Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Matt Roper <matthew.d.roper@intel.com>
To: intel-gfx@lists.freedesktop.org
Cc: Lucas De Marchi <lucas.demarchi@intel.com>,
	Matthew Auld <matthew.auld@intel.com>
Subject: [Intel-gfx] [PATCH v2 16/50] drm/i915/xehp: Changes to ss/eu definitions
Date: Tue, 13 Jul 2021 20:15:06 -0700	[thread overview]
Message-ID: <20210714031540.3539704-17-matthew.d.roper@intel.com> (raw)
In-Reply-To: <20210714031540.3539704-1-matthew.d.roper@intel.com>

From: Matthew Auld <matthew.auld@intel.com>

Xe_HP no longer has "slices" in the same way that old platforms did.
There are new concepts (gslices, cslices, mslices) that apply in various
contexts, but for the purposes of fusing slices no longer exist and we
just have one large pool of dual-subslices (DSS) to work with.
Furthermore, the meaning of the DSS fuse is inverted compared to past
platforms --- it now specifies which DSS are enabled rather than which
ones are disabled.

Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Prasad Nallani <prasad.nallani@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_sseu.c | 24 ++++++++++++++++++++----
 drivers/gpu/drm/i915/i915_getparam.c |  6 ++++--
 drivers/gpu/drm/i915/i915_reg.h      |  3 +++
 3 files changed, 27 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_sseu.c b/drivers/gpu/drm/i915/gt/intel_sseu.c
index bbed8e8625e1..5d1b7d06c96b 100644
--- a/drivers/gpu/drm/i915/gt/intel_sseu.c
+++ b/drivers/gpu/drm/i915/gt/intel_sseu.c
@@ -139,17 +139,33 @@ static void gen12_sseu_info_init(struct intel_gt *gt)
 	 * Gen12 has Dual-Subslices, which behave similarly to 2 gen11 SS.
 	 * Instead of splitting these, provide userspace with an array
 	 * of DSS to more closely represent the hardware resource.
+	 *
+	 * In addition, the concept of slice has been removed in Xe_HP.
+	 * To be compatible with prior generations, assume a single slice
+	 * across the entire device. Then calculate out the DSS for each
+	 * workload type within that software slice.
 	 */
 	intel_sseu_set_info(sseu, 1, 6, 16);
 
-	s_en = intel_uncore_read(uncore, GEN11_GT_SLICE_ENABLE) &
-		GEN11_GT_S_ENA_MASK;
+	/*
+	 * As mentioned above, Xe_HP does not have the concept of a slice.
+	 * Enable one for software backwards compatibility.
+	 */
+	if (GRAPHICS_VER_FULL(gt->i915) >= IP_VER(12, 50))
+		s_en = 0x1;
+	else
+		s_en = intel_uncore_read(uncore, GEN11_GT_SLICE_ENABLE) &
+		       GEN11_GT_S_ENA_MASK;
 
 	dss_en = intel_uncore_read(uncore, GEN12_GT_DSS_ENABLE);
 
 	/* one bit per pair of EUs */
-	eu_en_fuse = ~(intel_uncore_read(uncore, GEN11_EU_DISABLE) &
-		       GEN11_EU_DIS_MASK);
+	if (GRAPHICS_VER_FULL(gt->i915) >= IP_VER(12, 50))
+		eu_en_fuse = intel_uncore_read(uncore, XEHP_EU_ENABLE) & XEHP_EU_ENA_MASK;
+	else
+		eu_en_fuse = ~(intel_uncore_read(uncore, GEN11_EU_DISABLE) &
+			       GEN11_EU_DIS_MASK);
+
 	for (eu = 0; eu < sseu->max_eus_per_subslice / 2; eu++)
 		if (eu_en_fuse & BIT(eu))
 			eu_en |= BIT(eu * 2) | BIT(eu * 2 + 1);
diff --git a/drivers/gpu/drm/i915/i915_getparam.c b/drivers/gpu/drm/i915/i915_getparam.c
index 24e18219eb50..e289397d9178 100644
--- a/drivers/gpu/drm/i915/i915_getparam.c
+++ b/drivers/gpu/drm/i915/i915_getparam.c
@@ -15,7 +15,7 @@ int i915_getparam_ioctl(struct drm_device *dev, void *data,
 	struct pci_dev *pdev = to_pci_dev(dev->dev);
 	const struct sseu_dev_info *sseu = &i915->gt.info.sseu;
 	drm_i915_getparam_t *param = data;
-	int value;
+	int value = 0;
 
 	switch (param->param) {
 	case I915_PARAM_IRQ_ACTIVE:
@@ -150,7 +150,9 @@ 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 slice */
+		memcpy(&value, sseu->subslice_mask,
+		       min(sseu->ss_stride, (u8)sizeof(value)));
 		if (!value)
 			return -ENODEV;
 		break;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 47f015976966..4cabafa1c836 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -3151,6 +3151,9 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 
 #define GEN12_GT_DSS_ENABLE _MMIO(0x913C)
 
+#define XEHP_EU_ENABLE			_MMIO(0x9134)
+#define XEHP_EU_ENA_MASK		0xFF
+
 #define GEN6_BSD_SLEEP_PSMI_CONTROL	_MMIO(0x12050)
 #define   GEN6_BSD_SLEEP_MSG_DISABLE	(1 << 0)
 #define   GEN6_BSD_SLEEP_FLUSH_DISABLE	(1 << 2)
-- 
2.25.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  parent reply	other threads:[~2021-07-14  3:16 UTC|newest]

Thread overview: 88+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-14  3:14 [Intel-gfx] [PATCH v2 00/50] Begin enabling Xe_HP SDV and DG2 platforms Matt Roper
2021-07-14  3:14 ` [Intel-gfx] [PATCH v2 01/50] drm/i915: Add XE_HP initial definitions Matt Roper
2021-07-19 18:05   ` Souza, Jose
2021-07-14  3:14 ` [Intel-gfx] [PATCH v2 02/50] drm/i915: Fork DG1 interrupt handler Matt Roper
2021-07-19 21:18   ` Matt Atwood
2021-07-14  3:14 ` [Intel-gfx] [PATCH v2 03/50] drm/i915/xehp: VDBOX/VEBOX fusing registers are enable-based Matt Roper
2021-07-19 21:20   ` Matt Atwood
2021-07-14  3:14 ` [Intel-gfx] [PATCH v2 04/50] drm/i915/gen12: Use fuse info to enable SFC Matt Roper
2021-07-14  3:14 ` [Intel-gfx] [PATCH v2 05/50] drm/i915/selftests: Allow for larger engine counts Matt Roper
2021-07-14  3:14 ` [Intel-gfx] [PATCH v2 06/50] drm/i915/xehp: Extra media engines - Part 1 (engine definitions) Matt Roper
2021-07-20 23:03   ` Lucas De Marchi
2021-07-20 23:40     ` John Harrison
2021-07-20 23:49       ` Lucas De Marchi
2021-07-21 18:23   ` Lucas De Marchi
2021-07-14  3:14 ` [Intel-gfx] [PATCH v2 07/50] drm/i915/xehp: Extra media engines - Part 2 (interrupts) Matt Roper
2021-07-14  3:14 ` [Intel-gfx] [PATCH v2 08/50] drm/i915/xehp: Extra media engines - Part 3 (reset) Matt Roper
2021-07-19 18:13   ` Souza, Jose
2021-07-20 20:39   ` Matt Atwood
2021-07-14  3:14 ` [Intel-gfx] [PATCH v2 09/50] drm/i915/xehp: Xe_HP forcewake support Matt Roper
2021-07-20 20:57   ` Matt Atwood
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 10/50] drm/i915/xehp: Define multicast register ranges Matt Roper
2021-07-19 18:19   ` Souza, Jose
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 11/50] drm/i915/xehp: Handle new device context ID format Matt Roper
2021-07-20 21:39   ` Matt Atwood
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 12/50] drm/i915/xehp: New engine context offsets Matt Roper
2021-07-20 22:06   ` Matt Atwood
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 13/50] drm/i915/xehp: handle new steering options Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 14/50] drm/i915/xehp: Loop over all gslices for INSTDONE processing Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 15/50] drm/i915/xehpsdv: add initial XeHP SDV definitions Matt Roper
2021-07-19 18:20   ` Souza, Jose
2021-07-14  3:15 ` Matt Roper [this message]
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 17/50] drm/i915/xehpsdv: Add maximum sseu limits Matt Roper
2021-07-18 13:10   ` Yokoyama, Caz
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 18/50] drm/i915/xehpsdv: Add compute DSS type Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 19/50] drm/i915/xehpsdv: Define steering tables Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 20/50] drm/i915/xehpsdv: Define MOCS table for XeHP SDV Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 21/50] drm/i915/xehpsdv: factor out function to read RP_STATE_CAP Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 22/50] drm/i915/xehpsdv: Read correct RP_STATE_CAP register Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 23/50] drm/i915/dg2: add DG2 platform info Matt Roper
2021-07-19 18:21   ` Souza, Jose
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 24/50] drm/i915/dg2: DG2 uses the same sseu limits as XeHP SDV Matt Roper
2021-07-16 16:06   ` Yokoyama, Caz
2021-07-19 18:22   ` Souza, Jose
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 25/50] drm/i915/dg2: Add forcewake table Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 26/50] drm/i915/dg2: Update LNCF steering ranges Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 27/50] drm/i915/dg2: Add SQIDI steering Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 28/50] drm/i915/dg2: Add new LRI reg offsets Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 29/50] drm/i915/dg2: Maintain backward-compatible nested batch behavior Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 30/50] drm/i915/dg2: Report INSTDONE_GEOM values in error state Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 31/50] drm/i915/dg2: Define MOCS table for DG2 Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 32/50] drm/i915/dg2: Add fake PCH Matt Roper
2021-07-16 19:52   ` Souza, Jose
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 33/50] drm/i915/dg2: Add cdclk table and reference clock Matt Roper
2021-07-16 19:36   ` Souza, Jose
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 34/50] drm/i915/dg2: Skip shared DPLL handling Matt Roper
2021-07-16 19:38   ` Souza, Jose
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 35/50] drm/i915/dg2: Don't wait for AUX power well enable ACKs Matt Roper
2021-07-16 19:40   ` Souza, Jose
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 36/50] drm/i915/dg2: Setup display outputs Matt Roper
2021-07-16 19:40   ` Souza, Jose
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 37/50] drm/i915/dg2: Add dbuf programming Matt Roper
2021-07-16 19:45   ` Souza, Jose
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 38/50] drm/i915/dg2: Don't program BW_BUDDY registers Matt Roper
2021-07-16 19:47   ` Souza, Jose
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 39/50] drm/i915/dg2: Don't read DRAM info Matt Roper
2021-07-15 17:17   ` Srivatsa, Anusha
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 40/50] drm/i915/dg2: DG2 has fixed memory bandwidth Matt Roper
2021-07-21 17:42   ` Srivatsa, Anusha
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 41/50] drm/i915/dg2: Add MPLLB programming for SNPS PHY Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 42/50] drm/i915/dg2: Add MPLLB programming for HDMI Matt Roper
2021-07-16 21:13   ` Matt Atwood
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 43/50] drm/i915/dg2: Add vswing programming for SNPS phys Matt Roper
2021-07-16 21:38   ` Matt Atwood
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 44/50] drm/i915/dg2: Update modeset sequences Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 45/50] drm/i915/dg2: Classify DG2 PHY types Matt Roper
2021-07-16 20:50   ` Matt Atwood
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 46/50] drm/i915/dg2: Wait for SNPS PHY calibration during display init Matt Roper
2021-07-16 20:52   ` Matt Atwood
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 47/50] drm/i915/dg2: Update lane disable power state during PSR Matt Roper
2021-07-21 17:55   ` Srivatsa, Anusha
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 48/50] drm/i915/dg2: Add DG2 to the PSR2 defeature list Matt Roper
2021-07-21 17:56   ` Srivatsa, Anusha
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 49/50] drm/i915/dg2: Update to bigjoiner path Matt Roper
2021-07-14  3:15 ` [Intel-gfx] [PATCH v2 50/50] drm/i915/dg2: Configure PCON in DP pre-enable path Matt Roper
2021-07-14  4:23 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for Begin enabling Xe_HP SDV and DG2 platforms (rev5) Patchwork
2021-07-14  4:24 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2021-07-14  4:51 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2021-07-14 15:12 ` [Intel-gfx] ✓ Fi.CI.IGT: " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210714031540.3539704-17-matthew.d.roper@intel.com \
    --to=matthew.d.roper@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=lucas.demarchi@intel.com \
    --cc=matthew.auld@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox