drm/amdgpu/gfx10: Refine Cleaner Shader for GFX10.1.10

This patch updates the cleaner shader, which is responsible for
initializing GPU resources such as Local Data Share (LDS), Vector
General Purpose Registers (VGPRs), and Scalar General Purpose Registers
(SGPRs). Changes include adjustments to register clearing and shader
configuration.

- Updated GPU resource initialization addresses in the cleaner shader
from `be803080` to `be803000`.
- Simplified the logic in the SGPR clearing section, ensuring all SGPRs
are set to zero.

Fixes: 25961bad9212 ("drm/amdgpu/gfx10: Add cleaner shader for GFX10.1.10")
Cc: Christian König <[e-mail]>
Cc: Alex Deucher <[e-mail]>
Signed-off-by: Manu Rastogi <[e-mail]>
Signed-off-by: Vitaly Prosyak <[e-mail]>
Signed-off-by: Srinivasan Shanmugam <[e-mail]>
Acked-by: Alex Deucher <[e-mail]>
Signed-off-by: Alex Deucher <[e-mail]>

drivers/gpu/drm/amd/amdgpu/gfx_v10_0_cleaner_shader.h

···
 	0xd70f6a01, 0x000202ff,
 	0x00000400, 0x80828102,
 	0xbf84fff7, 0xbefc03ff,
-	0x00000068, 0xbe803080,
-	0xbe813080, 0xbe823080,
-	0xbe833080, 0x80fc847c,
+	0x00000068, 0xbe803000,
+	0xbe813000, 0xbe823000,
+	0xbe833000, 0x80fc847c,
 	0xbf84fffa, 0xbeea0480,
 	0xbeec0480, 0xbeee0480,
 	0xbef00480, 0xbef20480,

drivers/gpu/drm/amd/amdgpu/gfx_v10_1_10_cleaner_shader.asm

···
   type(CS)
   wave_size(32)
 // Note: original source code from SQ team
-
 //
 // Create 32 waves in a threadgroup (CS waves)
 // Each allocates 64 VGPRs
···
   s_sub_u32     s2, s2, 8
   s_cbranch_scc0  label_0005
   //
-  s_mov_b32     s2, 0x80000000                     // Bit31 is first_wave
-  s_and_b32     s2, s2, s0                                  // sgpr0 has tg_size (first_wave) term as in ucode only COMPUTE_PGM_RSRC2.tg_size_en is set
+  s_mov_b32     s2, 0x80000000                       // Bit31 is first_wave
+  s_and_b32     s2, s2, s1                           // sgpr0 has tg_size (first_wave) term as in ucode only COMPUTE_PGM_RSRC2.tg_size_en is set
   s_cbranch_scc0  label_0023                         // Clean LDS if its first wave of ThreadGroup/WorkGroup
   // CLEAR LDS
   //
···
 label_0023:
   s_mov_b32     m0, 0x00000068  // Loop 108/4=27 times  (loop unrolled for performance)
 label_sgpr_loop:
-  s_movreld_b32     s0, 0
-  s_movreld_b32     s1, 0
-  s_movreld_b32     s2, 0
-  s_movreld_b32     s3, 0
+  s_movreld_b32     s0, s0
+  s_movreld_b32     s1, s0
+  s_movreld_b32     s2, s0
+  s_movreld_b32     s3, s0
   s_sub_u32         m0, m0, 4
   s_cbranch_scc0  label_sgpr_loop