drm/radeon/kms: enable writeback (v2)

When writeback is enabled, the GPU shadows writes to certain
registers into a buffer in memory.  The driver can then read
the values from the shadow rather than reading back from the
register across the bus.  Writeback can be disabled by setting
the no_wb module param to 1.

On r6xx/r7xx/evergreen, the following registers are shadowed:
- CP scratch registers
- CP read pointer
- IH write pointer
On r1xx-rr5xx, the following registers are shadowed:
- CP scratch registers
- CP read pointer

v2:
- Combine wb patches for r6xx-evergreen and r1xx-r5xx
- Writeback is disabled on AGP boards since it tends to be
unreliable on AGP using the gart.
- Check radeon_wb_init return values properly.

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

1 files changed, 26 insertions, 10 deletions

diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 79082d4..e47d221 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -731,7 +731,7 @@ int evergreen_cp_resume(struct radeon_device *rdev)
 
 	/* Set ring buffer size */
 	rb_bufsz = drm_order(rdev->cp.ring_size / 8);
-	tmp = RB_NO_UPDATE | (drm_order(RADEON_GPU_PAGE_SIZE/8) << 8) | rb_bufsz;
+	tmp = (drm_order(RADEON_GPU_PAGE_SIZE/8) << 8) | rb_bufsz;
 #ifdef __BIG_ENDIAN
 	tmp |= BUF_SWAP_32BIT;
 #endif
@@ -745,8 +745,19 @@ int evergreen_cp_resume(struct radeon_device *rdev)
 	WREG32(CP_RB_CNTL, tmp | RB_RPTR_WR_ENA);
 	WREG32(CP_RB_RPTR_WR, 0);
 	WREG32(CP_RB_WPTR, 0);
-	WREG32(CP_RB_RPTR_ADDR, rdev->cp.gpu_addr & 0xFFFFFFFF);
-	WREG32(CP_RB_RPTR_ADDR_HI, upper_32_bits(rdev->cp.gpu_addr));
+
+	/* set the wb address wether it's enabled or not */
+	WREG32(CP_RB_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC);
+	WREG32(CP_RB_RPTR_ADDR_HI, upper_32_bits(rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFF);
+	WREG32(SCRATCH_ADDR, ((rdev->wb.gpu_addr + RADEON_WB_SCRATCH_OFFSET) >> 8) & 0xFFFFFFFF);
+
+	if (rdev->wb.enabled)
+		WREG32(SCRATCH_UMSK, 0xff);
+	else {
+		tmp |= RB_NO_UPDATE;
+		WREG32(SCRATCH_UMSK, 0);
+	}
+
 	mdelay(1);
 	WREG32(CP_RB_CNTL, tmp);
 
@@ -1759,8 +1770,10 @@ static inline u32 evergreen_get_ih_wptr(struct radeon_device *rdev)
 {
 	u32 wptr, tmp;
 
-	/* XXX use writeback */
-	wptr = RREG32(IH_RB_WPTR);
+	if (rdev->wb.enabled)
+		wptr = rdev->wb.wb[R600_WB_IH_WPTR_OFFSET/4];
+	else
+		wptr = RREG32(IH_RB_WPTR);
 
 	if (wptr & RB_OVERFLOW) {
 		/* When a ring buffer overflow happen start parsing interrupt
@@ -2068,6 +2081,11 @@ static int evergreen_startup(struct radeon_device *rdev)
 	}
 #endif
 
+	/* allocate wb buffer */
+	r = radeon_wb_init(rdev);
+	if (r)
+		return r;
+
 	/* Enable IRQ */
 	r = r600_irq_init(rdev);
 	if (r) {
@@ -2086,8 +2104,6 @@ static int evergreen_startup(struct radeon_device *rdev)
 	r = evergreen_cp_resume(rdev);
 	if (r)
 		return r;
-	/* write back buffer are not vital so don't worry about failure */
-	r600_wb_enable(rdev);
 
 	return 0;
 }
@@ -2128,7 +2144,7 @@ int evergreen_suspend(struct radeon_device *rdev)
 	r700_cp_stop(rdev);
 	rdev->cp.ready = false;
 	evergreen_irq_suspend(rdev);
-	r600_wb_disable(rdev);
+	radeon_wb_disable(rdev);
 	evergreen_pcie_gart_disable(rdev);
 #if 0
 	/* unpin shaders bo */
@@ -2245,8 +2261,8 @@ int evergreen_init(struct radeon_device *rdev)
 	if (r) {
 		dev_err(rdev->dev, "disabling GPU acceleration\n");
 		r700_cp_fini(rdev);
-		r600_wb_fini(rdev);
 		r600_irq_fini(rdev);
+		radeon_wb_fini(rdev);
 		radeon_irq_kms_fini(rdev);
 		evergreen_pcie_gart_fini(rdev);
 		rdev->accel_working = false;
@@ -2270,8 +2286,8 @@ void evergreen_fini(struct radeon_device *rdev)
 {
 	/*r600_blit_fini(rdev);*/
 	r700_cp_fini(rdev);
-	r600_wb_fini(rdev);
 	r600_irq_fini(rdev);
+	radeon_wb_fini(rdev);
 	radeon_irq_kms_fini(rdev);
 	evergreen_pcie_gart_fini(rdev);
 	radeon_gem_fini(rdev);