Skip to contentSkip to conversation

drm/i915/guc: Split GuC firmware xfer function into clear steps

PATCH Superseded Michal Wajdeczko 4 participants 10 messages Nov 3, 2017

Subsystems (2)

v2 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/4 drm/i915/guc: Split GuC firmware xfer function into clear steps

Transfer of GuC firmware requires few steps that currently are implemented in two large functions. Split this code into smaller functions to make these steps small and clear. Also be prepared for potential DMA xfer step failure.

v2: rename function steps (Sagar)

Cc Chris Wilson <[email protected]>
Cc Joonas Lahtinen <[email protected]>
Cc Sagar Arun Kamble <[email protected]>
---
 drivers/gpu/drm/i915/intel_guc_fw.c | 172 +++++++++++++++++++++---------------
 1 file changed, 103 insertions(+), 69 deletions(-)
drivers/gpu/drm/i915/intel_guc_fw.c +103 -69
index ef67a36..c4f4526 100644
@@ -97,23 +97,55 @@ int intel_guc_fw_select(struct intel_guc *guc)
 	return 0;
 }
 
-/*
- * Read the GuC status register (GUC_STATUS) and store it in the
- * specified location; then return a boolean indicating whether
- * the value matches either of two values representing completion
- * of the GuC boot process.
- *
- * This is used for polling the GuC status in a wait_for()
- * loop below.
- */
-static inline bool guc_ucode_response(struct drm_i915_private *dev_priv,
-				      u32 *status)
+static void guc_prepare_xfer(struct intel_guc *guc)
 {
-	u32 val = I915_READ(GUC_STATUS);
-	u32 uk_val = val & GS_UKERNEL_MASK;
-	*status = val;
-	return (uk_val == GS_UKERNEL_READY ||
-		((val & GS_MIA_CORE_STATE) && uk_val == GS_UKERNEL_LAPIC_DONE));
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+
+	/* Enable MIA caching. GuC clock gating is disabled. */
+	I915_WRITE(GUC_SHIM_CONTROL, GUC_SHIM_CONTROL_VALUE);
+
+	/* WaDisableMinuteIaClockGating:bxt */
+	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1)) {
+		I915_WRITE(GUC_SHIM_CONTROL, (I915_READ(GUC_SHIM_CONTROL) &
+					      ~GUC_ENABLE_MIA_CLOCK_GATING));
+	}
+
+	/* WaC6DisallowByGfxPause:bxt */
+	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_B0))
+		I915_WRITE(GEN6_GFXPAUSE, 0x30FFF);
+
+	if (IS_GEN9_LP(dev_priv))
+		I915_WRITE(GEN9LP_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
+	else
+		I915_WRITE(GEN9_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
+
+	if (IS_GEN9(dev_priv)) {
+		/* DOP Clock Gating Enable for GuC clocks */
+		I915_WRITE(GEN7_MISCCPCTL, (GEN8_DOP_CLOCK_GATE_GUC_ENABLE |
+					    I915_READ(GEN7_MISCCPCTL)));
+
+		/* allows for 5us (in 10ns units) before GT can go to RC6 */
+		I915_WRITE(GUC_ARAT_C6DIS, 0x1FF);
+	}
+}
+
+/* Copy RSA signature from the fw image to HW for verification */
+static int guc_xfer_rsa(struct intel_guc *guc, struct i915_vma *vma)
+{
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	struct intel_uc_fw *guc_fw = &guc->fw;
+	struct sg_table *sg = vma->pages;
+	u32 rsa[UOS_RSA_SCRATCH_MAX_COUNT];
+	int i;
+
+	if (sg_pcopy_to_buffer(sg->sgl, sg->nents, rsa, sizeof(rsa),
+			       guc_fw->rsa_offset) != sizeof(rsa))
+		return -EINVAL;
+
+	for (i = 0; i < UOS_RSA_SCRATCH_MAX_COUNT; i++)
+		I915_WRITE(UOS_RSA_SCRATCH(i), rsa[i]);
+
+	return 0;
 }
 
 /*
@@ -122,29 +154,17 @@ static inline bool guc_ucode_response(struct drm_i915_private *dev_priv,
  * Architecturally, the DMA engine is bidirectional, and can potentially even
  * transfer between GTT locations. This functionality is left out of the API
  * for now as there is no need for it.
- *
- * Note that GuC needs the CSS header plus uKernel code to be copied by the
- * DMA engine in one operation, whereas the RSA signature is loaded via MMIO.
  */
-static int guc_ucode_xfer_dma(struct drm_i915_private *dev_priv,
-			      struct i915_vma *vma)
+static int guc_xfer_ucode(struct intel_guc *guc, struct i915_vma *vma)
 {
-	struct intel_uc_fw *guc_fw = &dev_priv->guc.fw;
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	struct intel_uc_fw *guc_fw = &guc->fw;
 	unsigned long offset;
-	struct sg_table *sg = vma->pages;
-	u32 status, rsa[UOS_RSA_SCRATCH_MAX_COUNT];
-	int i, ret = 0;
-
-	/* where RSA signature starts */
-	offset = guc_fw->rsa_offset;
 
-	/* Copy RSA signature from the fw image to HW for verification */
-	sg_pcopy_to_buffer(sg->sgl, sg->nents, rsa, sizeof(rsa), offset);
-	for (i = 0; i < UOS_RSA_SCRATCH_MAX_COUNT; i++)
-		I915_WRITE(UOS_RSA_SCRATCH(i), rsa[i]);
-
-	/* The header plus uCode will be copied to WOPCM via DMA, excluding any
-	 * other components */
+	/*
+	 * The header plus uCode will be copied to WOPCM via DMA, excluding any
+	 * other components
+	 */
 	I915_WRITE(DMA_COPY_SIZE, guc_fw->header_size + guc_fw->ucode_size);
 
 	/* Set the source address for the new blob */
@@ -162,33 +182,57 @@ static int guc_ucode_xfer_dma(struct drm_i915_private *dev_priv,
 	/* Finally start the DMA */
 	I915_WRITE(DMA_CTRL, _MASKED_BIT_ENABLE(UOS_MOVE | START_DMA));
 
+	return 0;
+}
+
+/*
+ * Read the GuC status register (GUC_STATUS) and store it in the
+ * specified location; then return a boolean indicating whether
+ * the value matches either of two values representing completion
+ * of the GuC boot process.
+ *
+ * This is used for polling the GuC status in a wait_for()
+ * loop below.
+ */
+static inline bool guc_ready(struct intel_guc *guc, u32 *status)
+{
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	u32 val = I915_READ(GUC_STATUS);
+	u32 uk_val = val & GS_UKERNEL_MASK;
+
+	*status = val;
+	return (uk_val == GS_UKERNEL_READY) ||
+		((val & GS_MIA_CORE_STATE) && (uk_val == GS_UKERNEL_LAPIC_DONE));
+}
+
+static int guc_wait_ucode(struct intel_guc *guc)
+{
+	u32 status;
+	int ret;
+
 	/*
-	 * Wait for the DMA to complete & the GuC to start up.
+	 * Wait for the GuC to start up.
 	 * NB: Docs recommend not using the interrupt for completion.
 	 * Measurements indicate this should take no more than 20ms, so a
 	 * timeout here indicates that the GuC has failed and is unusable.
 	 * (Higher levels of the driver will attempt to fall back to
 	 * execlist mode if this happens.)
 	 */
-	ret = wait_for(guc_ucode_response(dev_priv, &status), 100);
-
-	DRM_DEBUG_DRIVER("DMA status 0x%x, GuC status 0x%x\n",
-			I915_READ(DMA_CTRL), status);
+	ret = wait_for(guc_ready(guc, &status), 100);
+	DRM_DEBUG_DRIVER("GuC status %#x\n", status);
 
 	if ((status & GS_BOOTROM_MASK) == GS_BOOTROM_RSA_FAILED) {
 		DRM_ERROR("GuC firmware signature verification failed\n");
 		ret = -ENOEXEC;
 	}
 
-	DRM_DEBUG_DRIVER("returning %d\n", ret);
-
 	return ret;
 }
 
 /*
  * Load the GuC firmware blob into the MinuteIA.
  */
-static int guc_ucode_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma)
+static int guc_fw_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma)
 {
 	struct intel_guc *guc = container_of(guc_fw, struct intel_guc, fw);
 	struct drm_i915_private *dev_priv = guc_to_i915(guc);
@@ -198,34 +242,24 @@ static int guc_ucode_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma)
 
 	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
 
-	/* Enable MIA caching. GuC clock gating is disabled. */
-	I915_WRITE(GUC_SHIM_CONTROL, GUC_SHIM_CONTROL_VALUE);
-
-	/* WaDisableMinuteIaClockGating:bxt */
-	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1)) {
-		I915_WRITE(GUC_SHIM_CONTROL, (I915_READ(GUC_SHIM_CONTROL) &
-					      ~GUC_ENABLE_MIA_CLOCK_GATING));
-	}
-
-	/* WaC6DisallowByGfxPause:bxt */
-	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_B0))
-		I915_WRITE(GEN6_GFXPAUSE, 0x30FFF);
+	guc_prepare_xfer(guc);
 
-	if (IS_GEN9_LP(dev_priv))
-		I915_WRITE(GEN9LP_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
-	else
-		I915_WRITE(GEN9_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
-
-	if (IS_GEN9(dev_priv)) {
-		/* DOP Clock Gating Enable for GuC clocks */
-		I915_WRITE(GEN7_MISCCPCTL, (GEN8_DOP_CLOCK_GATE_GUC_ENABLE |
-					    I915_READ(GEN7_MISCCPCTL)));
+	/*
+	 * Note that GuC needs the CSS header plus uKernel code to be copied
+	 * by the DMA engine in one operation, whereas the RSA signature is
+	 * loaded via MMIO.
+	 */
+	ret = guc_xfer_rsa(guc, vma);
+	if (ret)
+		DRM_WARN("GuC firmware signature xfer error %d\n", ret);
 
-		/* allows for 5us (in 10ns units) before GT can go to RC6 */
-		I915_WRITE(GUC_ARAT_C6DIS, 0x1FF);
-	}
+	ret = guc_xfer_ucode(guc, vma);
+	if (ret)
+		DRM_WARN("GuC firmware code xfer error %d\n", ret);
 
-	ret = guc_ucode_xfer_dma(dev_priv, vma);
+	ret = guc_wait_ucode(guc);
+	if (ret)
+		DRM_ERROR("GuC firmware xfer error %d\n", ret);
 
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 
@@ -247,5 +281,5 @@ static int guc_ucode_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma)
  */
 int intel_guc_fw_upload(struct intel_guc *guc)
 {
-	return intel_uc_fw_upload(&guc->fw, guc_ucode_xfer);
+	return intel_uc_fw_upload(&guc->fw, guc_fw_xfer);
 }
Signature · 6 lines
2.7.4

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

PATCH v2 2/4 drm/i915/guc: Wait for ucode DMA transfer completion

We silently assumed that DMA transfer will be completed within assumed timeout and thus we were waiting only at last step for GuC to become ready. Add intermediate wait to catch unexpected delays in DMA transfer.

Cc Chris Wilson <[email protected]>
Cc Joonas Lahtinen <[email protected]>
Cc Sagar Arun Kamble <[email protected]>
---
 drivers/gpu/drm/i915/intel_guc_fw.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/intel_guc_fw.c +8 -1
index c4f4526..74a61fe 100644
@@ -160,6 +160,8 @@ static int guc_xfer_ucode(struct intel_guc *guc, struct i915_vma *vma)
 	struct drm_i915_private *dev_priv = guc_to_i915(guc);
 	struct intel_uc_fw *guc_fw = &guc->fw;
 	unsigned long offset;
+	u32 status;
+	int ret;
 
 	/*
 	 * The header plus uCode will be copied to WOPCM via DMA, excluding any
@@ -182,7 +184,12 @@ static int guc_xfer_ucode(struct intel_guc *guc, struct i915_vma *vma)
 	/* Finally start the DMA */
 	I915_WRITE(DMA_CTRL, _MASKED_BIT_ENABLE(UOS_MOVE | START_DMA));
 
-	return 0;
+	/* Wait for DMA to finish */
+	ret = __intel_wait_for_register_fw(dev_priv, DMA_CTRL, START_DMA, 0,
+					   2, 100, &status);
+	DRM_DEBUG_DRIVER("GuC DMA status %#x\n", status);
+
+	return ret;
 }
 
 /*
Signature · 6 lines
2.7.4

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Reviewed-by on v2 2/4
#2 lore raw

Re: [PATCH v2 2/4] drm/i915/guc: Wait for ucode DMA transfer completion

Quoted 9 lines · Michal Wajdeczko

We silently assumed that DMA transfer will be completed within assumed timeout and thus we were waiting only at last step for GuC to become ready. Add intermediate wait to catch unexpected delays in DMA transfer.

Cc Chris Wilson <[email protected]>
Cc Joonas Lahtinen <[email protected]>
Cc Sagar Arun Kamble <[email protected]>
Quoted 21 lines
---
  drivers/gpu/drm/i915/intel_guc_fw.c | 9 ++++++++-
  1 file changed, 8 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/intel_guc_fw.c +2 -0
index c4f4526..74a61fe 100644
@@ -160,6 +160,8 @@ static int guc_xfer_ucode(struct intel_guc *guc, struct i915_vma *vma)
  	struct drm_i915_private *dev_priv = guc_to_i915(guc);
  	struct intel_uc_fw *guc_fw = &guc->fw;
  	unsigned long offset;
+	u32 status;
+	int ret;
 
  	/*
  	 * The header plus uCode will be copied to WOPCM via DMA, excluding any
@@ -182,7 +184,12 @@ static int guc_xfer_ucode(struct intel_guc *guc, struct i915_vma *vma)
  	/* Finally start the DMA */
  	I915_WRITE(DMA_CTRL, _MASKED_BIT_ENABLE(UOS_MOVE | START_DMA));
 

drivers/gpu/drm/i915/intel_guc_fw.c static int guc_xfer_ucode(struct intel_guc *guc, struct i915_vma *vma) · line 195

Quoted 10 lines
drivers/gpu/drm/i915/intel_guc_fw.c +6 -1
-	return 0;
+	/* Wait for DMA to finish */
+	ret = __intel_wait_for_register_fw(dev_priv, DMA_CTRL, START_DMA, 0,
+					   2, 100, &status);
+	DRM_DEBUG_DRIVER("GuC DMA status %#x\n", status);
+
+	return ret;
  }
 
  /*

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

#3 lore raw

PATCH v2 3/4 drm/i915/guc: Simplify programming of GUC_SHIM_CONTROL

In reply to Michal Wajdeczko

We unconditionally program GUC_SHIM_CONTROL register with all expected bits, but in case of required workaround we modify those bits by using additional READ/WRITE calls. Lets prepare correct value for this register prior to first register programming.

Cc Chris Wilson <[email protected]>
Cc Joonas Lahtinen <[email protected]>
Cc Sagar Arun Kamble <[email protected]>
---
 drivers/gpu/drm/i915/i915_guc_reg.h |  7 -------
 drivers/gpu/drm/i915/intel_guc_fw.c | 19 +++++++++++++------
 2 files changed, 13 insertions(+), 13 deletions(-)
drivers/gpu/drm/i915/i915_guc_reg.h +0 -7
index 35cf991..bc1ae7d 100644
@@ -102,13 +102,6 @@
 #define   GUC_ENABLE_MIA_CLOCK_GATING		(1<<15)
 #define   GUC_GEN10_SHIM_WC_ENABLE		(1<<21)
 
-#define GUC_SHIM_CONTROL_VALUE	(GUC_DISABLE_SRAM_INIT_TO_ZEROES	| \
-				 GUC_ENABLE_READ_CACHE_LOGIC		| \
-				 GUC_ENABLE_MIA_CACHING			| \
-				 GUC_ENABLE_READ_CACHE_FOR_SRAM_DATA	| \
-				 GUC_ENABLE_READ_CACHE_FOR_WOPCM_DATA	| \
-				 GUC_ENABLE_MIA_CLOCK_GATING)
-
 #define GUC_SEND_INTERRUPT		_MMIO(0xc4c8)
 #define   GUC_SEND_TRIGGER		  (1<<0)
 
drivers/gpu/drm/i915/intel_guc_fw.c +13 -6
index 74a61fe..3f013f7 100644
@@ -100,15 +100,22 @@ int intel_guc_fw_select(struct intel_guc *guc)
 static void guc_prepare_xfer(struct intel_guc *guc)
 {
 	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	u32 shim_control_value;
 
-	/* Enable MIA caching. GuC clock gating is disabled. */
-	I915_WRITE(GUC_SHIM_CONTROL, GUC_SHIM_CONTROL_VALUE);
+	/* Expected bits for normal operation */
+	shim_control_value = GUC_DISABLE_SRAM_INIT_TO_ZEROES |
+			     GUC_ENABLE_READ_CACHE_LOGIC |
+			     GUC_ENABLE_MIA_CACHING |
+			     GUC_ENABLE_READ_CACHE_FOR_SRAM_DATA |
+			     GUC_ENABLE_READ_CACHE_FOR_WOPCM_DATA |
+			     GUC_ENABLE_MIA_CLOCK_GATING;
 
 	/* WaDisableMinuteIaClockGating:bxt */
-	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1)) {
-		I915_WRITE(GUC_SHIM_CONTROL, (I915_READ(GUC_SHIM_CONTROL) &
-					      ~GUC_ENABLE_MIA_CLOCK_GATING));
-	}
+	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1))
+		shim_control_value &= ~GUC_ENABLE_MIA_CLOCK_GATING;
+
+	/* Must program this register before loading the ucode with DMA */
+	I915_WRITE(GUC_SHIM_CONTROL, shim_control_value);
 
 	/* WaC6DisallowByGfxPause:bxt */
 	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_B0))
Signature · 6 lines
2.7.4

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Reviewed-by on v2 3/4
#4 lore raw

Re: [PATCH v2 3/4] drm/i915/guc: Simplify programming of GUC_SHIM_CONTROL

Quoted 10 lines · Michal Wajdeczko

We unconditionally program GUC_SHIM_CONTROL register with all expected bits, but in case of required workaround we modify those bits by using additional READ/WRITE calls. Lets prepare correct value for this register prior to first register programming.

Cc Chris Wilson <[email protected]>
Cc Joonas Lahtinen <[email protected]>
Cc Sagar Arun Kamble <[email protected]>

As per bspec, WaDisableMinuteIaClockGating is applicable to SKL A0/B0 as well. Should fix for those platforms unless we dont want to carry early pre-production workarounds.

Quoted 46 lines
---
  drivers/gpu/drm/i915/i915_guc_reg.h |  7 -------
  drivers/gpu/drm/i915/intel_guc_fw.c | 19 +++++++++++++------
  2 files changed, 13 insertions(+), 13 deletions(-)
drivers/gpu/drm/i915/i915_guc_reg.h +0 -7
index 35cf991..bc1ae7d 100644
@@ -102,13 +102,6 @@
  #define   GUC_ENABLE_MIA_CLOCK_GATING		(1<<15)
  #define   GUC_GEN10_SHIM_WC_ENABLE		(1<<21)
 
-#define GUC_SHIM_CONTROL_VALUE	(GUC_DISABLE_SRAM_INIT_TO_ZEROES	| \
-				 GUC_ENABLE_READ_CACHE_LOGIC		| \
-				 GUC_ENABLE_MIA_CACHING			| \
-				 GUC_ENABLE_READ_CACHE_FOR_SRAM_DATA	| \
-				 GUC_ENABLE_READ_CACHE_FOR_WOPCM_DATA	| \
-				 GUC_ENABLE_MIA_CLOCK_GATING)
-
  #define GUC_SEND_INTERRUPT		_MMIO(0xc4c8)
  #define   GUC_SEND_TRIGGER		  (1<<0)
 
drivers/gpu/drm/i915/intel_guc_fw.c +8 -4
index 74a61fe..3f013f7 100644
@@ -100,15 +100,22 @@ int intel_guc_fw_select(struct intel_guc *guc)
  static void guc_prepare_xfer(struct intel_guc *guc)
  {
  	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	u32 shim_control_value;
 
-	/* Enable MIA caching. GuC clock gating is disabled. */
-	I915_WRITE(GUC_SHIM_CONTROL, GUC_SHIM_CONTROL_VALUE);
+	/* Expected bits for normal operation */
+	shim_control_value = GUC_DISABLE_SRAM_INIT_TO_ZEROES |
+			     GUC_ENABLE_READ_CACHE_LOGIC |
+			     GUC_ENABLE_MIA_CACHING |
+			     GUC_ENABLE_READ_CACHE_FOR_SRAM_DATA |
+			     GUC_ENABLE_READ_CACHE_FOR_WOPCM_DATA |
+			     GUC_ENABLE_MIA_CLOCK_GATING;
 
  	/* WaDisableMinuteIaClockGating:bxt */
-	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1)) {
-		I915_WRITE(GUC_SHIM_CONTROL, (I915_READ(GUC_SHIM_CONTROL) &

drivers/gpu/drm/i915/intel_guc_fw.c int intel_guc_fw_select(struct intel_guc *guc) · line 121

Quoted 10 lines
drivers/gpu/drm/i915/intel_guc_fw.c +5 -2
-					      ~GUC_ENABLE_MIA_CLOCK_GATING));
-	}
+	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1))
+		shim_control_value &= ~GUC_ENABLE_MIA_CLOCK_GATING;
+
+	/* Must program this register before loading the ucode with DMA */
+	I915_WRITE(GUC_SHIM_CONTROL, shim_control_value);
 
  	/* WaC6DisallowByGfxPause:bxt */
  	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_B0))

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

#5 lore raw

Re: [PATCH v2 3/4] drm/i915/guc: Simplify programming of GUC_SHIM_CONTROL

Quoted 22 lines · Michal Wajdeczko
Quoted 21 lines

We unconditionally program GUC_SHIM_CONTROL register with

drivers/gpu/drm/i915/intel_guc_fw.c +8 -2
index 74a61fe..3f013f7 100644
@@ -100,15 +100,22 @@ int intel_guc_fw_select(struct intel_guc *guc)
  static void guc_prepare_xfer(struct intel_guc *guc)
  {
      struct drm_i915_private *dev_priv = guc_to_i915(guc);
+     u32 shim_control_value;
 
-     /* Enable MIA caching. GuC clock gating is disabled. */
-     I915_WRITE(GUC_SHIM_CONTROL, GUC_SHIM_CONTROL_VALUE);
+     /* Expected bits for normal operation */
+     shim_control_value = GUC_DISABLE_SRAM_INIT_TO_ZEROES |
+                          GUC_ENABLE_READ_CACHE_LOGIC |
+                          GUC_ENABLE_MIA_CACHING |
+                          GUC_ENABLE_READ_CACHE_FOR_SRAM_DATA |
+                          GUC_ENABLE_READ_CACHE_FOR_WOPCM_DATA |
+                          GUC_ENABLE_MIA_CLOCK_GATING;
 
Quoted 8 lines

drivers/gpu/drm/i915/intel_guc_fw.c int intel_guc_fw_select(struct intel_guc *guc) · line 116

Quoted 8 lines
drivers/gpu/drm/i915/intel_guc_fw.c +3 -4
      /* WaDisableMinuteIaClockGating:bxt */
-     if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1)) {
-             I915_WRITE(GUC_SHIM_CONTROL, (I915_READ(GUC_SHIM_CONTROL) &
-                                           ~GUC_ENABLE_MIA_CLOCK_GATING));
-     }
+     if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1))
+             shim_control_value &= ~GUC_ENABLE_MIA_CLOCK_GATING;
+

Do you want to before changing this process, rip out the preproduction steps?
-Chris
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#6 lore raw

PATCH v2 4/4 HAX enable GuC submission for CI

In reply to Michal Wajdeczko

Also revert ("drm/i915/guc: Assert that we switch between known ggtt->invalidate functions"

---
 drivers/gpu/drm/i915/i915_gem_gtt.c | 8 ++------
 drivers/gpu/drm/i915/i915_params.h  | 4 ++--
 2 files changed, 4 insertions(+), 8 deletions(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +2 -6
index 0684d5d..a351ddf 100644
@@ -3562,17 +3562,13 @@ int i915_ggtt_enable_hw(struct drm_i915_private *dev_priv)
 
 void i915_ggtt_enable_guc(struct drm_i915_private *i915)
 {
-	GEM_BUG_ON(i915->ggtt.invalidate != gen6_ggtt_invalidate);
-
 	i915->ggtt.invalidate = guc_ggtt_invalidate;
 }
 
 void i915_ggtt_disable_guc(struct drm_i915_private *i915)
 {
-	/* We should only be called after i915_ggtt_enable_guc() */
-	GEM_BUG_ON(i915->ggtt.invalidate != guc_ggtt_invalidate);
-
-	i915->ggtt.invalidate = gen6_ggtt_invalidate;
+	if (i915->ggtt.invalidate == guc_ggtt_invalidate)
+		i915->ggtt.invalidate = gen6_ggtt_invalidate;
 }
 
 void i915_gem_restore_gtt_mappings(struct drm_i915_private *dev_priv)
drivers/gpu/drm/i915/i915_params.h +2 -2
index c729226..c38cef0 100644
@@ -44,8 +44,8 @@
 	param(int, disable_power_well, -1) \
 	param(int, enable_ips, 1) \
 	param(int, invert_brightness, 0) \
-	param(int, enable_guc_loading, 0) \
-	param(int, enable_guc_submission, 0) \
+	param(int, enable_guc_loading, 1) \
+	param(int, enable_guc_submission, 1) \
 	param(int, guc_log_level, -1) \
 	param(char *, guc_firmware_path, NULL) \
 	param(char *, huc_firmware_path, NULL) \
Signature · 6 lines
2.7.4

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Reviewed-by on v2 1/4
#7 lore raw
In reply to Michal Wajdeczko
Quoted 11 lines · Michal Wajdeczko

Transfer of GuC firmware requires few steps that currently are implemented in two large functions. Split this code into smaller functions to make these steps small and clear. Also be prepared for potential DMA xfer step failure.

v2: rename function steps (Sagar)

Cc Chris Wilson <[email protected]>
Cc Joonas Lahtinen <[email protected]>
Cc Sagar Arun Kamble <[email protected]>
Quoted 234 lines
---
  drivers/gpu/drm/i915/intel_guc_fw.c | 172 +++++++++++++++++++++---------------
  1 file changed, 103 insertions(+), 69 deletions(-)
drivers/gpu/drm/i915/intel_guc_fw.c +102 -68
index ef67a36..c4f4526 100644
@@ -97,23 +97,55 @@ int intel_guc_fw_select(struct intel_guc *guc)
  	return 0;
  }
 
-/*
- * Read the GuC status register (GUC_STATUS) and store it in the
- * specified location; then return a boolean indicating whether
- * the value matches either of two values representing completion
- * of the GuC boot process.
- *
- * This is used for polling the GuC status in a wait_for()
- * loop below.
- */
-static inline bool guc_ucode_response(struct drm_i915_private *dev_priv,
-				      u32 *status)
+static void guc_prepare_xfer(struct intel_guc *guc)
  {
-	u32 val = I915_READ(GUC_STATUS);
-	u32 uk_val = val & GS_UKERNEL_MASK;
-	*status = val;
-	return (uk_val == GS_UKERNEL_READY ||
-		((val & GS_MIA_CORE_STATE) && uk_val == GS_UKERNEL_LAPIC_DONE));
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+
+	/* Enable MIA caching. GuC clock gating is disabled. */
+	I915_WRITE(GUC_SHIM_CONTROL, GUC_SHIM_CONTROL_VALUE);
+
+	/* WaDisableMinuteIaClockGating:bxt */
+	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1)) {
+		I915_WRITE(GUC_SHIM_CONTROL, (I915_READ(GUC_SHIM_CONTROL) &
+					      ~GUC_ENABLE_MIA_CLOCK_GATING));
+	}
+
+	/* WaC6DisallowByGfxPause:bxt */
+	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_B0))
+		I915_WRITE(GEN6_GFXPAUSE, 0x30FFF);
+
+	if (IS_GEN9_LP(dev_priv))
+		I915_WRITE(GEN9LP_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
+	else
+		I915_WRITE(GEN9_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
+
+	if (IS_GEN9(dev_priv)) {
+		/* DOP Clock Gating Enable for GuC clocks */
+		I915_WRITE(GEN7_MISCCPCTL, (GEN8_DOP_CLOCK_GATE_GUC_ENABLE |
+					    I915_READ(GEN7_MISCCPCTL)));
+
+		/* allows for 5us (in 10ns units) before GT can go to RC6 */
+		I915_WRITE(GUC_ARAT_C6DIS, 0x1FF);
+	}
+}
+
+/* Copy RSA signature from the fw image to HW for verification */
+static int guc_xfer_rsa(struct intel_guc *guc, struct i915_vma *vma)
+{
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	struct intel_uc_fw *guc_fw = &guc->fw;
+	struct sg_table *sg = vma->pages;
+	u32 rsa[UOS_RSA_SCRATCH_MAX_COUNT];
+	int i;
+
+	if (sg_pcopy_to_buffer(sg->sgl, sg->nents, rsa, sizeof(rsa),
+			       guc_fw->rsa_offset) != sizeof(rsa))
+		return -EINVAL;
+
+	for (i = 0; i < UOS_RSA_SCRATCH_MAX_COUNT; i++)
+		I915_WRITE(UOS_RSA_SCRATCH(i), rsa[i]);
+
+	return 0;
  }
 
  /*
@@ -122,29 +154,17 @@ static inline bool guc_ucode_response(struct drm_i915_private *dev_priv,
   * Architecturally, the DMA engine is bidirectional, and can potentially even
   * transfer between GTT locations. This functionality is left out of the API
   * for now as there is no need for it.
- *
- * Note that GuC needs the CSS header plus uKernel code to be copied by the
- * DMA engine in one operation, whereas the RSA signature is loaded via MMIO.
   */
-static int guc_ucode_xfer_dma(struct drm_i915_private *dev_priv,
-			      struct i915_vma *vma)
+static int guc_xfer_ucode(struct intel_guc *guc, struct i915_vma *vma)
  {
-	struct intel_uc_fw *guc_fw = &dev_priv->guc.fw;
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	struct intel_uc_fw *guc_fw = &guc->fw;
  	unsigned long offset;
-	struct sg_table *sg = vma->pages;
-	u32 status, rsa[UOS_RSA_SCRATCH_MAX_COUNT];
-	int i, ret = 0;
-
-	/* where RSA signature starts */
-	offset = guc_fw->rsa_offset;
 
-	/* Copy RSA signature from the fw image to HW for verification */
-	sg_pcopy_to_buffer(sg->sgl, sg->nents, rsa, sizeof(rsa), offset);
-	for (i = 0; i < UOS_RSA_SCRATCH_MAX_COUNT; i++)
-		I915_WRITE(UOS_RSA_SCRATCH(i), rsa[i]);
-
-	/* The header plus uCode will be copied to WOPCM via DMA, excluding any
-	 * other components */
+	/*
+	 * The header plus uCode will be copied to WOPCM via DMA, excluding any
+	 * other components
+	 */
  	I915_WRITE(DMA_COPY_SIZE, guc_fw->header_size + guc_fw->ucode_size);
 
  	/* Set the source address for the new blob */
@@ -162,33 +182,57 @@ static int guc_ucode_xfer_dma(struct drm_i915_private *dev_priv,
  	/* Finally start the DMA */
  	I915_WRITE(DMA_CTRL, _MASKED_BIT_ENABLE(UOS_MOVE | START_DMA));
 
+	return 0;
+}
+
+/*
+ * Read the GuC status register (GUC_STATUS) and store it in the
+ * specified location; then return a boolean indicating whether
+ * the value matches either of two values representing completion
+ * of the GuC boot process.
+ *
+ * This is used for polling the GuC status in a wait_for()
+ * loop below.
+ */
+static inline bool guc_ready(struct intel_guc *guc, u32 *status)
+{
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	u32 val = I915_READ(GUC_STATUS);
+	u32 uk_val = val & GS_UKERNEL_MASK;
+
+	*status = val;
+	return (uk_val == GS_UKERNEL_READY) ||
+		((val & GS_MIA_CORE_STATE) && (uk_val == GS_UKERNEL_LAPIC_DONE));
+}
+
+static int guc_wait_ucode(struct intel_guc *guc)
+{
+	u32 status;
+	int ret;
+
  	/*
-	 * Wait for the DMA to complete & the GuC to start up.
+	 * Wait for the GuC to start up.
  	 * NB: Docs recommend not using the interrupt for completion.
  	 * Measurements indicate this should take no more than 20ms, so a
  	 * timeout here indicates that the GuC has failed and is unusable.
  	 * (Higher levels of the driver will attempt to fall back to
  	 * execlist mode if this happens.)
  	 */
-	ret = wait_for(guc_ucode_response(dev_priv, &status), 100);
-
-	DRM_DEBUG_DRIVER("DMA status 0x%x, GuC status 0x%x\n",
-			I915_READ(DMA_CTRL), status);
+	ret = wait_for(guc_ready(guc, &status), 100);
+	DRM_DEBUG_DRIVER("GuC status %#x\n", status);
 
  	if ((status & GS_BOOTROM_MASK) == GS_BOOTROM_RSA_FAILED) {
  		DRM_ERROR("GuC firmware signature verification failed\n");
  		ret = -ENOEXEC;
  	}
 
-	DRM_DEBUG_DRIVER("returning %d\n", ret);
-
  	return ret;
  }
 
  /*
   * Load the GuC firmware blob into the MinuteIA.
   */
-static int guc_ucode_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma)
+static int guc_fw_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma)
  {
  	struct intel_guc *guc = container_of(guc_fw, struct intel_guc, fw);
  	struct drm_i915_private *dev_priv = guc_to_i915(guc);
@@ -198,34 +242,24 @@ static int guc_ucode_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma)
 
  	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
 
-	/* Enable MIA caching. GuC clock gating is disabled. */
-	I915_WRITE(GUC_SHIM_CONTROL, GUC_SHIM_CONTROL_VALUE);
-
-	/* WaDisableMinuteIaClockGating:bxt */
-	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_A1)) {
-		I915_WRITE(GUC_SHIM_CONTROL, (I915_READ(GUC_SHIM_CONTROL) &
-					      ~GUC_ENABLE_MIA_CLOCK_GATING));
-	}
-
-	/* WaC6DisallowByGfxPause:bxt */
-	if (IS_BXT_REVID(dev_priv, 0, BXT_REVID_B0))
-		I915_WRITE(GEN6_GFXPAUSE, 0x30FFF);
+	guc_prepare_xfer(guc);
 
-	if (IS_GEN9_LP(dev_priv))
-		I915_WRITE(GEN9LP_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
-	else
-		I915_WRITE(GEN9_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
-
-	if (IS_GEN9(dev_priv)) {
-		/* DOP Clock Gating Enable for GuC clocks */
-		I915_WRITE(GEN7_MISCCPCTL, (GEN8_DOP_CLOCK_GATE_GUC_ENABLE |
-					    I915_READ(GEN7_MISCCPCTL)));
+	/*
+	 * Note that GuC needs the CSS header plus uKernel code to be copied
+	 * by the DMA engine in one operation, whereas the RSA signature is
+	 * loaded via MMIO.
+	 */
+	ret = guc_xfer_rsa(guc, vma);
+	if (ret)
+		DRM_WARN("GuC firmware signature xfer error %d\n", ret);
 
-		/* allows for 5us (in 10ns units) before GT can go to RC6 */
-		I915_WRITE(GUC_ARAT_C6DIS, 0x1FF);
-	}
+	ret = guc_xfer_ucode(guc, vma);
+	if (ret)
+		DRM_WARN("GuC firmware code xfer error %d\n", ret);
 
-	ret = guc_ucode_xfer_dma(dev_priv, vma);
+	ret = guc_wait_ucode(guc);
+	if (ret)
+		DRM_ERROR("GuC firmware xfer error %d\n", ret);
 
  	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 

drivers/gpu/drm/i915/intel_guc_fw.c static int guc_ucode_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma) · line 285

Quoted 6 lines
drivers/gpu/drm/i915/intel_guc_fw.c +1 -1
@@ -247,5 +281,5 @@ static int guc_ucode_xfer(struct intel_uc_fw *guc_fw, struct i915_vma *vma)
   */
  int intel_guc_fw_upload(struct intel_guc *guc)
  {
-	return intel_uc_fw_upload(&guc->fw, guc_ucode_xfer);
+	return intel_uc_fw_upload(&guc->fw, guc_fw_xfer);
  }

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

#8 lore raw

✓ Fi.CI.BAT: success for series starting with [v2,1/4] drm/i915/guc: Split GuC firmware xfer function into clear steps

In reply to Michal Wajdeczko
== Series Details ==

== Series Details ==

Series: series starting with [v2,1/4] drm/i915/guc: Split GuC firmware xfer function into clear steps
URL : https://patchwork.freedesktop.org/series/33112/
State : success

== Summary ==

Series 33112v1 series starting with [v2,1/4] drm/i915/guc: Split GuC firmware xfer function into clear steps
https://patchwork.freedesktop.org/api/1.0/series/33112/revisions/1/mbox/

Test chamelium:

        Subgroup dp-crc-fast:
                incomplete -> PASS       (fi-kbl-7500u) fdo#102514
Test gem_exec_reloc:
        Subgroup basic-cpu-gtt-active:
                pass       -> FAIL       (fi-gdg-551) fdo#102582 +1

fdo#102514 https://bugs.freedesktop.org/show_bug.cgi?id=102514
fdo#102582 https://bugs.freedesktop.org/show_bug.cgi?id=102582

fi-bdw-5557u     total:289  pass:268  dwarn:0   dfail:0   fail:0   skip:21  time:450s
fi-blb-e6850     total:289  pass:223  dwarn:1   dfail:0   fail:0   skip:65  time:380s
fi-bsw-n3050     total:289  pass:243  dwarn:0   dfail:0   fail:0   skip:46  time:533s
fi-bwr-2160      total:289  pass:183  dwarn:0   dfail:0   fail:0   skip:106 time:275s
fi-bxt-dsi       total:289  pass:259  dwarn:0   dfail:0   fail:0   skip:30  time:512s
fi-bxt-j4205     total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:511s
fi-byt-j1900     total:289  pass:253  dwarn:1   dfail:0   fail:0   skip:35  time:504s
fi-byt-n2820     total:289  pass:249  dwarn:1   dfail:0   fail:0   skip:39  time:498s
fi-cfl-s         total:289  pass:254  dwarn:3   dfail:0   fail:0   skip:32  time:565s
fi-elk-e7500     total:289  pass:229  dwarn:0   dfail:0   fail:0   skip:60  time:419s
fi-gdg-551       total:289  pass:176  dwarn:1   dfail:0   fail:3   skip:109 time:261s
fi-glk-1         total:289  pass:261  dwarn:0   dfail:0   fail:0   skip:28  time:590s
fi-glk-dsi       total:289  pass:258  dwarn:0   dfail:0   fail:1   skip:30  time:489s
fi-hsw-4770      total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:436s
fi-hsw-4770r     total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:430s
fi-ilk-650       total:289  pass:228  dwarn:0   dfail:0   fail:0   skip:61  time:425s
fi-ivb-3770      total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:461s
fi-kbl-7500u     total:289  pass:264  dwarn:1   dfail:0   fail:0   skip:24  time:491s
fi-kbl-7560u     total:289  pass:270  dwarn:0   dfail:0   fail:0   skip:19  time:574s
fi-kbl-7567u     total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:481s
fi-kbl-r         total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:581s
fi-pnv-d510      total:289  pass:222  dwarn:1   dfail:0   fail:0   skip:66  time:572s
fi-skl-6260u     total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:451s
fi-skl-6600u     total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:591s
fi-skl-6700hq    total:289  pass:263  dwarn:0   dfail:0   fail:0   skip:26  time:648s
fi-skl-6700k     total:289  pass:265  dwarn:0   dfail:0   fail:0   skip:24  time:525s
fi-skl-6770hq    total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:506s
fi-skl-gvtdvm    total:289  pass:266  dwarn:0   dfail:0   fail:0   skip:23  time:462s
fi-snb-2520m     total:289  pass:250  dwarn:0   dfail:0   fail:0   skip:39  time:570s
fi-snb-2600      total:289  pass:249  dwarn:0   dfail:0   fail:0   skip:40  time:426s

fi-bdw-gvtdvm failed to connect after reboot

2f945d948d05de1dd4bd1ca14eac40ff563d9d77 drm-tip: 2017y-11m-03d-09h-28m-20s UTC integration manifest f13cf7f4d241 HAX enable GuC submission for CI
5d7b95ce7782 drm/i915/guc: Simplify programming of GUC_SHIM_CONTROL
997bafa93e44 drm/i915/guc: Wait for ucode DMA transfer completion
959ce04b0ab9 drm/i915/guc: Split GuC firmware xfer function into clear steps

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_6936/
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#9 lore raw

✗ Fi.CI.IGT: warning for series starting with [v2,1/4] drm/i915/guc: Split GuC firmware xfer function into clear steps

In reply to Michal Wajdeczko
== Series Details ==

== Series Details ==

Series: series starting with [v2,1/4] drm/i915/guc: Split GuC firmware xfer function into clear steps
URL : https://patchwork.freedesktop.org/series/33112/
State : warning

== Summary ==

Test kms_setmode:

        Subgroup basic:
                fail       -> PASS       (shard-hsw) fdo#99912
Test kms_plane_lowres:
        Subgroup pipe-C-tiling-none:
                pass       -> DMESG-WARN (shard-hsw)

fdo#99912 https://bugs.freedesktop.org/show_bug.cgi?id=99912

shard-hsw total:2539 pass:1433 dwarn:1 dfail:1 fail:7 skip:1097 time:9490s

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_6936/shards.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx