[PATCH] amdgpu/dc: remove wait_reg/wait_reg_func interfaces.
Dave Airlie
airlied at gmail.com
Thu Sep 28 03:29:32 UTC 2017
From: Dave Airlie <airlied at redhat.com>
These aren't used in the tree anywhere, and there is a TODO.
Signed-off-by: Dave Airlie <airlied at redhat.com>
---
drivers/gpu/drm/amd/display/dc/dm_services.h | 46 ----------------------------
1 file changed, 46 deletions(-)
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h
index c976e2a..8166027 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services.h
@@ -236,52 +236,6 @@ unsigned int generic_reg_wait(const struct dc_context *ctx,
block ## reg_num ## _ ## reg_name ## __ ## reg_field ## _MASK,\
block ## reg_num ## _ ## reg_name ## __ ## reg_field ## __SHIFT)
-/* TODO get rid of this pos*/
-static inline bool wait_reg_func(
- const struct dc_context *ctx,
- uint32_t addr,
- uint32_t mask,
- uint8_t shift,
- uint32_t condition_value,
- unsigned int interval_us,
- unsigned int timeout_us)
-{
- uint32_t field_value;
- uint32_t reg_val;
- unsigned int count = 0;
-
- if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
- timeout_us *= 655; /* 6553 give about 30 second before time out */
-
- do {
- /* try once without sleeping */
- if (count > 0) {
- if (interval_us >= 1000)
- msleep(interval_us/1000);
- else
- udelay(interval_us);
- }
- reg_val = dm_read_reg(ctx, addr);
- field_value = get_reg_field_value_ex(reg_val, mask, shift);
- count += interval_us;
-
- } while (field_value != condition_value && count <= timeout_us);
-
- ASSERT(count <= timeout_us);
-
- return count <= timeout_us;
-}
-
-#define wait_reg(ctx, inst_offset, reg_name, reg_field, condition_value)\
- wait_reg_func(\
- ctx,\
- mm##reg_name + inst_offset + DCE_BASE.instance[0].segment[mm##reg_name##_BASE_IDX],\
- reg_name ## __ ## reg_field ## _MASK,\
- reg_name ## __ ## reg_field ## __SHIFT,\
- condition_value,\
- 20000,\
- 200000)
-
/**************************************
* Power Play (PP) interfaces
**************************************/
--
2.9.4
More information about the amd-gfx
mailing list