Skip to contentSkip to conversation

Fix for SKL partial EU enablement

PATCH In review [email protected] 3 participants 6 messages Feb 13, 2015 – Feb 23, 2015

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

cover Fix for SKL partial EU enablement

From: Jeff McGee <[email protected]>

The exit from SKL render power gating may not fully restore slice and EU components. We have to explicitly restore them to full enablement through the Render Power Clock State register.

Jeff McGee (3):
drm/i915/skl: Determine SKL slice/subslice/EU info drm/i915/skl: Add SKL HW status to SSEU status drm/i915: Request full SSEU enablement on Gen9

 drivers/gpu/drm/i915/i915_debugfs.c | 80 +++++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_dma.c     | 73 +++++++++++++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_drv.h     | 11 ++++-
 drivers/gpu/drm/i915/i915_reg.h     | 44 ++++++++++++++++++++
 drivers/gpu/drm/i915/intel_lrc.c    | 47 +++++++++++++++++++++-
 5 files changed, 252 insertions(+), 3 deletions(-)
Signature · 6 lines
2.2.2

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

PATCH 1/3 drm/i915/skl: Determine SKL slice/subslice/EU info

From: Jeff McGee <[email protected]>

Read fuse registers to determine the available slice total, subslice total, subslice per slice, EU total, and EU per subslice counts of the SKL device. The EU per subslice attribute is more precisely defined as the maximum EU available on any one subslice, since available EU counts may vary across subslices due to fusing. Set flags indicating the SKL device's slice/subslice/EU (SSEU) power gating capability. Make all values available via debugfs entry 'i915_sseu_status'.

v2: Several small clean-ups suggested by Damien. Most notably,
used smaller types for the new device info fields to reduce memory usage and improved the clarity/readability of the method used to extract attribute values from the fuse registers.

---
 drivers/gpu/drm/i915/i915_debugfs.c | 30 +++++++++++++++
 drivers/gpu/drm/i915/i915_dma.c     | 73 +++++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_drv.h     | 11 +++++-
 drivers/gpu/drm/i915/i915_reg.h     | 11 ++++++
 4 files changed, 124 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/i915_debugfs.c +30 -0
index 164fa82..255054c 100644
@@ -4370,6 +4370,35 @@ DEFINE_SIMPLE_ATTRIBUTE(i915_cache_sharing_fops,
 			i915_cache_sharing_get, i915_cache_sharing_set,
 			"%llu\n");
 
+static int i915_sseu_status(struct seq_file *m, void *unused)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_device *dev = node->minor->dev;
+
+	if (INTEL_INFO(dev)->gen < 9)
+		return -ENODEV;
+
+	seq_puts(m, "SSEU Device Info\n");
+	seq_printf(m, "  Available Slice Total: %u\n",
+		   INTEL_INFO(dev)->slice_total);
+	seq_printf(m, "  Available Subslice Total: %u\n",
+		   INTEL_INFO(dev)->subslice_total);
+	seq_printf(m, "  Available Subslice Per Slice: %u\n",
+		   INTEL_INFO(dev)->subslice_per_slice);
+	seq_printf(m, "  Available EU Total: %u\n",
+		   INTEL_INFO(dev)->eu_total);
+	seq_printf(m, "  Available EU Per Subslice: %u\n",
+		   INTEL_INFO(dev)->eu_per_subslice);
+	seq_printf(m, "  Has Slice Power Gating: %s\n",
+		   yesno(INTEL_INFO(dev)->has_slice_pg));
+	seq_printf(m, "  Has Subslice Power Gating: %s\n",
+		   yesno(INTEL_INFO(dev)->has_subslice_pg));
+	seq_printf(m, "  Has EU Power Gating: %s\n",
+		   yesno(INTEL_INFO(dev)->has_eu_pg));
+
+	return 0;
+}
+
 static int i915_forcewake_open(struct inode *inode, struct file *file)
 {
 	struct drm_device *dev = inode->i_private;
@@ -4483,6 +4512,7 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_dp_mst_info", i915_dp_mst_info, 0},
 	{"i915_wa_registers", i915_wa_registers, 0},
 	{"i915_ddb_info", i915_ddb_info, 0},
+	{"i915_sseu_status", i915_sseu_status, 0},
 };
 #define I915_DEBUGFS_ENTRIES ARRAY_SIZE(i915_debugfs_list)
 
drivers/gpu/drm/i915/i915_dma.c +73 -0

86 lines not on this page · raw message

drivers/gpu/drm/i915/i915_drv.h +10 -1
index 2dedd43..de07474 100644
@@ -693,7 +693,16 @@ struct intel_device_info {
 	int trans_offsets[I915_MAX_TRANSCODERS];
 	int palette_offsets[I915_MAX_PIPES];
 	int cursor_offsets[I915_MAX_PIPES];
-	unsigned int eu_total;
+
+	/* Slice/subslice/EU info */
+	u8 slice_total;
+	u8 subslice_total;
+	u8 subslice_per_slice;
+	u8 eu_total;
+	u8 eu_per_subslice;
+	u8 has_slice_pg:1;
+	u8 has_subslice_pg:1;
+	u8 has_eu_pg:1;
 };
 
 #undef DEFINE_FLAG
drivers/gpu/drm/i915/i915_reg.h +11 -0
index 1dc91de..15914d7 100644
@@ -1506,6 +1506,17 @@ enum skl_disp_power_wells {
 #define   CHV_FGT_EU_DIS_SS1_R1_SHIFT	28
 #define   CHV_FGT_EU_DIS_SS1_R1_MASK	(0xf << CHV_FGT_EU_DIS_SS1_R1_SHIFT)
 
+#define GEN8_FUSE2			0x9120
+#define   GEN8_F2_S_ENA_SHIFT		25
+#define   GEN8_F2_S_ENA_MASK		(0x7 << GEN8_F2_S_ENA_SHIFT)
+
+#define   GEN9_F2_SS_DIS_SHIFT		20
+#define   GEN9_F2_SS_DIS_MASK		(0xf << GEN9_F2_SS_DIS_SHIFT)
+
+#define GEN8_EU_DISABLE0		0x9134
+#define GEN8_EU_DISABLE1		0x9138
+#define GEN8_EU_DISABLE2		0x913c
+
 #define GEN6_BSD_SLEEP_PSMI_CONTROL	0x12050
 #define   GEN6_BSD_SLEEP_MSG_DISABLE	(1 << 0)
 #define   GEN6_BSD_SLEEP_FLUSH_DISABLE	(1 << 2)
Signature · 6 lines
2.2.2

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

PATCH 2/3 drm/i915/skl: Add SKL HW status to SSEU status

In reply to [email protected]

From: Jeff McGee <[email protected]>

Add a new section to the 'i915_sseu_status' debugfs entry to report the currently enabled counts of slice, subslice, and execution units on the device. The count of enabled subslice per slice represents the most enabled subslice on any one slice for devices where imbalances may exist. Similarly, the count of enabled EU per subslice represents the most enabled EU on any one subslice.

Collect this device status for Skylake by reading the Gen9 power gate control ack message registers. Power gate control operates on EU in pairs, therefore our reported counts of enabled EU can be overestimated by one for each pair in which one EU is fused-off.

---
 drivers/gpu/drm/i915/i915_debugfs.c | 50 +++++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_reg.h     | 20 +++++++++++++++
 2 files changed, 70 insertions(+)
drivers/gpu/drm/i915/i915_debugfs.c +50 -0
index 255054c..e930a31d 100644
@@ -4374,6 +4374,8 @@ static int i915_sseu_status(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned int s_tot = 0, ss_tot = 0, ss_per = 0, eu_tot = 0, eu_per = 0;
 
 	if (INTEL_INFO(dev)->gen < 9)
 		return -ENODEV;
@@ -4396,6 +4398,54 @@ static int i915_sseu_status(struct seq_file *m, void *unused)
 	seq_printf(m, "  Has EU Power Gating: %s\n",
 		   yesno(INTEL_INFO(dev)->has_eu_pg));
 
+	seq_puts(m, "SSEU Device Status\n");
+	if (IS_SKYLAKE(dev)) {
+		const int s_max = 3, ss_max = 4;
+		int s, ss;
+		u32 s_reg[s_max], eu_reg[2*s_max], eu_mask[2];
+
+		s_reg[0] = I915_READ(GEN9_SLICE0_PGCTL_ACK);
+		s_reg[1] = I915_READ(GEN9_SLICE1_PGCTL_ACK);
+		s_reg[2] = I915_READ(GEN9_SLICE2_PGCTL_ACK);
+		eu_reg[0] = I915_READ(GEN9_SLICE0_SS01_EU_PGCTL_ACK);
+		eu_reg[1] = I915_READ(GEN9_SLICE0_SS23_EU_PGCTL_ACK);
+		eu_reg[2] = I915_READ(GEN9_SLICE1_SS01_EU_PGCTL_ACK);
+		eu_reg[3] = I915_READ(GEN9_SLICE1_SS23_EU_PGCTL_ACK);
+		eu_reg[4] = I915_READ(GEN9_SLICE2_SS01_EU_PGCTL_ACK);
+		eu_reg[5] = I915_READ(GEN9_SLICE2_SS23_EU_PGCTL_ACK);
+		eu_mask[0] = GEN9_PGCTL_SSA_EU08_ACK |
+			     GEN9_PGCTL_SSA_EU19_ACK |
+			     GEN9_PGCTL_SSA_EU210_ACK |
+			     GEN9_PGCTL_SSA_EU311_ACK;
+		eu_mask[1] = GEN9_PGCTL_SSB_EU08_ACK |
+			     GEN9_PGCTL_SSB_EU19_ACK |
+			     GEN9_PGCTL_SSB_EU210_ACK |
+			     GEN9_PGCTL_SSB_EU311_ACK;
+
+		for (s = 0; s < s_max; s++) {
+			if ((s_reg[s] & GEN9_PGCTL_SLICE_ACK) == 0)
+				/* skip disabled slice */
+				continue;
+
+			s_tot++;
+			ss_per = INTEL_INFO(dev)->subslice_per_slice;
+			ss_tot += ss_per;
+			for (ss = 0; ss < ss_max; ss++) {
+				unsigned int eu_cnt;
+
+				eu_cnt = 2 * hweight32(eu_reg[2*s + ss/2] &
+						       eu_mask[ss%2]);
+				eu_tot += eu_cnt;
+				eu_per = max(eu_per, eu_cnt);
+			}
+		}
+	}
+	seq_printf(m, "  Enabled Slice Total: %u\n", s_tot);
+	seq_printf(m, "  Enabled Subslice Total: %u\n", ss_tot);
+	seq_printf(m, "  Enabled Subslice Per Slice: %u\n", ss_per);
+	seq_printf(m, "  Enabled EU Total: %u\n", eu_tot);
+	seq_printf(m, "  Enabled EU Per Subslice: %u\n", eu_per);
+
 	return 0;
 }
 
drivers/gpu/drm/i915/i915_reg.h +20 -0
index 15914d7..0c9d1a8 100644
@@ -6198,6 +6198,26 @@ enum skl_disp_power_wells {
 #define   GEN6_RC6			3
 #define   GEN6_RC7			4
 
+#define GEN9_SLICE0_PGCTL_ACK		0x804c
+#define GEN9_SLICE1_PGCTL_ACK		0x8050
+#define GEN9_SLICE2_PGCTL_ACK		0x8054
+#define   GEN9_PGCTL_SLICE_ACK		(1 << 0)
+
+#define GEN9_SLICE0_SS01_EU_PGCTL_ACK	0x805c
+#define GEN9_SLICE0_SS23_EU_PGCTL_ACK	0x8060
+#define GEN9_SLICE1_SS01_EU_PGCTL_ACK	0x8064
+#define GEN9_SLICE1_SS23_EU_PGCTL_ACK	0x8068
+#define GEN9_SLICE2_SS01_EU_PGCTL_ACK	0x806c
+#define GEN9_SLICE2_SS23_EU_PGCTL_ACK	0x8070
+#define   GEN9_PGCTL_SSA_EU08_ACK	(1 << 0)
+#define   GEN9_PGCTL_SSA_EU19_ACK	(1 << 2)
+#define   GEN9_PGCTL_SSA_EU210_ACK	(1 << 4)
+#define   GEN9_PGCTL_SSA_EU311_ACK	(1 << 6)
+#define   GEN9_PGCTL_SSB_EU08_ACK	(1 << 8)
+#define   GEN9_PGCTL_SSB_EU19_ACK	(1 << 10)
+#define   GEN9_PGCTL_SSB_EU210_ACK	(1 << 12)
+#define   GEN9_PGCTL_SSB_EU311_ACK	(1 << 14)
+
 #define GEN7_MISCCPCTL			(0x9424)
 #define   GEN7_DOP_CLOCK_GATE_ENABLE	(1<<0)
 
Signature · 6 lines
2.2.2

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

PATCH 3/3 drm/i915: Request full SSEU enablement on Gen9

In reply to [email protected]

From: Jeff McGee <[email protected]>

On Gen9 the render power gating can leave slice/subslice/EU in a partially enabled state. We must make an explicit request for full SSEU enablement through the Render Power Clock State register when resuming render work. This register is save/ restored in the logical ring context image for execlist submission mode. Initialize its value in each LRC image to request full enablement according to the device SSEU config.

Thanks to Sharma Ankitprasad and Akash Goel for highlighting the issue and proposing the initial fix on which this patch is based.

v2: Adjusted the names of the power gating support flags to fit
update of an earlier patch.

---
 drivers/gpu/drm/i915/i915_reg.h  | 13 +++++++++++
 drivers/gpu/drm/i915/intel_lrc.c | 47 ++++++++++++++++++++++++++++++++++++++--
 2 files changed, 58 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/i915_reg.h +13 -0
index 0c9d1a8..3bba101 100644
@@ -139,6 +139,19 @@
 #define GEN8_RING_PDP_UDW(ring, n)	((ring)->mmio_base+0x270 + ((n) * 8 + 4))
 #define GEN8_RING_PDP_LDW(ring, n)	((ring)->mmio_base+0x270 + (n) * 8)
 
+#define GEN8_R_PWR_CLK_STATE		0x20C8
+#define   GEN8_RPCS_ENABLE		(1 << 31)
+#define   GEN8_RPCS_S_CNT_ENABLE	(1 << 18)
+#define   GEN8_RPCS_S_CNT_SHIFT		15
+#define   GEN8_RPCS_S_CNT_MASK		(0x7 << GEN8_RPCS_S_CNT_SHIFT)
+#define   GEN8_RPCS_SS_CNT_ENABLE	(1 << 11)
+#define   GEN8_RPCS_SS_CNT_SHIFT	8
+#define   GEN8_RPCS_SS_CNT_MASK		(0x7 << GEN8_RPCS_SS_CNT_SHIFT)
+#define   GEN8_RPCS_EU_MAX_SHIFT	4
+#define   GEN8_RPCS_EU_MAX_MASK		(0xf << GEN8_RPCS_EU_MAX_SHIFT)
+#define   GEN8_RPCS_EU_MIN_SHIFT	0
+#define   GEN8_RPCS_EU_MIN_MASK		(0xf << GEN8_RPCS_EU_MIN_SHIFT)
+
 #define GAM_ECOCHK			0x4090
 #define   BDW_DISABLE_HDC_INVALIDATION	(1<<25)
 #define   ECOCHK_SNB_BIT		(1<<10)
drivers/gpu/drm/i915/intel_lrc.c +45 -2

59 lines not on this page · raw message

Signature · 6 lines
2.2.2

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Tested-by on v1 3/3
#4 lore raw

Re: [PATCH 3/3] drm/i915: Request full SSEU enablement on Gen9

Task id: 5774 -------------------------------------Summary-------------------------------------
Platform Delta drm-intel-nightly Series Applied
PNV -1 281/281 280/281

ILK                                  313/313              313/313
SNB                                  309/309              309/309
IVB                                  382/382              382/382
BYT                                  296/296              296/296
HSW                                  426/426              426/426

BDW -1 318/318 317/318
-------------------------------------Detailed-------------------------------------
Platform Test drm-intel-nightly Series Applied
*PNV igt_gen3_render_mixed_blits PASS(1) CRASH(1)
*BDW igt_gem_gtt_hog PASS(1) DMESG_WARN(1)
Note: You need to pay more attention to line start with '*'
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#5 lore raw
In reply to [email protected]
Quoted 10 lines · On Fri, Feb 13, 2015 at 10:27:53AM -0600, [email protected] wrote:

From: Jeff McGee <[email protected]>

The exit from SKL render power gating may not fully restore slice and EU components. We have to explicitly restore them to full enablement through the Render Power Clock State register.

Jeff McGee (3):
drm/i915/skl: Determine SKL slice/subslice/EU info drm/i915/skl: Add SKL HW status to SSEU status drm/i915: Request full SSEU enablement on Gen9

All merged to dinq, thanks.
-Daniel

Quoted 15 lines
 drivers/gpu/drm/i915/i915_debugfs.c | 80 +++++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_dma.c     | 73 +++++++++++++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_drv.h     | 11 ++++-
 drivers/gpu/drm/i915/i915_reg.h     | 44 ++++++++++++++++++++
 drivers/gpu/drm/i915/intel_lrc.c    | 47 +++++++++++++++++++++-
 5 files changed, 252 insertions(+), 3 deletions(-)
Signature · 6 lines
2.2.2

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 7 lines
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx