aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/gpu
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu')
-rw-r--r--drivers/gpu/drm/nouveau/core/core/falcon.c7
-rw-r--r--drivers/gpu/drm/nouveau/core/core/subdev.c2
-rw-r--r--drivers/gpu/drm/nouveau/core/include/core/object.h7
-rw-r--r--drivers/gpu/drm/nouveau/core/subdev/fb/base.c4
-rw-r--r--drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c5
-rw-r--r--drivers/gpu/drm/nouveau/nouveau_bo.c1
-rw-r--r--drivers/gpu/drm/nouveau/nouveau_drm.c3
-rw-r--r--drivers/gpu/drm/radeon/evergreen.c27
-rw-r--r--drivers/gpu/drm/radeon/evergreen_cs.c86
-rw-r--r--drivers/gpu/drm/radeon/r600.c7
-rw-r--r--drivers/gpu/drm/radeon/r600_cs.c38
-rw-r--r--drivers/gpu/drm/radeon/radeon_asic.c6
-rw-r--r--drivers/gpu/drm/radeon/radeon_combios.c8
-rw-r--r--drivers/gpu/drm/radeon/radeon_display.c4
-rw-r--r--drivers/gpu/drm/radeon/radeon_ring.c3
-rw-r--r--drivers/gpu/drm/radeon/radeon_ttm.c1
-rw-r--r--drivers/gpu/drm/radeon/reg_srcs/cayman1
-rw-r--r--drivers/gpu/drm/radeon/rv515.c2
-rw-r--r--drivers/gpu/drm/ttm/ttm_bo_util.c13
19 files changed, 140 insertions, 85 deletions
diff --git a/drivers/gpu/drm/nouveau/core/core/falcon.c b/drivers/gpu/drm/nouveau/core/core/falcon.c
index 6b0843c3387..e05c1577758 100644
--- a/drivers/gpu/drm/nouveau/core/core/falcon.c
+++ b/drivers/gpu/drm/nouveau/core/core/falcon.c
@@ -73,8 +73,11 @@ _nouveau_falcon_init(struct nouveau_object *object)
nv_debug(falcon, "data limit: %d\n", falcon->data.limit);
/* wait for 'uc halted' to be signalled before continuing */
- if (falcon->secret) {
- nv_wait(falcon, 0x008, 0x00000010, 0x00000010);
+ if (falcon->secret && falcon->version < 4) {
+ if (!falcon->version)
+ nv_wait(falcon, 0x008, 0x00000010, 0x00000010);
+ else
+ nv_wait(falcon, 0x180, 0x80000000, 0);
nv_wo32(falcon, 0x004, 0x00000010);
}
diff --git a/drivers/gpu/drm/nouveau/core/core/subdev.c b/drivers/gpu/drm/nouveau/core/core/subdev.c
index f74c30aa33a..48f06378d3f 100644
--- a/drivers/gpu/drm/nouveau/core/core/subdev.c
+++ b/drivers/gpu/drm/nouveau/core/core/subdev.c
@@ -99,7 +99,7 @@ nouveau_subdev_create_(struct nouveau_object *parent,
if (ret)
return ret;
- mutex_init(&subdev->mutex);
+ __mutex_init(&subdev->mutex, subname, &oclass->lock_class_key);
subdev->name = subname;
if (parent) {
diff --git a/drivers/gpu/drm/nouveau/core/include/core/object.h b/drivers/gpu/drm/nouveau/core/include/core/object.h
index 5982935ee23..106bb19fdd9 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/object.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/object.h
@@ -50,10 +50,13 @@ int nouveau_object_fini(struct nouveau_object *, bool suspend);
extern struct nouveau_ofuncs nouveau_object_ofuncs;
+/* Don't allocate dynamically, because lockdep needs lock_class_keys to be in
+ * ".data". */
struct nouveau_oclass {
u32 handle;
- struct nouveau_ofuncs *ofuncs;
- struct nouveau_omthds *omthds;
+ struct nouveau_ofuncs * const ofuncs;
+ struct nouveau_omthds * const omthds;
+ struct lock_class_key lock_class_key;
};
#define nv_oclass(o) nv_object(o)->oclass
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/base.c b/drivers/gpu/drm/nouveau/core/subdev/fb/base.c
index d6d16007ec1..d62045f454b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/base.c
@@ -86,8 +86,8 @@ nouveau_fb_preinit(struct nouveau_fb *pfb)
return ret;
}
- if (!nouveau_mm_initialised(&pfb->tags) && tags) {
- ret = nouveau_mm_init(&pfb->tags, 0, ++tags, 1);
+ if (!nouveau_mm_initialised(&pfb->tags)) {
+ ret = nouveau_mm_init(&pfb->tags, 0, tags ? ++tags : 0, 1);
if (ret)
return ret;
}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c
index 487cb8c6c20..eac236ed19b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv50.c
@@ -99,7 +99,7 @@ nv50_fb_vram_init(struct nouveau_fb *pfb)
struct nouveau_bios *bios = nouveau_bios(device);
const u32 rsvd_head = ( 256 * 1024) >> 12; /* vga memory */
const u32 rsvd_tail = (1024 * 1024) >> 12; /* vbios etc */
- u32 size;
+ u32 size, tags = 0;
int ret;
pfb->ram.size = nv_rd32(pfb, 0x10020c);
@@ -140,10 +140,11 @@ nv50_fb_vram_init(struct nouveau_fb *pfb)
return ret;
pfb->ram.ranks = (nv_rd32(pfb, 0x100200) & 0x4) ? 2 : 1;
+ tags = nv_rd32(pfb, 0x100320);
break;
}
- return nv_rd32(pfb, 0x100320);
+ return tags;
}
static int
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 69d7b1d0b9d..1699a9083a2 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -28,6 +28,7 @@
*/
#include <core/engine.h>
+#include <linux/swiotlb.h>
#include <subdev/fb.h>
#include <subdev/vm.h>
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 8b090f1eb51..5e7aef23825 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -245,6 +245,8 @@ static int nouveau_drm_probe(struct pci_dev *pdev,
return 0;
}
+static struct lock_class_key drm_client_lock_class_key;
+
static int
nouveau_drm_load(struct drm_device *dev, unsigned long flags)
{
@@ -256,6 +258,7 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags)
ret = nouveau_cli_create(pdev, "DRM", sizeof(*drm), (void**)&drm);
if (ret)
return ret;
+ lockdep_set_class(&drm->client.mutex, &drm_client_lock_class_key);
dev->dev_private = drm;
drm->dev = dev;
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 4d0e60adbc6..a2d478e8692 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -1313,14 +1313,18 @@ void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *sav
if (!(tmp & EVERGREEN_CRTC_BLANK_DATA_EN)) {
radeon_wait_for_vblank(rdev, i);
tmp |= EVERGREEN_CRTC_BLANK_DATA_EN;
+ WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 1);
WREG32(EVERGREEN_CRTC_BLANK_CONTROL + crtc_offsets[i], tmp);
+ WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 0);
}
} else {
tmp = RREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i]);
if (!(tmp & EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE)) {
radeon_wait_for_vblank(rdev, i);
tmp |= EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE;
+ WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 1);
WREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i], tmp);
+ WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 0);
}
}
/* wait for the next frame */
@@ -1345,6 +1349,8 @@ void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *sav
blackout &= ~BLACKOUT_MODE_MASK;
WREG32(MC_SHARED_BLACKOUT_CNTL, blackout | 1);
}
+ /* wait for the MC to settle */
+ udelay(100);
}
void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *save)
@@ -1378,11 +1384,15 @@ void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *s
if (ASIC_IS_DCE6(rdev)) {
tmp = RREG32(EVERGREEN_CRTC_BLANK_CONTROL + crtc_offsets[i]);
tmp |= EVERGREEN_CRTC_BLANK_DATA_EN;
+ WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 1);
WREG32(EVERGREEN_CRTC_BLANK_CONTROL + crtc_offsets[i], tmp);
+ WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 0);
} else {
tmp = RREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i]);
tmp &= ~EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE;
+ WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 1);
WREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i], tmp);
+ WREG32(EVERGREEN_CRTC_UPDATE_LOCK + crtc_offsets[i], 0);
}
/* wait for the next frame */
frame_count = radeon_get_vblank_counter(rdev, i);
@@ -2036,9 +2046,20 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
WREG32(HDP_ADDR_CONFIG, gb_addr_config);
WREG32(DMA_TILING_CONFIG, gb_addr_config);
- tmp = gb_addr_config & NUM_PIPES_MASK;
- tmp = r6xx_remap_render_backend(rdev, tmp, rdev->config.evergreen.max_backends,
- EVERGREEN_MAX_BACKENDS, disabled_rb_mask);
+ if ((rdev->config.evergreen.max_backends == 1) &&
+ (rdev->flags & RADEON_IS_IGP)) {
+ if ((disabled_rb_mask & 3) == 1) {
+ /* RB0 disabled, RB1 enabled */
+ tmp = 0x11111111;
+ } else {
+ /* RB1 disabled, RB0 enabled */
+ tmp = 0x00000000;
+ }
+ } else {
+ tmp = gb_addr_config & NUM_PIPES_MASK;
+ tmp = r6xx_remap_render_backend(rdev, tmp, rdev->config.evergreen.max_backends,
+ EVERGREEN_MAX_BACKENDS, disabled_rb_mask);
+ }
WREG32(GB_BACKEND_MAP, tmp);
WREG32(CGTS_SYS_TCC_DISABLE, 0);
diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index 7a445666e71..ee4cff534f1 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -2909,14 +2909,14 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
return -EINVAL;
}
if (tiled) {
- dst_offset = ib[idx+1];
+ dst_offset = radeon_get_ib_value(p, idx+1);
dst_offset <<= 8;
ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
p->idx += count + 7;
} else {
- dst_offset = ib[idx+1];
- dst_offset |= ((u64)(ib[idx+2] & 0xff)) << 32;
+ dst_offset = radeon_get_ib_value(p, idx+1);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32;
ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
ib[idx+2] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
@@ -2954,12 +2954,12 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
DRM_ERROR("bad L2T, frame to fields DMA_PACKET_COPY\n");
return -EINVAL;
}
- dst_offset = ib[idx+1];
+ dst_offset = radeon_get_ib_value(p, idx+1);
dst_offset <<= 8;
- dst2_offset = ib[idx+2];
+ dst2_offset = radeon_get_ib_value(p, idx+2);
dst2_offset <<= 8;
- src_offset = ib[idx+8];
- src_offset |= ((u64)(ib[idx+9] & 0xff)) << 32;
+ src_offset = radeon_get_ib_value(p, idx+8);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
dev_warn(p->dev, "DMA L2T, frame to fields src buffer too small (%llu %lu)\n",
src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
@@ -3014,12 +3014,12 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
return -EINVAL;
}
- dst_offset = ib[idx+1];
+ dst_offset = radeon_get_ib_value(p, idx+1);
dst_offset <<= 8;
- dst2_offset = ib[idx+2];
+ dst2_offset = radeon_get_ib_value(p, idx+2);
dst2_offset <<= 8;
- src_offset = ib[idx+8];
- src_offset |= ((u64)(ib[idx+9] & 0xff)) << 32;
+ src_offset = radeon_get_ib_value(p, idx+8);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
dev_warn(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
@@ -3046,22 +3046,22 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
/* detile bit */
if (idx_value & (1 << 31)) {
/* tiled src, linear dst */
- src_offset = ib[idx+1];
+ src_offset = radeon_get_ib_value(p, idx+1);
src_offset <<= 8;
ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
- dst_offset = ib[idx+7];
- dst_offset |= ((u64)(ib[idx+8] & 0xff)) << 32;
+ dst_offset = radeon_get_ib_value(p, idx+7);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
ib[idx+7] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
ib[idx+8] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
} else {
/* linear src, tiled dst */
- src_offset = ib[idx+7];
- src_offset |= ((u64)(ib[idx+8] & 0xff)) << 32;
+ src_offset = radeon_get_ib_value(p, idx+7);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
ib[idx+7] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
ib[idx+8] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
- dst_offset = ib[idx+1];
+ dst_offset = radeon_get_ib_value(p, idx+1);
dst_offset <<= 8;
ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
}
@@ -3098,12 +3098,12 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
return -EINVAL;
}
- dst_offset = ib[idx+1];
+ dst_offset = radeon_get_ib_value(p, idx+1);
dst_offset <<= 8;
- dst2_offset = ib[idx+2];
+ dst2_offset = radeon_get_ib_value(p, idx+2);
dst2_offset <<= 8;
- src_offset = ib[idx+8];
- src_offset |= ((u64)(ib[idx+9] & 0xff)) << 32;
+ src_offset = radeon_get_ib_value(p, idx+8);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
dev_warn(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
@@ -3135,22 +3135,22 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
/* detile bit */
if (idx_value & (1 << 31)) {
/* tiled src, linear dst */
- src_offset = ib[idx+1];
+ src_offset = radeon_get_ib_value(p, idx+1);
src_offset <<= 8;
ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
- dst_offset = ib[idx+7];
- dst_offset |= ((u64)(ib[idx+8] & 0xff)) << 32;
+ dst_offset = radeon_get_ib_value(p, idx+7);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
ib[idx+7] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
ib[idx+8] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
} else {
/* linear src, tiled dst */
- src_offset = ib[idx+7];
- src_offset |= ((u64)(ib[idx+8] & 0xff)) << 32;
+ src_offset = radeon_get_ib_value(p, idx+7);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+8) & 0xff)) << 32;
ib[idx+7] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
ib[idx+8] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
- dst_offset = ib[idx+1];
+ dst_offset = radeon_get_ib_value(p, idx+1);
dst_offset <<= 8;
ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
}
@@ -3176,10 +3176,10 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
switch (misc) {
case 0:
/* L2L, byte */
- src_offset = ib[idx+2];
- src_offset |= ((u64)(ib[idx+4] & 0xff)) << 32;
- dst_offset = ib[idx+1];
- dst_offset |= ((u64)(ib[idx+3] & 0xff)) << 32;
+ src_offset = radeon_get_ib_value(p, idx+2);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
+ dst_offset = radeon_get_ib_value(p, idx+1);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff)) << 32;
if ((src_offset + count) > radeon_bo_size(src_reloc->robj)) {
dev_warn(p->dev, "DMA L2L, byte src buffer too small (%llu %lu)\n",
src_offset + count, radeon_bo_size(src_reloc->robj));
@@ -3216,12 +3216,12 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
DRM_ERROR("bad L2L, dw, broadcast DMA_PACKET_COPY\n");
return -EINVAL;
}
- dst_offset = ib[idx+1];
- dst_offset |= ((u64)(ib[idx+4] & 0xff)) << 32;
- dst2_offset = ib[idx+2];
- dst2_offset |= ((u64)(ib[idx+5] & 0xff)) << 32;
- src_offset = ib[idx+3];
- src_offset |= ((u64)(ib[idx+6] & 0xff)) << 32;
+ dst_offset = radeon_get_ib_value(p, idx+1);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
+ dst2_offset = radeon_get_ib_value(p, idx+2);
+ dst2_offset |= ((u64)(radeon_get_ib_value(p, idx+5) & 0xff)) << 32;
+ src_offset = radeon_get_ib_value(p, idx+3);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+6) & 0xff)) << 32;
if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
dev_warn(p->dev, "DMA L2L, dw, broadcast src buffer too small (%llu %lu)\n",
src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
@@ -3251,10 +3251,10 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
}
} else {
/* L2L, dw */
- src_offset = ib[idx+2];
- src_offset |= ((u64)(ib[idx+4] & 0xff)) << 32;
- dst_offset = ib[idx+1];
- dst_offset |= ((u64)(ib[idx+3] & 0xff)) << 32;
+ src_offset = radeon_get_ib_value(p, idx+2);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
+ dst_offset = radeon_get_ib_value(p, idx+1);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff)) << 32;
if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
dev_warn(p->dev, "DMA L2L, dw src buffer too small (%llu %lu)\n",
src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
@@ -3279,8 +3279,8 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
DRM_ERROR("bad DMA_PACKET_CONSTANT_FILL\n");
return -EINVAL;
}
- dst_offset = ib[idx+1];
- dst_offset |= ((u64)(ib[idx+3] & 0x00ff0000)) << 16;
+ dst_offset = radeon_get_ib_value(p, idx+1);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0x00ff0000)) << 16;
if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
dev_warn(p->dev, "DMA constant fill buffer too small (%llu %lu)\n",
dst_offset, radeon_bo_size(dst_reloc->robj));
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index bc2540b17c5..becb03e8b32 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -1462,12 +1462,15 @@ u32 r6xx_remap_render_backend(struct radeon_device *rdev,
u32 disabled_rb_mask)
{
u32 rendering_pipe_num, rb_num_width, req_rb_num;
- u32 pipe_rb_ratio, pipe_rb_remain;
+ u32 pipe_rb_ratio, pipe_rb_remain, tmp;
u32 data = 0, mask = 1 << (max_rb_num - 1);
unsigned i, j;
/* mask out the RBs that don't exist on that asic */
- disabled_rb_mask |= (0xff << max_rb_num) & 0xff;
+ tmp = disabled_rb_mask | ((0xff << max_rb_num) & 0xff);
+ /* make sure at least one RB is available */
+ if ((tmp & 0xff) != 0xff)
+ disabled_rb_mask = tmp;
rendering_pipe_num = 1 << tiling_pipe_num;
req_rb_num = total_max_rb_num - r600_count_pipe_bits(disabled_rb_mask);
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 69ec24ab8d6..9b2512bf1a4 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -2623,14 +2623,14 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p)
return -EINVAL;
}
if (tiled) {
- dst_offset = ib[idx+1];
+ dst_offset = radeon_get_ib_value(p, idx+1);
dst_offset <<= 8;
ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
p->idx += count + 5;
} else {
- dst_offset = ib[idx+1];
- dst_offset |= ((u64)(ib[idx+2] & 0xff)) << 32;
+ dst_offset = radeon_get_ib_value(p, idx+1);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32;
ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
ib[idx+2] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
@@ -2658,32 +2658,32 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p)
/* detile bit */
if (idx_value & (1 << 31)) {
/* tiled src, linear dst */
- src_offset = ib[idx+1];
+ src_offset = radeon_get_ib_value(p, idx+1);
src_offset <<= 8;
ib[idx+1] += (u32)(src_reloc->lobj.gpu_offset >> 8);
- dst_offset = ib[idx+5];
- dst_offset |= ((u64)(ib[idx+6] & 0xff)) << 32;
+ dst_offset = radeon_get_ib_value(p, idx+5);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+6) & 0xff)) << 32;
ib[idx+5] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
ib[idx+6] += upper_32_bits(dst_reloc->lobj.gpu_offset) & 0xff;
} else {
/* linear src, tiled dst */
- src_offset = ib[idx+5];
- src_offset |= ((u64)(ib[idx+6] & 0xff)) << 32;
+ src_offset = radeon_get_ib_value(p, idx+5);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+6) & 0xff)) << 32;
ib[idx+5] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
ib[idx+6] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
- dst_offset = ib[idx+1];
+ dst_offset = radeon_get_ib_value(p, idx+1);
dst_offset <<= 8;
ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset >> 8);
}
p->idx += 7;
} else {
if (p->family >= CHIP_RV770) {
- src_offset = ib[idx+2];
- src_offset |= ((u64)(ib[idx+4] & 0xff)) << 32;
- dst_offset = ib[idx+1];
- dst_offset |= ((u64)(ib[idx+3] & 0xff)) << 32;
+ src_offset = radeon_get_ib_value(p, idx+2);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
+ dst_offset = radeon_get_ib_value(p, idx+1);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff)) << 32;
ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
ib[idx+2] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
@@ -2691,10 +2691,10 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p)
ib[idx+4] += upper_32_bits(src_reloc->lobj.gpu_offset) & 0xff;
p->idx += 5;
} else {
- src_offset = ib[idx+2];
- src_offset |= ((u64)(ib[idx+3] & 0xff)) << 32;
- dst_offset = ib[idx+1];
- dst_offset |= ((u64)(ib[idx+3] & 0xff0000)) << 16;
+ src_offset = radeon_get_ib_value(p, idx+2);
+ src_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff)) << 32;
+ dst_offset = radeon_get_ib_value(p, idx+1);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff0000)) << 16;
ib[idx+1] += (u32)(dst_reloc->lobj.gpu_offset & 0xfffffffc);
ib[idx+2] += (u32)(src_reloc->lobj.gpu_offset & 0xfffffffc);
@@ -2724,8 +2724,8 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p)
DRM_ERROR("bad DMA_PACKET_WRITE\n");
return -EINVAL;
}
- dst_offset = ib[idx+1];
- dst_offset |= ((u64)(ib[idx+3] & 0x00ff0000)) << 16;
+ dst_offset = radeon_get_ib_value(p, idx+1);
+ dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0x00ff0000)) << 16;
if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
dev_warn(p->dev, "DMA constant fill buffer too small (%llu %lu)\n",
dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 9056fafb00e..0b202c07fe5 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -1445,7 +1445,7 @@ static struct radeon_asic cayman_asic = {
.vm = {
.init = &cayman_vm_init,
.fini = &cayman_vm_fini,
- .pt_ring_index = R600_RING_TYPE_DMA_INDEX,
+ .pt_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.set_page = &cayman_vm_set_page,
},
.ring = {
@@ -1572,7 +1572,7 @@ static struct radeon_asic trinity_asic = {
.vm = {
.init = &cayman_vm_init,
.fini = &cayman_vm_fini,
- .pt_ring_index = R600_RING_TYPE_DMA_INDEX,
+ .pt_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.set_page = &cayman_vm_set_page,
},
.ring = {
@@ -1699,7 +1699,7 @@ static struct radeon_asic si_asic = {
.vm = {
.init = &si_vm_init,
.fini = &si_vm_fini,
- .pt_ring_index = R600_RING_TYPE_DMA_INDEX,
+ .pt_ring_index = RADEON_RING_TYPE_GFX_INDEX,
.set_page = &si_vm_set_page,
},
.ring = {
diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c
index 33a56a09ff1..3e403bdda58 100644
--- a/drivers/gpu/drm/radeon/radeon_combios.c
+++ b/drivers/gpu/drm/radeon/radeon_combios.c
@@ -2470,6 +2470,14 @@ bool radeon_get_legacy_connector_info_from_bios(struct drm_device *dev)
1),
ATOM_DEVICE_CRT1_SUPPORT);
}
+ /* RV100 board with external TDMS bit mis-set.
+ * Actually uses internal TMDS, clear the bit.
+ */
+ if (dev->pdev->device == 0x5159 &&
+ dev->pdev->subsystem_vendor == 0x1014 &&
+ dev->pdev->subsystem_device == 0x029A) {
+ tmp &= ~(1 << 4);
+ }
if ((tmp >> 4) & 0x1) {
devices |= ATOM_DEVICE_DFP2_SUPPORT;
radeon_add_legacy_encoder(dev,
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index ff3def78461..05c96fa0b05 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -1115,8 +1115,10 @@ radeon_user_framebuffer_create(struct drm_device *dev,
}
radeon_fb = kzalloc(sizeof(*radeon_fb), GFP_KERNEL);
- if (radeon_fb == NULL)
+ if (radeon_fb == NULL) {
+ drm_gem_object_unreference_unlocked(obj);
return ERR_PTR(-ENOMEM);
+ }
ret = radeon_framebuffer_init(dev, radeon_fb, mode_cmd, obj);
if (ret) {
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index 2430d80b187..cd72062d5a9 100644
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -377,6 +377,9 @@ int radeon_ring_alloc(struct radeon_device *rdev, struct radeon_ring *ring, unsi
{
int r;
+ /* make sure we aren't trying to allocate more space than there is on the ring */
+ if (ndw > (ring->ring_size / 4))
+ return -ENOMEM;
/* Align requested size with padding so unlock_commit can
* pad safely */
ndw = (ndw + ring->align_mask) & ~ring->align_mask;
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 1d8ff2f850b..93f760e27a9 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -38,6 +38,7 @@
#include <drm/radeon_drm.h>
#include <linux/seq_file.h>
#include <linux/slab.h>
+#include <linux/swiotlb.h>
#include "radeon_reg.h"
#include "radeon.h"
diff --git a/drivers/gpu/drm/radeon/reg_srcs/cayman b/drivers/gpu/drm/radeon/reg_srcs/cayman
index 0f656b111c1..a072fa8c46b 100644
--- a/drivers/gpu/drm/radeon/reg_srcs/cayman
+++ b/drivers/gpu/drm/radeon/reg_srcs/cayman
@@ -1,5 +1,6 @@
cayman 0x9400
0x0000802C GRBM_GFX_INDEX
+0x00008040 WAIT_UNTIL
0x000084FC CP_STRMOUT_CNTL
0x000085F0 CP_COHER_CNTL
0x000085F4 CP_COHER_SIZE
diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 2bb6d0e84b3..435ed355136 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -336,6 +336,8 @@ void rv515_mc_stop(struct radeon_device *rdev, struct rv515_mc_save *save)
WREG32(R600_CITF_CNTL, blackout);
}
}
+ /* wait for the MC to settle */
+ udelay(100);
}
void rv515_mc_resume(struct radeon_device *rdev, struct rv515_mc_save *save)
diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index 44420fca7df..8be35c809c7 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -429,7 +429,7 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo,
struct ttm_bo_device *bdev = bo->bdev;
struct ttm_bo_driver *driver = bdev->driver;
- fbo = kzalloc(sizeof(*fbo), GFP_KERNEL);
+ fbo = kmalloc(sizeof(*fbo), GFP_KERNEL);
if (!fbo)
return -ENOMEM;
@@ -448,7 +448,12 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo,
fbo->vm_node = NULL;
atomic_set(&fbo->cpu_writers, 0);
- fbo->sync_obj = driver->sync_obj_ref(bo->sync_obj);
+ spin_lock(&bdev->fence_lock);
+ if (bo->sync_obj)
+ fbo->sync_obj = driver->sync_obj_ref(bo->sync_obj);
+ else
+ fbo->sync_obj = NULL;
+ spin_unlock(&bdev->fence_lock);
kref_init(&fbo->list_kref);
kref_init(&fbo->kref);
fbo->destroy = &ttm_transfered_destroy;
@@ -661,13 +666,11 @@ int ttm_bo_move_accel_cleanup(struct ttm_buffer_object *bo,
*/
set_bit(TTM_BO_PRIV_FLAG_MOVING, &bo->priv_flags);
-
- /* ttm_buffer_object_transfer accesses bo->sync_obj */
- ret = ttm_buffer_object_transfer(bo, &ghost_obj);
spin_unlock(&bdev->fence_lock);
if (tmp_obj)
driver->sync_obj_unref(&tmp_obj);
+ ret = ttm_buffer_object_transfer(bo, &ghost_obj);
if (ret)
return ret;
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-29 10:24:45 +0000