drm/amd/pp: Memory Latency is always 25us on Vega10
authorRex Zhu <Rex.Zhu@amd.com>
Mon, 18 Jun 2018 10:49:07 +0000 (18:49 +0800)
committerAlex Deucher <alexander.deucher@amd.com>
Thu, 5 Jul 2018 21:39:55 +0000 (16:39 -0500)
For HBM, 25us latency is enough for memory clock switch.

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c

index 37c2e5c15d62172991c71eff9bc9ac125dcf795a..7fdd9ede96c79b40123b2bba019f2fa223cfa451 100644 (file)
 
 static const uint32_t channel_number[] = {1, 2, 0, 4, 0, 8, 0, 16, 2};
 
-#define MEM_FREQ_LOW_LATENCY        25000
-#define MEM_FREQ_HIGH_LATENCY       80000
-#define MEM_LATENCY_HIGH            245
-#define MEM_LATENCY_LOW             35
-#define MEM_LATENCY_ERR             0xFFFF
-
 #define mmDF_CS_AON0_DramBaseAddress0                                                                  0x0044
 #define mmDF_CS_AON0_DramBaseAddress0_BASE_IDX                                                         0
 
@@ -4075,18 +4069,6 @@ static void vega10_get_sclks(struct pp_hwmgr *hwmgr,
 
 }
 
-static uint32_t vega10_get_mem_latency(struct pp_hwmgr *hwmgr,
-               uint32_t clock)
-{
-       if (clock >= MEM_FREQ_LOW_LATENCY &&
-                       clock < MEM_FREQ_HIGH_LATENCY)
-               return MEM_LATENCY_HIGH;
-       else if (clock >= MEM_FREQ_HIGH_LATENCY)
-               return MEM_LATENCY_LOW;
-       else
-               return MEM_LATENCY_ERR;
-}
-
 static void vega10_get_memclocks(struct pp_hwmgr *hwmgr,
                struct pp_clock_levels_with_latency *clocks)
 {
@@ -4100,14 +4082,13 @@ static void vega10_get_memclocks(struct pp_hwmgr *hwmgr,
 
        for (i = 0; i < dep_table->count; i++) {
                if (dep_table->entries[i].clk) {
+
                        clocks->data[j].clocks_in_khz =
                                                dep_table->entries[i].clk * 10;
                        data->mclk_latency_table.entries[j].frequency =
                                                        dep_table->entries[i].clk;
                        clocks->data[j].latency_in_us =
-                               data->mclk_latency_table.entries[j].latency =
-                                               vega10_get_mem_latency(hwmgr,
-                                                       dep_table->entries[i].clk);
+                               data->mclk_latency_table.entries[j].latency = 25;
                        j++;
                }
        }