drm/amdkfd: Unify gfx9/gfx10 context save area layouts

Add some padding before the MODE register in the HWREGs block to
preserve the same layout as gfx9. This simplifies implementation of a
user-mode debugger.
Signed-off-by: Laurent Morichetti <laurent.morichetti@amd.com>
Reviewed-by: Jay Cornwall <jay.cornwall@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm

@@ -407,6 +407,7 @@ L_SAVE_HWREG:
 
 	// Not used on Sienna_Cichlid but keep layout same for debugger.
 	write_hwreg_to_mem(s_save_xnack_mask, s_save_buf_rsrc0, s_save_mem_offset)
+	s_add_u32	s_save_mem_offset, s_save_mem_offset, 4
 
 	s_getreg_b32	s_save_m0, hwreg(HW_REG_MODE)
 	write_hwreg_to_mem(s_save_m0, s_save_buf_rsrc0, s_save_mem_offset)
@@ -912,6 +913,7 @@ L_RESTORE_HWREG:
 	read_hwreg_from_mem(s_restore_status, s_restore_buf_rsrc0, s_restore_mem_offset)
 	read_hwreg_from_mem(s_restore_trapsts, s_restore_buf_rsrc0, s_restore_mem_offset)
 	read_hwreg_from_mem(s_restore_xnack_mask, s_restore_buf_rsrc0, s_restore_mem_offset)
+	s_add_u32	s_restore_mem_offset, s_restore_mem_offset, 4
 	read_hwreg_from_mem(s_restore_mode, s_restore_buf_rsrc0, s_restore_mem_offset)
 	read_hwreg_from_mem(s_restore_flat_scratch, s_restore_buf_rsrc0, s_restore_mem_offset)
 	s_waitcnt	lgkmcnt(0)