info.c +0 -1
-					sseu_set_eus(sseu, s,

ss,
eu_en);

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+					intel_sseu_set_eus(sseu

, s, ss,

eu_en);
   			}
   		}
   	}
drivers/gpu/drm/i915/intel_device_info.c +1 -1
@@ -222,32 +222,32 @@ static void gen10_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	/* Slice0 */
   	eu_en = ~I915_READ(GEN8_EU_DISABLE0);
   	for (ss = 0; ss < sseu->max_subslices; ss++)
-		sseu_set_eus(sseu, 0, ss, (eu_en >> (8 * ss)) &
 eu_mask);
+		intel_sseu_set_eus(sseu, 0, ss, (eu_en >> (8 *

ss)) &

eu_mask);
   	/* Slice1 */
drivers/gpu/drm/i915/intel_device_info.c +1 -1
-	sseu_set_eus(sseu, 1, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 1, 0, (eu_en >> 24) &
eu_mask);
   	eu_en = ~I915_READ(GEN8_EU_DISABLE1);
drivers/gpu/drm/i915/intel_device_info.c +3 -3
-	sseu_set_eus(sseu, 1, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 1, 1, eu_en & eu_mask);
   	/* Slice2 */
-	sseu_set_eus(sseu, 2, 0, (eu_en >> 8) & eu_mask);
-	sseu_set_eus(sseu, 2, 1, (eu_en >> 16) & eu_mask);
+	intel_sseu_set_eus(sseu, 2, 0, (eu_en >> 8) & eu_mask);
+	intel_sseu_set_eus(sseu, 2, 1, (eu_en >> 16) &
eu_mask);
   	/* Slice3 */
drivers/gpu/drm/i915/intel_device_info.c +1 -1
-	sseu_set_eus(sseu, 3, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 3, 0, (eu_en >> 24) &
eu_mask);
   	eu_en = ~I915_READ(GEN8_EU_DISABLE2);
drivers/gpu/drm/i915/intel_device_info.c +3 -3
-	sseu_set_eus(sseu, 3, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 3, 1, eu_en & eu_mask);
   	/* Slice4 */
-	sseu_set_eus(sseu, 4, 0, (eu_en >> 8) & eu_mask);
-	sseu_set_eus(sseu, 4, 1, (eu_en >> 16) & eu_mask);
+	intel_sseu_set_eus(sseu, 4, 0, (eu_en >> 8) & eu_mask);
+	intel_sseu_set_eus(sseu, 4, 1, (eu_en >> 16) &
eu_mask);
   	/* Slice5 */
drivers/gpu/drm/i915/intel_device_info.c +1 -1
-	sseu_set_eus(sseu, 5, 0, (eu_en >> 24) & eu_mask);
+	intel_sseu_set_eus(sseu, 5, 0, (eu_en >> 24) &
eu_mask);
   	eu_en = ~I915_READ(GEN10_EU_DISABLE3);
drivers/gpu/drm/i915/intel_device_info.c +1 -1
-	sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
+	intel_sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
 
   	/* Do a second pass where we mark the subslices

disabled if all
their
* eus are off.
*/

   	for (s = 0; s < sseu->max_slices; s++) {
   		for (ss = 0; ss < sseu->max_subslices; ss++) {
drivers/gpu/drm/i915/intel_device_info.c +1 -1
-			if (sseu_get_eus(sseu, s, ss) == 0)
+			if (intel_sseu_get_eus(sseu, s, ss) ==

0)
sseu->subslice_mask[s] &=

~BIT(ss);
   		}
   	}
drivers/gpu/drm/i915/intel_device_info.c +1 -1
@@ -260,9 +260,10 @@ static void gen10_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	 * EU in any one subslice may be fused off for die
   	 * recovery.
   	 */
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu)

?
DIV_ROUND_UP(sseu->eu_total,

drivers/gpu/drm/i915/intel_device_info.c +0 -1
-					     sseu_subslice_tota

l(sseu))
: 0;

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+					     intel_sseu_subslic

e_total(
sseu)) :

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+				0;
 
   	/* No restrictions on Power Gating */
   	sseu->has_slice_pg = 1;
@@ -290,7 +291,7 @@ static void

cherryview_sseu_info_init(struct
drm_i915_private *dev_priv)

   			  CHV_FGT_EU_DIS_SS0_R1_SHIFT) << 4);
   
   		sseu->subslice_mask[0] |= BIT(0);
drivers/gpu/drm/i915/intel_device_info.c +1 -1
-		sseu_set_eus(sseu, 0, 0, ~disabled_mask);
+		intel_sseu_set_eus(sseu, 0, 0, ~disabled_mask);
   	}
 
   	if (!(fuse & CHV_FGT_DISABLE_SS1)) {
@@ -301,7 +302,7 @@ static void

cherryview_sseu_info_init(struct
drm_i915_private *dev_priv)

   			  CHV_FGT_EU_DIS_SS1_R1_SHIFT) << 4);
   
   		sseu->subslice_mask[0] |= BIT(1);
drivers/gpu/drm/i915/intel_device_info.c +1 -1
-		sseu_set_eus(sseu, 0, 1, ~disabled_mask);
+		intel_sseu_set_eus(sseu, 0, 1, ~disabled_mask);
   	}
 
   	sseu->eu_total = compute_eu_total(sseu);
@@ -310,8 +311,8 @@ static void

cherryview_sseu_info_init(struct
drm_i915_private *dev_priv)
* CHV expected to always have a uniform distribution
of EU
* across subslices.
*/

drivers/gpu/drm/i915/intel_device_info.c +0 -2
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
-				sseu->eu_total /

sseu_subslice_total(sseu) :

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu)

?

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+				sseu->eu_total /

intel_sseu_subslice_total(sseu) :

   				0;
   	/*
   	 * CHV supports subslice power gating on devices with

more than

drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -319,7 +320,7 @@ static void

cherryview_sseu_info_init(struct
drm_i915_private *dev_priv)
* more than one EU pair per subslice.
*/
sseu->has_slice_pg = 0;

drivers/gpu/drm/i915/intel_device_info.c +1 -1
-	sseu->has_subslice_pg = sseu_subslice_total(sseu) > 1;
+	sseu->has_subslice_pg = intel_sseu_subslice_total(sseu)
Quoted 1 line

1;

   	sseu->has_eu_pg = (sseu->eu_per_subslice > 2);
   }
drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -369,7 +370,7 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
 
   			eu_disabled_mask = (eu_disable >> (ss *

8)) &
eu_mask;

drivers/gpu/drm/i915/intel_device_info.c +0 -1
-			sseu_set_eus(sseu, s, ss,

~eu_disabled_mask);

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+			intel_sseu_set_eus(sseu, s, ss,
~eu_disabled_mask);
   
   			eu_per_ss = sseu->max_eus_per_subslice
drivers/gpu/drm/i915/intel_device_info.c +0 -1
-
   				hweight8(eu_disabled_mask);
@@ -393,9 +394,10 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	 * recovery. BXT is expected to be perfectly uniform in

EU
* distribution.
*/

drivers/gpu/drm/i915/intel_device_info.c +1 -1
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu)

?
DIV_ROUND_UP(sseu->eu_total,

drivers/gpu/drm/i915/intel_device_info.c +0 -1
-					     sseu_subslice_tota

l(sseu))
: 0;

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+					     intel_sseu_subslic

e_total(
sseu)) :

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+				0;
   	/*
   	 * SKL+ supports slice power gating on devices with

more than
* one slice, and supports EU power gating on devices
with

drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -407,7 +409,7 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	sseu->has_slice_pg =
   		!IS_GEN9_LP(dev_priv) && hweight8(sseu-
Quoted 1 line

slice_mask) >

1;
   	sseu->has_subslice_pg =
drivers/gpu/drm/i915/intel_device_info.c +0 -1
-		IS_GEN9_LP(dev_priv) &&

sseu_subslice_total(sseu) > 1;

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+		IS_GEN9_LP(dev_priv) &&

intel_sseu_subslice_total(sseu)

Quoted 1 line

1;

   	sseu->has_eu_pg = sseu->eu_per_subslice > 2;
   
   	if (IS_GEN9_LP(dev_priv)) {
drivers/gpu/drm/i915/intel_device_info.c +0 -1
@@ -477,7 +479,7 @@ static void broadwell_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			eu_disabled_mask =
   				eu_disable[s] >> (ss * sseu-
 > max_eus_per_subslice);
 
 
-			sseu_set_eus(sseu, s, ss,

~eu_disabled_mask);

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+			intel_sseu_set_eus(sseu, s, ss,

~eu_disabled_mask);

n_disabled =
hweight8(eu_disabled_mask);

drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -496,9 +498,10 @@ static void

broadwell_sseu_info_init(struct
drm_i915_private *dev_priv)
* subslices with the exception that any one EU in any
one
subslice may
* be fused off for die recovery.
*/

drivers/gpu/drm/i915/intel_device_info.c +1 -1
-	sseu->eu_per_subslice = sseu_subslice_total(sseu) ?
+	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu)

?
DIV_ROUND_UP(sseu->eu_total,

drivers/gpu/drm/i915/intel_device_info.c +0 -1
-					     sseu_subslice_tota

l(sseu))
: 0;

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+					     intel_sseu_subslic

e_total(
sseu)) :

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+				0;
 
   	/*
   	 * BDW supports slice power gating on devices with more

than

drivers/gpu/drm/i915/intel_device_info.c +0 -2
@@ -561,8 +564,8 @@ static void haswell_sseu_info_init(struct
 drm_i915_private *dev_priv)
 
   	for (s = 0; s < sseu->max_slices; s++) {
   		for (ss = 0; ss < sseu->max_subslices; ss++) {
-			sseu_set_eus(sseu, s, ss,
-				     (1UL << sseu-
Quoted 1 line

eu_per_subslice) -

1);

drivers/gpu/drm/i915/intel_device_info.c +2 -0
+			intel_sseu_set_eus(sseu, s, ss,
+					   (1UL << sseu-
Quoted 1 line

eu_per_subslice) - 1);

   		}
   	}

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#26 lore raw

Re: [PATCH 6/6] drm/i915: Expand subslice mask

In reply to Stuart Summers
Quoted 14 lines · Stuart Summers

Currently, the subslice_mask runtime parameter is stored as an array of subslices per slice. Expand the subslice mask array to better match what is presented to userspace through the I915_QUERY_TOPOLOGY_INFO ioctl. The index into this array is then calculated:
slice * subslice stride + subslice index / 8

v2: fix spacing in set_sseu_info args
use set_sseu_info to initialize sseu data when building device status in debugfs
rename variables in intel_engine_types.h to avoid checkpatch warnings
v3: update headers in intel_sseu.h
v4: add const to some sseu_dev_info variables

Quoted 4 lines
     use sseu->eu_stride for EU stride calculations
Cc Daniele Ceraolo Spurio <[email protected]>

Can you also get an ack from Lionel, to make sure this all fits with the expected reporting?

Quoted 19 lines
---
  drivers/gpu/drm/i915/gt/intel_engine_cs.c    |   6 +-
  drivers/gpu/drm/i915/gt/intel_engine_types.h |  32 +++--
  drivers/gpu/drm/i915/gt/intel_hangcheck.c    |   3 +-
  drivers/gpu/drm/i915/gt/intel_sseu.c         |  49 +++++--
  drivers/gpu/drm/i915/gt/intel_sseu.h         |  16 ++-
  drivers/gpu/drm/i915/gt/intel_workarounds.c  |   2 +-
  drivers/gpu/drm/i915/i915_debugfs.c          |  44 +++---
  drivers/gpu/drm/i915/i915_drv.c              |   6 +-
  drivers/gpu/drm/i915/i915_gpu_error.c        |   5 +-
  drivers/gpu/drm/i915/i915_query.c            |  10 +-
  drivers/gpu/drm/i915/intel_device_info.c     | 142 +++++++++++--------
  11 files changed, 198 insertions(+), 117 deletions(-)
drivers/gpu/drm/i915/gt/intel_engine_cs.c +0 -0
index 6e40f8ea9a6a..8f7967cc9a50 100644

drivers/gpu/drm/i915/gt/intel_engine_cs.c u32 intel_calculate_mcr_s_ss_select(struct drm_i915_private *dev_priv) · line 917

Quoted 5 lines
drivers/gpu/drm/i915/gt/intel_engine_cs.c +1 -1
@@ -914,7 +914,7 @@ u32 intel_calculate_mcr_s_ss_select(struct drm_i915_private *dev_priv)
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	u32 mcr_s_ss_select;
  	u32 slice = fls(sseu->slice_mask);
-	u32 subslice = fls(sseu->subslice_mask[slice]);
+	u32 subslice = fls(sseu->subslice_mask[slice * sseu->ss_stride]);

This (and the registers we use below) only works if ss_stride = 1. Can we add a:

	GEM_BUG_ON(sseu->ss_stride > 1);

to catch the fact that this function will need updating to handle that case if/when we get it?

Quoted 37 lines
drivers/gpu/drm/i915/gt/intel_engine_cs.c +3 -1
  	if (IS_GEN(dev_priv, 10))
  		mcr_s_ss_select = GEN8_MCR_SLICE(slice) |
@@ -990,6 +990,7 @@ void intel_engine_get_instdone(struct intel_engine_cs *engine,
  			       struct intel_instdone *instdone)
  {
  	struct drm_i915_private *dev_priv = engine->i915;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	struct intel_uncore *uncore = engine->uncore;
  	u32 mmio_base = engine->mmio_base;
  	int slice;
@@ -1007,7 +1008,8 @@ void intel_engine_get_instdone(struct intel_engine_cs *engine,
 
  		instdone->slice_common =
  			intel_uncore_read(uncore, GEN7_SC_INSTDONE);
-		for_each_instdone_slice_subslice(dev_priv, slice, subslice) {
+		for_each_instdone_slice_subslice(dev_priv, sseu, slice,
+						 subslice) {
  			instdone->sampler[slice][subslice] =
  				read_subslice_reg(dev_priv, slice, subslice,
  						  GEN7_SAMPLER_INSTDONE);
drivers/gpu/drm/i915/gt/intel_engine_types.h +0 -8
index 9d64e33f8427..1710546a2446 100644
@@ -534,20 +534,22 @@ intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine)
  	return engine->flags & I915_ENGINE_NEEDS_BREADCRUMB_TASKLET;
  }
 
-#define instdone_slice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.slice_mask)
-
-#define instdone_subslice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.subslice_mask[0])
-

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 539

Quoted 10 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +3 -7
-#define for_each_instdone_slice_subslice(dev_priv__, slice__, subslice__) \
-	for ((slice__) = 0, (subslice__) = 0; \
-	     (slice__) < I915_MAX_SLICES; \
-	     (subslice__) = ((subslice__) + 1) < I915_MAX_SUBSLICES ? (subslice__) + 1 : 0, \
-	       (slice__) += ((subslice__) == 0)) \
-		for_each_if((BIT(slice__) & instdone_slice_mask(dev_priv__)) && \
-			    (BIT(subslice__) & instdone_subslice_mask(dev_priv__)))
+#define instdone_has_slice(dev_priv___, sseu___, slice___) \
+	((IS_GEN(dev_priv___, 7) ? \
+	  1 : (sseu___)->slice_mask) & \

I'd put the ternary op on the same line here for readability

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 540

Quoted 1 line
drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
+	BIT(slice___)) \

no need for "\" here (and below).

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 542

Quoted 2 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +2 -0
+
+#define instdone_has_subslice(dev_priv__, sseu__, slice__, subslice__) \

need some more parenthesis in this macro to fix the MACRO_ARG_PRECEDENCE warning in checkpatch.

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 545

Quoted 3 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +3 -0
+	((IS_GEN(dev_priv__, 7) ? \
+	  1 : (sseu__)->subslice_mask[slice__ * (sseu__)->ss_stride + \
+				      subslice__ / BITS_PER_BYTE]) & \

The calculation to get the correct subslice u8 entry:

	sseu->subslice_mask[s * sseu->ss_stride + ss / BITS_PER_BYTE]

seems to be repeated a few times in this patch, so it might be worth moving it to its own inline function. looks like you always ultimately want a bool, so we could also go a bit further and have something like:

static inline bool intel_sseu_has_subslice(sseu, s, ss)

{
	u8 mask = sseu->subslice_mask[s * sseu->ss_stride +
				      ss / BITS_PER_BYTE];

	return mask & BIT(ss % BITS_PER_BYTE);
}

and then do:

#define instdone_has_subslice(dev_priv__, sseu__, slice__, subslice__) \
	((IS_GEN(dev_priv__, 7) ? subslice__ == 0 : \
		intel_sseu_has_subslice(...))

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 550

Quoted 5 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +5 -0
+	 BIT(subslice__ % BITS_PER_BYTE)) \
+
+#define for_each_instdone_slice_subslice(dev_priv_, sseu_, slice_, subslice_) \
+	for ((slice_) = 0, (subslice_) = 0; (slice_) < I915_MAX_SLICES; \
+	     (subslice_) = ((subslice_) + 1) < I915_MAX_SUBSLICES ? (subslice_) + 1 : 0, \

This ternary op should be simplifiable as:

	(subslice_) = ((subslice_) + 1) % I915_MAX_SUBSLICES,

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 552

Quoted 2 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +2 -0
+	       (slice_) += ((subslice_) == 0)) \
+		for_each_if(instdone_has_slice(dev_priv_, sseu_, slice) && \

missing the "_" after "slice"

Quoted 47 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
+			    instdone_has_subslice(dev_priv_, sseu_, slice_, subslice_)) \
 
  #endif /* __INTEL_ENGINE_TYPES_H__ */
drivers/gpu/drm/i915/gt/intel_hangcheck.c +2 -1
index e5eaa06fe74d..53c1c98161e1 100644
@@ -50,6 +50,7 @@ static bool instdone_unchanged(u32 current_instdone, u32 *old_instdone)
  static bool subunits_stuck(struct intel_engine_cs *engine)
  {
  	struct drm_i915_private *dev_priv = engine->i915;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	struct intel_instdone instdone;
  	struct intel_instdone *accu_instdone = &engine->hangcheck.instdone;
  	bool stuck;
@@ -71,7 +72,7 @@ static bool subunits_stuck(struct intel_engine_cs *engine)
  	stuck &= instdone_unchanged(instdone.slice_common,
  				    &accu_instdone->slice_common);
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice) {
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice, subslice) {
  		stuck &= instdone_unchanged(instdone.sampler[slice][subslice],
  					    &accu_instdone->sampler[slice][subslice]);
  		stuck &= instdone_unchanged(instdone.row[slice][subslice],
drivers/gpu/drm/i915/gt/intel_sseu.c +11 -0
index 4a0b82fc108c..49316b7ef074 100644
@@ -8,6 +8,17 @@
  #include "intel_lrc_reg.h"
  #include "intel_sseu.h"
 
+void intel_sseu_set_info(struct sseu_dev_info *sseu, u8 max_slices,
+			 u8 max_subslices, u8 max_eus_per_subslice)
+{
+	sseu->max_slices = max_slices;
+	sseu->max_subslices = max_subslices;
+	sseu->max_eus_per_subslice = max_eus_per_subslice;
+
+	sseu->ss_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
+	sseu->eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
+}
+
  unsigned int
  intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
  {

drivers/gpu/drm/i915/gt/intel_sseu.c intel_sseu_subslice_total(const struct sseu_dev_info *sseu) · line 34

Quoted 2 lines
drivers/gpu/drm/i915/gt/intel_sseu.c +0 -0
@@ -22,17 +33,39 @@ intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
  unsigned int
  intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice)

Here we pass slice as u8, but below we use int. Any reason for the difference?

drivers/gpu/drm/i915/gt/intel_sseu.c intel_sseu_subslice_total(const struct sseu_dev_info *sseu) · line 46

Quoted 13 lines
drivers/gpu/drm/i915/gt/intel_sseu.c +11 -1
  {
-	return hweight8(sseu->subslice_mask[slice]);
+	unsigned int i, total = 0;
+
+	for (i = 0; i < sseu->ss_stride; i++)
+		total += hweight8(sseu->subslice_mask[slice * sseu->ss_stride +
+						      i]);
+
+	return total;
+}
+
+void intel_sseu_copy_subslices(const struct sseu_dev_info *sseu, int slice,
+			       u8 *to_mask, const u8 *from_mask)

You always use sseu->subslice_mask has a from_mask, can't we just get that from the sseu param and avoid the from_mask?

Quoted 110 lines
drivers/gpu/drm/i915/gt/intel_sseu.c +18 -8
+{
+	int offset = slice * sseu->ss_stride;
+
+	memcpy(&to_mask[offset], &from_mask[offset], sseu->ss_stride);
+}
+
+void intel_sseu_set_subslices(struct sseu_dev_info *sseu, int slice,
+			      u32 ss_mask)
+{
+	int i, offset = slice * sseu->ss_stride;
+
+	for (i = 0; i < sseu->ss_stride; i++)
+		sseu->subslice_mask[offset + i] =
+			(ss_mask >> (BITS_PER_BYTE * i)) & 0xff;
  }
 
  static int intel_sseu_eu_idx(const struct sseu_dev_info *sseu, int slice,
  			     int subslice)
  {
-	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
-					   BITS_PER_BYTE);
-	int slice_stride = sseu->max_subslices * subslice_stride;
+	int slice_stride = sseu->max_subslices * sseu->eu_stride;
 
-	return slice * slice_stride + subslice * subslice_stride;
+	return slice * slice_stride + subslice * sseu->eu_stride;
  }
 
  u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int slice,
@@ -41,8 +74,7 @@ u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int slice,
  	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
  	u16 eu_mask = 0;
 
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+	for (i = 0; i < sseu->eu_stride; i++) {
  		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
  			(i * BITS_PER_BYTE);
  	}
@@ -55,8 +87,7 @@ void intel_sseu_set_eus(struct sseu_dev_info *sseu, int slice, int subslice,
  {
  	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
 
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice, BITS_PER_BYTE); i++) {
+	for (i = 0; i < sseu->eu_stride; i++) {
  		sseu->eu_mask[offset + i] =
  			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
  	}
drivers/gpu/drm/i915/gt/intel_sseu.h +15 -1
index 56e3721ae83f..bf01f338a8cc 100644
@@ -9,16 +9,18 @@
 
  #include <linux/types.h>
  #include <linux/kernel.h>
+#include <linux/string.h>
 
  struct drm_i915_private;
 
  #define GEN_MAX_SLICES		(6) /* CNL upper bound */
  #define GEN_MAX_SUBSLICES	(8) /* ICL upper bound */
  #define GEN_SSEU_STRIDE(bits) DIV_ROUND_UP(bits, BITS_PER_BYTE)
+#define GEN_MAX_SUBSLICE_STRIDE GEN_SSEU_STRIDE(GEN_MAX_SUBSLICES)
 
  struct sseu_dev_info {
  	u8 slice_mask;
-	u8 subslice_mask[GEN_MAX_SLICES];
+	u8 subslice_mask[GEN_MAX_SLICES * GEN_MAX_SUBSLICE_STRIDE];
  	u16 eu_total;
  	u8 eu_per_subslice;
  	u8 min_eu_in_pool;
@@ -33,6 +35,9 @@ struct sseu_dev_info {
  	u8 max_subslices;
  	u8 max_eus_per_subslice;
 
+	u8 ss_stride;
+	u8 eu_stride;
+
  	/* We don't have more than 8 eus per subslice at the moment and as we
  	 * store eus enabled using bits, no need to multiply by eus per
  	 * subslice.
@@ -63,12 +68,21 @@ intel_sseu_from_device_info(const struct sseu_dev_info *sseu)
  	return value;
  }
 
+void intel_sseu_set_info(struct sseu_dev_info *sseu, u8 max_slices,
+			 u8 max_subslices, u8 max_eus_per_subslice);
+
  unsigned int
  intel_sseu_subslice_total(const struct sseu_dev_info *sseu);
 
  unsigned int
  intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu, u8 slice);
 
+void intel_sseu_copy_subslices(const struct sseu_dev_info *sseu, int slice,
+			       u8 *to_mask, const u8 *from_mask);
+
+void intel_sseu_set_subslices(struct sseu_dev_info *sseu, int slice,
+			      u32 ss_mask);
+
  u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int slice,
  		       int subslice);
 
drivers/gpu/drm/i915/gt/intel_workarounds.c +0 -0
index 43e290306551..7c7e9556c1c5 100644

drivers/gpu/drm/i915/gt/intel_workarounds.c wa_init_mcr(struct drm_i915_private *i915, struct i915_wa_list *wal) · line 770

Quoted 5 lines
drivers/gpu/drm/i915/gt/intel_workarounds.c +1 -1
@@ -767,7 +767,7 @@ wa_init_mcr(struct drm_i915_private *i915, struct i915_wa_list *wal)
  		u32 slice = fls(sseu->slice_mask);
  		u32 fuse3 =
  			intel_uncore_read(&i915->uncore, GEN10_MIRROR_FUSE3);
-		u8 ss_mask = sseu->subslice_mask[slice];
+		u8 ss_mask = sseu->subslice_mask[slice * sseu->ss_stride];

could use a

	GEM_BUG_ON(sseu->ss_stride > 1);

here as well to remind us this will need changes in that case

Quoted 140 lines
drivers/gpu/drm/i915/gt/intel_workarounds.c +0 -0
  		u8 enabled_mask = (ss_mask | ss_mask >>
  				   GEN10_L3BANK_PAIR_COUNT) & GEN10_L3BANK_MASK;
drivers/gpu/drm/i915/i915_debugfs.c +25 -19
index 3f3ee83ac315..08089c24db25 100644
@@ -1257,6 +1257,7 @@ static void i915_instdone_info(struct drm_i915_private *dev_priv,
  			       struct seq_file *m,
  			       struct intel_instdone *instdone)
  {
+	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	int slice;
  	int subslice;
 
@@ -1272,11 +1273,11 @@ static void i915_instdone_info(struct drm_i915_private *dev_priv,
  	if (INTEL_GEN(dev_priv) <= 6)
  		return;
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice)
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice, subslice)
  		seq_printf(m, "\t\tSAMPLER_INSTDONE[%d][%d]: 0x%08x\n",
  			   slice, subslice, instdone->sampler[slice][subslice]);
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice)
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice, subslice)
  		seq_printf(m, "\t\tROW_INSTDONE[%d][%d]: 0x%08x\n",
  			   slice, subslice, instdone->row[slice][subslice]);
  }
@@ -4066,7 +4067,9 @@ static void gen10_sseu_device_status(struct drm_i915_private *dev_priv,
  			continue;
 
  		sseu->slice_mask |= BIT(s);
-		sseu->subslice_mask[s] = info->sseu.subslice_mask[s];
+		intel_sseu_copy_subslices(&info->sseu, s,
+					  sseu->subslice_mask,
+					  info->sseu.subslice_mask);
 
  		for (ss = 0; ss < info->sseu.max_subslices; ss++) {
  			unsigned int eu_cnt;
@@ -4117,18 +4120,22 @@ static void gen9_sseu_device_status(struct drm_i915_private *dev_priv,
  		sseu->slice_mask |= BIT(s);
 
  		if (IS_GEN9_BC(dev_priv))
-			sseu->subslice_mask[s] =
-				RUNTIME_INFO(dev_priv)->sseu.subslice_mask[s];
+			intel_sseu_copy_subslices(&info->sseu, s,
+						  sseu->subslice_mask,
+						  info->sseu.subslice_mask);
 
  		for (ss = 0; ss < info->sseu.max_subslices; ss++) {
  			unsigned int eu_cnt;
+			u8 ss_idx = s * info->sseu.ss_stride +
+				    ss / BITS_PER_BYTE;
 
  			if (IS_GEN9_LP(dev_priv)) {
  				if (!(s_reg[s] & (GEN9_PGCTL_SS_ACK(ss))))
  					/* skip disabled subslice */
  					continue;
 
-				sseu->subslice_mask[s] |= BIT(ss);
+				sseu->subslice_mask[ss_idx] |=
+					BIT(ss % BITS_PER_BYTE);
  			}
 
  			eu_cnt = 2 * hweight32(eu_reg[2*s + ss/2] &
@@ -4145,25 +4152,24 @@ static void gen9_sseu_device_status(struct drm_i915_private *dev_priv,
  static void broadwell_sseu_device_status(struct drm_i915_private *dev_priv,
  					 struct sseu_dev_info *sseu)
  {
+	struct intel_runtime_info *info = RUNTIME_INFO(dev_priv);
  	u32 slice_info = I915_READ(GEN8_GT_SLICE_INFO);
  	int s;
 
  	sseu->slice_mask = slice_info & GEN8_LSLICESTAT_MASK;
 
  	if (sseu->slice_mask) {
-		sseu->eu_per_subslice =
-			RUNTIME_INFO(dev_priv)->sseu.eu_per_subslice;
-		for (s = 0; s < fls(sseu->slice_mask); s++) {
-			sseu->subslice_mask[s] =
-				RUNTIME_INFO(dev_priv)->sseu.subslice_mask[s];
-		}
+		sseu->eu_per_subslice = info->sseu.eu_per_subslice;
+		for (s = 0; s < fls(sseu->slice_mask); s++)
+			intel_sseu_copy_subslices(&info->sseu, s,
+						  sseu->subslice_mask,
+						  info->sseu.subslice_mask);
  		sseu->eu_total = sseu->eu_per_subslice *
  				 intel_sseu_subslice_total(sseu);
 
  		/* subtract fused off EU(s) from enabled slice(s) */
  		for (s = 0; s < fls(sseu->slice_mask); s++) {
-			u8 subslice_7eu =
-				RUNTIME_INFO(dev_priv)->sseu.subslice_7eu[s];
+			u8 subslice_7eu = info->sseu.subslice_7eu[s];
 
  			sseu->eu_total -= hweight8(subslice_7eu);
  		}
@@ -4210,6 +4216,7 @@ static void i915_print_sseu_info(struct seq_file *m, bool is_available_info,
  static int i915_sseu_status(struct seq_file *m, void *unused)
  {
  	struct drm_i915_private *dev_priv = node_to_i915(m->private);
+	const struct intel_runtime_info *info = RUNTIME_INFO(dev_priv);
  	struct sseu_dev_info sseu;
  	intel_wakeref_t wakeref;
 
@@ -4217,14 +4224,13 @@ static int i915_sseu_status(struct seq_file *m, void *unused)
  		return -ENODEV;
 
  	seq_puts(m, "SSEU Device Info\n");
-	i915_print_sseu_info(m, true, &RUNTIME_INFO(dev_priv)->sseu);
+	i915_print_sseu_info(m, true, &info->sseu);
 
  	seq_puts(m, "SSEU Device Status\n");
  	memset(&sseu, 0, sizeof(sseu));
-	sseu.max_slices = RUNTIME_INFO(dev_priv)->sseu.max_slices;
-	sseu.max_subslices = RUNTIME_INFO(dev_priv)->sseu.max_subslices;
-	sseu.max_eus_per_subslice =
-		RUNTIME_INFO(dev_priv)->sseu.max_eus_per_subslice;
+	intel_sseu_set_info(&sseu, info->sseu.max_slices,
+			    info->sseu.max_subslices,
+			    info->sseu.max_eus_per_subslice);
 
  	with_intel_runtime_pm(dev_priv, wakeref) {
  		if (IS_CHERRYVIEW(dev_priv))
drivers/gpu/drm/i915/i915_drv.c +1 -1
index 130c5140db0d..6afe4e3afea4 100644
@@ -326,7 +326,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
  	struct pci_dev *pdev = dev_priv->drm.pdev;
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	drm_i915_getparam_t *param = data;
-	int value;
+	int value = 0;
 
  	switch (param->param) {
  	case I915_PARAM_IRQ_ACTIVE:

drivers/gpu/drm/i915/i915_drv.c static int i915_getparam_ioctl(struct drm_device *dev, void *data, · line 458

Quoted 5 lines
drivers/gpu/drm/i915/i915_drv.c +1 -1
@@ -455,7 +455,9 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data,
  			return -ENODEV;
  		break;
  	case I915_PARAM_SUBSLICE_MASK:
-		value = sseu->subslice_mask[0];
+		/* Only copy bits from the first subslice */

s/subslice/slice/ ?

Quoted 158 lines
drivers/gpu/drm/i915/i915_drv.c +2 -0
+		memcpy(&value, sseu->subslice_mask,
+		       min(sseu->ss_stride, (u8)sizeof(value)));
  		if (!value)
  			return -ENODEV;
  		break;
drivers/gpu/drm/i915/i915_gpu_error.c +3 -2
index e1b858bd1d32..140918dd9b7d 100644
@@ -407,6 +407,7 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
  static void error_print_instdone(struct drm_i915_error_state_buf *m,
  				 const struct drm_i915_error_engine *ee)
  {
+	struct sseu_dev_info *sseu = &RUNTIME_INFO(m->i915)->sseu;
  	int slice;
  	int subslice;
 
@@ -422,12 +423,12 @@ static void error_print_instdone(struct drm_i915_error_state_buf *m,
  	if (INTEL_GEN(m->i915) <= 6)
  		return;
 
-	for_each_instdone_slice_subslice(m->i915, slice, subslice)
+	for_each_instdone_slice_subslice(m->i915, sseu, slice, subslice)
  		err_printf(m, "  SAMPLER_INSTDONE[%d][%d]: 0x%08x\n",
  			   slice, subslice,
  			   ee->instdone.sampler[slice][subslice]);
 
-	for_each_instdone_slice_subslice(m->i915, slice, subslice)
+	for_each_instdone_slice_subslice(m->i915, sseu, slice, subslice)
  		err_printf(m, "  ROW_INSTDONE[%d][%d]: 0x%08x\n",
  			   slice, subslice,
  			   ee->instdone.row[slice][subslice]);
drivers/gpu/drm/i915/i915_query.c +4 -6
index 7c1708c22811..000dcb145ce0 100644
@@ -37,8 +37,6 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	struct drm_i915_query_topology_info topo;
  	u32 slice_length, subslice_length, eu_length, total_length;
-	u8 subslice_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
-	u8 eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
  	int ret;
 
  	if (query_item->flags != 0)
@@ -50,8 +48,8 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
  	BUILD_BUG_ON(sizeof(u8) != sizeof(sseu->slice_mask));
 
  	slice_length = sizeof(sseu->slice_mask);
-	subslice_length = sseu->max_slices * subslice_stride;
-	eu_length = sseu->max_slices * sseu->max_subslices * eu_stride;
+	subslice_length = sseu->max_slices * sseu->ss_stride;
+	eu_length = sseu->max_slices * sseu->max_subslices * sseu->eu_stride;
  	total_length = sizeof(topo) + slice_length + subslice_length +
  		       eu_length;
 
@@ -69,9 +67,9 @@ static int query_topology_info(struct drm_i915_private *dev_priv,
  	topo.max_eus_per_subslice = sseu->max_eus_per_subslice;
 
  	topo.subslice_offset = slice_length;
-	topo.subslice_stride = subslice_stride;
+	topo.subslice_stride = sseu->ss_stride;
  	topo.eu_offset = slice_length + subslice_length;
-	topo.eu_stride = eu_stride;
+	topo.eu_stride = sseu->eu_stride;
 
  	if (__copy_to_user(u64_to_user_ptr(query_item->data_ptr),
  			   &topo, sizeof(topo)))
drivers/gpu/drm/i915/intel_device_info.c +34 -13
index e1dbccf04cd9..bbbc0a8c2183 100644
@@ -84,17 +84,42 @@ void intel_device_info_dump_flags(const struct intel_device_info *info,
  #undef PRINT_FLAG
  }
 
+#define SS_STR_MAX_SIZE (GEN_MAX_SUBSLICE_STRIDE * 2)
+
+static u8 *
+subslice_per_slice_str(u8 *buf, const struct sseu_dev_info *sseu, u8 slice)
+{
+	int i;
+	u8 ss_offset = slice * sseu->ss_stride;
+
+	GEM_BUG_ON(slice >= sseu->max_slices);
+
+	memset(buf, 0, SS_STR_MAX_SIZE);
+
+	/*
+	 * Print subslice information in reverse order to match
+	 * userspace expectations.
+	 */
+	for (i = 0; i < sseu->ss_stride; i++)
+		sprintf(&buf[i * 2], "%02x",
+			sseu->subslice_mask[ss_offset + sseu->ss_stride -
+					    (i + 1)]);
+
+	return buf;
+}
+
  static void sseu_dump(const struct sseu_dev_info *sseu, struct drm_printer *p)
  {
  	int s;
+	u8 buf[SS_STR_MAX_SIZE];
 
  	drm_printf(p, "slice total: %u, mask=%04x\n",
  		   hweight8(sseu->slice_mask), sseu->slice_mask);
  	drm_printf(p, "subslice total: %u\n", intel_sseu_subslice_total(sseu));
  	for (s = 0; s < sseu->max_slices; s++) {
-		drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
+		drm_printf(p, "slice%d: %u subslices, mask=%s\n",
  			   s, intel_sseu_subslices_per_slice(sseu, s),
-			   sseu->subslice_mask[s]);
+			   subslice_per_slice_str(buf, sseu, s));
  	}
  	drm_printf(p, "EU total: %u\n", sseu->eu_total);
  	drm_printf(p, "EU per subslice: %u\n", sseu->eu_per_subslice);
@@ -118,6 +143,7 @@ void intel_device_info_dump_topology(const struct sseu_dev_info *sseu,
  				     struct drm_printer *p)
  {
  	int s, ss;
+	u8 buf[SS_STR_MAX_SIZE];
 
  	if (sseu->max_slices == 0) {
  		drm_printf(p, "Unavailable\n");
@@ -125,9 +151,9 @@ void intel_device_info_dump_topology(const struct sseu_dev_info *sseu,
  	}
 
  	for (s = 0; s < sseu->max_slices; s++) {
-		drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
+		drm_printf(p, "slice%d: %u subslice(s) (0x%s):\n",
  			   s, intel_sseu_subslices_per_slice(sseu, s),
-			   sseu->subslice_mask[s]);
+			   subslice_per_slice_str(buf, sseu, s));
 
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			u16 enabled_eus = intel_sseu_get_eus(sseu, s, ss);
@@ -156,15 +182,10 @@ static void gen11_sseu_info_init(struct drm_i915_private *dev_priv)
  	u8 eu_en;
  	int s;
 
-	if (IS_ELKHARTLAKE(dev_priv)) {
-		sseu->max_slices = 1;
-		sseu->max_subslices = 4;
-		sseu->max_eus_per_subslice = 8;
-	} else {
-		sseu->max_slices = 1;
-		sseu->max_subslices = 8;
-		sseu->max_eus_per_subslice = 8;
-	}
+	if (IS_ELKHARTLAKE(dev_priv))
+		intel_sseu_set_info(sseu, 1, 4, 8);
+	else
+		intel_sseu_set_info(sseu, 1, 8, 8);
 
  	s_en = I915_READ(GEN11_GT_SLICE_ENABLE) & GEN11_GT_S_ENA_MASK;
  	ss_en = ~I915_READ(GEN11_GT_SUBSLICE_DISABLE);

drivers/gpu/drm/i915/intel_device_info.c static void gen11_sseu_info_init(struct drm_i915_private *dev_priv) · line 202

Quoted 6 lines
drivers/gpu/drm/i915/intel_device_info.c +2 -1
@@ -177,9 +198,11 @@ static void gen11_sseu_info_init(struct drm_i915_private *dev_priv)
  			int ss;
 
  			sseu->slice_mask |= BIT(s);
-			sseu->subslice_mask[s] = (ss_en >> ss_idx) & ss_en_mask;
+			sseu->subslice_mask[s * sseu->ss_stride] =
+				(ss_en >> ss_idx) & ss_en_mask;

Shouldn't this just call intel_sseu_set_subslices() instead of doing the setting locally?

drivers/gpu/drm/i915/intel_device_info.c static void gen11_sseu_info_init(struct drm_i915_private *dev_priv) · line 205

Quoted 4 lines
drivers/gpu/drm/i915/intel_device_info.c +2 -1
  			for (ss = 0; ss < sseu->max_subslices; ss++) {
-				if (sseu->subslice_mask[s] & BIT(ss))
+				if (sseu->subslice_mask[s * sseu->ss_stride] &
+				    BIT(ss))

This culd use the intel_sseu_has_subslice() suggested earlier, otherwise it needs to consider ss_stride > 1

Quoted 207 lines
drivers/gpu/drm/i915/intel_device_info.c +39 -43
  					intel_sseu_set_eus(sseu, s, ss, eu_en);
  			}
  		}
@@ -201,23 +224,10 @@ static void gen10_sseu_info_init(struct drm_i915_private *dev_priv)
  	const int eu_mask = 0xff;
  	u32 subslice_mask, eu_en;
 
+	intel_sseu_set_info(sseu, 6, 4, 8);
+
  	sseu->slice_mask = (fuse2 & GEN10_F2_S_ENA_MASK) >>
  			    GEN10_F2_S_ENA_SHIFT;
-	sseu->max_slices = 6;
-	sseu->max_subslices = 4;
-	sseu->max_eus_per_subslice = 8;
-
-	subslice_mask = (1 << 4) - 1;
-	subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
-			   GEN10_F2_SS_DIS_SHIFT);
-
-	/*
-	 * Slice0 can have up to 3 subslices, but there are only 2 in
-	 * slice1/2.
-	 */
-	sseu->subslice_mask[0] = subslice_mask;
-	for (s = 1; s < sseu->max_slices; s++)
-		sseu->subslice_mask[s] = subslice_mask & 0x3;
 
  	/* Slice0 */
  	eu_en = ~I915_READ(GEN8_EU_DISABLE0);
@@ -242,14 +252,22 @@ static void gen10_sseu_info_init(struct drm_i915_private *dev_priv)
  	eu_en = ~I915_READ(GEN10_EU_DISABLE3);
  	intel_sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
 
-	/* Do a second pass where we mark the subslices disabled if all their
-	 * eus are off.
-	 */
+	subslice_mask = (1 << 4) - 1;
+	subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
+			   GEN10_F2_SS_DIS_SHIFT);
+
  	for (s = 0; s < sseu->max_slices; s++) {
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			if (intel_sseu_get_eus(sseu, s, ss) == 0)
-				sseu->subslice_mask[s] &= ~BIT(ss);
+				subslice_mask &= ~BIT(ss);
  		}
+
+		/*
+		 * Slice0 can have up to 3 subslices, but there are only 2 in
+		 * slice1/2.
+		 */
+		intel_sseu_set_subslices(sseu, s, s == 0 ? subslice_mask :
+							   subslice_mask & 0x3);
  	}
 
  	sseu->eu_total = compute_eu_total(sseu);
@@ -275,13 +293,12 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  {
  	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	u32 fuse;
+	u8 subslice_mask;
 
  	fuse = I915_READ(CHV_FUSE_GT);
 
  	sseu->slice_mask = BIT(0);
-	sseu->max_slices = 1;
-	sseu->max_subslices = 2;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, 1, 2, 8);
 
  	if (!(fuse & CHV_FGT_DISABLE_SS0)) {
  		u8 disabled_mask =
@@ -290,7 +307,7 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  			(((fuse & CHV_FGT_EU_DIS_SS0_R1_MASK) >>
  			  CHV_FGT_EU_DIS_SS0_R1_SHIFT) << 4);
 
-		sseu->subslice_mask[0] |= BIT(0);
+		subslice_mask |= BIT(0);
  		intel_sseu_set_eus(sseu, 0, 0, ~disabled_mask);
  	}
 
@@ -301,10 +318,12 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  			(((fuse & CHV_FGT_EU_DIS_SS1_R1_MASK) >>
  			  CHV_FGT_EU_DIS_SS1_R1_SHIFT) << 4);
 
-		sseu->subslice_mask[0] |= BIT(1);
+		subslice_mask |= BIT(1);
  		intel_sseu_set_eus(sseu, 0, 1, ~disabled_mask);
  	}
 
+	intel_sseu_set_subslices(sseu, 0, subslice_mask);
+
  	sseu->eu_total = compute_eu_total(sseu);
 
  	/*
@@ -312,7 +331,8 @@ static void cherryview_sseu_info_init(struct drm_i915_private *dev_priv)
  	 * across subslices.
  	*/
  	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
-				sseu->eu_total / intel_sseu_subslice_total(sseu) :
+				sseu->eu_total /
+					intel_sseu_subslice_total(sseu) :
  				0;
  	/*
  	 * CHV supports subslice power gating on devices with more than
@@ -336,9 +356,8 @@ static void gen9_sseu_info_init(struct drm_i915_private *dev_priv)
  	sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >> GEN8_F2_S_ENA_SHIFT;
 
  	/* BXT has a single slice and at most 3 subslices. */
-	sseu->max_slices = IS_GEN9_LP(dev_priv) ? 1 : 3;
-	sseu->max_subslices = IS_GEN9_LP(dev_priv) ? 3 : 4;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, IS_GEN9_LP(dev_priv) ? 1 : 3,
+			    IS_GEN9_LP(dev_priv) ? 3 : 4, 8);
 
  	/*
  	 * The subslice disable field is global, i.e. it applies
@@ -357,14 +376,16 @@ static void gen9_sseu_info_init(struct drm_i915_private *dev_priv)
  			/* skip disabled slice */
  			continue;
 
-		sseu->subslice_mask[s] = subslice_mask;
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
 
  		eu_disable = I915_READ(GEN9_EU_DISABLE(s));
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			int eu_per_ss;
  			u8 eu_disabled_mask;
+			u8 ss_idx = s * sseu->ss_stride + ss / BITS_PER_BYTE;
 
-			if (!(sseu->subslice_mask[s] & BIT(ss)))
+			if (!(sseu->subslice_mask[ss_idx] &
+			      BIT(ss % BITS_PER_BYTE)))
  				/* skip disabled subslice */
  				continue;
 
@@ -437,9 +458,7 @@ static void broadwell_sseu_info_init(struct drm_i915_private *dev_priv)
 
  	fuse2 = I915_READ(GEN8_FUSE2);
  	sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >> GEN8_F2_S_ENA_SHIFT;
-	sseu->max_slices = 3;
-	sseu->max_subslices = 3;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, 3, 3, 8);
 
  	/*
  	 * The subslice disable field is global, i.e. it applies
@@ -466,18 +485,21 @@ static void broadwell_sseu_info_init(struct drm_i915_private *dev_priv)
  			/* skip disabled slice */
  			continue;
 
-		sseu->subslice_mask[s] = subslice_mask;
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
 
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			u8 eu_disabled_mask;
+			u8 ss_idx = s * sseu->ss_stride + ss / BITS_PER_BYTE;
  			u32 n_disabled;
 
-			if (!(sseu->subslice_mask[s] & BIT(ss)))
+			if (!(sseu->subslice_mask[ss_idx] &
+			      BIT(ss % BITS_PER_BYTE)))
  				/* skip disabled subslice */
  				continue;
 
  			eu_disabled_mask =
-				eu_disable[s] >> (ss * sseu->max_eus_per_subslice);
+				eu_disable[s] >>
+					(ss * sseu->max_eus_per_subslice);
 
  			intel_sseu_set_eus(sseu, s, ss, ~eu_disabled_mask);
 
@@ -517,6 +539,7 @@ static void haswell_sseu_info_init(struct drm_i915_private *dev_priv)
  	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	u32 fuse1;
  	int s, ss;
+	u32 subslice_mask;
 
  	/*
  	 * There isn't a register to tell us how many slices/subslices. We
@@ -528,22 +551,18 @@ static void haswell_sseu_info_init(struct drm_i915_private *dev_priv)
  		/* fall through */
  	case 1:
  		sseu->slice_mask = BIT(0);
-		sseu->subslice_mask[0] = BIT(0);
+		subslice_mask = BIT(0);
  		break;
  	case 2:
  		sseu->slice_mask = BIT(0);
-		sseu->subslice_mask[0] = BIT(0) | BIT(1);
+		subslice_mask = BIT(0) | BIT(1);
  		break;
  	case 3:
  		sseu->slice_mask = BIT(0) | BIT(1);
-		sseu->subslice_mask[0] = BIT(0) | BIT(1);
-		sseu->subslice_mask[1] = BIT(0) | BIT(1);
+		subslice_mask = BIT(0) | BIT(1);
  		break;
  	}
 
-	sseu->max_slices = hweight8(sseu->slice_mask);
-	sseu->max_subslices = hweight8(sseu->subslice_mask[0]);
-
  	fuse1 = I915_READ(HSW_PAVP_FUSE1);
  	switch ((fuse1 & HSW_F1_EU_DIS_MASK) >> HSW_F1_EU_DIS_SHIFT) {
  	default:

drivers/gpu/drm/i915/intel_device_info.c static void haswell_sseu_info_init(struct drm_i915_private *dev_priv) · line 585

Quoted 8 lines
drivers/gpu/drm/i915/intel_device_info.c +4 -1
@@ -560,9 +579,14 @@ static void haswell_sseu_info_init(struct drm_i915_private *dev_priv)
  		sseu->eu_per_subslice = 6;
  		break;
  	}
-	sseu->max_eus_per_subslice = sseu->eu_per_subslice;
+
+	intel_sseu_set_info(sseu, hweight8(sseu->slice_mask),
+			    hweight8(subslice_mask),
+			    sseu->eu_per_subslice);

Personal preference: could use a local variable for eu_per_subslice above to avoid setting it to itself here.

Daniele

drivers/gpu/drm/i915/intel_device_info.c static void haswell_sseu_info_init(struct drm_i915_private *dev_priv)

Quoted 8 lines
drivers/gpu/drm/i915/intel_device_info.c +2 -0
  	for (s = 0; s < sseu->max_slices; s++) {
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
+
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			intel_sseu_set_eus(sseu, s, ss,
  					   (1UL << sseu->eu_per_subslice) - 1);
 

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

maintainer Acked-by
#18 lore raw

Re: [PATCH 5/6] drm/i915: Remove inline from sseu helper functions

In reply to Summers, Stuart
Quoted 16 lines · Stuart
Quoted 15 lines · Daniele Ceraolo Spurio
Quoted 12 lines · Stuart
Quoted 5 lines · Daniele Ceraolo Spurio

Can you elaborate a bit more on what's the rationale for this? do you just want to avoid having too many inlines since the paths they're used in are not critical, or do you have some more functional reason? This is not a critic to the patch, I just want to understand where you're coming from ;)

This was a request from Jani Nikula in a previous series update. I don't have a strong preference either way personally. If you don't have any major concerns, I'd prefer to keep the series as-is to prevent too much thrash here, but let me know.

Quoted 5 lines
Quoted 3 lines

No concerns, just please update the commit message to explain that we're moving them because there is no need for them to be inline since they're not on a critical path where we need preformance.

Sounds great.

I've become critical of superfluous inlines. They break the abstraction by exposing the internals in the header, and make the interdependencies of headers harder to resolve.

As the driver keeps growing and more people contribute to it, I think we need to pay more attention on how we structure the source. To this end we've added new gt/ subdir, are about to add gem/ and likely display/ too before long, and we've significantly split off the monster i915_drv.h and intel_drv.h headers.

Obviously inlines have their place and purpose, but I think we sprinkle them a bit too eagerly without paying attention.

I like the patch.

Signature · 5 lines
Jani Nikula, Intel Open Source Graphics Center
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#33 lore raw

✓ Fi.CI.IGT: success for Refactor to expand subslice mask (rev7)

In reply to Stuart Summers
== Series Details ==

== Series Details ==

Series: Refactor to expand subslice mask (rev7)
URL : https://patchwork.freedesktop.org/series/59742/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_6021_full -> Patchwork_12926_full
====================================================

Summary
-------

**SUCCESS**

No regressions found.

Known issues
------------

Here are the changes found in Patchwork_12926_full that come from known issues:

### IGT changes ###

#### Issues hit ####

* igt@i915_suspend@debugfs-reader:
- shard-kbl: [PASS][1] -> [DMESG-WARN][2] ([fdo#108566]) +1 similar issue
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-kbl4/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-kbl7/igt@[email protected]

* igt@i915_suspend@fence-restore-untiled:
- shard-apl: [PASS][3] -> [DMESG-WARN][4] ([fdo#108566]) +4 similar issues
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-apl4/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-apl2/igt@[email protected]

* igt@kms_cursor_crc@cursor-64x21-sliding:
- shard-skl: [PASS][5] -> [INCOMPLETE][6] ([fdo#110581]) +2 similar issues
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl6/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl2/igt@[email protected]

* igt@kms_draw_crc@draw-method-xrgb8888-render-xtiled:
- shard-snb: [PASS][7] -> [SKIP][8] ([fdo#109271]) +1 similar issue
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-snb1/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-snb4/igt@[email protected]

* igt@kms_flip@flip-vs-expired-vblank:
- shard-skl: [PASS][9] -> [FAIL][10] ([fdo#105363])
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl4/igt@[email protected]
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl8/igt@[email protected]

* igt@kms_flip@flip-vs-expired-vblank-interruptible:
- shard-iclb: [PASS][11] -> [INCOMPLETE][12] ([fdo#107713] / [fdo#110581]) +1 similar issue
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb2/igt@[email protected]
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb6/igt@[email protected]

* igt@kms_flip@flip-vs-suspend:
- shard-hsw: [PASS][13] -> [INCOMPLETE][14] ([fdo#103540] / [fdo#110581])
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-hsw6/igt@[email protected]
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-hsw1/igt@[email protected]
- shard-skl: [PASS][15] -> [INCOMPLETE][16] ([fdo#107773] / [fdo#109507] / [fdo#110581])
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl9/igt@[email protected]
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl9/igt@[email protected]

* igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-blt:
- shard-iclb: [PASS][17] -> [FAIL][18] ([fdo#103167]) +4 similar issues
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb3/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-blt.html
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb2/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-blt.html

* igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-draw-render:
- shard-iclb: [PASS][19] -> [INCOMPLETE][20] ([fdo#106978] / [fdo#107713] / [fdo#110581])
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb1/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-draw-render.html
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb5/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-draw-render.html

* igt@kms_plane@pixel-format-pipe-b-planes:
- shard-glk: [PASS][21] -> [SKIP][22] ([fdo#109271])
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-glk9/igt@[email protected]
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-glk4/igt@[email protected]

* igt@kms_plane_alpha_blend@pipe-a-constant-alpha-min:
- shard-skl: [PASS][23] -> [FAIL][24] ([fdo#108145])
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl5/igt@[email protected]
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl5/igt@[email protected]

* igt@kms_plane_scaling@pipe-c-scaler-with-pixel-format:
- shard-glk: [PASS][25] -> [SKIP][26] ([fdo#109271] / [fdo#109278])
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-glk9/igt@[email protected]
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-glk4/igt@[email protected]

* igt@kms_psr2_su@frontbuffer:
- shard-iclb: [PASS][27] -> [SKIP][28] ([fdo#109642])
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb2/igt@[email protected]
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb1/igt@[email protected]

* igt@kms_psr@psr2_sprite_plane_move:
- shard-iclb: [PASS][29] -> [SKIP][30] ([fdo#109441]) +2 similar issues
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb2/igt@kms_psr@psr2_sprite_plane_move.html
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb1/igt@kms_psr@psr2_sprite_plane_move.html

* igt@perf_pmu@enable-race-rcs0:
- shard-glk: [PASS][31] -> [INCOMPLETE][32] ([fdo#103359] / [fdo#110581] / [k.org#198133])
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-glk9/igt@[email protected]
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-glk4/igt@[email protected]

* igt@perf_pmu@rc6:
- shard-kbl: [PASS][33] -> [SKIP][34] ([fdo#109271])
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-kbl7/igt@[email protected]
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-kbl5/igt@[email protected]

* igt@tools_test@tools_test:
- shard-iclb: [PASS][35] -> [SKIP][36] ([fdo#109352])
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb7/igt@tools_test@tools_test.html
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb8/igt@tools_test@tools_test.html

#### Possible fixes ####

* igt@gem_ctx_switch@basic-all-light:
- shard-iclb: [INCOMPLETE][37] ([fdo#107713] / [fdo#109100]) -> [PASS][38]
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb1/igt@[email protected]
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb7/igt@[email protected]

* igt@gem_exec_blt@dumb-buf-min:
- shard-hsw: [INCOMPLETE][39] ([fdo#103540]) -> [PASS][40]
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-hsw6/igt@[email protected]
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-hsw8/igt@[email protected]

* igt@gem_exec_schedule@preempt-hang-vebox:
- shard-skl: [INCOMPLETE][41] ([fdo#110581]) -> [PASS][42] +1 similar issue
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl8/igt@[email protected]
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl9/igt@[email protected]

* igt@gem_workarounds@suspend-resume:
- shard-apl: [DMESG-WARN][43] ([fdo#108566]) -> [PASS][44] +3 similar issues
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-apl8/igt@[email protected]
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-apl7/igt@[email protected]

* igt@i915_pm_rpm@system-suspend-execbuf:
- shard-skl: [INCOMPLETE][45] ([fdo#104108] / [fdo#107773] / [fdo#107807]) -> [PASS][46]
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl10/igt@[email protected]
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl3/igt@[email protected]

* igt@i915_selftest@mock_requests:
- shard-skl: [INCOMPLETE][47] ([fdo#110550]) -> [PASS][48]
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl5/igt@i915_selftest@mock_requests.html
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl6/igt@i915_selftest@mock_requests.html

* igt@kms_cursor_crc@cursor-128x128-suspend:
- shard-skl: [INCOMPLETE][49] ([fdo#104108] / [fdo#107773]) -> [PASS][50]
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl5/igt@[email protected]
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl5/igt@[email protected]

* igt@kms_cursor_crc@cursor-256x256-suspend:
- shard-skl: [INCOMPLETE][51] ([fdo#104108]) -> [PASS][52]
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl2/igt@[email protected]
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl2/igt@[email protected]

* igt@kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-mmap-gtt:
- shard-iclb: [FAIL][53] ([fdo#103167]) -> [PASS][54] +1 similar issue
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb2/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-mmap-gtt.html
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb6/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-mmap-gtt.html

* igt@kms_plane@pixel-format-pipe-c-planes:
- shard-glk: [SKIP][55] ([fdo#109271]) -> [PASS][56]
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-glk4/igt@[email protected]
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-glk9/igt@[email protected]

* igt@kms_plane_lowres@pipe-a-tiling-y:
- shard-iclb: [FAIL][57] ([fdo#103166]) -> [PASS][58]
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb6/igt@[email protected]
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb8/igt@[email protected]

* igt@kms_plane_scaling@pipe-b-scaler-with-clipping-clamping:
- shard-glk: [SKIP][59] ([fdo#109271] / [fdo#109278]) -> [PASS][60] +1 similar issue
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-glk4/igt@[email protected]
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-glk9/igt@[email protected]

* igt@kms_psr@no_drrs:
- shard-iclb: [FAIL][61] ([fdo#108341]) -> [PASS][62]
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb1/igt@kms_psr@no_drrs.html
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb5/igt@kms_psr@no_drrs.html

* igt@kms_psr@psr2_primary_mmap_gtt:
- shard-iclb: [SKIP][63] ([fdo#109441]) -> [PASS][64]
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb4/igt@kms_psr@psr2_primary_mmap_gtt.html
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb2/igt@kms_psr@psr2_primary_mmap_gtt.html

* igt@kms_setmode@basic:
- shard-apl: [FAIL][65] ([fdo#99912]) -> [PASS][66]
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-apl3/igt@[email protected]
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-apl1/igt@[email protected]

* igt@kms_vblank@pipe-a-wait-busy:
- shard-iclb: [INCOMPLETE][67] ([fdo#107713]) -> [PASS][68]
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-iclb3/igt@[email protected]
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-iclb1/igt@[email protected]

#### Warnings ####

* igt@kms_atomic_transition@5x-modeset-transitions-nonblocking-fencing:
- shard-hsw: [SKIP][69] ([fdo#109271] / [fdo#109278]) -> [INCOMPLETE][70] ([fdo#103540] / [fdo#110581])
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-hsw4/igt@kms_atomic_transition@5x-modeset-transitions-nonblocking-fencing.html
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-hsw2/igt@kms_atomic_transition@5x-modeset-transitions-nonblocking-fencing.html
- shard-snb: [SKIP][71] ([fdo#109271] / [fdo#109278]) -> [SKIP][72] ([fdo#109271])
[71]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-snb1/igt@kms_atomic_transition@5x-modeset-transitions-nonblocking-fencing.html
[72]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-snb4/igt@kms_atomic_transition@5x-modeset-transitions-nonblocking-fencing.html

* igt@kms_flip@2x-flip-vs-suspend:
- shard-glk: [INCOMPLETE][73] ([fdo#103359] / [k.org#198133]) -> [INCOMPLETE][74] ([fdo#103359] / [fdo#110581] / [k.org#198133])
[73]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-glk9/igt@[email protected]
[74]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-glk4/igt@[email protected]

* igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-indfb-plflip-blt:
- shard-skl: [FAIL][75] ([fdo#108040]) -> [FAIL][76] ([fdo#103167])
[75]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl10/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-indfb-plflip-blt.html
[76]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl6/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-indfb-plflip-blt.html

* igt@kms_plane_scaling@pipe-b-scaler-with-clipping-clamping:
- shard-skl: [INCOMPLETE][77] ([fdo#108972]) -> [INCOMPLETE][78] ([fdo#108972] / [fdo#110581]) +1 similar issue
[77]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6021/shard-skl2/igt@[email protected]
[78]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/shard-skl2/igt@[email protected]

[fdo#103166]: https://bugs.freedesktop.org/show_bug.cgi?id=103166
[fdo#103167]: https://bugs.freedesktop.org/show_bug.cgi?id=103167
[fdo#103359]: https://bugs.freedesktop.org/show_bug.cgi?id=103359
[fdo#103540]: https://bugs.freedesktop.org/show_bug.cgi?id=103540
[fdo#104108]: https://bugs.freedesktop.org/show_bug.cgi?id=104108
[fdo#105363]: https://bugs.freedesktop.org/show_bug.cgi?id=105363
[fdo#106978]: https://bugs.freedesktop.org/show_bug.cgi?id=106978
[fdo#107713]: https://bugs.freedesktop.org/show_bug.cgi?id=107713
[fdo#107773]: https://bugs.freedesktop.org/show_bug.cgi?id=107773
[fdo#107807]: https://bugs.freedesktop.org/show_bug.cgi?id=107807
[fdo#108040]: https://bugs.freedesktop.org/show_bug.cgi?id=108040
[fdo#108145]: https://bugs.freedesktop.org/show_bug.cgi?id=108145
[fdo#108341]: https://bugs.freedesktop.org/show_bug.cgi?id=108341
[fdo#108566]: https://bugs.freedesktop.org/show_bug.cgi?id=108566
[fdo#108972]: https://bugs.freedesktop.org/show_bug.cgi?id=108972
[fdo#109100]: https://bugs.freedesktop.org/show_bug.cgi?id=109100
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#109278]: https://bugs.freedesktop.org/show_bug.cgi?id=109278
[fdo#109352]: https://bugs.freedesktop.org/show_bug.cgi?id=109352
[fdo#109441]: https://bugs.freedesktop.org/show_bug.cgi?id=109441
[fdo#109507]: https://bugs.freedesktop.org/show_bug.cgi?id=109507
[fdo#109642]: https://bugs.freedesktop.org/show_bug.cgi?id=109642
[fdo#110550]: https://bugs.freedesktop.org/show_bug.cgi?id=110550
[fdo#110581]: https://bugs.freedesktop.org/show_bug.cgi?id=110581
[fdo#99912]: https://bugs.freedesktop.org/show_bug.cgi?id=99912
[k.org#198133]: https://bugzilla.kernel.org/show_bug.cgi?id=198133

Participating hosts (10 -> 10)
------------------------------

No changes in participating hosts

Build changes
-------------

* Linux: CI_DRM_6021 -> Patchwork_12926

CI_DRM_6021: 850aa4220e8bf7609b03bf89bce146305704bec6 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4971: fc5e0467eb6913d21ad932aa8a31c77fdb5a9c77 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_12926: 3416f0a72f5df37142ab06ae57f55bff003ecf5d @ git://anongit.freedesktop.org/gfx-ci/linux
piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_12926/
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#27 lore raw

Re: [PATCH 6/6] drm/i915: Expand subslice mask

In reply to Daniele Ceraolo Spurio
Quoted 18 lines · Daniele Ceraolo Spurio
Quoted 16 lines · Stuart Summers

Currently, the subslice_mask runtime parameter is stored as an array of subslices per slice. Expand the subslice mask array to better match what is presented to userspace through the I915_QUERY_TOPOLOGY_INFO ioctl. The index into this array is then calculated:
slice * subslice stride + subslice index / 8

v2: fix spacing in set_sseu_info args
use set_sseu_info to initialize sseu data when building device status in debugfs
rename variables in intel_engine_types.h to avoid checkpatch warnings
v3: update headers in intel_sseu.h
v4: add const to some sseu_dev_info variables

     use sseu->eu_stride for EU stride calculations
Quoted 6 lines
Quoted 2 lines
Cc Daniele Ceraolo Spurio <[email protected]>

Can you also get an ack from Lionel, to make sure this all fits with the
expected reporting?

Cc Lionel Landwerlin <[email protected]>
Quoted 35 lines
Quoted 22 lines
---
  drivers/gpu/drm/i915/gt/intel_engine_cs.c    |   6 +-
  drivers/gpu/drm/i915/gt/intel_engine_types.h |  32 +++--
  drivers/gpu/drm/i915/gt/intel_hangcheck.c    |   3 +-
  drivers/gpu/drm/i915/gt/intel_sseu.c         |  49 +++++--
  drivers/gpu/drm/i915/gt/intel_sseu.h         |  16 ++-
  drivers/gpu/drm/i915/gt/intel_workarounds.c  |   2 +-
  drivers/gpu/drm/i915/i915_debugfs.c          |  44 +++---
  drivers/gpu/drm/i915/i915_drv.c              |   6 +-
  drivers/gpu/drm/i915/i915_gpu_error.c        |   5 +-
  drivers/gpu/drm/i915/i915_query.c            |  10 +-
  drivers/gpu/drm/i915/intel_device_info.c     | 142 +++++++++++---
-----
  11 files changed, 198 insertions(+), 117 deletions(-)
drivers/gpu/drm/i915/gt/intel_engine_cs.c +0 -0
index 6e40f8ea9a6a..8f7967cc9a50 100644

drivers/gpu/drm/i915/gt/intel_engine_cs.c u32 intel_calculate_mcr_s_ss_select(struct · line 920

Quoted 8 lines
drivers/gpu/drm/i915/gt/intel_engine_cs.c +1 -1
@@ -914,7 +914,7 @@ u32 intel_calculate_mcr_s_ss_select(struct
 drm_i915_private *dev_priv)
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 >sseu;
  	u32 mcr_s_ss_select;
  	u32 slice = fls(sseu->slice_mask);
-	u32 subslice = fls(sseu->subslice_mask[slice]);
+	u32 subslice = fls(sseu->subslice_mask[slice * sseu-
 >ss_stride]);

This (and the registers we use below) only works if ss_stride = 1. Can
we add a:

Quoted 5 lines
	GEM_BUG_ON(sseu->ss_stride > 1);

to catch the fact that this function will need updating to handle that
case if/when we get it?

I'll rework this and post an update.

Quoted 54 lines
Quoted 54 lines

if (IS_GEN(dev_priv, 10))
mcr_s_ss_select = GEN8_MCR_SLICE(slice) |

drivers/gpu/drm/i915/gt/intel_engine_cs.c +1 -0
@@ -990,6 +990,7 @@ void intel_engine_get_instdone(struct
 intel_engine_cs *engine,
  			       struct intel_instdone *instdone)
  {
  	struct drm_i915_private *dev_priv = engine->i915;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 >sseu;
  	struct intel_uncore *uncore = engine->uncore;
  	u32 mmio_base = engine->mmio_base;
  	int slice;
drivers/gpu/drm/i915/gt/intel_engine_cs.c +2 -1
@@ -1007,7 +1008,8 @@ void intel_engine_get_instdone(struct
 intel_engine_cs *engine,
 
  		instdone->slice_common =
  			intel_uncore_read(uncore, GEN7_SC_INSTDONE);
-		for_each_instdone_slice_subslice(dev_priv, slice,
 subslice) {
+		for_each_instdone_slice_subslice(dev_priv, sseu, slice,
+						 subslice) {
  			instdone->sampler[slice][subslice] =

read_subslice_reg(dev_priv, slice,
subslice,
GEN7_SAMPLER_INSTDONE
);

drivers/gpu/drm/i915/gt/intel_engine_types.h +0 -14
index 9d64e33f8427..1710546a2446 100644
@@ -534,20 +534,22 @@ intel_engine_needs_breadcrumb_tasklet(const
 struct intel_engine_cs *engine)
  	return engine->flags & I915_ENGINE_NEEDS_BREADCRUMB_TASKLET;
  }
 
-#define instdone_slice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.slice_mask)
-
-#define instdone_subslice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.subslice_mask[0])
-
-#define for_each_instdone_slice_subslice(dev_priv__, slice__,
 subslice__) \
-	for ((slice__) = 0, (subslice__) = 0; \
-	     (slice__) < I915_MAX_SLICES; \
-	     (subslice__) = ((subslice__) + 1) < I915_MAX_SUBSLICES ?
 (subslice__) + 1 : 0, \
-	       (slice__) += ((subslice__) == 0)) \
-		for_each_if((BIT(slice__) &
Quoted 8 lines

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 539

Quoted 6 lines

instdone_slice_mask(dev_priv__)) && \

drivers/gpu/drm/i915/gt/intel_engine_types.h +0 -1
-			    (BIT(subslice__) &

instdone_subslice_mask(dev_priv__)))

drivers/gpu/drm/i915/gt/intel_engine_types.h +3 -0
+#define instdone_has_slice(dev_priv___, sseu___, slice___) \
+	((IS_GEN(dev_priv___, 7) ? \
+	  1 : (sseu___)->slice_mask) & \

I'd put the ternary op on the same line here for readability

Yeah good point.

Quoted 4 lines

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 540

Quoted 2 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
+	BIT(slice___)) \

no need for "\" here (and below).

Ok.

Quoted 8 lines
Quoted 4 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +2 -0
+
+#define instdone_has_subslice(dev_priv__, sseu__, slice__,

subslice__) \

need some more parenthesis in this macro to fix the MACRO_ARG_PRECEDENCE
warning in checkpatch.

Thanks, I'll fix this.

Quoted 23 lines

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 545

Quoted 4 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +3 -0
+	((IS_GEN(dev_priv__, 7) ? \
+	  1 : (sseu__)->subslice_mask[slice__ * (sseu__)->ss_stride + \
+				      subslice__ / BITS_PER_BYTE]) & \

The calculation to get the correct subslice u8 entry:

	sseu->subslice_mask[s * sseu->ss_stride + ss / BITS_PER_BYTE]

seems to be repeated a few times in this patch, so it might be worth moving it to its own inline function. looks like you always ultimately
want a bool, so we could also go a bit further and have something like:

static inline bool intel_sseu_has_subslice(sseu, s, ss)

{
	u8 mask = sseu->subslice_mask[s * sseu->ss_stride +
				      ss / BITS_PER_BYTE];

	return mask & BIT(ss % BITS_PER_BYTE);
}
Quoted 7 lines

and then do:

#define instdone_has_subslice(dev_priv__, sseu__, slice__,

subslice__) \
	((IS_GEN(dev_priv__, 7) ? subslice__ == 0 : \
		intel_sseu_has_subslice(...))

Good point and good suggeMakes sense. stion. I'll clean this up.

Quoted 11 lines
Quoted 8 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +3 -0
+	 BIT(subslice__ % BITS_PER_BYTE)) \
+
+#define for_each_instdone_slice_subslice(dev_priv_, sseu_, slice_,

subslice_) \

drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
+	for ((slice_) = 0, (subslice_) = 0; (slice_) < I915_MAX_SLICES;

\

drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
+	     (subslice_) = ((subslice_) + 1) < I915_MAX_SUBSLICES ?

(subslice_) + 1 : 0, \

This ternary op should be simplifiable as:

Quoted 1 line
	(subslice_) = ((subslice_) + 1) % I915_MAX_SUBSLICES,

I had been carrying this forward as-is as much as possible. Your suggestion makes sense though. I'll take a look.

Quoted 6 lines
Quoted 4 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +2 -0
+	       (slice_) += ((subslice_) == 0)) \
+		for_each_if(instdone_has_slice(dev_priv_, sseu_, slice)

&& \

missing the "_" after "slice"

Good catch, thanks!

Quoted 59 lines
Quoted 59 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
+			    instdone_has_subslice(dev_priv_, sseu_,

slice_, subslice_)) \

  #endif /* __INTEL_ENGINE_TYPES_H__ */
drivers/gpu/drm/i915/gt/intel_hangcheck.c +1 -0
index e5eaa06fe74d..53c1c98161e1 100644
@@ -50,6 +50,7 @@ static bool instdone_unchanged(u32
 current_instdone, u32 *old_instdone)
  static bool subunits_stuck(struct intel_engine_cs *engine)
  {
  	struct drm_i915_private *dev_priv = engine->i915;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 >sseu;
  	struct intel_instdone instdone;

struct intel_instdone *accu_instdone = &engine-

Quoted 1 line

hangcheck.instdone;

bool stuck;

drivers/gpu/drm/i915/gt/intel_hangcheck.c +1 -1
@@ -71,7 +72,7 @@ static bool subunits_stuck(struct intel_engine_cs
 *engine)
  	stuck &= instdone_unchanged(instdone.slice_common,
  				    &accu_instdone->slice_common);
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice) {
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice,
 subslice) {
  		stuck &=

instdone_unchanged(instdone.sampler[slice][subslice],
&accu_instdone-

Quoted 1 line

sampler[slice][subslice]);

stuck &=
instdone_unchanged(instdone.row[slice][subslice],

drivers/gpu/drm/i915/gt/intel_sseu.c +11 -0
index 4a0b82fc108c..49316b7ef074 100644
@@ -8,6 +8,17 @@
  #include "intel_lrc_reg.h"
  #include "intel_sseu.h"
 
+void intel_sseu_set_info(struct sseu_dev_info *sseu, u8
 max_slices,
+			 u8 max_subslices, u8 max_eus_per_subslice)
+{
+	sseu->max_slices = max_slices;
+	sseu->max_subslices = max_subslices;
+	sseu->max_eus_per_subslice = max_eus_per_subslice;
+
+	sseu->ss_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
+	sseu->eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
+}
+
  unsigned int
  intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
  {
Quoted 8 lines

drivers/gpu/drm/i915/gt/intel_sseu.c intel_sseu_subslice_total(const struct · line 36

Quoted 5 lines
drivers/gpu/drm/i915/gt/intel_sseu.c +0 -0
@@ -22,17 +33,39 @@ intel_sseu_subslice_total(const struct
 sseu_dev_info *sseu)
  unsigned int
  intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu,
 u8 slice)

Here we pass slice as u8, but below we use int. Any reason for the difference?

No good reason. I'll fix it.

Quoted 8 lines
Quoted 8 lines
drivers/gpu/drm/i915/gt/intel_sseu.c +4 -1
  {
-	return hweight8(sseu->subslice_mask[slice]);
+	unsigned int i, total = 0;
+
+	for (i = 0; i < sseu->ss_stride; i++)
+		total += hweight8(sseu->subslice_mask[slice * sseu-
Quoted 1 line

ss_stride +

Quoted 12 lines

drivers/gpu/drm/i915/gt/intel_sseu.c intel_sseu_subslice_total(const struct sseu_dev_info *sseu) · line 46

Quoted 8 lines
drivers/gpu/drm/i915/gt/intel_sseu.c +6 -0
+						      i]);
+
+	return total;
+}
+
+void intel_sseu_copy_subslices(const struct sseu_dev_info *sseu,

int slice,

drivers/gpu/drm/i915/gt/intel_sseu.c +1 -0
+			       u8 *to_mask, const u8 *from_mask)

You always use sseu->subslice_mask has a from_mask, can't we just get
that from the sseu param and avoid the from_mask?

I wanted to make this a little more generic, but I agree maybe that's overkill. I'll rework this.

Quoted 134 lines
Quoted 134 lines
drivers/gpu/drm/i915/gt/intel_sseu.c +7 -0
+{
+	int offset = slice * sseu->ss_stride;
+
+	memcpy(&to_mask[offset], &from_mask[offset], sseu->ss_stride);
+}
+
+void intel_sseu_set_subslices(struct sseu_dev_info *sseu, int

slice,

drivers/gpu/drm/i915/gt/intel_sseu.c +7 -0
+			      u32 ss_mask)
+{
+	int i, offset = slice * sseu->ss_stride;
+
+	for (i = 0; i < sseu->ss_stride; i++)
+		sseu->subslice_mask[offset + i] =
+			(ss_mask >> (BITS_PER_BYTE * i)) & 0xff;
  }
 
  static int intel_sseu_eu_idx(const struct sseu_dev_info *sseu,

int slice,
int subslice)
{

drivers/gpu/drm/i915/gt/intel_sseu.c +2 -4
-	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
-					   BITS_PER_BYTE);
-	int slice_stride = sseu->max_subslices * subslice_stride;
+	int slice_stride = sseu->max_subslices * sseu->eu_stride;
 
-	return slice * slice_stride + subslice * subslice_stride;
+	return slice * slice_stride + subslice * sseu->eu_stride;
  }
 
  u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int

slice,

drivers/gpu/drm/i915/gt/intel_sseu.c +1 -2
@@ -41,8 +74,7 @@ u16 intel_sseu_get_eus(const struct sseu_dev_info
 *sseu, int slice,
  	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
  	u16 eu_mask = 0;
 
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,
 BITS_PER_BYTE); i++) {
+	for (i = 0; i < sseu->eu_stride; i++) {
  		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
  			(i * BITS_PER_BYTE);
  	}
drivers/gpu/drm/i915/gt/intel_sseu.c +1 -2
@@ -55,8 +87,7 @@ void intel_sseu_set_eus(struct sseu_dev_info
 *sseu, int slice, int subslice,
  {
  	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
 
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,
 BITS_PER_BYTE); i++) {
+	for (i = 0; i < sseu->eu_stride; i++) {
  		sseu->eu_mask[offset + i] =
  			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
  	}
drivers/gpu/drm/i915/gt/intel_sseu.h +6 -1
index 56e3721ae83f..bf01f338a8cc 100644
@@ -9,16 +9,18 @@
 
  #include <linux/types.h>
  #include <linux/kernel.h>
+#include <linux/string.h>
 
  struct drm_i915_private;
 
  #define GEN_MAX_SLICES		(6) /* CNL upper bound */
  #define GEN_MAX_SUBSLICES	(8) /* ICL upper bound */
  #define GEN_SSEU_STRIDE(bits) DIV_ROUND_UP(bits, BITS_PER_BYTE)
+#define GEN_MAX_SUBSLICE_STRIDE GEN_SSEU_STRIDE(GEN_MAX_SUBSLICES)
 
  struct sseu_dev_info {
  	u8 slice_mask;
-	u8 subslice_mask[GEN_MAX_SLICES];
+	u8 subslice_mask[GEN_MAX_SLICES * GEN_MAX_SUBSLICE_STRIDE];
  	u16 eu_total;
  	u8 eu_per_subslice;
  	u8 min_eu_in_pool;
@@ -33,6 +35,9 @@ struct sseu_dev_info {
  	u8 max_subslices;
  	u8 max_eus_per_subslice;
 
+	u8 ss_stride;
+	u8 eu_stride;
+
  	/* We don't have more than 8 eus per subslice at the moment and

as we
* store eus enabled using bits, no need to multiply by eus per
* subslice.

drivers/gpu/drm/i915/gt/intel_sseu.h +4 -0
@@ -63,12 +68,21 @@ intel_sseu_from_device_info(const struct
 sseu_dev_info *sseu)
  	return value;
  }
 
+void intel_sseu_set_info(struct sseu_dev_info *sseu, u8
 max_slices,
+			 u8 max_subslices, u8 max_eus_per_subslice);
+
  unsigned int
  intel_sseu_subslice_total(const struct sseu_dev_info *sseu);
 
  unsigned int
  intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu,
 u8 slice);
 
+void intel_sseu_copy_subslices(const struct sseu_dev_info *sseu,

int slice,

drivers/gpu/drm/i915/gt/intel_sseu.h +3 -0
+			       u8 *to_mask, const u8 *from_mask);
+
+void intel_sseu_set_subslices(struct sseu_dev_info *sseu, int

slice,

drivers/gpu/drm/i915/gt/intel_sseu.h +2 -0
+			      u32 ss_mask);
+
  u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int

slice,
int subslice);

drivers/gpu/drm/i915/gt/intel_workarounds.c +1 -1
index 43e290306551..7c7e9556c1c5 100644
@@ -767,7 +767,7 @@ wa_init_mcr(struct drm_i915_private *i915,
 struct i915_wa_list *wal)
  		u32 slice = fls(sseu->slice_mask);
  		u32 fuse3 =
  			intel_uncore_read(&i915->uncore,
 GEN10_MIRROR_FUSE3);
-		u8 ss_mask = sseu->subslice_mask[slice];
+		u8 ss_mask = sseu->subslice_mask[slice * sseu-
Quoted 7 lines

drivers/gpu/drm/i915/gt/intel_workarounds.c wa_init_mcr(struct drm_i915_private *i915, · line 773

Quoted 1 line
drivers/gpu/drm/i915/gt/intel_workarounds.c +0 -0
 >ss_stride];

could use a

	GEM_BUG_ON(sseu->ss_stride > 1);

here as well to remind us this will need changes in that case

Ok.

Quoted 162 lines
Quoted 162 lines

u8 enabled_mask = (ss_mask | ss_mask >>
GEN10_L3BANK_PAIR_COUNT) &
GEN10_L3BANK_MASK;

drivers/gpu/drm/i915/i915_debugfs.c +3 -2
index 3f3ee83ac315..08089c24db25 100644
@@ -1257,6 +1257,7 @@ static void i915_instdone_info(struct
 drm_i915_private *dev_priv,
  			       struct seq_file *m,
  			       struct intel_instdone *instdone)
  {
+	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	int slice;
  	int subslice;
@@ -1272,11 +1273,11 @@ static void i915_instdone_info(struct
 drm_i915_private *dev_priv,
  	if (INTEL_GEN(dev_priv) <= 6)
  		return;
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice)
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice,
 subslice)
  		seq_printf(m, "\t\tSAMPLER_INSTDONE[%d][%d]: 0x%08x\n",
  			   slice, subslice, instdone-
 >sampler[slice][subslice]);
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice)
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice,

subslice)
seq_printf(m, "\t\tROW_INSTDONE[%d][%d]: 0x%08x\n",
slice, subslice, instdone-

Quoted 1 line

row[slice][subslice]);

}

drivers/gpu/drm/i915/i915_debugfs.c +3 -1
@@ -4066,7 +4067,9 @@ static void gen10_sseu_device_status(struct
 drm_i915_private *dev_priv,
  			continue;
 
  		sseu->slice_mask |= BIT(s);
-		sseu->subslice_mask[s] = info->sseu.subslice_mask[s];
+		intel_sseu_copy_subslices(&info->sseu, s,
+					  sseu->subslice_mask,
+					  info->sseu.subslice_mask);
 
  		for (ss = 0; ss < info->sseu.max_subslices; ss++) {

unsigned int eu_cnt;

drivers/gpu/drm/i915/i915_debugfs.c +13 -11
@@ -4117,18 +4120,22 @@ static void gen9_sseu_device_status(struct
 drm_i915_private *dev_priv,
  		sseu->slice_mask |= BIT(s);
 
  		if (IS_GEN9_BC(dev_priv))
-			sseu->subslice_mask[s] =
-				RUNTIME_INFO(dev_priv)-
 >sseu.subslice_mask[s];
+			intel_sseu_copy_subslices(&info->sseu, s,
+						  sseu->subslice_mask,
+						  info-
 >sseu.subslice_mask);
 
  		for (ss = 0; ss < info->sseu.max_subslices; ss++) {
  			unsigned int eu_cnt;
+			u8 ss_idx = s * info->sseu.ss_stride +
+				    ss / BITS_PER_BYTE;
 
  			if (IS_GEN9_LP(dev_priv)) {
  				if (!(s_reg[s] &
 (GEN9_PGCTL_SS_ACK(ss))))
  					/* skip disabled subslice */
  					continue;
 
-				sseu->subslice_mask[s] |= BIT(ss);
+				sseu->subslice_mask[ss_idx] |=
+					BIT(ss % BITS_PER_BYTE);
  			}
 
  			eu_cnt = 2 * hweight32(eu_reg[2*s + ss/2] &
@@ -4145,25 +4152,24 @@ static void gen9_sseu_device_status(struct
 drm_i915_private *dev_priv,
  static void broadwell_sseu_device_status(struct drm_i915_private
 *dev_priv,
  					 struct sseu_dev_info *sseu)
  {
+	struct intel_runtime_info *info = RUNTIME_INFO(dev_priv);
  	u32 slice_info = I915_READ(GEN8_GT_SLICE_INFO);
  	int s;
 
  	sseu->slice_mask = slice_info & GEN8_LSLICESTAT_MASK;
 
  	if (sseu->slice_mask) {
-		sseu->eu_per_subslice =
-			RUNTIME_INFO(dev_priv)->sseu.eu_per_subslice;
-		for (s = 0; s < fls(sseu->slice_mask); s++) {
-			sseu->subslice_mask[s] =
-				RUNTIME_INFO(dev_priv)-
 >sseu.subslice_mask[s];
-		}
+		sseu->eu_per_subslice = info->sseu.eu_per_subslice;
+		for (s = 0; s < fls(sseu->slice_mask); s++)
+			intel_sseu_copy_subslices(&info->sseu, s,
+						  sseu->subslice_mask,
+						  info-
 >sseu.subslice_mask);
  		sseu->eu_total = sseu->eu_per_subslice *
  				 intel_sseu_subslice_total(sseu);
 
  		/* subtract fused off EU(s) from enabled slice(s) */
  		for (s = 0; s < fls(sseu->slice_mask); s++) {
-			u8 subslice_7eu =
-				RUNTIME_INFO(dev_priv)-
Quoted 1 line

sseu.subslice_7eu[s];

drivers/gpu/drm/i915/i915_debugfs.c +6 -5
+			u8 subslice_7eu = info->sseu.subslice_7eu[s];
 
  			sseu->eu_total -= hweight8(subslice_7eu);
  		}
@@ -4210,6 +4216,7 @@ static void i915_print_sseu_info(struct
 seq_file *m, bool is_available_info,
  static int i915_sseu_status(struct seq_file *m, void *unused)
  {
  	struct drm_i915_private *dev_priv = node_to_i915(m->private);
+	const struct intel_runtime_info *info = RUNTIME_INFO(dev_priv);
  	struct sseu_dev_info sseu;
  	intel_wakeref_t wakeref;
@@ -4217,14 +4224,13 @@ static int i915_sseu_status(struct seq_file
 *m, void *unused)
  		return -ENODEV;
 
  	seq_puts(m, "SSEU Device Info\n");
-	i915_print_sseu_info(m, true, &RUNTIME_INFO(dev_priv)->sseu);
+	i915_print_sseu_info(m, true, &info->sseu);
 
  	seq_puts(m, "SSEU Device Status\n");
  	memset(&sseu, 0, sizeof(sseu));
-	sseu.max_slices = RUNTIME_INFO(dev_priv)->sseu.max_slices;
-	sseu.max_subslices = RUNTIME_INFO(dev_priv)-
 >sseu.max_subslices;
-	sseu.max_eus_per_subslice =
-		RUNTIME_INFO(dev_priv)->sseu.max_eus_per_subslice;
+	intel_sseu_set_info(&sseu, info->sseu.max_slices,
+			    info->sseu.max_subslices,
+			    info->sseu.max_eus_per_subslice);
 
  	with_intel_runtime_pm(dev_priv, wakeref) {
  		if (IS_CHERRYVIEW(dev_priv))
drivers/gpu/drm/i915/i915_drv.c +1 -1
index 130c5140db0d..6afe4e3afea4 100644
@@ -326,7 +326,7 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
  	struct pci_dev *pdev = dev_priv->drm.pdev;
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 >sseu;
  	drm_i915_getparam_t *param = data;
-	int value;
+	int value = 0;
 
Quoted 11 lines

drivers/gpu/drm/i915/i915_drv.c static int i915_getparam_ioctl(struct · line 459

Quoted 9 lines
  	switch (param->param) {
  	case I915_PARAM_IRQ_ACTIVE:
drivers/gpu/drm/i915/i915_drv.c +1 -1
@@ -455,7 +455,9 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
  			return -ENODEV;
  		break;
  	case I915_PARAM_SUBSLICE_MASK:
-		value = sseu->subslice_mask[0];
+		/* Only copy bits from the first subslice */

s/subslice/slice/ ?

True, thanks.

Quoted 185 lines
Quoted 185 lines
drivers/gpu/drm/i915/i915_drv.c +2 -0
+		memcpy(&value, sseu->subslice_mask,
+		       min(sseu->ss_stride, (u8)sizeof(value)));
  		if (!value)
  			return -ENODEV;

break;

drivers/gpu/drm/i915/i915_gpu_error.c +1 -0
index e1b858bd1d32..140918dd9b7d 100644
@@ -407,6 +407,7 @@ static void print_error_buffers(struct
 drm_i915_error_state_buf *m,
  static void error_print_instdone(struct drm_i915_error_state_buf
 *m,
  				 const struct drm_i915_error_engine
 *ee)
  {
+	struct sseu_dev_info *sseu = &RUNTIME_INFO(m->i915)->sseu;

int slice;
int subslice;

drivers/gpu/drm/i915/i915_gpu_error.c +2 -2
@@ -422,12 +423,12 @@ static void error_print_instdone(struct
 drm_i915_error_state_buf *m,
  	if (INTEL_GEN(m->i915) <= 6)
  		return;
 
-	for_each_instdone_slice_subslice(m->i915, slice, subslice)
+	for_each_instdone_slice_subslice(m->i915, sseu, slice,
 subslice)
  		err_printf(m, "  SAMPLER_INSTDONE[%d][%d]: 0x%08x\n",
  			   slice, subslice,
  			   ee->instdone.sampler[slice][subslice]);
 
-	for_each_instdone_slice_subslice(m->i915, slice, subslice)
+	for_each_instdone_slice_subslice(m->i915, sseu, slice,
 subslice)

err_printf(m, " ROW_INSTDONE[%d][%d]: 0x%08x\n",
slice, subslice, ee->instdone.row[slice][subslice]);

drivers/gpu/drm/i915/i915_query.c +0 -2
index 7c1708c22811..000dcb145ce0 100644
@@ -37,8 +37,6 @@ static int query_topology_info(struct
 drm_i915_private *dev_priv,
  	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 >sseu;
  	struct drm_i915_query_topology_info topo;
  	u32 slice_length, subslice_length, eu_length, total_length;
-	u8 subslice_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
-	u8 eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
  	int ret;

if (query_item->flags != 0)

drivers/gpu/drm/i915/i915_query.c +2 -2
@@ -50,8 +48,8 @@ static int query_topology_info(struct
 drm_i915_private *dev_priv,
  	BUILD_BUG_ON(sizeof(u8) != sizeof(sseu->slice_mask));
 
  	slice_length = sizeof(sseu->slice_mask);
-	subslice_length = sseu->max_slices * subslice_stride;
-	eu_length = sseu->max_slices * sseu->max_subslices * eu_stride;
+	subslice_length = sseu->max_slices * sseu->ss_stride;
+	eu_length = sseu->max_slices * sseu->max_subslices * sseu-
 >eu_stride;
  	total_length = sizeof(topo) + slice_length + subslice_length +

eu_length;

drivers/gpu/drm/i915/i915_query.c +2 -2
@@ -69,9 +67,9 @@ static int query_topology_info(struct
 drm_i915_private *dev_priv,
  	topo.max_eus_per_subslice = sseu->max_eus_per_subslice;
 
  	topo.subslice_offset = slice_length;
-	topo.subslice_stride = subslice_stride;
+	topo.subslice_stride = sseu->ss_stride;
  	topo.eu_offset = slice_length + subslice_length;
-	topo.eu_stride = eu_stride;
+	topo.eu_stride = sseu->eu_stride;
 
  	if (__copy_to_user(u64_to_user_ptr(query_item->data_ptr),

&topo, sizeof(topo)))

drivers/gpu/drm/i915/intel_device_info.c +28 -3
index e1dbccf04cd9..bbbc0a8c2183 100644
@@ -84,17 +84,42 @@ void intel_device_info_dump_flags(const struct
 intel_device_info *info,
  #undef PRINT_FLAG
  }
 
+#define SS_STR_MAX_SIZE (GEN_MAX_SUBSLICE_STRIDE * 2)
+
+static u8 *
+subslice_per_slice_str(u8 *buf, const struct sseu_dev_info *sseu,
 u8 slice)
+{
+	int i;
+	u8 ss_offset = slice * sseu->ss_stride;
+
+	GEM_BUG_ON(slice >= sseu->max_slices);
+
+	memset(buf, 0, SS_STR_MAX_SIZE);
+
+	/*
+	 * Print subslice information in reverse order to match
+	 * userspace expectations.
+	 */
+	for (i = 0; i < sseu->ss_stride; i++)
+		sprintf(&buf[i * 2], "%02x",
+			sseu->subslice_mask[ss_offset + sseu->ss_stride 
-
+					    (i + 1)]);
+
+	return buf;
+}
+
  static void sseu_dump(const struct sseu_dev_info *sseu, struct
 drm_printer *p)
  {
  	int s;
+	u8 buf[SS_STR_MAX_SIZE];
 
  	drm_printf(p, "slice total: %u, mask=%04x\n",
  		   hweight8(sseu->slice_mask), sseu->slice_mask);
  	drm_printf(p, "subslice total: %u\n",
 intel_sseu_subslice_total(sseu));
  	for (s = 0; s < sseu->max_slices; s++) {
-		drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
+		drm_printf(p, "slice%d: %u subslices, mask=%s\n",
  			   s, intel_sseu_subslices_per_slice(sseu, s),
-			   sseu->subslice_mask[s]);
+			   subslice_per_slice_str(buf, sseu, s));
  	}
  	drm_printf(p, "EU total: %u\n", sseu->eu_total);
  	drm_printf(p, "EU per subslice: %u\n", sseu->eu_per_subslice);
@@ -118,6 +143,7 @@ void intel_device_info_dump_topology(const
 struct sseu_dev_info *sseu,
  				     struct drm_printer *p)
  {
  	int s, ss;
+	u8 buf[SS_STR_MAX_SIZE];
 
  	if (sseu->max_slices == 0) {
  		drm_printf(p, "Unavailable\n");
drivers/gpu/drm/i915/intel_device_info.c +2 -2
@@ -125,9 +151,9 @@ void intel_device_info_dump_topology(const
 struct sseu_dev_info *sseu,
  	}
 
  	for (s = 0; s < sseu->max_slices; s++) {
-		drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
+		drm_printf(p, "slice%d: %u subslice(s) (0x%s):\n",
  			   s, intel_sseu_subslices_per_slice(sseu, s),
-			   sseu->subslice_mask[s]);
+			   subslice_per_slice_str(buf, sseu, s));
 
  		for (ss = 0; ss < sseu->max_subslices; ss++) {

u16 enabled_eus = intel_sseu_get_eus(sseu, s,
ss);

drivers/gpu/drm/i915/intel_device_info.c +4 -9
@@ -156,15 +182,10 @@ static void gen11_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	u8 eu_en;
  	int s;
 
-	if (IS_ELKHARTLAKE(dev_priv)) {
-		sseu->max_slices = 1;
-		sseu->max_subslices = 4;
-		sseu->max_eus_per_subslice = 8;
-	} else {
-		sseu->max_slices = 1;
-		sseu->max_subslices = 8;
-		sseu->max_eus_per_subslice = 8;
-	}
+	if (IS_ELKHARTLAKE(dev_priv))
+		intel_sseu_set_info(sseu, 1, 4, 8);
+	else
+		intel_sseu_set_info(sseu, 1, 8, 8);
 
  	s_en = I915_READ(GEN11_GT_SLICE_ENABLE) & GEN11_GT_S_ENA_MASK;
  	ss_en = ~I915_READ(GEN11_GT_SUBSLICE_DISABLE);
drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -177,9 +198,11 @@ static void gen11_sseu_info_init(struct
 drm_i915_private *dev_priv)
  			int ss;
 
Quoted 9 lines

drivers/gpu/drm/i915/intel_device_info.c static void gen11_sseu_info_init(struct drm_i915_private *dev_priv) · line 202

Quoted 5 lines
drivers/gpu/drm/i915/intel_device_info.c +2 -1
  			sseu->slice_mask |= BIT(s);
-			sseu->subslice_mask[s] = (ss_en >> ss_idx) &
 ss_en_mask;
+			sseu->subslice_mask[s * sseu->ss_stride] =
+				(ss_en >> ss_idx) & ss_en_mask;

Shouldn't this just call intel_sseu_set_subslices() instead of doing the
setting locally?

Yes, let me fix this.

Quoted 10 lines

drivers/gpu/drm/i915/intel_device_info.c static void gen11_sseu_info_init(struct drm_i915_private *dev_priv) · line 205

Quoted 6 lines
drivers/gpu/drm/i915/intel_device_info.c +1 -1
  			for (ss = 0; ss < sseu->max_subslices; ss++) {
-				if (sseu->subslice_mask[s] & BIT(ss))
+				if (sseu->subslice_mask[s * sseu-
Quoted 1 line

ss_stride] &

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+				    BIT(ss))

This culd use the intel_sseu_has_subslice() suggested earlier, otherwise
it needs to consider ss_stride > 1

Ok.

Quoted 239 lines
Quoted 239 lines

intel_sseu_set_eus(sseu, s, ss,
eu_en);
}
}

drivers/gpu/drm/i915/intel_device_info.c +2 -15
@@ -201,23 +224,10 @@ static void gen10_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	const int eu_mask = 0xff;
  	u32 subslice_mask, eu_en;
 
+	intel_sseu_set_info(sseu, 6, 4, 8);
+
  	sseu->slice_mask = (fuse2 & GEN10_F2_S_ENA_MASK) >>
  			    GEN10_F2_S_ENA_SHIFT;
-	sseu->max_slices = 6;
-	sseu->max_subslices = 4;
-	sseu->max_eus_per_subslice = 8;
-
-	subslice_mask = (1 << 4) - 1;
-	subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
-			   GEN10_F2_SS_DIS_SHIFT);
-
-	/*
-	 * Slice0 can have up to 3 subslices, but there are only 2 in
-	 * slice1/2.
-	 */
-	sseu->subslice_mask[0] = subslice_mask;
-	for (s = 1; s < sseu->max_slices; s++)
-		sseu->subslice_mask[s] = subslice_mask & 0x3;
 
  	/* Slice0 */
  	eu_en = ~I915_READ(GEN8_EU_DISABLE0);
drivers/gpu/drm/i915/intel_device_info.c +0 -1
@@ -242,14 +252,22 @@ static void gen10_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	eu_en = ~I915_READ(GEN10_EU_DISABLE3);
  	intel_sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
 
-	/* Do a second pass where we mark the subslices disabled if all

their

drivers/gpu/drm/i915/intel_device_info.c +8 -3
-	 * eus are off.
-	 */
+	subslice_mask = (1 << 4) - 1;
+	subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
+			   GEN10_F2_SS_DIS_SHIFT);
+
  	for (s = 0; s < sseu->max_slices; s++) {
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			if (intel_sseu_get_eus(sseu, s, ss) == 0)
-				sseu->subslice_mask[s] &= ~BIT(ss);
+				subslice_mask &= ~BIT(ss);
  		}
+
+		/*
+		 * Slice0 can have up to 3 subslices, but there are

only 2 in

drivers/gpu/drm/i915/intel_device_info.c +3 -0
+		 * slice1/2.
+		 */
+		intel_sseu_set_subslices(sseu, s, s == 0 ?

subslice_mask :

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+							   subslice_mas

k & 0x3);
}

  	sseu->eu_total = compute_eu_total(sseu);
drivers/gpu/drm/i915/intel_device_info.c +2 -3
@@ -275,13 +293,12 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
  {
  	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	u32 fuse;
+	u8 subslice_mask;
 
  	fuse = I915_READ(CHV_FUSE_GT);
 
  	sseu->slice_mask = BIT(0);
-	sseu->max_slices = 1;
-	sseu->max_subslices = 2;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, 1, 2, 8);
 
  	if (!(fuse & CHV_FGT_DISABLE_SS0)) {

u8 disabled_mask =

drivers/gpu/drm/i915/intel_device_info.c +4 -2
@@ -290,7 +307,7 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
  			(((fuse & CHV_FGT_EU_DIS_SS0_R1_MASK) >>
  			  CHV_FGT_EU_DIS_SS0_R1_SHIFT) << 4);
 
-		sseu->subslice_mask[0] |= BIT(0);
+		subslice_mask |= BIT(0);
  		intel_sseu_set_eus(sseu, 0, 0, ~disabled_mask);
  	}
@@ -301,10 +318,12 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
  			(((fuse & CHV_FGT_EU_DIS_SS1_R1_MASK) >>
  			  CHV_FGT_EU_DIS_SS1_R1_SHIFT) << 4);
 
-		sseu->subslice_mask[0] |= BIT(1);
+		subslice_mask |= BIT(1);
  		intel_sseu_set_eus(sseu, 0, 1, ~disabled_mask);
  	}
 
+	intel_sseu_set_subslices(sseu, 0, subslice_mask);
+
  	sseu->eu_total = compute_eu_total(sseu);
 

/*

drivers/gpu/drm/i915/intel_device_info.c +0 -1
@@ -312,7 +331,8 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	 * across subslices.
  	*/
  	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
-				sseu->eu_total /

intel_sseu_subslice_total(sseu) :

drivers/gpu/drm/i915/intel_device_info.c +2 -0
+				sseu->eu_total /
+					intel_sseu_subslice_total(sseu)

:
0;
/*
* CHV supports subslice power gating on devices with more than

drivers/gpu/drm/i915/intel_device_info.c +2 -3
@@ -336,9 +356,8 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >>
 GEN8_F2_S_ENA_SHIFT;
 
  	/* BXT has a single slice and at most 3 subslices. */
-	sseu->max_slices = IS_GEN9_LP(dev_priv) ? 1 : 3;
-	sseu->max_subslices = IS_GEN9_LP(dev_priv) ? 3 : 4;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, IS_GEN9_LP(dev_priv) ? 1 : 3,
+			    IS_GEN9_LP(dev_priv) ? 3 : 4, 8);
 

/*
* The subslice disable field is global, i.e. it applies

drivers/gpu/drm/i915/intel_device_info.c +4 -2
@@ -357,14 +376,16 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
  			/* skip disabled slice */
  			continue;
 
-		sseu->subslice_mask[s] = subslice_mask;
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
 
  		eu_disable = I915_READ(GEN9_EU_DISABLE(s));
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			int eu_per_ss;
  			u8 eu_disabled_mask;
+			u8 ss_idx = s * sseu->ss_stride + ss /
 BITS_PER_BYTE;
 
-			if (!(sseu->subslice_mask[s] & BIT(ss)))
+			if (!(sseu->subslice_mask[ss_idx] &
+			      BIT(ss % BITS_PER_BYTE)))
  				/* skip disabled subslice */

continue;

drivers/gpu/drm/i915/intel_device_info.c +1 -3
@@ -437,9 +458,7 @@ static void broadwell_sseu_info_init(struct
 drm_i915_private *dev_priv)
 
  	fuse2 = I915_READ(GEN8_FUSE2);
  	sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >>
 GEN8_F2_S_ENA_SHIFT;
-	sseu->max_slices = 3;
-	sseu->max_subslices = 3;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, 3, 3, 8);
 

/*
* The subslice disable field is global, i.e. it applies

drivers/gpu/drm/i915/intel_device_info.c +6 -3
@@ -466,18 +485,21 @@ static void broadwell_sseu_info_init(struct
 drm_i915_private *dev_priv)
  			/* skip disabled slice */
  			continue;
 
-		sseu->subslice_mask[s] = subslice_mask;
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
 
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			u8 eu_disabled_mask;
+			u8 ss_idx = s * sseu->ss_stride + ss /
 BITS_PER_BYTE;
  			u32 n_disabled;
 
-			if (!(sseu->subslice_mask[s] & BIT(ss)))
+			if (!(sseu->subslice_mask[ss_idx] &
+			      BIT(ss % BITS_PER_BYTE)))
  				/* skip disabled subslice */
  				continue;
 
  			eu_disabled_mask =
-				eu_disable[s] >> (ss * sseu-
 >max_eus_per_subslice);
+				eu_disable[s] >>
+					(ss * sseu-
Quoted 1 line

max_eus_per_subslice);

intel_sseu_set_eus(sseu, s, ss,
~eu_disabled_mask);

drivers/gpu/drm/i915/intel_device_info.c +1 -0
@@ -517,6 +539,7 @@ static void haswell_sseu_info_init(struct
 drm_i915_private *dev_priv)
  	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
  	u32 fuse1;
  	int s, ss;
+	u32 subslice_mask;
 
  	/*

* There isn't a register to tell us how many slices/subslices.
We

drivers/gpu/drm/i915/intel_device_info.c +3 -7
@@ -528,22 +551,18 @@ static void haswell_sseu_info_init(struct
 drm_i915_private *dev_priv)
  		/* fall through */
  	case 1:
  		sseu->slice_mask = BIT(0);
-		sseu->subslice_mask[0] = BIT(0);
+		subslice_mask = BIT(0);
  		break;
  	case 2:
  		sseu->slice_mask = BIT(0);
-		sseu->subslice_mask[0] = BIT(0) | BIT(1);
+		subslice_mask = BIT(0) | BIT(1);
  		break;
  	case 3:
  		sseu->slice_mask = BIT(0) | BIT(1);
-		sseu->subslice_mask[0] = BIT(0) | BIT(1);
-		sseu->subslice_mask[1] = BIT(0) | BIT(1);
+		subslice_mask = BIT(0) | BIT(1);
  		break;
  	}
 
-	sseu->max_slices = hweight8(sseu->slice_mask);
-	sseu->max_subslices = hweight8(sseu->subslice_mask[0]);
-
  	fuse1 = I915_READ(HSW_PAVP_FUSE1);
  	switch ((fuse1 & HSW_F1_EU_DIS_MASK) >> HSW_F1_EU_DIS_SHIFT) {

default:

drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -560,9 +579,14 @@ static void haswell_sseu_info_init(struct
 drm_i915_private *dev_priv)
  		sseu->eu_per_subslice = 6;
  		break;
  	}
Quoted 8 lines

drivers/gpu/drm/i915/intel_device_info.c static void haswell_sseu_info_init(struct drm_i915_private *dev_priv) · line 585

Quoted 5 lines
drivers/gpu/drm/i915/intel_device_info.c +4 -1
-	sseu->max_eus_per_subslice = sseu->eu_per_subslice;
+
+	intel_sseu_set_info(sseu, hweight8(sseu->slice_mask),
+			    hweight8(subslice_mask),
+			    sseu->eu_per_subslice);

Personal preference: could use a local variable for eu_per_subslice above to avoid setting it to itself here.

Yeah this is a bit ugly... I'll change it.

Thanks for the feedback!
Stuart

Quoted 12 lines

Daniele

Quoted 9 lines
drivers/gpu/drm/i915/intel_device_info.c +2 -0
  	for (s = 0; s < sseu->max_slices; s++) {
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
+
  		for (ss = 0; ss < sseu->max_subslices; ss++) {
  			intel_sseu_set_eus(sseu, s, ss,
  					   (1UL << sseu-
Quoted 1 line

eu_per_subslice) - 1);

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#19 lore raw

Re: [PATCH 5/6] drm/i915: Remove inline from sseu helper functions

In reply to Jani Nikula
Quoted 44 lines · Jani Nikula
Quoted 21 lines · Stuart
Quoted 20 lines · Daniele Ceraolo Spurio
Quoted 17 lines · Stuart
Quoted 7 lines · Daniele Ceraolo Spurio

Can you elaborate a bit more on what's the rationale for this? do
you just want to avoid having too many inlines since the paths
they're used in are not critical, or do you have some more functional reason? This is not a critic to the patch, I just want to understand where you're coming from ;)

This was a request from Jani Nikula in a previous series update. I
don't have a strong preference either way personally. If you don't
have any major concerns, I'd prefer to keep the series as-is to prevent too much thrash here, but let me know.

Quoted 6 lines
Quoted 4 lines

No concerns, just please update the commit message to explain that
we're moving them because there is no need for them to be inline since they're not on a critical path where we need preformance.

Sounds great.

I've become critical of superfluous inlines. They break the abstraction
by exposing the internals in the header, and make the interdependencies
of headers harder to resolve.

As the driver keeps growing and more people contribute to it, I think we
need to pay more attention on how we structure the source. To this end
we've added new gt/ subdir, are about to add gem/ and likely display/ too before long, and we've significantly split off the monster i915_drv.h and intel_drv.h headers.

Quoted 7 lines

Obviously inlines have their place and purpose, but I think we sprinkle
them a bit too eagerly without paying attention.

I like the patch.

Jani, based on Daniele's feedback, I'm planning on squashing this patch with the patch that moves these helper functions to intel_sseu.h. Any issue keeping your Ack here?

-Stuart

Quoted 2 lines

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

maintainer
#20 lore raw

Re: [PATCH 5/6] drm/i915: Remove inline from sseu helper functions

Quoted 6 lines · Stuart
Quoted 1 line · Jani Nikula

Jani, based on Daniele's feedback, I'm planning on squashing this patch with the patch that moves these helper functions to intel_sseu.h. Any issue keeping your Ack here?

None.

BR,
Jani.

Signature · 5 lines
Jani Nikula, Intel Open Source Graphics Center
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#21 lore raw

Re: [PATCH 5/6] drm/i915: Remove inline from sseu helper functions

Quoted 5 lines · Jani Nikula
Quoted 3 lines · Stuart
Quoted 1 line · Jani Nikula
Quoted 7 lines
Quoted 5 lines

Jani, based on Daniele's feedback, I'm planning on squashing this patch
with the patch that moves these helper functions to intel_sseu.h. Any
issue keeping your Ack here?

None.

Thanks for the Ack!

-Stuart

Quoted 4 lines

BR,
Jani.

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#28 lore raw

Re: [PATCH 6/6] drm/i915: Expand subslice mask

In reply to Summers, Stuart
Quoted 18 lines · Stuart
Quoted 17 lines · Daniele Ceraolo Spurio
Quoted 16 lines · Stuart Summers

Currently, the subslice_mask runtime parameter is stored as an array of subslices per slice. Expand the subslice mask array to better match what is presented to userspace through the I915_QUERY_TOPOLOGY_INFO ioctl. The index into this array is then calculated:
slice * subslice stride + subslice index / 8

v2: fix spacing in set_sseu_info args
use set_sseu_info to initialize sseu data when building device status in debugfs
rename variables in intel_engine_types.h to avoid checkpatch warnings
v3: update headers in intel_sseu.h
v4: add const to some sseu_dev_info variables

      use sseu->eu_stride for EU stride calculations
Quoted 6 lines
Quoted 5 lines
Quoted 2 lines
Cc Daniele Ceraolo Spurio <[email protected]>

Can you also get an ack from Lionel, to make sure this all fits with the
expected reporting?

Cc Lionel Landwerlin <[email protected]>

The change makes sense to me but I haven't had time to verify every single bit in this patch :)

The i915_query IGT tests should be able to catch some issues on HSW (which can have 10EUs per subslice).

Quoted 1,016 lines
Quoted 34 lines
Quoted 22 lines
---
   drivers/gpu/drm/i915/gt/intel_engine_cs.c    |   6 +-
   drivers/gpu/drm/i915/gt/intel_engine_types.h |  32 +++--
   drivers/gpu/drm/i915/gt/intel_hangcheck.c    |   3 +-
   drivers/gpu/drm/i915/gt/intel_sseu.c         |  49 +++++--
   drivers/gpu/drm/i915/gt/intel_sseu.h         |  16 ++-
   drivers/gpu/drm/i915/gt/intel_workarounds.c  |   2 +-
   drivers/gpu/drm/i915/i915_debugfs.c          |  44 +++---
   drivers/gpu/drm/i915/i915_drv.c              |   6 +-
   drivers/gpu/drm/i915/i915_gpu_error.c        |   5 +-
   drivers/gpu/drm/i915/i915_query.c            |  10 +-
   drivers/gpu/drm/i915/intel_device_info.c     | 142 +++++++++++---
-----
   11 files changed, 198 insertions(+), 117 deletions(-)
drivers/gpu/drm/i915/gt/intel_engine_cs.c +0 -0
index 6e40f8ea9a6a..8f7967cc9a50 100644

drivers/gpu/drm/i915/gt/intel_engine_cs.c u32 intel_calculate_mcr_s_ss_select(struct · line 920

Quoted 8 lines
drivers/gpu/drm/i915/gt/intel_engine_cs.c +1 -1
@@ -914,7 +914,7 @@ u32 intel_calculate_mcr_s_ss_select(struct
 drm_i915_private *dev_priv)
   	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 > sseu;
   	u32 mcr_s_ss_select;
   	u32 slice = fls(sseu->slice_mask);
-	u32 subslice = fls(sseu->subslice_mask[slice]);
+	u32 subslice = fls(sseu->subslice_mask[slice * sseu-
 > ss_stride]);

This (and the registers we use below) only works if ss_stride = 1. Can
we add a:

Quoted 5 lines
	GEM_BUG_ON(sseu->ss_stride > 1);

to catch the fact that this function will need updating to handle that
case if/when we get it?

I'll rework this and post an update.

Quoted 52 lines
Quoted 52 lines

if (IS_GEN(dev_priv, 10))
mcr_s_ss_select = GEN8_MCR_SLICE(slice) |

drivers/gpu/drm/i915/gt/intel_engine_cs.c +1 -0
@@ -990,6 +990,7 @@ void intel_engine_get_instdone(struct
 intel_engine_cs *engine,
   			       struct intel_instdone *instdone)
   {
   	struct drm_i915_private *dev_priv = engine->i915;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 > sseu;
   	struct intel_uncore *uncore = engine->uncore;
   	u32 mmio_base = engine->mmio_base;
   	int slice;
drivers/gpu/drm/i915/gt/intel_engine_cs.c +2 -1
@@ -1007,7 +1008,8 @@ void intel_engine_get_instdone(struct
 intel_engine_cs *engine,
 
   		instdone->slice_common =
   			intel_uncore_read(uncore, GEN7_SC_INSTDONE);
-		for_each_instdone_slice_subslice(dev_priv, slice,
 subslice) {
+		for_each_instdone_slice_subslice(dev_priv, sseu, slice,
+						 subslice) {
   			instdone->sampler[slice][subslice] =

read_subslice_reg(dev_priv, slice,
subslice,
GEN7_SAMPLER_INSTDONE
);

drivers/gpu/drm/i915/gt/intel_engine_types.h +0 -13
index 9d64e33f8427..1710546a2446 100644
@@ -534,20 +534,22 @@ intel_engine_needs_breadcrumb_tasklet(const
 struct intel_engine_cs *engine)
   	return engine->flags & I915_ENGINE_NEEDS_BREADCRUMB_TASKLET;
   }
 
-#define instdone_slice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.slice_mask)
-
-#define instdone_subslice_mask(dev_priv__) \
-	(IS_GEN(dev_priv__, 7) ? \
-	 1 : RUNTIME_INFO(dev_priv__)->sseu.subslice_mask[0])
-
-#define for_each_instdone_slice_subslice(dev_priv__, slice__,
 subslice__) \
-	for ((slice__) = 0, (subslice__) = 0; \
-	     (slice__) < I915_MAX_SLICES; \
-	     (subslice__) = ((subslice__) + 1) < I915_MAX_SUBSLICES ?
 (subslice__) + 1 : 0, \
-	       (slice__) += ((subslice__) == 0)) \
Quoted 8 lines

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 539

Quoted 7 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +0 -1
-		for_each_if((BIT(slice__) &

instdone_slice_mask(dev_priv__)) && \

drivers/gpu/drm/i915/gt/intel_engine_types.h +0 -1
-			    (BIT(subslice__) &

instdone_subslice_mask(dev_priv__)))

drivers/gpu/drm/i915/gt/intel_engine_types.h +3 -0
+#define instdone_has_slice(dev_priv___, sseu___, slice___) \
+	((IS_GEN(dev_priv___, 7) ? \
+	  1 : (sseu___)->slice_mask) & \

I'd put the ternary op on the same line here for readability

Yeah good point.

Quoted 2 lines

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 540

Quoted 1 line
drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
+	BIT(slice___)) \

no need for "\" here (and below).

Ok.

Quoted 6 lines
Quoted 3 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +2 -0
+
+#define instdone_has_subslice(dev_priv__, sseu__, slice__,

subslice__) \

need some more parenthesis in this macro to fix the MACRO_ARG_PRECEDENCE
warning in checkpatch.

Thanks, I'll fix this.

Quoted 21 lines

drivers/gpu/drm/i915/gt/intel_engine_types.h intel_engine_needs_breadcrumb_tasklet(const struct intel_engine_cs *engine) · line 545

Quoted 3 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +3 -0
+	((IS_GEN(dev_priv__, 7) ? \
+	  1 : (sseu__)->subslice_mask[slice__ * (sseu__)->ss_stride + \
+				      subslice__ / BITS_PER_BYTE]) & \

The calculation to get the correct subslice u8 entry:

	sseu->subslice_mask[s * sseu->ss_stride + ss / BITS_PER_BYTE]

seems to be repeated a few times in this patch, so it might be worth moving it to its own inline function. looks like you always ultimately
want a bool, so we could also go a bit further and have something like:

static inline bool intel_sseu_has_subslice(sseu, s, ss)

{
	u8 mask = sseu->subslice_mask[s * sseu->ss_stride +
				      ss / BITS_PER_BYTE];

	return mask & BIT(ss % BITS_PER_BYTE);
}
Quoted 7 lines

and then do:

#define instdone_has_subslice(dev_priv__, sseu__, slice__,

subslice__) \
	((IS_GEN(dev_priv__, 7) ? subslice__ == 0 : \
		intel_sseu_has_subslice(...))

Good point and good suggeMakes sense. stion. I'll clean this up.

Quoted 11 lines
Quoted 8 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +3 -0
+	 BIT(subslice__ % BITS_PER_BYTE)) \
+
+#define for_each_instdone_slice_subslice(dev_priv_, sseu_, slice_,

subslice_) \

drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
+	for ((slice_) = 0, (subslice_) = 0; (slice_) < I915_MAX_SLICES;

\

drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
+	     (subslice_) = ((subslice_) + 1) < I915_MAX_SUBSLICES ?

(subslice_) + 1 : 0, \

This ternary op should be simplifiable as:

	(subslice_) = ((subslice_) + 1) % I915_MAX_SUBSLICES,

I had been carrying this forward as-is as much as possible. Your suggestion makes sense though. I'll take a look.

Quoted 4 lines
Quoted 3 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +2 -0
+	       (slice_) += ((subslice_) == 0)) \
+		for_each_if(instdone_has_slice(dev_priv_, sseu_, slice)

&& \

missing the "_" after "slice"

Good catch, thanks!

Quoted 57 lines
Quoted 57 lines
drivers/gpu/drm/i915/gt/intel_engine_types.h +1 -0
+			    instdone_has_subslice(dev_priv_, sseu_,
slice_, subslice_)) \
   
   #endif /* __INTEL_ENGINE_TYPES_H__ */
drivers/gpu/drm/i915/gt/intel_hangcheck.c +1 -0
index e5eaa06fe74d..53c1c98161e1 100644
@@ -50,6 +50,7 @@ static bool instdone_unchanged(u32
 current_instdone, u32 *old_instdone)
   static bool subunits_stuck(struct intel_engine_cs *engine)
   {
   	struct drm_i915_private *dev_priv = engine->i915;
+	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 > sseu;
   	struct intel_instdone instdone;

struct intel_instdone *accu_instdone = &engine-

Quoted 1 line

hangcheck.instdone;

bool stuck;

drivers/gpu/drm/i915/gt/intel_hangcheck.c +1 -1
@@ -71,7 +72,7 @@ static bool subunits_stuck(struct intel_engine_cs
 *engine)
   	stuck &= instdone_unchanged(instdone.slice_common,
   				    &accu_instdone->slice_common);
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice) {
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice,
 subslice) {
   		stuck &=

instdone_unchanged(instdone.sampler[slice][subslice],
&accu_instdone-

Quoted 1 line

sampler[slice][subslice]);

stuck &=
instdone_unchanged(instdone.row[slice][subslice],

drivers/gpu/drm/i915/gt/intel_sseu.c +11 -0
index 4a0b82fc108c..49316b7ef074 100644
@@ -8,6 +8,17 @@
   #include "intel_lrc_reg.h"
   #include "intel_sseu.h"
 
+void intel_sseu_set_info(struct sseu_dev_info *sseu, u8
 max_slices,
+			 u8 max_subslices, u8 max_eus_per_subslice)
+{
+	sseu->max_slices = max_slices;
+	sseu->max_subslices = max_subslices;
+	sseu->max_eus_per_subslice = max_eus_per_subslice;
+
+	sseu->ss_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
+	sseu->eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
+}
+
   unsigned int
   intel_sseu_subslice_total(const struct sseu_dev_info *sseu)
Quoted 8 lines

drivers/gpu/drm/i915/gt/intel_sseu.c intel_sseu_subslice_total(const struct · line 36

Quoted 6 lines
   {
drivers/gpu/drm/i915/gt/intel_sseu.c +0 -0
@@ -22,17 +33,39 @@ intel_sseu_subslice_total(const struct
 sseu_dev_info *sseu)
   unsigned int
   intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu,
 u8 slice)

Here we pass slice as u8, but below we use int. Any reason for the difference?

No good reason. I'll fix it.

Quoted 7 lines
Quoted 7 lines
drivers/gpu/drm/i915/gt/intel_sseu.c +4 -1
   {
-	return hweight8(sseu->subslice_mask[slice]);
+	unsigned int i, total = 0;
+
+	for (i = 0; i < sseu->ss_stride; i++)
+		total += hweight8(sseu->subslice_mask[slice * sseu-
Quoted 1 line

ss_stride +

Quoted 11 lines

drivers/gpu/drm/i915/gt/intel_sseu.c intel_sseu_subslice_total(const struct sseu_dev_info *sseu) · line 46

Quoted 8 lines
drivers/gpu/drm/i915/gt/intel_sseu.c +6 -0
+						      i]);
+
+	return total;
+}
+
+void intel_sseu_copy_subslices(const struct sseu_dev_info *sseu,

int slice,

drivers/gpu/drm/i915/gt/intel_sseu.c +1 -0
+			       u8 *to_mask, const u8 *from_mask)

You always use sseu->subslice_mask has a from_mask, can't we just get
that from the sseu param and avoid the from_mask?

I wanted to make this a little more generic, but I agree maybe that's overkill. I'll rework this.

Quoted 133 lines
Quoted 133 lines
drivers/gpu/drm/i915/gt/intel_sseu.c +7 -0
+{
+	int offset = slice * sseu->ss_stride;
+
+	memcpy(&to_mask[offset], &from_mask[offset], sseu->ss_stride);
+}
+
+void intel_sseu_set_subslices(struct sseu_dev_info *sseu, int

slice,

drivers/gpu/drm/i915/gt/intel_sseu.c +7 -0
+			      u32 ss_mask)
+{
+	int i, offset = slice * sseu->ss_stride;
+
+	for (i = 0; i < sseu->ss_stride; i++)
+		sseu->subslice_mask[offset + i] =
+			(ss_mask >> (BITS_PER_BYTE * i)) & 0xff;
   }
 
   static int intel_sseu_eu_idx(const struct sseu_dev_info *sseu,

int slice,
int subslice)
{

drivers/gpu/drm/i915/gt/intel_sseu.c +2 -4
-	int subslice_stride = DIV_ROUND_UP(sseu->max_eus_per_subslice,
-					   BITS_PER_BYTE);
-	int slice_stride = sseu->max_subslices * subslice_stride;
+	int slice_stride = sseu->max_subslices * sseu->eu_stride;
 
-	return slice * slice_stride + subslice * subslice_stride;
+	return slice * slice_stride + subslice * sseu->eu_stride;
   }
 
   u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int

slice,

drivers/gpu/drm/i915/gt/intel_sseu.c +1 -2
@@ -41,8 +74,7 @@ u16 intel_sseu_get_eus(const struct sseu_dev_info
 *sseu, int slice,
   	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
   	u16 eu_mask = 0;
 
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,
 BITS_PER_BYTE); i++) {
+	for (i = 0; i < sseu->eu_stride; i++) {
   		eu_mask |= ((u16)sseu->eu_mask[offset + i]) <<
   			(i * BITS_PER_BYTE);
   	}
drivers/gpu/drm/i915/gt/intel_sseu.c +1 -2
@@ -55,8 +87,7 @@ void intel_sseu_set_eus(struct sseu_dev_info
 *sseu, int slice, int subslice,
   {
   	int i, offset = intel_sseu_eu_idx(sseu, slice, subslice);
 
-	for (i = 0;
-	     i < DIV_ROUND_UP(sseu->max_eus_per_subslice,
 BITS_PER_BYTE); i++) {
+	for (i = 0; i < sseu->eu_stride; i++) {
   		sseu->eu_mask[offset + i] =
   			(eu_mask >> (BITS_PER_BYTE * i)) & 0xff;
   	}
drivers/gpu/drm/i915/gt/intel_sseu.h +6 -1
index 56e3721ae83f..bf01f338a8cc 100644
@@ -9,16 +9,18 @@
 
   #include <linux/types.h>
   #include <linux/kernel.h>
+#include <linux/string.h>
 
   struct drm_i915_private;
 
   #define GEN_MAX_SLICES		(6) /* CNL upper bound */
   #define GEN_MAX_SUBSLICES	(8) /* ICL upper bound */
   #define GEN_SSEU_STRIDE(bits) DIV_ROUND_UP(bits, BITS_PER_BYTE)
+#define GEN_MAX_SUBSLICE_STRIDE GEN_SSEU_STRIDE(GEN_MAX_SUBSLICES)
 
   struct sseu_dev_info {
   	u8 slice_mask;
-	u8 subslice_mask[GEN_MAX_SLICES];
+	u8 subslice_mask[GEN_MAX_SLICES * GEN_MAX_SUBSLICE_STRIDE];
   	u16 eu_total;
   	u8 eu_per_subslice;
   	u8 min_eu_in_pool;
@@ -33,6 +35,9 @@ struct sseu_dev_info {
   	u8 max_subslices;
   	u8 max_eus_per_subslice;
 
+	u8 ss_stride;
+	u8 eu_stride;
+
   	/* We don't have more than 8 eus per subslice at the moment and

as we
* store eus enabled using bits, no need to multiply by eus per
* subslice.

drivers/gpu/drm/i915/gt/intel_sseu.h +4 -0
@@ -63,12 +68,21 @@ intel_sseu_from_device_info(const struct
 sseu_dev_info *sseu)
   	return value;
   }
 
+void intel_sseu_set_info(struct sseu_dev_info *sseu, u8
 max_slices,
+			 u8 max_subslices, u8 max_eus_per_subslice);
+
   unsigned int
   intel_sseu_subslice_total(const struct sseu_dev_info *sseu);
 
   unsigned int
   intel_sseu_subslices_per_slice(const struct sseu_dev_info *sseu,
 u8 slice);
 
+void intel_sseu_copy_subslices(const struct sseu_dev_info *sseu,

int slice,

drivers/gpu/drm/i915/gt/intel_sseu.h +3 -0
+			       u8 *to_mask, const u8 *from_mask);
+
+void intel_sseu_set_subslices(struct sseu_dev_info *sseu, int

slice,

drivers/gpu/drm/i915/gt/intel_sseu.h +2 -0
+			      u32 ss_mask);
+
   u16 intel_sseu_get_eus(const struct sseu_dev_info *sseu, int

slice,
int subslice);

drivers/gpu/drm/i915/gt/intel_workarounds.c +1 -1
index 43e290306551..7c7e9556c1c5 100644
@@ -767,7 +767,7 @@ wa_init_mcr(struct drm_i915_private *i915,
 struct i915_wa_list *wal)
   		u32 slice = fls(sseu->slice_mask);
   		u32 fuse3 =
   			intel_uncore_read(&i915->uncore,
 GEN10_MIRROR_FUSE3);
-		u8 ss_mask = sseu->subslice_mask[slice];
+		u8 ss_mask = sseu->subslice_mask[slice * sseu-
Quoted 6 lines

drivers/gpu/drm/i915/gt/intel_workarounds.c wa_init_mcr(struct drm_i915_private *i915, · line 773

Quoted 1 line
drivers/gpu/drm/i915/gt/intel_workarounds.c +0 -0
 > ss_stride];

could use a

	GEM_BUG_ON(sseu->ss_stride > 1);

here as well to remind us this will need changes in that case

Ok.

Quoted 161 lines
Quoted 161 lines

u8 enabled_mask = (ss_mask | ss_mask >>
GEN10_L3BANK_PAIR_COUNT) &
GEN10_L3BANK_MASK;

drivers/gpu/drm/i915/i915_debugfs.c +3 -2
index 3f3ee83ac315..08089c24db25 100644
@@ -1257,6 +1257,7 @@ static void i915_instdone_info(struct
 drm_i915_private *dev_priv,
   			       struct seq_file *m,
   			       struct intel_instdone *instdone)
   {
+	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
   	int slice;
   	int subslice;
@@ -1272,11 +1273,11 @@ static void i915_instdone_info(struct
 drm_i915_private *dev_priv,
   	if (INTEL_GEN(dev_priv) <= 6)
   		return;
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice)
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice,
 subslice)
   		seq_printf(m, "\t\tSAMPLER_INSTDONE[%d][%d]: 0x%08x\n",
   			   slice, subslice, instdone-
 > sampler[slice][subslice]);
 
-	for_each_instdone_slice_subslice(dev_priv, slice, subslice)
+	for_each_instdone_slice_subslice(dev_priv, sseu, slice,

subslice)
seq_printf(m, "\t\tROW_INSTDONE[%d][%d]: 0x%08x\n",
slice, subslice, instdone-

Quoted 1 line

row[slice][subslice]);

}

drivers/gpu/drm/i915/i915_debugfs.c +3 -1
@@ -4066,7 +4067,9 @@ static void gen10_sseu_device_status(struct
 drm_i915_private *dev_priv,
   			continue;
 
   		sseu->slice_mask |= BIT(s);
-		sseu->subslice_mask[s] = info->sseu.subslice_mask[s];
+		intel_sseu_copy_subslices(&info->sseu, s,
+					  sseu->subslice_mask,
+					  info->sseu.subslice_mask);
 
   		for (ss = 0; ss < info->sseu.max_subslices; ss++) {
   			unsigned int eu_cnt;
drivers/gpu/drm/i915/i915_debugfs.c +13 -11
@@ -4117,18 +4120,22 @@ static void gen9_sseu_device_status(struct
 drm_i915_private *dev_priv,
   		sseu->slice_mask |= BIT(s);
 
   		if (IS_GEN9_BC(dev_priv))
-			sseu->subslice_mask[s] =
-				RUNTIME_INFO(dev_priv)-
 > sseu.subslice_mask[s];
+			intel_sseu_copy_subslices(&info->sseu, s,
+						  sseu->subslice_mask,
+						  info-
 > sseu.subslice_mask);
 
   		for (ss = 0; ss < info->sseu.max_subslices; ss++) {
   			unsigned int eu_cnt;
+			u8 ss_idx = s * info->sseu.ss_stride +
+				    ss / BITS_PER_BYTE;
 
   			if (IS_GEN9_LP(dev_priv)) {
   				if (!(s_reg[s] &
 (GEN9_PGCTL_SS_ACK(ss))))
   					/* skip disabled subslice */
   					continue;
 
-				sseu->subslice_mask[s] |= BIT(ss);
+				sseu->subslice_mask[ss_idx] |=
+					BIT(ss % BITS_PER_BYTE);
   			}
 
   			eu_cnt = 2 * hweight32(eu_reg[2*s + ss/2] &
@@ -4145,25 +4152,24 @@ static void gen9_sseu_device_status(struct
 drm_i915_private *dev_priv,
   static void broadwell_sseu_device_status(struct drm_i915_private
 *dev_priv,
   					 struct sseu_dev_info *sseu)
   {
+	struct intel_runtime_info *info = RUNTIME_INFO(dev_priv);
   	u32 slice_info = I915_READ(GEN8_GT_SLICE_INFO);
   	int s;
 
   	sseu->slice_mask = slice_info & GEN8_LSLICESTAT_MASK;
 
   	if (sseu->slice_mask) {
-		sseu->eu_per_subslice =
-			RUNTIME_INFO(dev_priv)->sseu.eu_per_subslice;
-		for (s = 0; s < fls(sseu->slice_mask); s++) {
-			sseu->subslice_mask[s] =
-				RUNTIME_INFO(dev_priv)-
 > sseu.subslice_mask[s];
-		}
+		sseu->eu_per_subslice = info->sseu.eu_per_subslice;
+		for (s = 0; s < fls(sseu->slice_mask); s++)
+			intel_sseu_copy_subslices(&info->sseu, s,
+						  sseu->subslice_mask,
+						  info-
 > sseu.subslice_mask);
   		sseu->eu_total = sseu->eu_per_subslice *
   				 intel_sseu_subslice_total(sseu);
 
   		/* subtract fused off EU(s) from enabled slice(s) */
   		for (s = 0; s < fls(sseu->slice_mask); s++) {
-			u8 subslice_7eu =
-				RUNTIME_INFO(dev_priv)-
Quoted 1 line

sseu.subslice_7eu[s];

drivers/gpu/drm/i915/i915_debugfs.c +6 -5
+			u8 subslice_7eu = info->sseu.subslice_7eu[s];
 
   			sseu->eu_total -= hweight8(subslice_7eu);
   		}
@@ -4210,6 +4216,7 @@ static void i915_print_sseu_info(struct
 seq_file *m, bool is_available_info,
   static int i915_sseu_status(struct seq_file *m, void *unused)
   {
   	struct drm_i915_private *dev_priv = node_to_i915(m->private);
+	const struct intel_runtime_info *info = RUNTIME_INFO(dev_priv);
   	struct sseu_dev_info sseu;
   	intel_wakeref_t wakeref;
@@ -4217,14 +4224,13 @@ static int i915_sseu_status(struct seq_file
 *m, void *unused)
   		return -ENODEV;
 
   	seq_puts(m, "SSEU Device Info\n");
-	i915_print_sseu_info(m, true, &RUNTIME_INFO(dev_priv)->sseu);
+	i915_print_sseu_info(m, true, &info->sseu);
 
   	seq_puts(m, "SSEU Device Status\n");
   	memset(&sseu, 0, sizeof(sseu));
-	sseu.max_slices = RUNTIME_INFO(dev_priv)->sseu.max_slices;
-	sseu.max_subslices = RUNTIME_INFO(dev_priv)-
 > sseu.max_subslices;
-	sseu.max_eus_per_subslice =
-		RUNTIME_INFO(dev_priv)->sseu.max_eus_per_subslice;
+	intel_sseu_set_info(&sseu, info->sseu.max_slices,
+			    info->sseu.max_subslices,
+			    info->sseu.max_eus_per_subslice);
 
   	with_intel_runtime_pm(dev_priv, wakeref) {
   		if (IS_CHERRYVIEW(dev_priv))
drivers/gpu/drm/i915/i915_drv.c +1 -1
index 130c5140db0d..6afe4e3afea4 100644
@@ -326,7 +326,7 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
   	struct pci_dev *pdev = dev_priv->drm.pdev;
   	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 > sseu;
   	drm_i915_getparam_t *param = data;
-	int value;
+	int value = 0;
 
Quoted 10 lines

drivers/gpu/drm/i915/i915_drv.c static int i915_getparam_ioctl(struct · line 459

Quoted 9 lines
   	switch (param->param) {
   	case I915_PARAM_IRQ_ACTIVE:
drivers/gpu/drm/i915/i915_drv.c +1 -1
@@ -455,7 +455,9 @@ static int i915_getparam_ioctl(struct
 drm_device *dev, void *data,
   			return -ENODEV;
   		break;
   	case I915_PARAM_SUBSLICE_MASK:
-		value = sseu->subslice_mask[0];
+		/* Only copy bits from the first subslice */

s/subslice/slice/ ?

True, thanks.

Quoted 184 lines
Quoted 184 lines
drivers/gpu/drm/i915/i915_drv.c +2 -0
+		memcpy(&value, sseu->subslice_mask,
+		       min(sseu->ss_stride, (u8)sizeof(value)));
   		if (!value)
   			return -ENODEV;
   		break;
drivers/gpu/drm/i915/i915_gpu_error.c +1 -0
index e1b858bd1d32..140918dd9b7d 100644
@@ -407,6 +407,7 @@ static void print_error_buffers(struct
 drm_i915_error_state_buf *m,
   static void error_print_instdone(struct drm_i915_error_state_buf
 *m,
   				 const struct drm_i915_error_engine
 *ee)
   {
+	struct sseu_dev_info *sseu = &RUNTIME_INFO(m->i915)->sseu;
   	int slice;
   	int subslice;
drivers/gpu/drm/i915/i915_gpu_error.c +2 -2
@@ -422,12 +423,12 @@ static void error_print_instdone(struct
 drm_i915_error_state_buf *m,
   	if (INTEL_GEN(m->i915) <= 6)
   		return;
 
-	for_each_instdone_slice_subslice(m->i915, slice, subslice)
+	for_each_instdone_slice_subslice(m->i915, sseu, slice,
 subslice)
   		err_printf(m, "  SAMPLER_INSTDONE[%d][%d]: 0x%08x\n",
   			   slice, subslice,
   			   ee->instdone.sampler[slice][subslice]);
 
-	for_each_instdone_slice_subslice(m->i915, slice, subslice)
+	for_each_instdone_slice_subslice(m->i915, sseu, slice,
 subslice)
   		err_printf(m, "  ROW_INSTDONE[%d][%d]: 0x%08x\n",
   			   slice, subslice,
   			   ee->instdone.row[slice][subslice]);
drivers/gpu/drm/i915/i915_query.c +0 -2
index 7c1708c22811..000dcb145ce0 100644
@@ -37,8 +37,6 @@ static int query_topology_info(struct
 drm_i915_private *dev_priv,
   	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)-
 > sseu;
   	struct drm_i915_query_topology_info topo;
   	u32 slice_length, subslice_length, eu_length, total_length;
-	u8 subslice_stride = GEN_SSEU_STRIDE(sseu->max_subslices);
-	u8 eu_stride = GEN_SSEU_STRIDE(sseu->max_eus_per_subslice);
   	int ret;
   
   	if (query_item->flags != 0)
drivers/gpu/drm/i915/i915_query.c +2 -2
@@ -50,8 +48,8 @@ static int query_topology_info(struct
 drm_i915_private *dev_priv,
   	BUILD_BUG_ON(sizeof(u8) != sizeof(sseu->slice_mask));
 
   	slice_length = sizeof(sseu->slice_mask);
-	subslice_length = sseu->max_slices * subslice_stride;
-	eu_length = sseu->max_slices * sseu->max_subslices * eu_stride;
+	subslice_length = sseu->max_slices * sseu->ss_stride;
+	eu_length = sseu->max_slices * sseu->max_subslices * sseu-
 > eu_stride;
   	total_length = sizeof(topo) + slice_length + subslice_length +
   		       eu_length;
drivers/gpu/drm/i915/i915_query.c +2 -2
@@ -69,9 +67,9 @@ static int query_topology_info(struct
 drm_i915_private *dev_priv,
   	topo.max_eus_per_subslice = sseu->max_eus_per_subslice;
 
   	topo.subslice_offset = slice_length;
-	topo.subslice_stride = subslice_stride;
+	topo.subslice_stride = sseu->ss_stride;
   	topo.eu_offset = slice_length + subslice_length;
-	topo.eu_stride = eu_stride;
+	topo.eu_stride = sseu->eu_stride;
 
   	if (__copy_to_user(u64_to_user_ptr(query_item->data_ptr),
   			   &topo, sizeof(topo)))
drivers/gpu/drm/i915/intel_device_info.c +28 -3
index e1dbccf04cd9..bbbc0a8c2183 100644
@@ -84,17 +84,42 @@ void intel_device_info_dump_flags(const struct
 intel_device_info *info,
   #undef PRINT_FLAG
   }
 
+#define SS_STR_MAX_SIZE (GEN_MAX_SUBSLICE_STRIDE * 2)
+
+static u8 *
+subslice_per_slice_str(u8 *buf, const struct sseu_dev_info *sseu,
 u8 slice)
+{
+	int i;
+	u8 ss_offset = slice * sseu->ss_stride;
+
+	GEM_BUG_ON(slice >= sseu->max_slices);
+
+	memset(buf, 0, SS_STR_MAX_SIZE);
+
+	/*
+	 * Print subslice information in reverse order to match
+	 * userspace expectations.
+	 */
+	for (i = 0; i < sseu->ss_stride; i++)
+		sprintf(&buf[i * 2], "%02x",
+			sseu->subslice_mask[ss_offset + sseu->ss_stride
-
+					    (i + 1)]);
+
+	return buf;
+}
+
   static void sseu_dump(const struct sseu_dev_info *sseu, struct
 drm_printer *p)
   {
   	int s;
+	u8 buf[SS_STR_MAX_SIZE];
 
   	drm_printf(p, "slice total: %u, mask=%04x\n",
   		   hweight8(sseu->slice_mask), sseu->slice_mask);
   	drm_printf(p, "subslice total: %u\n",
 intel_sseu_subslice_total(sseu));
   	for (s = 0; s < sseu->max_slices; s++) {
-		drm_printf(p, "slice%d: %u subslices, mask=%04x\n",
+		drm_printf(p, "slice%d: %u subslices, mask=%s\n",
   			   s, intel_sseu_subslices_per_slice(sseu, s),
-			   sseu->subslice_mask[s]);
+			   subslice_per_slice_str(buf, sseu, s));
   	}
   	drm_printf(p, "EU total: %u\n", sseu->eu_total);
   	drm_printf(p, "EU per subslice: %u\n", sseu->eu_per_subslice);
@@ -118,6 +143,7 @@ void intel_device_info_dump_topology(const
 struct sseu_dev_info *sseu,
   				     struct drm_printer *p)
   {
   	int s, ss;
+	u8 buf[SS_STR_MAX_SIZE];
 
   	if (sseu->max_slices == 0) {
   		drm_printf(p, "Unavailable\n");
drivers/gpu/drm/i915/intel_device_info.c +2 -2
@@ -125,9 +151,9 @@ void intel_device_info_dump_topology(const
 struct sseu_dev_info *sseu,
   	}
 
   	for (s = 0; s < sseu->max_slices; s++) {
-		drm_printf(p, "slice%d: %u subslice(s) (0x%hhx):\n",
+		drm_printf(p, "slice%d: %u subslice(s) (0x%s):\n",
   			   s, intel_sseu_subslices_per_slice(sseu, s),
-			   sseu->subslice_mask[s]);
+			   subslice_per_slice_str(buf, sseu, s));
 
   		for (ss = 0; ss < sseu->max_subslices; ss++) {

u16 enabled_eus = intel_sseu_get_eus(sseu, s,
ss);

drivers/gpu/drm/i915/intel_device_info.c +4 -9
@@ -156,15 +182,10 @@ static void gen11_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	u8 eu_en;
   	int s;
 
-	if (IS_ELKHARTLAKE(dev_priv)) {
-		sseu->max_slices = 1;
-		sseu->max_subslices = 4;
-		sseu->max_eus_per_subslice = 8;
-	} else {
-		sseu->max_slices = 1;
-		sseu->max_subslices = 8;
-		sseu->max_eus_per_subslice = 8;
-	}
+	if (IS_ELKHARTLAKE(dev_priv))
+		intel_sseu_set_info(sseu, 1, 4, 8);
+	else
+		intel_sseu_set_info(sseu, 1, 8, 8);
 
   	s_en = I915_READ(GEN11_GT_SLICE_ENABLE) & GEN11_GT_S_ENA_MASK;
   	ss_en = ~I915_READ(GEN11_GT_SUBSLICE_DISABLE);
drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -177,9 +198,11 @@ static void gen11_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			int ss;
 
Quoted 8 lines

drivers/gpu/drm/i915/intel_device_info.c static void gen11_sseu_info_init(struct drm_i915_private *dev_priv) · line 202

Quoted 5 lines
drivers/gpu/drm/i915/intel_device_info.c +2 -1
   			sseu->slice_mask |= BIT(s);
-			sseu->subslice_mask[s] = (ss_en >> ss_idx) &
 ss_en_mask;
+			sseu->subslice_mask[s * sseu->ss_stride] =
+				(ss_en >> ss_idx) & ss_en_mask;

Shouldn't this just call intel_sseu_set_subslices() instead of doing the
setting locally?

Yes, let me fix this.

Quoted 8 lines

drivers/gpu/drm/i915/intel_device_info.c static void gen11_sseu_info_init(struct drm_i915_private *dev_priv) · line 205

Quoted 5 lines
drivers/gpu/drm/i915/intel_device_info.c +1 -1
   			for (ss = 0; ss < sseu->max_subslices; ss++) {
-				if (sseu->subslice_mask[s] & BIT(ss))
+				if (sseu->subslice_mask[s * sseu-
Quoted 1 line

ss_stride] &

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+				    BIT(ss))

This culd use the intel_sseu_has_subslice() suggested earlier, otherwise
it needs to consider ss_stride > 1

Ok.

Quoted 237 lines
Quoted 237 lines

intel_sseu_set_eus(sseu, s, ss,

eu_en);
   			}
   		}
drivers/gpu/drm/i915/intel_device_info.c +2 -15
@@ -201,23 +224,10 @@ static void gen10_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	const int eu_mask = 0xff;
   	u32 subslice_mask, eu_en;
 
+	intel_sseu_set_info(sseu, 6, 4, 8);
+
   	sseu->slice_mask = (fuse2 & GEN10_F2_S_ENA_MASK) >>
   			    GEN10_F2_S_ENA_SHIFT;
-	sseu->max_slices = 6;
-	sseu->max_subslices = 4;
-	sseu->max_eus_per_subslice = 8;
-
-	subslice_mask = (1 << 4) - 1;
-	subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
-			   GEN10_F2_SS_DIS_SHIFT);
-
-	/*
-	 * Slice0 can have up to 3 subslices, but there are only 2 in
-	 * slice1/2.
-	 */
-	sseu->subslice_mask[0] = subslice_mask;
-	for (s = 1; s < sseu->max_slices; s++)
-		sseu->subslice_mask[s] = subslice_mask & 0x3;
 
   	/* Slice0 */
   	eu_en = ~I915_READ(GEN8_EU_DISABLE0);
drivers/gpu/drm/i915/intel_device_info.c +0 -1
@@ -242,14 +252,22 @@ static void gen10_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	eu_en = ~I915_READ(GEN10_EU_DISABLE3);
   	intel_sseu_set_eus(sseu, 5, 1, eu_en & eu_mask);
 
-	/* Do a second pass where we mark the subslices disabled if all

their

drivers/gpu/drm/i915/intel_device_info.c +8 -3
-	 * eus are off.
-	 */
+	subslice_mask = (1 << 4) - 1;
+	subslice_mask &= ~((fuse2 & GEN10_F2_SS_DIS_MASK) >>
+			   GEN10_F2_SS_DIS_SHIFT);
+
   	for (s = 0; s < sseu->max_slices; s++) {
   		for (ss = 0; ss < sseu->max_subslices; ss++) {
   			if (intel_sseu_get_eus(sseu, s, ss) == 0)
-				sseu->subslice_mask[s] &= ~BIT(ss);
+				subslice_mask &= ~BIT(ss);
   		}
+
+		/*
+		 * Slice0 can have up to 3 subslices, but there are

only 2 in

drivers/gpu/drm/i915/intel_device_info.c +3 -0
+		 * slice1/2.
+		 */
+		intel_sseu_set_subslices(sseu, s, s == 0 ?

subslice_mask :

drivers/gpu/drm/i915/intel_device_info.c +1 -0
+							   subslice_mas
k & 0x3);
   	}
   
   	sseu->eu_total = compute_eu_total(sseu);
drivers/gpu/drm/i915/intel_device_info.c +2 -3
@@ -275,13 +293,12 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
   {
   	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
   	u32 fuse;
+	u8 subslice_mask;
 
   	fuse = I915_READ(CHV_FUSE_GT);
 
   	sseu->slice_mask = BIT(0);
-	sseu->max_slices = 1;
-	sseu->max_subslices = 2;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, 1, 2, 8);
 
   	if (!(fuse & CHV_FGT_DISABLE_SS0)) {

u8 disabled_mask =

drivers/gpu/drm/i915/intel_device_info.c +4 -2
@@ -290,7 +307,7 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			(((fuse & CHV_FGT_EU_DIS_SS0_R1_MASK) >>
   			  CHV_FGT_EU_DIS_SS0_R1_SHIFT) << 4);
 
-		sseu->subslice_mask[0] |= BIT(0);
+		subslice_mask |= BIT(0);
   		intel_sseu_set_eus(sseu, 0, 0, ~disabled_mask);
   	}
@@ -301,10 +318,12 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			(((fuse & CHV_FGT_EU_DIS_SS1_R1_MASK) >>
   			  CHV_FGT_EU_DIS_SS1_R1_SHIFT) << 4);
 
-		sseu->subslice_mask[0] |= BIT(1);
+		subslice_mask |= BIT(1);
   		intel_sseu_set_eus(sseu, 0, 1, ~disabled_mask);
   	}
 
+	intel_sseu_set_subslices(sseu, 0, subslice_mask);
+
   	sseu->eu_total = compute_eu_total(sseu);
 
   	/*
drivers/gpu/drm/i915/intel_device_info.c +0 -1
@@ -312,7 +331,8 @@ static void cherryview_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	 * across subslices.
   	*/
   	sseu->eu_per_subslice = intel_sseu_subslice_total(sseu) ?
-				sseu->eu_total /

intel_sseu_subslice_total(sseu) :

drivers/gpu/drm/i915/intel_device_info.c +2 -0
+				sseu->eu_total /
+					intel_sseu_subslice_total(sseu)

:

   				0;
   	/*
   	 * CHV supports subslice power gating on devices with more than
drivers/gpu/drm/i915/intel_device_info.c +2 -3
@@ -336,9 +356,8 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >>
 GEN8_F2_S_ENA_SHIFT;
 
   	/* BXT has a single slice and at most 3 subslices. */
-	sseu->max_slices = IS_GEN9_LP(dev_priv) ? 1 : 3;
-	sseu->max_subslices = IS_GEN9_LP(dev_priv) ? 3 : 4;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, IS_GEN9_LP(dev_priv) ? 1 : 3,
+			    IS_GEN9_LP(dev_priv) ? 3 : 4, 8);
 

/*
* The subslice disable field is global, i.e. it applies

drivers/gpu/drm/i915/intel_device_info.c +4 -2
@@ -357,14 +376,16 @@ static void gen9_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			/* skip disabled slice */
   			continue;
 
-		sseu->subslice_mask[s] = subslice_mask;
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
 
   		eu_disable = I915_READ(GEN9_EU_DISABLE(s));
   		for (ss = 0; ss < sseu->max_subslices; ss++) {
   			int eu_per_ss;
   			u8 eu_disabled_mask;
+			u8 ss_idx = s * sseu->ss_stride + ss /
 BITS_PER_BYTE;
 
-			if (!(sseu->subslice_mask[s] & BIT(ss)))
+			if (!(sseu->subslice_mask[ss_idx] &
+			      BIT(ss % BITS_PER_BYTE)))
   				/* skip disabled subslice */
   				continue;
drivers/gpu/drm/i915/intel_device_info.c +1 -3
@@ -437,9 +458,7 @@ static void broadwell_sseu_info_init(struct
 drm_i915_private *dev_priv)
 
   	fuse2 = I915_READ(GEN8_FUSE2);
   	sseu->slice_mask = (fuse2 & GEN8_F2_S_ENA_MASK) >>
 GEN8_F2_S_ENA_SHIFT;
-	sseu->max_slices = 3;
-	sseu->max_subslices = 3;
-	sseu->max_eus_per_subslice = 8;
+	intel_sseu_set_info(sseu, 3, 3, 8);
 

/*
* The subslice disable field is global, i.e. it applies

drivers/gpu/drm/i915/intel_device_info.c +6 -3
@@ -466,18 +485,21 @@ static void broadwell_sseu_info_init(struct
 drm_i915_private *dev_priv)
   			/* skip disabled slice */
   			continue;
 
-		sseu->subslice_mask[s] = subslice_mask;
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
 
   		for (ss = 0; ss < sseu->max_subslices; ss++) {
   			u8 eu_disabled_mask;
+			u8 ss_idx = s * sseu->ss_stride + ss /
 BITS_PER_BYTE;
   			u32 n_disabled;
 
-			if (!(sseu->subslice_mask[s] & BIT(ss)))
+			if (!(sseu->subslice_mask[ss_idx] &
+			      BIT(ss % BITS_PER_BYTE)))
   				/* skip disabled subslice */
   				continue;
 
   			eu_disabled_mask =
-				eu_disable[s] >> (ss * sseu-
 > max_eus_per_subslice);
+				eu_disable[s] >>
+					(ss * sseu-
Quoted 1 line

max_eus_per_subslice);

intel_sseu_set_eus(sseu, s, ss,
~eu_disabled_mask);

drivers/gpu/drm/i915/intel_device_info.c +1 -0
@@ -517,6 +539,7 @@ static void haswell_sseu_info_init(struct
 drm_i915_private *dev_priv)
   	struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
   	u32 fuse1;
   	int s, ss;
+	u32 subslice_mask;
 
   	/*

* There isn't a register to tell us how many slices/subslices.
We

drivers/gpu/drm/i915/intel_device_info.c +3 -7
@@ -528,22 +551,18 @@ static void haswell_sseu_info_init(struct
 drm_i915_private *dev_priv)
   		/* fall through */
   	case 1:
   		sseu->slice_mask = BIT(0);
-		sseu->subslice_mask[0] = BIT(0);
+		subslice_mask = BIT(0);
   		break;
   	case 2:
   		sseu->slice_mask = BIT(0);
-		sseu->subslice_mask[0] = BIT(0) | BIT(1);
+		subslice_mask = BIT(0) | BIT(1);
   		break;
   	case 3:
   		sseu->slice_mask = BIT(0) | BIT(1);
-		sseu->subslice_mask[0] = BIT(0) | BIT(1);
-		sseu->subslice_mask[1] = BIT(0) | BIT(1);
+		subslice_mask = BIT(0) | BIT(1);
   		break;
   	}
 
-	sseu->max_slices = hweight8(sseu->slice_mask);
-	sseu->max_subslices = hweight8(sseu->subslice_mask[0]);
-
   	fuse1 = I915_READ(HSW_PAVP_FUSE1);
   	switch ((fuse1 & HSW_F1_EU_DIS_MASK) >> HSW_F1_EU_DIS_SHIFT) {

default:

drivers/gpu/drm/i915/intel_device_info.c +0 -0
@@ -560,9 +579,14 @@ static void haswell_sseu_info_init(struct
 drm_i915_private *dev_priv)
   		sseu->eu_per_subslice = 6;
   		break;
Quoted 8 lines

drivers/gpu/drm/i915/intel_device_info.c static void haswell_sseu_info_init(struct drm_i915_private *dev_priv) · line 585

Quoted 6 lines
drivers/gpu/drm/i915/intel_device_info.c +4 -1
   	}
-	sseu->max_eus_per_subslice = sseu->eu_per_subslice;
+
+	intel_sseu_set_info(sseu, hweight8(sseu->slice_mask),
+			    hweight8(subslice_mask),
+			    sseu->eu_per_subslice);

Personal preference: could use a local variable for eu_per_subslice above to avoid setting it to itself here.

Quoted 15 lines

Yeah this is a bit ugly... I'll change it.

Thanks for the feedback!
Stuart

Quoted 10 lines

Daniele

Quoted 8 lines
drivers/gpu/drm/i915/intel_device_info.c +2 -0
   	for (s = 0; s < sseu->max_slices; s++) {
+		intel_sseu_set_subslices(sseu, s, subslice_mask);
+
   		for (ss = 0; ss < sseu->max_subslices; ss++) {
   			intel_sseu_set_eus(sseu, s, ss,
   					   (1UL << sseu-
Quoted 1 line

eu_per_subslice) - 1);

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx