drm/radeon: make all functions work with multiple rings.
Give all asic and radeon_ring_* functions a
radeon_cp parameter, so they know the ring to work with.
Signed-off-by: Christian König <deathsimple@vodafone.de>
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 2f18163..271cee7 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -811,30 +811,33 @@
void r100_fence_ring_emit(struct radeon_device *rdev,
struct radeon_fence *fence)
{
+ struct radeon_cp *cp = &rdev->cp;
+
/* We have to make sure that caches are flushed before
* CPU might read something from VRAM. */
- radeon_ring_write(rdev, PACKET0(RADEON_RB3D_DSTCACHE_CTLSTAT, 0));
- radeon_ring_write(rdev, RADEON_RB3D_DC_FLUSH_ALL);
- radeon_ring_write(rdev, PACKET0(RADEON_RB3D_ZCACHE_CTLSTAT, 0));
- radeon_ring_write(rdev, RADEON_RB3D_ZC_FLUSH_ALL);
+ radeon_ring_write(cp, PACKET0(RADEON_RB3D_DSTCACHE_CTLSTAT, 0));
+ radeon_ring_write(cp, RADEON_RB3D_DC_FLUSH_ALL);
+ radeon_ring_write(cp, PACKET0(RADEON_RB3D_ZCACHE_CTLSTAT, 0));
+ radeon_ring_write(cp, RADEON_RB3D_ZC_FLUSH_ALL);
/* Wait until IDLE & CLEAN */
- radeon_ring_write(rdev, PACKET0(RADEON_WAIT_UNTIL, 0));
- radeon_ring_write(rdev, RADEON_WAIT_2D_IDLECLEAN | RADEON_WAIT_3D_IDLECLEAN);
- radeon_ring_write(rdev, PACKET0(RADEON_HOST_PATH_CNTL, 0));
- radeon_ring_write(rdev, rdev->config.r100.hdp_cntl |
+ radeon_ring_write(cp, PACKET0(RADEON_WAIT_UNTIL, 0));
+ radeon_ring_write(cp, RADEON_WAIT_2D_IDLECLEAN | RADEON_WAIT_3D_IDLECLEAN);
+ radeon_ring_write(cp, PACKET0(RADEON_HOST_PATH_CNTL, 0));
+ radeon_ring_write(cp, rdev->config.r100.hdp_cntl |
RADEON_HDP_READ_BUFFER_INVALIDATE);
- radeon_ring_write(rdev, PACKET0(RADEON_HOST_PATH_CNTL, 0));
- radeon_ring_write(rdev, rdev->config.r100.hdp_cntl);
+ radeon_ring_write(cp, PACKET0(RADEON_HOST_PATH_CNTL, 0));
+ radeon_ring_write(cp, rdev->config.r100.hdp_cntl);
/* Emit fence sequence & fire IRQ */
- radeon_ring_write(rdev, PACKET0(rdev->fence_drv[fence->ring].scratch_reg, 0));
- radeon_ring_write(rdev, fence->seq);
- radeon_ring_write(rdev, PACKET0(RADEON_GEN_INT_STATUS, 0));
- radeon_ring_write(rdev, RADEON_SW_INT_FIRE);
+ radeon_ring_write(cp, PACKET0(rdev->fence_drv[fence->ring].scratch_reg, 0));
+ radeon_ring_write(cp, fence->seq);
+ radeon_ring_write(cp, PACKET0(RADEON_GEN_INT_STATUS, 0));
+ radeon_ring_write(cp, RADEON_SW_INT_FIRE);
}
void r100_semaphore_ring_emit(struct radeon_device *rdev,
+ struct radeon_cp *cp,
struct radeon_semaphore *semaphore,
- unsigned ring, bool emit_wait)
+ bool emit_wait)
{
/* Unused on older asics, since we don't have semaphores or multiple rings */
BUG();
@@ -846,6 +849,7 @@
unsigned num_gpu_pages,
struct radeon_fence *fence)
{
+ struct radeon_cp *cp = &rdev->cp;
uint32_t cur_pages;
uint32_t stride_bytes = RADEON_GPU_PAGE_SIZE;
uint32_t pitch;
@@ -863,7 +867,7 @@
/* Ask for enough room for blit + flush + fence */
ndw = 64 + (10 * num_loops);
- r = radeon_ring_lock(rdev, ndw);
+ r = radeon_ring_lock(rdev, cp, ndw);
if (r) {
DRM_ERROR("radeon: moving bo (%d) asking for %u dw.\n", r, ndw);
return -EINVAL;
@@ -877,8 +881,8 @@
/* pages are in Y direction - height
page width in X direction - width */
- radeon_ring_write(rdev, PACKET3(PACKET3_BITBLT_MULTI, 8));
- radeon_ring_write(rdev,
+ radeon_ring_write(cp, PACKET3(PACKET3_BITBLT_MULTI, 8));
+ radeon_ring_write(cp,
RADEON_GMC_SRC_PITCH_OFFSET_CNTL |
RADEON_GMC_DST_PITCH_OFFSET_CNTL |
RADEON_GMC_SRC_CLIPPING |
@@ -890,26 +894,26 @@
RADEON_DP_SRC_SOURCE_MEMORY |
RADEON_GMC_CLR_CMP_CNTL_DIS |
RADEON_GMC_WR_MSK_DIS);
- radeon_ring_write(rdev, (pitch << 22) | (src_offset >> 10));
- radeon_ring_write(rdev, (pitch << 22) | (dst_offset >> 10));
- radeon_ring_write(rdev, (0x1fff) | (0x1fff << 16));
- radeon_ring_write(rdev, 0);
- radeon_ring_write(rdev, (0x1fff) | (0x1fff << 16));
- radeon_ring_write(rdev, num_gpu_pages);
- radeon_ring_write(rdev, num_gpu_pages);
- radeon_ring_write(rdev, cur_pages | (stride_pixels << 16));
+ radeon_ring_write(cp, (pitch << 22) | (src_offset >> 10));
+ radeon_ring_write(cp, (pitch << 22) | (dst_offset >> 10));
+ radeon_ring_write(cp, (0x1fff) | (0x1fff << 16));
+ radeon_ring_write(cp, 0);
+ radeon_ring_write(cp, (0x1fff) | (0x1fff << 16));
+ radeon_ring_write(cp, num_gpu_pages);
+ radeon_ring_write(cp, num_gpu_pages);
+ radeon_ring_write(cp, cur_pages | (stride_pixels << 16));
}
- radeon_ring_write(rdev, PACKET0(RADEON_DSTCACHE_CTLSTAT, 0));
- radeon_ring_write(rdev, RADEON_RB2D_DC_FLUSH_ALL);
- radeon_ring_write(rdev, PACKET0(RADEON_WAIT_UNTIL, 0));
- radeon_ring_write(rdev,
+ radeon_ring_write(cp, PACKET0(RADEON_DSTCACHE_CTLSTAT, 0));
+ radeon_ring_write(cp, RADEON_RB2D_DC_FLUSH_ALL);
+ radeon_ring_write(cp, PACKET0(RADEON_WAIT_UNTIL, 0));
+ radeon_ring_write(cp,
RADEON_WAIT_2D_IDLECLEAN |
RADEON_WAIT_HOST_IDLECLEAN |
RADEON_WAIT_DMA_GUI_IDLE);
if (fence) {
r = radeon_fence_emit(rdev, fence);
}
- radeon_ring_unlock_commit(rdev);
+ radeon_ring_unlock_commit(rdev, cp);
return r;
}
@@ -930,19 +934,20 @@
void r100_ring_start(struct radeon_device *rdev)
{
+ struct radeon_cp *cp = &rdev->cp;
int r;
- r = radeon_ring_lock(rdev, 2);
+ r = radeon_ring_lock(rdev, cp, 2);
if (r) {
return;
}
- radeon_ring_write(rdev, PACKET0(RADEON_ISYNC_CNTL, 0));
- radeon_ring_write(rdev,
+ radeon_ring_write(cp, PACKET0(RADEON_ISYNC_CNTL, 0));
+ radeon_ring_write(cp,
RADEON_ISYNC_ANY2D_IDLE3D |
RADEON_ISYNC_ANY3D_IDLE2D |
RADEON_ISYNC_WAIT_IDLEGUI |
RADEON_ISYNC_CPSCRATCH_IDLEGUI);
- radeon_ring_unlock_commit(rdev);
+ radeon_ring_unlock_commit(rdev, cp);
}
@@ -1043,6 +1048,7 @@
int r100_cp_init(struct radeon_device *rdev, unsigned ring_size)
{
+ struct radeon_cp *cp = &rdev->cp;
unsigned rb_bufsz;
unsigned rb_blksz;
unsigned max_fetch;
@@ -1068,7 +1074,7 @@
rb_bufsz = drm_order(ring_size / 8);
ring_size = (1 << (rb_bufsz + 1)) * 4;
r100_cp_load_microcode(rdev);
- r = radeon_ring_init(rdev, ring_size);
+ r = radeon_ring_init(rdev, cp, ring_size);
if (r) {
return r;
}
@@ -1077,7 +1083,7 @@
rb_blksz = 9;
/* cp will read 128bytes at a time (4 dwords) */
max_fetch = 1;
- rdev->cp.align_mask = 16 - 1;
+ cp->align_mask = 16 - 1;
/* Write to CP_RB_WPTR will be delayed for pre_write_timer clocks */
pre_write_timer = 64;
/* Force CP_RB_WPTR write if written more than one time before the
@@ -1107,13 +1113,13 @@
WREG32(RADEON_CP_RB_CNTL, tmp | RADEON_RB_NO_UPDATE);
/* Set ring address */
- DRM_INFO("radeon: ring at 0x%016lX\n", (unsigned long)rdev->cp.gpu_addr);
- WREG32(RADEON_CP_RB_BASE, rdev->cp.gpu_addr);
+ DRM_INFO("radeon: ring at 0x%016lX\n", (unsigned long)cp->gpu_addr);
+ WREG32(RADEON_CP_RB_BASE, cp->gpu_addr);
/* Force read & write ptr to 0 */
WREG32(RADEON_CP_RB_CNTL, tmp | RADEON_RB_RPTR_WR_ENA | RADEON_RB_NO_UPDATE);
WREG32(RADEON_CP_RB_RPTR_WR, 0);
- rdev->cp.wptr = 0;
- WREG32(RADEON_CP_RB_WPTR, rdev->cp.wptr);
+ cp->wptr = 0;
+ WREG32(RADEON_CP_RB_WPTR, cp->wptr);
/* set the wb address whether it's enabled or not */
WREG32(R_00070C_CP_RB_RPTR_ADDR,
@@ -1129,7 +1135,7 @@
WREG32(RADEON_CP_RB_CNTL, tmp);
udelay(10);
- rdev->cp.rptr = RREG32(RADEON_CP_RB_RPTR);
+ cp->rptr = RREG32(RADEON_CP_RB_RPTR);
/* Set cp mode to bus mastering & enable cp*/
WREG32(RADEON_CP_CSQ_MODE,
REG_SET(RADEON_INDIRECT2_START, indirect2_start) |
@@ -1138,12 +1144,12 @@
WREG32(RADEON_CP_CSQ_MODE, 0x00004D4D);
WREG32(RADEON_CP_CSQ_CNTL, RADEON_CSQ_PRIBM_INDBM);
radeon_ring_start(rdev);
- r = radeon_ring_test(rdev);
+ r = radeon_ring_test(rdev, cp);
if (r) {
DRM_ERROR("radeon: cp isn't working (%d).\n", r);
return r;
}
- rdev->cp.ready = true;
+ cp->ready = true;
radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size);
return 0;
}
@@ -1155,7 +1161,7 @@
}
/* Disable ring */
r100_cp_disable(rdev);
- radeon_ring_fini(rdev);
+ radeon_ring_fini(rdev, &rdev->cp);
DRM_INFO("radeon: cp finalized\n");
}
@@ -1173,9 +1179,9 @@
}
}
-void r100_cp_commit(struct radeon_device *rdev)
+void r100_cp_commit(struct radeon_device *rdev, struct radeon_cp *cp)
{
- WREG32(RADEON_CP_RB_WPTR, rdev->cp.wptr);
+ WREG32(RADEON_CP_RB_WPTR, cp->wptr);
(void)RREG32(RADEON_CP_RB_WPTR);
}
@@ -2160,26 +2166,26 @@
return false;
}
-bool r100_gpu_is_lockup(struct radeon_device *rdev)
+bool r100_gpu_is_lockup(struct radeon_device *rdev, struct radeon_cp *cp)
{
u32 rbbm_status;
int r;
rbbm_status = RREG32(R_000E40_RBBM_STATUS);
if (!G_000E40_GUI_ACTIVE(rbbm_status)) {
- r100_gpu_lockup_update(&rdev->config.r100.lockup, &rdev->cp);
+ r100_gpu_lockup_update(&rdev->config.r100.lockup, cp);
return false;
}
/* force CP activities */
- r = radeon_ring_lock(rdev, 2);
+ r = radeon_ring_lock(rdev, cp, 2);
if (!r) {
/* PACKET2 NOP */
- radeon_ring_write(rdev, 0x80000000);
- radeon_ring_write(rdev, 0x80000000);
- radeon_ring_unlock_commit(rdev);
+ radeon_ring_write(cp, 0x80000000);
+ radeon_ring_write(cp, 0x80000000);
+ radeon_ring_unlock_commit(rdev, cp);
}
- rdev->cp.rptr = RREG32(RADEON_CP_RB_RPTR);
- return r100_gpu_cp_is_lockup(rdev, &rdev->config.r100.lockup, &rdev->cp);
+ cp->rptr = RREG32(RADEON_CP_RB_RPTR);
+ return r100_gpu_cp_is_lockup(rdev, &rdev->config.r100.lockup, cp);
}
void r100_bm_disable(struct radeon_device *rdev)
@@ -2587,21 +2593,22 @@
struct drm_info_node *node = (struct drm_info_node *) m->private;
struct drm_device *dev = node->minor->dev;
struct radeon_device *rdev = dev->dev_private;
+ struct radeon_cp *cp = &rdev->cp;
uint32_t rdp, wdp;
unsigned count, i, j;
- radeon_ring_free_size(rdev);
+ radeon_ring_free_size(rdev, cp);
rdp = RREG32(RADEON_CP_RB_RPTR);
wdp = RREG32(RADEON_CP_RB_WPTR);
- count = (rdp + rdev->cp.ring_size - wdp) & rdev->cp.ptr_mask;
+ count = (rdp + cp->ring_size - wdp) & cp->ptr_mask;
seq_printf(m, "CP_STAT 0x%08x\n", RREG32(RADEON_CP_STAT));
seq_printf(m, "CP_RB_WPTR 0x%08x\n", wdp);
seq_printf(m, "CP_RB_RPTR 0x%08x\n", rdp);
- seq_printf(m, "%u free dwords in ring\n", rdev->cp.ring_free_dw);
+ seq_printf(m, "%u free dwords in ring\n", cp->ring_free_dw);
seq_printf(m, "%u dwords in ring\n", count);
for (j = 0; j <= count; j++) {
- i = (rdp + j) & rdev->cp.ptr_mask;
- seq_printf(m, "r[%04d]=0x%08x\n", i, rdev->cp.ring[i]);
+ i = (rdp + j) & cp->ptr_mask;
+ seq_printf(m, "r[%04d]=0x%08x\n", i, cp->ring[i]);
}
return 0;
}
@@ -3643,7 +3650,7 @@
}
}
-int r100_ring_test(struct radeon_device *rdev)
+int r100_ring_test(struct radeon_device *rdev, struct radeon_cp *cp)
{
uint32_t scratch;
uint32_t tmp = 0;
@@ -3656,15 +3663,15 @@
return r;
}
WREG32(scratch, 0xCAFEDEAD);
- r = radeon_ring_lock(rdev, 2);
+ r = radeon_ring_lock(rdev, cp, 2);
if (r) {
DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
radeon_scratch_free(rdev, scratch);
return r;
}
- radeon_ring_write(rdev, PACKET0(scratch, 0));
- radeon_ring_write(rdev, 0xDEADBEEF);
- radeon_ring_unlock_commit(rdev);
+ radeon_ring_write(cp, PACKET0(scratch, 0));
+ radeon_ring_write(cp, 0xDEADBEEF);
+ radeon_ring_unlock_commit(rdev, cp);
for (i = 0; i < rdev->usec_timeout; i++) {
tmp = RREG32(scratch);
if (tmp == 0xDEADBEEF) {
@@ -3685,9 +3692,11 @@
void r100_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
{
- radeon_ring_write(rdev, PACKET0(RADEON_CP_IB_BASE, 1));
- radeon_ring_write(rdev, ib->gpu_addr);
- radeon_ring_write(rdev, ib->length_dw);
+ struct radeon_cp *cp = &rdev->cp;
+
+ radeon_ring_write(cp, PACKET0(RADEON_CP_IB_BASE, 1));
+ radeon_ring_write(cp, ib->gpu_addr);
+ radeon_ring_write(cp, ib->length_dw);
}
int r100_ib_test(struct radeon_device *rdev)
@@ -3704,7 +3713,7 @@
return r;
}
WREG32(scratch, 0xCAFEDEAD);
- r = radeon_ib_get(rdev, &ib);
+ r = radeon_ib_get(rdev, RADEON_RING_TYPE_GFX_INDEX, &ib);
if (r) {
return r;
}