Maarten Maathuis
2010-Feb-01  18:34 UTC
[Nouveau] [PATCH 1/4] drm/nv50: align size of buffer object to the right boundaries.
- Depth and stencil buffers are supposed to be large enough in general.
Signed-off-by: Maarten Maathuis <madman2003 at gmail.com>
---
 drivers/gpu/drm/nouveau/nouveau_bo.c |   10 +++++-----
 1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c
b/drivers/gpu/drm/nouveau/nouveau_bo.c
index db0ed4c..028719f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -65,8 +65,10 @@ nouveau_bo_fixup_align(struct drm_device *dev,
 
 	/*
 	 * Some of the tile_flags have a periodic structure of N*4096 bytes,
-	 * align to to that as well as the page size. Overallocate memory to
-	 * avoid corruption of other buffer objects.
+	 * align to to that as well as the page size. Align the size to the
+	 * appropriate boundaries. This does imply that sizes are rounded up
+	 * 3-7 pages, so be aware of this and do not waste memory by allocating
+	 * many small buffers.
 	 */
 	if (dev_priv->card_type == NV_50) {
 		uint32_t block_size = nouveau_mem_fb_amount(dev) >> 15;
@@ -77,22 +79,20 @@ nouveau_bo_fixup_align(struct drm_device *dev,
 		case 0x2800:
 		case 0x4800:
 		case 0x7a00:
-			*size = roundup(*size, block_size);
 			if (is_power_of_2(block_size)) {
-				*size += 3 * block_size;
 				for (i = 1; i < 10; i++) {
 					*align = 12 * i * block_size;
 					if (!(*align % 65536))
 						break;
 				}
 			} else {
-				*size += 6 * block_size;
 				for (i = 1; i < 10; i++) {
 					*align = 8 * i * block_size;
 					if (!(*align % 65536))
 						break;
 				}
 			}
+			*size = roundup(*size, *align);
 			break;
 		default:
 			break;
-- 
1.6.6.1
Maarten Maathuis
2010-Feb-01  18:34 UTC
[Nouveau] [PATCH 2/4] drm/nv50: make the pgraph irq handler loop like the pre-nv50 version
Signed-off-by: Maarten Maathuis <madman2003 at gmail.com>
---
 drivers/gpu/drm/nouveau/nouveau_irq.c |  140 ++++++++++++++++++---------------
 1 files changed, 76 insertions(+), 64 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nouveau_irq.c
b/drivers/gpu/drm/nouveau/nouveau_irq.c
index baa9b3e..475d0f2 100644
--- a/drivers/gpu/drm/nouveau/nouveau_irq.c
+++ b/drivers/gpu/drm/nouveau/nouveau_irq.c
@@ -580,83 +580,95 @@ nouveau_pgraph_irq_handler(struct drm_device *dev)
 static void
 nv50_pgraph_irq_handler(struct drm_device *dev)
 {
-	uint32_t status, nsource;
+	uint32_t status;
 
-	status = nv_rd32(dev, NV03_PGRAPH_INTR);
-	nsource = nv_rd32(dev, NV03_PGRAPH_NSOURCE);
+	while ((status = nv_rd32(dev, NV03_PGRAPH_INTR))) {
+		uint32_t nsource = nv_rd32(dev, NV03_PGRAPH_NSOURCE);
 
-	if (status & 0x00000001) {
-		nouveau_pgraph_intr_notify(dev, nsource);
-		status &= ~0x00000001;
-		nv_wr32(dev, NV03_PGRAPH_INTR, 0x00000001);
-	}
+		if (status & 0x00000001) {
+			nouveau_pgraph_intr_notify(dev, nsource);
+			status &= ~0x00000001;
+			nv_wr32(dev, NV03_PGRAPH_INTR, 0x00000001);
+		}
 
-	if (status & 0x00000010) {
-		nouveau_pgraph_intr_error(dev, nsource |
-					  NV03_PGRAPH_NSOURCE_ILLEGAL_MTHD);
+		if (status & 0x00000010) {
+			nouveau_pgraph_intr_error(dev, nsource |
+						  NV03_PGRAPH_NSOURCE_ILLEGAL_MTHD);
 
-		status &= ~0x00000010;
-		nv_wr32(dev, NV03_PGRAPH_INTR, 0x00000010);
-	}
+			status &= ~0x00000010;
+			nv_wr32(dev, NV03_PGRAPH_INTR, 0x00000010);
+		}
 
-	if (status & 0x00001000) {
-		nv_wr32(dev, 0x400500, 0x00000000);
-		nv_wr32(dev, NV03_PGRAPH_INTR, NV_PGRAPH_INTR_CONTEXT_SWITCH);
-		nv_wr32(dev, NV40_PGRAPH_INTR_EN, nv_rd32(dev,
-			NV40_PGRAPH_INTR_EN) & ~NV_PGRAPH_INTR_CONTEXT_SWITCH);
-		nv_wr32(dev, 0x400500, 0x00010001);
+		if (status & 0x00001000) {
+			nv_wr32(dev, 0x400500, 0x00000000);
+			nv_wr32(dev, NV03_PGRAPH_INTR, NV_PGRAPH_INTR_CONTEXT_SWITCH);
+			nv_wr32(dev, NV40_PGRAPH_INTR_EN, nv_rd32(dev,
+				NV40_PGRAPH_INTR_EN) & ~NV_PGRAPH_INTR_CONTEXT_SWITCH);
+			nv_wr32(dev, 0x400500, 0x00010001);
 
-		nv50_graph_context_switch(dev);
+			nv50_graph_context_switch(dev);
 
-		status &= ~NV_PGRAPH_INTR_CONTEXT_SWITCH;
-	}
+			status &= ~NV_PGRAPH_INTR_CONTEXT_SWITCH;
+		}
 
-	if (status & 0x00100000) {
-		nouveau_pgraph_intr_error(dev, nsource |
-					  NV03_PGRAPH_NSOURCE_DATA_ERROR);
+		if (status & 0x00100000) {
+			nouveau_pgraph_intr_error(dev, nsource |
+						  NV03_PGRAPH_NSOURCE_DATA_ERROR);
 
-		status &= ~0x00100000;
-		nv_wr32(dev, NV03_PGRAPH_INTR, 0x00100000);
-	}
+			status &= ~0x00100000;
+			nv_wr32(dev, NV03_PGRAPH_INTR, 0x00100000);
+		}
 
-	if (status & 0x00200000) {
-		int r;
-
-		nouveau_pgraph_intr_error(dev, nsource |
-					  NV03_PGRAPH_NSOURCE_PROTECTION_ERROR);
-
-		NV_ERROR(dev, "magic set 1:\n");
-		for (r = 0x408900; r <= 0x408910; r += 4)
-			NV_ERROR(dev, "\t0x%08x: 0x%08x\n", r, nv_rd32(dev, r));
-		nv_wr32(dev, 0x408900, nv_rd32(dev, 0x408904) | 0xc0000000);
-		for (r = 0x408e08; r <= 0x408e24; r += 4)
-			NV_ERROR(dev, "\t0x%08x: 0x%08x\n", r, nv_rd32(dev, r));
-		nv_wr32(dev, 0x408e08, nv_rd32(dev, 0x408e08) | 0xc0000000);
-
-		NV_ERROR(dev, "magic set 2:\n");
-		for (r = 0x409900; r <= 0x409910; r += 4)
-			NV_ERROR(dev, "\t0x%08x: 0x%08x\n", r, nv_rd32(dev, r));
-		nv_wr32(dev, 0x409900, nv_rd32(dev, 0x409904) | 0xc0000000);
-		for (r = 0x409e08; r <= 0x409e24; r += 4)
-			NV_ERROR(dev, "\t0x%08x: 0x%08x\n", r, nv_rd32(dev, r));
-		nv_wr32(dev, 0x409e08, nv_rd32(dev, 0x409e08) | 0xc0000000);
-
-		status &= ~0x00200000;
-		nv_wr32(dev, NV03_PGRAPH_NSOURCE, nsource);
-		nv_wr32(dev, NV03_PGRAPH_INTR, 0x00200000);
-	}
+		if (status & 0x00200000) {
+			int r;
+
+			nouveau_pgraph_intr_error(dev, nsource |
+					NV03_PGRAPH_NSOURCE_PROTECTION_ERROR);
+
+			NV_ERROR(dev, "magic set 1:\n");
+			for (r = 0x408900; r <= 0x408910; r += 4)
+				NV_ERROR(dev, "\t0x%08x: 0x%08x\n", r,
+					nv_rd32(dev, r));
+			nv_wr32(dev, 0x408900,
+				nv_rd32(dev, 0x408904) | 0xc0000000);
+			for (r = 0x408e08; r <= 0x408e24; r += 4)
+				NV_ERROR(dev, "\t0x%08x: 0x%08x\n", r,
+							nv_rd32(dev, r));
+			nv_wr32(dev, 0x408e08,
+				nv_rd32(dev, 0x408e08) | 0xc0000000);
+
+			NV_ERROR(dev, "magic set 2:\n");
+			for (r = 0x409900; r <= 0x409910; r += 4)
+				NV_ERROR(dev, "\t0x%08x: 0x%08x\n", r,
+					nv_rd32(dev, r));
+			nv_wr32(dev, 0x409900,
+				nv_rd32(dev, 0x409904) | 0xc0000000);
+			for (r = 0x409e08; r <= 0x409e24; r += 4)
+				NV_ERROR(dev, "\t0x%08x: 0x%08x\n", r,
+					nv_rd32(dev, r));
+			nv_wr32(dev, 0x409e08,
+				nv_rd32(dev, 0x409e08) | 0xc0000000);
+
+			status &= ~0x00200000;
+			nv_wr32(dev, NV03_PGRAPH_NSOURCE, nsource);
+			nv_wr32(dev, NV03_PGRAPH_INTR, 0x00200000);
+		}
 
-	if (status) {
-		NV_INFO(dev, "Unhandled PGRAPH_INTR - 0x%08x\n", status);
-		nv_wr32(dev, NV03_PGRAPH_INTR, status);
-	}
+		if (status) {
+			NV_INFO(dev, "Unhandled PGRAPH_INTR - 0x%08x\n",
+				status);
+			nv_wr32(dev, NV03_PGRAPH_INTR, status);
+		}
 
-	{
-		const int isb = (1 << 16) | (1 << 0);
+		{
+			const int isb = (1 << 16) | (1 << 0);
 
-		if ((nv_rd32(dev, 0x400500) & isb) != isb)
-			nv_wr32(dev, 0x400500, nv_rd32(dev, 0x400500) | isb);
-		nv_wr32(dev, 0x400824, nv_rd32(dev, 0x400824) & ~(1 << 31));
+			if ((nv_rd32(dev, 0x400500) & isb) != isb)
+				nv_wr32(dev, 0x400500,
+					nv_rd32(dev, 0x400500) | isb);
+			nv_wr32(dev, 0x400824,
+				nv_rd32(dev, 0x400824) & ~(1 << 31));
+		}
 	}
 
 	nv_wr32(dev, NV03_PMC_INTR_0, NV_PMC_INTR_0_PGRAPH_PENDING);
-- 
1.6.6.1