Skip to contentSkip to conversation

Fix stack usage of DML

PATCH In review Harry Wentland 5 participants 11 messages Sep 9, 2021 – Sep 15, 2021

Subsystems (4)
lore
Share & tools
lore

Harry Wentland

author maintainer
#0 lore raw

cover Fix stack usage of DML

With the '-Werror' enablement patch the amdgpu build was failing on clang builds because a bunch of functions were blowing past the 1024 byte stack frame default. Due to this we also noticed that a lot of functions were passing large structs by value instead of by pointer.

This series attempts to fix this.

There is still one remaining function that blows the 1024 limit by 40 bytes:

drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn21/display_mode_vba_21.c:3397:6:

error: stack frame size of 1064 bytes in function 'dml21_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than=]

This will be a slightly more challenging fix but I'll see if we can get it below 1024 by breaking it into smaller functions.

With this series I can build amdgpu with CC=clang and a stack frame limit of 1064.

This series boots on a Radeon RX 5500 XT.

Harry Wentland (4):
drm/amd/display: Pass display_pipe_params_st as const in DML drm/amd/display: Pass all structs in display_rq_dlg_helpers by pointer drm/amd/display: Fix rest of pass-by-value structs in DML drm/amd/display: Allocate structs needed by dcn_bw_calc_rq_dlg_ttu in
pipe_ctx

 .../gpu/drm/amd/display/dc/calcs/dcn_calcs.c  |  55 ++--
 .../drm/amd/display/dc/dcn20/dcn20_resource.c |   2 +-
 .../dc/dml/dcn20/display_rq_dlg_calc_20.c     | 158 +++++------
 .../dc/dml/dcn20/display_rq_dlg_calc_20.h     |   4 +-
 .../dc/dml/dcn20/display_rq_dlg_calc_20v2.c   | 156 +++++------
 .../dc/dml/dcn20/display_rq_dlg_calc_20v2.h   |   4 +-
 .../dc/dml/dcn21/display_rq_dlg_calc_21.c     | 156 +++++------
 .../dc/dml/dcn21/display_rq_dlg_calc_21.h     |   4 +-
 .../dc/dml/dcn30/display_rq_dlg_calc_30.c     | 132 ++++-----
 .../dc/dml/dcn30/display_rq_dlg_calc_30.h     |   4 +-
 .../dc/dml/dcn31/display_rq_dlg_calc_31.c     | 166 ++++++------
 .../dc/dml/dcn31/display_rq_dlg_calc_31.h     |   4 +-
 .../drm/amd/display/dc/dml/display_mode_lib.h |   4 +-
 .../display/dc/dml/display_rq_dlg_helpers.c   | 256 +++++++++---------
 .../display/dc/dml/display_rq_dlg_helpers.h   |  20 +-
 .../display/dc/dml/dml1_display_rq_dlg_calc.c | 246 ++++++++---------
 .../display/dc/dml/dml1_display_rq_dlg_calc.h |  10 +-
 .../gpu/drm/amd/display/dc/inc/core_types.h   |   3 +
 18 files changed, 695 insertions(+), 689 deletions(-)
Signature · 1 line
2.33.0
author maintainer
#1 lore raw

PATCH 1/4 drm/amd/display: Pass display_pipe_params_st as const in DML

[Why]
This neither needs to be on the stack nor passed by value to each function call. In fact, when building with clang it seems to break the Linux's default 1024 byte stack frame limit.

[How]
We can simply pass this as a const pointer.

Fixes 3fe617ccafd6 ("Enable '-Werror' by default for all kernel builds")
Cc Nick Desaulniers <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Linux Kernel Mailing List <[email protected]>
Cc Arnd Bergmann <[email protected]>
Cc Leo Li <[email protected]>
Cc Alex Deucher <[email protected]>
Cc Christian König <[email protected]>
Cc Xinhui Pan <[email protected]>
Cc Nathan Chancellor <[email protected]>
Cc Guenter Roeck <[email protected]>
---
 .../drm/amd/display/dc/dcn20/dcn20_resource.c |  2 +-
 .../dc/dml/dcn20/display_rq_dlg_calc_20.c     |  6 +-
 .../dc/dml/dcn20/display_rq_dlg_calc_20.h     |  4 +-
 .../dc/dml/dcn20/display_rq_dlg_calc_20v2.c   |  6 +-
 .../dc/dml/dcn20/display_rq_dlg_calc_20v2.h   |  4 +-
 .../dc/dml/dcn21/display_rq_dlg_calc_21.c     | 62 ++++++++--------
 .../dc/dml/dcn21/display_rq_dlg_calc_21.h     |  4 +-
 .../dc/dml/dcn30/display_rq_dlg_calc_30.c     | 72 +++++++++----------
 .../dc/dml/dcn30/display_rq_dlg_calc_30.h     |  4 +-
 .../dc/dml/dcn31/display_rq_dlg_calc_31.c     | 68 +++++++++---------
 .../dc/dml/dcn31/display_rq_dlg_calc_31.h     |  4 +-
 .../drm/amd/display/dc/dml/display_mode_lib.h |  4 +-
 12 files changed, 120 insertions(+), 120 deletions(-)
drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +1 -1
index e3e01b17c164..4389b36f0760 100644
@@ -3152,7 +3152,7 @@ void dcn20_calculate_dlg_params(
 
 		context->bw_ctx.dml.funcs.rq_dlg_get_rq_reg(&context->bw_ctx.dml,
 				&context->res_ctx.pipe_ctx[i].rq_regs,
-				pipes[pipe_idx].pipe);
+				&pipes[pipe_idx].pipe);
 		pipe_idx++;
 	}
 }
drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c +3 -3
index 2091dd8c252d..8c168f348a27 100644
@@ -768,12 +768,12 @@ static void dml20_rq_dlg_get_rq_params(struct display_mode_lib *mode_lib,
 
 void dml20_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
 		display_rq_regs_st *rq_regs,
-		const display_pipe_params_st pipe_param)
+		const display_pipe_params_st *pipe_param)
 {
 	display_rq_params_st rq_param = {0};
 
 	memset(rq_regs, 0, sizeof(*rq_regs));
-	dml20_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param.src);
+	dml20_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param->src);
 	extract_rq_regs(mode_lib, rq_regs, rq_param);
 
 	print__rq_regs_st(mode_lib, *rq_regs);
@@ -1549,7 +1549,7 @@ static void dml20_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib,
 void dml20_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib,
 		display_dlg_regs_st *dlg_regs,
 		display_ttu_regs_st *ttu_regs,
-		display_e2e_pipe_params_st *e2e_pipe_param,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
 		const unsigned int num_pipes,
 		const unsigned int pipe_idx,
 		const bool cstate_en,
drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.h +2 -2
index d0b90947f540..8b23867e97c1 100644
@@ -43,7 +43,7 @@ struct display_mode_lib;
 void dml20_rq_dlg_get_rq_reg(
 		struct display_mode_lib *mode_lib,
 		display_rq_regs_st *rq_regs,
-		const display_pipe_params_st pipe_param);
+		const display_pipe_params_st *pipe_param);
 
 
 // Function: dml_rq_dlg_get_dlg_reg
@@ -61,7 +61,7 @@ void dml20_rq_dlg_get_dlg_reg(
 		struct display_mode_lib *mode_lib,
 		display_dlg_regs_st *dlg_regs,
 		display_ttu_regs_st *ttu_regs,
-		display_e2e_pipe_params_st *e2e_pipe_param,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
 		const unsigned int num_pipes,
 		const unsigned int pipe_idx,
 		const bool cstate_en,
drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c +3 -3
index 1a0c14e465fa..26ececfd40cd 100644
@@ -768,12 +768,12 @@ static void dml20v2_rq_dlg_get_rq_params(struct display_mode_lib *mode_lib,
 
 void dml20v2_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
 		display_rq_regs_st *rq_regs,
-		const display_pipe_params_st pipe_param)
+		const display_pipe_params_st *pipe_param)
 {
 	display_rq_params_st rq_param = {0};
 
 	memset(rq_regs, 0, sizeof(*rq_regs));
-	dml20v2_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param.src);
+	dml20v2_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param->src);
 	extract_rq_regs(mode_lib, rq_regs, rq_param);
 
 	print__rq_regs_st(mode_lib, *rq_regs);
@@ -1550,7 +1550,7 @@ static void dml20v2_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib,
 void dml20v2_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib,
 		display_dlg_regs_st *dlg_regs,
 		display_ttu_regs_st *ttu_regs,
-		display_e2e_pipe_params_st *e2e_pipe_param,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
 		const unsigned int num_pipes,
 		const unsigned int pipe_idx,
 		const bool cstate_en,
drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.h +2 -2
index 27cf8bed9376..2b4e46ea1c3d 100644
@@ -43,7 +43,7 @@ struct display_mode_lib;
 void dml20v2_rq_dlg_get_rq_reg(
 		struct display_mode_lib *mode_lib,
 		display_rq_regs_st *rq_regs,
-		const display_pipe_params_st pipe_param);
+		const display_pipe_params_st *pipe_param);
 
 
 // Function: dml_rq_dlg_get_dlg_reg
@@ -61,7 +61,7 @@ void dml20v2_rq_dlg_get_dlg_reg(
 		struct display_mode_lib *mode_lib,
 		display_dlg_regs_st *dlg_regs,
 		display_ttu_regs_st *ttu_regs,
-		display_e2e_pipe_params_st *e2e_pipe_param,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
 		const unsigned int num_pipes,
 		const unsigned int pipe_idx,
 		const bool cstate_en,
drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c +31 -31

133 lines not on this page · raw message

drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.h +2 -2
index e8f7785e3fc6..af6ad0ca9cf8 100644
@@ -44,7 +44,7 @@ struct display_mode_lib;
 void dml21_rq_dlg_get_rq_reg(
 		struct display_mode_lib *mode_lib,
 		display_rq_regs_st *rq_regs,
-		const display_pipe_params_st pipe_param);
+		const display_pipe_params_st *pipe_param);
 
 // Function: dml_rq_dlg_get_dlg_reg
 //   Calculate and return DLG and TTU register struct given the system setting
@@ -61,7 +61,7 @@ void dml21_rq_dlg_get_dlg_reg(
 		struct display_mode_lib *mode_lib,
 		display_dlg_regs_st *dlg_regs,
 		display_ttu_regs_st *ttu_regs,
-		display_e2e_pipe_params_st *e2e_pipe_param,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
 		const unsigned int num_pipes,
 		const unsigned int pipe_idx,
 		const bool cstate_en,
drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c +36 -36

144 lines not on this page · raw message

drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.h +2 -2
index c04965cceff3..625e41f8d575 100644
@@ -41,7 +41,7 @@ struct display_mode_lib;
 //            See also: <display_rq_regs_st>
 void dml30_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
 		display_rq_regs_st *rq_regs,
-		const display_pipe_params_st pipe_param);
+		const display_pipe_params_st *pipe_param);
 
 // Function: dml_rq_dlg_get_dlg_reg
 //   Calculate and return DLG and TTU register struct given the system setting
@@ -57,7 +57,7 @@ void dml30_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
 void dml30_rq_dlg_get_dlg_reg(struct display_mode_lib             *mode_lib,
 		display_dlg_regs_st          *dlg_regs,
 		display_ttu_regs_st          *ttu_regs,
-		display_e2e_pipe_params_st   *e2e_pipe_param,
+		const display_e2e_pipe_params_st   *e2e_pipe_param,
 		const unsigned int            num_pipes,
 		const unsigned int            pipe_idx,
 		const bool                    cstate_en,
drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c +34 -34

133 lines not on this page · raw message

drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.h +2 -2
index adf8518f761f..8ee991351699 100644
@@ -41,7 +41,7 @@ struct display_mode_lib;
 //            See also: <display_rq_regs_st>
 void dml31_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
 		display_rq_regs_st *rq_regs,
-		const display_pipe_params_st pipe_param);
+		const display_pipe_params_st *pipe_param);
 
 // Function: dml_rq_dlg_get_dlg_reg
 //   Calculate and return DLG and TTU register struct given the system setting
@@ -57,7 +57,7 @@ void dml31_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
 void dml31_rq_dlg_get_dlg_reg(struct display_mode_lib             *mode_lib,
 		display_dlg_regs_st          *dlg_regs,
 		display_ttu_regs_st          *ttu_regs,
-		display_e2e_pipe_params_st   *e2e_pipe_param,
+		const display_e2e_pipe_params_st *e2e_pipe_param,
 		const unsigned int            num_pipes,
 		const unsigned int            pipe_idx,
 		const bool                    cstate_en,
drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h +2 -2
index d42a0aeca6be..72b1957022aa 100644
@@ -49,7 +49,7 @@ struct dml_funcs {
 			struct display_mode_lib *mode_lib,
 			display_dlg_regs_st *dlg_regs,
 			display_ttu_regs_st *ttu_regs,
-			display_e2e_pipe_params_st *e2e_pipe_param,
+			const display_e2e_pipe_params_st *e2e_pipe_param,
 			const unsigned int num_pipes,
 			const unsigned int pipe_idx,
 			const bool cstate_en,
@@ -60,7 +60,7 @@ struct dml_funcs {
 	void (*rq_dlg_get_rq_reg)(
 		struct display_mode_lib *mode_lib,
 		display_rq_regs_st *rq_regs,
-		const display_pipe_params_st pipe_param);
+		const display_pipe_params_st *pipe_param);
 	void (*recalculate)(struct display_mode_lib *mode_lib);
 	void (*validate)(struct display_mode_lib *mode_lib);
 };
Signature · 1 line
2.33.0
#2 lore raw

Re: [1/4] drm/amd/display: Pass display_pipe_params_st as const in DML

Hi Harry,

This patch fixes the following CID's. Thanks.

Addresses-Coverity-ID: 1424031: ("Big parameter passed by value")
Addresses-Coverity-ID: 1423970: ("Big parameter passed by value")
Addresses-Coverity-ID: 1423941: ("Big parameter passed by value")
Addresses-Coverity-ID: 1451742: ("Big parameter passed by value")
Addresses-Coverity-ID: 1451887: ("Big parameter passed by value")
Addresses-Coverity-ID: 1454146: ("Big parameter passed by value")
Addresses-Coverity-ID: 1454152: ("Big parameter passed by value")
Addresses-Coverity-ID: 1454413: ("Big parameter passed by value")
Addresses-Coverity-ID: 1466144: ("Big parameter passed by value")
Addresses-Coverity-ID: 1487237: ("Big parameter passed by value")

-- Anson

author maintainer
#3 lore raw

PATCH 2/4 drm/amd/display: Pass all structs in display_rq_dlg_helpers by pointer

In reply to Harry Wentland

Passing structs adds a lot of overhead. We don't ever want to pass anything bigger than primitives by value.

Cc Nick Desaulniers <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Linux Kernel Mailing List <[email protected]>
Cc Arnd Bergmann <[email protected]>
Cc Leo Li <[email protected]>
Cc Alex Deucher <[email protected]>
Cc Christian König <[email protected]>
Cc Xinhui Pan <[email protected]>
Cc Nathan Chancellor <[email protected]>
Cc Guenter Roeck <[email protected]>
---
 .../dc/dml/dcn20/display_rq_dlg_calc_20.c     |  32 +--
 .../dc/dml/dcn20/display_rq_dlg_calc_20v2.c   |  32 +--
 .../dc/dml/dcn21/display_rq_dlg_calc_21.c     |  32 +--
 .../dc/dml/dcn30/display_rq_dlg_calc_30.c     |  32 +--
 .../dc/dml/dcn31/display_rq_dlg_calc_31.c     |  32 +--
 .../display/dc/dml/display_rq_dlg_helpers.c   | 256 +++++++++---------
 .../display/dc/dml/display_rq_dlg_helpers.h   |  20 +-
 .../display/dc/dml/dml1_display_rq_dlg_calc.c |  28 +-
 8 files changed, 232 insertions(+), 232 deletions(-)
drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c +16 -16
index 8c168f348a27..6a4f479c5aab 100644
@@ -164,26 +164,26 @@ static unsigned int get_blk_size_bytes(const enum source_macro_tile_size tile_si
 
 static void extract_rq_sizing_regs(struct display_mode_lib *mode_lib,
 		display_data_rq_regs_st *rq_regs,
-		const display_data_rq_sizing_params_st rq_sizing)
+		const display_data_rq_sizing_params_st *rq_sizing)
 {
 	dml_print("DML_DLG: %s: rq_sizing param\n", __func__);
 	print__data_rq_sizing_params_st(mode_lib, rq_sizing);
 
-	rq_regs->chunk_size = dml_log2(rq_sizing.chunk_bytes) - 10;
+	rq_regs->chunk_size = dml_log2(rq_sizing->chunk_bytes) - 10;
 
-	if (rq_sizing.min_chunk_bytes == 0)
+	if (rq_sizing->min_chunk_bytes == 0)
 		rq_regs->min_chunk_size = 0;
 	else
-		rq_regs->min_chunk_size = dml_log2(rq_sizing.min_chunk_bytes) - 8 + 1;
+		rq_regs->min_chunk_size = dml_log2(rq_sizing->min_chunk_bytes) - 8 + 1;
 
-	rq_regs->meta_chunk_size = dml_log2(rq_sizing.meta_chunk_bytes) - 10;
-	if (rq_sizing.min_meta_chunk_bytes == 0)
+	rq_regs->meta_chunk_size = dml_log2(rq_sizing->meta_chunk_bytes) - 10;
+	if (rq_sizing->min_meta_chunk_bytes == 0)
 		rq_regs->min_meta_chunk_size = 0;
 	else
-		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing.min_meta_chunk_bytes) - 6 + 1;
+		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing->min_meta_chunk_bytes) - 6 + 1;
 
-	rq_regs->dpte_group_size = dml_log2(rq_sizing.dpte_group_bytes) - 6;
-	rq_regs->mpte_group_size = dml_log2(rq_sizing.mpte_group_bytes) - 6;
+	rq_regs->dpte_group_size = dml_log2(rq_sizing->dpte_group_bytes) - 6;
+	rq_regs->mpte_group_size = dml_log2(rq_sizing->mpte_group_bytes) - 6;
 }
 
 static void extract_rq_regs(struct display_mode_lib *mode_lib,
@@ -193,13 +193,13 @@ static void extract_rq_regs(struct display_mode_lib *mode_lib,
 	unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024;
 	unsigned int detile_buf_plane1_addr = 0;
 
-	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), rq_param.sizing.rq_l);
+	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), &rq_param.sizing.rq_l);
 
 	rq_regs->rq_regs_l.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_l.dpte_row_height),
 			1) - 3;
 
 	if (rq_param.yuv420) {
-		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), rq_param.sizing.rq_c);
+		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), &rq_param.sizing.rq_c);
 		rq_regs->rq_regs_c.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_c.dpte_row_height),
 				1) - 3;
 	}
@@ -763,7 +763,7 @@ static void dml20_rq_dlg_get_rq_params(struct display_mode_lib *mode_lib,
 
 	// calculate how to split the det buffer space between luma and chroma
 	handle_det_buf_split(mode_lib, rq_param, pipe_src_param);
-	print__rq_params_st(mode_lib, *rq_param);
+	print__rq_params_st(mode_lib, rq_param);
 }
 
 void dml20_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
@@ -776,7 +776,7 @@ void dml20_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
 	dml20_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param->src);
 	extract_rq_regs(mode_lib, rq_regs, rq_param);
 
-	print__rq_regs_st(mode_lib, *rq_regs);
+	print__rq_regs_st(mode_lib, rq_regs);
 }
 
 // Note: currently taken in as is.
@@ -1542,8 +1542,8 @@ static void dml20_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib,
 	disp_ttu_regs->min_ttu_vblank = min_ttu_vblank * refclk_freq_in_mhz;
 	ASSERT(disp_ttu_regs->min_ttu_vblank < dml_pow(2, 24));
 
-	print__ttu_regs_st(mode_lib, *disp_ttu_regs);
-	print__dlg_regs_st(mode_lib, *disp_dlg_regs);
+	print__ttu_regs_st(mode_lib, disp_ttu_regs);
+	print__dlg_regs_st(mode_lib, disp_dlg_regs);
 }
 
 void dml20_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib,
@@ -1579,7 +1579,7 @@ void dml20_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib,
 	dlg_sys_param.t_srx_delay_us = mode_lib->ip.dcfclk_cstate_latency
 			/ dlg_sys_param.deepsleep_dcfclk_mhz; // TODO: Deprecated
 
-	print__dlg_sys_params_st(mode_lib, dlg_sys_param);
+	print__dlg_sys_params_st(mode_lib, &dlg_sys_param);
 
 	// system parameter calculation done
 
drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c +16 -16
index 26ececfd40cd..b5f33f05c8e8 100644
@@ -164,26 +164,26 @@ static unsigned int get_blk_size_bytes(const enum source_macro_tile_size tile_si
 
 static void extract_rq_sizing_regs(struct display_mode_lib *mode_lib,
 		display_data_rq_regs_st *rq_regs,
-		const display_data_rq_sizing_params_st rq_sizing)
+		const display_data_rq_sizing_params_st *rq_sizing)
 {
 	dml_print("DML_DLG: %s: rq_sizing param\n", __func__);
 	print__data_rq_sizing_params_st(mode_lib, rq_sizing);
 
-	rq_regs->chunk_size = dml_log2(rq_sizing.chunk_bytes) - 10;
+	rq_regs->chunk_size = dml_log2(rq_sizing->chunk_bytes) - 10;
 
-	if (rq_sizing.min_chunk_bytes == 0)
+	if (rq_sizing->min_chunk_bytes == 0)
 		rq_regs->min_chunk_size = 0;
 	else
-		rq_regs->min_chunk_size = dml_log2(rq_sizing.min_chunk_bytes) - 8 + 1;
+		rq_regs->min_chunk_size = dml_log2(rq_sizing->min_chunk_bytes) - 8 + 1;
 
-	rq_regs->meta_chunk_size = dml_log2(rq_sizing.meta_chunk_bytes) - 10;
-	if (rq_sizing.min_meta_chunk_bytes == 0)
+	rq_regs->meta_chunk_size = dml_log2(rq_sizing->meta_chunk_bytes) - 10;
+	if (rq_sizing->min_meta_chunk_bytes == 0)
 		rq_regs->min_meta_chunk_size = 0;
 	else
-		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing.min_meta_chunk_bytes) - 6 + 1;
+		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing->min_meta_chunk_bytes) - 6 + 1;
 
-	rq_regs->dpte_group_size = dml_log2(rq_sizing.dpte_group_bytes) - 6;
-	rq_regs->mpte_group_size = dml_log2(rq_sizing.mpte_group_bytes) - 6;
+	rq_regs->dpte_group_size = dml_log2(rq_sizing->dpte_group_bytes) - 6;
+	rq_regs->mpte_group_size = dml_log2(rq_sizing->mpte_group_bytes) - 6;
 }
 
 static void extract_rq_regs(struct display_mode_lib *mode_lib,
@@ -193,13 +193,13 @@ static void extract_rq_regs(struct display_mode_lib *mode_lib,
 	unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024;
 	unsigned int detile_buf_plane1_addr = 0;
 
-	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), rq_param.sizing.rq_l);
+	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), &rq_param.sizing.rq_l);
 
 	rq_regs->rq_regs_l.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_l.dpte_row_height),
 			1) - 3;
 
 	if (rq_param.yuv420) {
-		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), rq_param.sizing.rq_c);
+		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), &rq_param.sizing.rq_c);
 		rq_regs->rq_regs_c.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_c.dpte_row_height),
 				1) - 3;
 	}
@@ -763,7 +763,7 @@ static void dml20v2_rq_dlg_get_rq_params(struct display_mode_lib *mode_lib,
 
 	// calculate how to split the det buffer space between luma and chroma
 	handle_det_buf_split(mode_lib, rq_param, pipe_src_param);
-	print__rq_params_st(mode_lib, *rq_param);
+	print__rq_params_st(mode_lib, rq_param);
 }
 
 void dml20v2_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
@@ -776,7 +776,7 @@ void dml20v2_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
 	dml20v2_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param->src);
 	extract_rq_regs(mode_lib, rq_regs, rq_param);
 
-	print__rq_regs_st(mode_lib, *rq_regs);
+	print__rq_regs_st(mode_lib, rq_regs);
 }
 
 // Note: currently taken in as is.
@@ -1543,8 +1543,8 @@ static void dml20v2_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib,
 	disp_ttu_regs->min_ttu_vblank = min_ttu_vblank * refclk_freq_in_mhz;
 	ASSERT(disp_ttu_regs->min_ttu_vblank < dml_pow(2, 24));
 
-	print__ttu_regs_st(mode_lib, *disp_ttu_regs);
-	print__dlg_regs_st(mode_lib, *disp_dlg_regs);
+	print__ttu_regs_st(mode_lib, disp_ttu_regs);
+	print__dlg_regs_st(mode_lib, disp_dlg_regs);
 }
 
 void dml20v2_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib,
@@ -1580,7 +1580,7 @@ void dml20v2_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib,
 	dlg_sys_param.t_srx_delay_us = mode_lib->ip.dcfclk_cstate_latency
 			/ dlg_sys_param.deepsleep_dcfclk_mhz; // TODO: Deprecated
 
-	print__dlg_sys_params_st(mode_lib, dlg_sys_param);
+	print__dlg_sys_params_st(mode_lib, &dlg_sys_param);
 
 	// system parameter calculation done
 
drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c +16 -16
index 736978c4d40a..661fbb622824 100644
@@ -141,26 +141,26 @@ static unsigned int get_blk_size_bytes(const enum source_macro_tile_size tile_si
 static void extract_rq_sizing_regs(
 		struct display_mode_lib *mode_lib,
 		display_data_rq_regs_st *rq_regs,
-		const display_data_rq_sizing_params_st rq_sizing)
+		const display_data_rq_sizing_params_st *rq_sizing)
 {
 	dml_print("DML_DLG: %s: rq_sizing param\n", __func__);
 	print__data_rq_sizing_params_st(mode_lib, rq_sizing);
 
-	rq_regs->chunk_size = dml_log2(rq_sizing.chunk_bytes) - 10;
+	rq_regs->chunk_size = dml_log2(rq_sizing->chunk_bytes) - 10;
 
-	if (rq_sizing.min_chunk_bytes == 0)
+	if (rq_sizing->min_chunk_bytes == 0)
 		rq_regs->min_chunk_size = 0;
 	else
-		rq_regs->min_chunk_size = dml_log2(rq_sizing.min_chunk_bytes) - 8 + 1;
+		rq_regs->min_chunk_size = dml_log2(rq_sizing->min_chunk_bytes) - 8 + 1;
 
-	rq_regs->meta_chunk_size = dml_log2(rq_sizing.meta_chunk_bytes) - 10;
-	if (rq_sizing.min_meta_chunk_bytes == 0)
+	rq_regs->meta_chunk_size = dml_log2(rq_sizing->meta_chunk_bytes) - 10;
+	if (rq_sizing->min_meta_chunk_bytes == 0)
 		rq_regs->min_meta_chunk_size = 0;
 	else
-		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing.min_meta_chunk_bytes) - 6 + 1;
+		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing->min_meta_chunk_bytes) - 6 + 1;
 
-	rq_regs->dpte_group_size = dml_log2(rq_sizing.dpte_group_bytes) - 6;
-	rq_regs->mpte_group_size = dml_log2(rq_sizing.mpte_group_bytes) - 6;
+	rq_regs->dpte_group_size = dml_log2(rq_sizing->dpte_group_bytes) - 6;
+	rq_regs->mpte_group_size = dml_log2(rq_sizing->mpte_group_bytes) - 6;
 }
 
 static void extract_rq_regs(
@@ -171,14 +171,14 @@ static void extract_rq_regs(
 	unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024;
 	unsigned int detile_buf_plane1_addr = 0;
 
-	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), rq_param.sizing.rq_l);
+	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), &rq_param.sizing.rq_l);
 
 	rq_regs->rq_regs_l.pte_row_height_linear = dml_floor(
 			dml_log2(rq_param.dlg.rq_l.dpte_row_height),
 			1) - 3;
 
 	if (rq_param.yuv420) {
-		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), rq_param.sizing.rq_c);
+		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), &rq_param.sizing.rq_c);
 		rq_regs->rq_regs_c.pte_row_height_linear = dml_floor(
 				dml_log2(rq_param.dlg.rq_c.dpte_row_height),
 				1) - 3;
@@ -807,7 +807,7 @@ static void dml_rq_dlg_get_rq_params(
 
 	// calculate how to split the det buffer space between luma and chroma
 	handle_det_buf_split(mode_lib, rq_param, pipe_param->src);
-	print__rq_params_st(mode_lib, *rq_param);
+	print__rq_params_st(mode_lib, rq_param);
 }
 
 void dml21_rq_dlg_get_rq_reg(
@@ -821,7 +821,7 @@ void dml21_rq_dlg_get_rq_reg(
 	dml_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param);
 	extract_rq_regs(mode_lib, rq_regs, rq_param);
 
-	print__rq_regs_st(mode_lib, *rq_regs);
+	print__rq_regs_st(mode_lib, rq_regs);
 }
 
 // Note: currently taken in as is.
@@ -1650,8 +1650,8 @@ static void dml_rq_dlg_get_dlg_params(
 	disp_ttu_regs->min_ttu_vblank = min_ttu_vblank * refclk_freq_in_mhz;
 	ASSERT(disp_ttu_regs->min_ttu_vblank < dml_pow(2, 24));
 
-	print__ttu_regs_st(mode_lib, *disp_ttu_regs);
-	print__dlg_regs_st(mode_lib, *disp_dlg_regs);
+	print__ttu_regs_st(mode_lib, disp_ttu_regs);
+	print__dlg_regs_st(mode_lib, disp_dlg_regs);
 }
 
 void dml21_rq_dlg_get_dlg_reg(
@@ -1691,7 +1691,7 @@ void dml21_rq_dlg_get_dlg_reg(
 	dlg_sys_param.t_srx_delay_us = mode_lib->ip.dcfclk_cstate_latency
 			/ dlg_sys_param.deepsleep_dcfclk_mhz; // TODO: Deprecated
 
-	print__dlg_sys_params_st(mode_lib, dlg_sys_param);
+	print__dlg_sys_params_st(mode_lib, &dlg_sys_param);
 
 	// system parameter calculation done
 
drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c +16 -16
index 2120e0941a09..1535b01a8c5d 100644
@@ -89,26 +89,26 @@ static unsigned int get_blk_size_bytes(const enum source_macro_tile_size tile_si
 
 static void extract_rq_sizing_regs(struct display_mode_lib *mode_lib,
 	display_data_rq_regs_st *rq_regs,
-	const display_data_rq_sizing_params_st rq_sizing)
+	const display_data_rq_sizing_params_st *rq_sizing)
 {
 	dml_print("DML_DLG: %s: rq_sizing param\n", __func__);
 	print__data_rq_sizing_params_st(mode_lib, rq_sizing);
 
-	rq_regs->chunk_size = dml_log2(rq_sizing.chunk_bytes) - 10;
+	rq_regs->chunk_size = dml_log2(rq_sizing->chunk_bytes) - 10;
 
-	if (rq_sizing.min_chunk_bytes == 0)
+	if (rq_sizing->min_chunk_bytes == 0)
 		rq_regs->min_chunk_size = 0;
 	else
-		rq_regs->min_chunk_size = dml_log2(rq_sizing.min_chunk_bytes) - 8 + 1;
+		rq_regs->min_chunk_size = dml_log2(rq_sizing->min_chunk_bytes) - 8 + 1;
 
-	rq_regs->meta_chunk_size = dml_log2(rq_sizing.meta_chunk_bytes) - 10;
-	if (rq_sizing.min_meta_chunk_bytes == 0)
+	rq_regs->meta_chunk_size = dml_log2(rq_sizing->meta_chunk_bytes) - 10;
+	if (rq_sizing->min_meta_chunk_bytes == 0)
 		rq_regs->min_meta_chunk_size = 0;
 	else
-		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing.min_meta_chunk_bytes) - 6 + 1;
+		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing->min_meta_chunk_bytes) - 6 + 1;
 
-	rq_regs->dpte_group_size = dml_log2(rq_sizing.dpte_group_bytes) - 6;
-	rq_regs->mpte_group_size = dml_log2(rq_sizing.mpte_group_bytes) - 6;
+	rq_regs->dpte_group_size = dml_log2(rq_sizing->dpte_group_bytes) - 6;
+	rq_regs->mpte_group_size = dml_log2(rq_sizing->mpte_group_bytes) - 6;
 }
 
 static void extract_rq_regs(struct display_mode_lib *mode_lib,
@@ -118,13 +118,13 @@ static void extract_rq_regs(struct display_mode_lib *mode_lib,
 	unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024;
 	unsigned int detile_buf_plane1_addr = 0;
 
-	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), rq_param.sizing.rq_l);
+	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), &rq_param.sizing.rq_l);
 
 	rq_regs->rq_regs_l.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_l.dpte_row_height),
 		1) - 3;
 
 	if (rq_param.yuv420) {
-		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), rq_param.sizing.rq_c);
+		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), &rq_param.sizing.rq_c);
 		rq_regs->rq_regs_c.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_c.dpte_row_height),
 			1) - 3;
 	}
@@ -872,7 +872,7 @@ static void dml_rq_dlg_get_rq_params(struct display_mode_lib *mode_lib,
 
 	// calculate how to split the det buffer space between luma and chroma
 	handle_det_buf_split(mode_lib, rq_param, pipe_param->src);
-	print__rq_params_st(mode_lib, *rq_param);
+	print__rq_params_st(mode_lib, rq_param);
 }
 
 void dml30_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
@@ -885,7 +885,7 @@ void dml30_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
 	dml_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param);
 	extract_rq_regs(mode_lib, rq_regs, rq_param);
 
-	print__rq_regs_st(mode_lib, *rq_regs);
+	print__rq_regs_st(mode_lib, rq_regs);
 }
 
 static void calculate_ttu_cursor(struct display_mode_lib *mode_lib,
@@ -1824,8 +1824,8 @@ static void dml_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib,
 	disp_ttu_regs->min_ttu_vblank = min_ttu_vblank * refclk_freq_in_mhz;
 	ASSERT(disp_ttu_regs->min_ttu_vblank < dml_pow(2, 24));
 
-	print__ttu_regs_st(mode_lib, *disp_ttu_regs);
-	print__dlg_regs_st(mode_lib, *disp_dlg_regs);
+	print__ttu_regs_st(mode_lib, disp_ttu_regs);
+	print__dlg_regs_st(mode_lib, disp_dlg_regs);
 }
 
 void dml30_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib,
@@ -1861,7 +1861,7 @@ void dml30_rq_dlg_get_dlg_reg(struct display_mode_lib *mode_lib,
 	dlg_sys_param.t_srx_delay_us = mode_lib->ip.dcfclk_cstate_latency
 		/ dlg_sys_param.deepsleep_dcfclk_mhz; // TODO: Deprecated
 
-	print__dlg_sys_params_st(mode_lib, dlg_sys_param);
+	print__dlg_sys_params_st(mode_lib, &dlg_sys_param);
 
 	// system parameter calculation done
 
drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c +16 -16
index 57bd4e3f8a82..442abae4bc8c 100644
@@ -175,25 +175,25 @@ static unsigned int get_blk_size_bytes(const enum source_macro_tile_size tile_si
 		return (4 * 1024);
 }
 
-static void extract_rq_sizing_regs(struct display_mode_lib *mode_lib, display_data_rq_regs_st *rq_regs, const display_data_rq_sizing_params_st rq_sizing)
+static void extract_rq_sizing_regs(struct display_mode_lib *mode_lib, display_data_rq_regs_st *rq_regs, const display_data_rq_sizing_params_st *rq_sizing)
 {
 	print__data_rq_sizing_params_st(mode_lib, rq_sizing);
 
-	rq_regs->chunk_size = dml_log2(rq_sizing.chunk_bytes) - 10;
+	rq_regs->chunk_size = dml_log2(rq_sizing->chunk_bytes) - 10;
 
-	if (rq_sizing.min_chunk_bytes == 0)
+	if (rq_sizing->min_chunk_bytes == 0)
 		rq_regs->min_chunk_size = 0;
 	else
-		rq_regs->min_chunk_size = dml_log2(rq_sizing.min_chunk_bytes) - 8 + 1;
+		rq_regs->min_chunk_size = dml_log2(rq_sizing->min_chunk_bytes) - 8 + 1;
 
-	rq_regs->meta_chunk_size = dml_log2(rq_sizing.meta_chunk_bytes) - 10;
-	if (rq_sizing.min_meta_chunk_bytes == 0)
+	rq_regs->meta_chunk_size = dml_log2(rq_sizing->meta_chunk_bytes) - 10;
+	if (rq_sizing->min_meta_chunk_bytes == 0)
 		rq_regs->min_meta_chunk_size = 0;
 	else
-		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing.min_meta_chunk_bytes) - 6 + 1;
+		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing->min_meta_chunk_bytes) - 6 + 1;
 
-	rq_regs->dpte_group_size = dml_log2(rq_sizing.dpte_group_bytes) - 6;
-	rq_regs->mpte_group_size = dml_log2(rq_sizing.mpte_group_bytes) - 6;
+	rq_regs->dpte_group_size = dml_log2(rq_sizing->dpte_group_bytes) - 6;
+	rq_regs->mpte_group_size = dml_log2(rq_sizing->mpte_group_bytes) - 6;
 }
 
 static void extract_rq_regs(struct display_mode_lib *mode_lib, display_rq_regs_st *rq_regs, const display_rq_params_st rq_param)
@@ -201,12 +201,12 @@ static void extract_rq_regs(struct display_mode_lib *mode_lib, display_rq_regs_s
 	unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024;
 	unsigned int detile_buf_plane1_addr = 0;
 
-	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), rq_param.sizing.rq_l);
+	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), &rq_param.sizing.rq_l);
 
 	rq_regs->rq_regs_l.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_l.dpte_row_height), 1) - 3;
 
 	if (rq_param.yuv420) {
-		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), rq_param.sizing.rq_c);
+		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), &rq_param.sizing.rq_c);
 		rq_regs->rq_regs_c.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_c.dpte_row_height), 1) - 3;
 	}
 
@@ -850,7 +850,7 @@ static void dml_rq_dlg_get_rq_params(struct display_mode_lib *mode_lib, display_
 
 	// calculate how to split the det buffer space between luma and chroma
 	handle_det_buf_split(mode_lib, rq_param, pipe_param->src);
-	print__rq_params_st(mode_lib, *rq_param);
+	print__rq_params_st(mode_lib, rq_param);
 }
 
 void dml31_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib, display_rq_regs_st *rq_regs, const display_pipe_params_st *pipe_param)
@@ -861,7 +861,7 @@ void dml31_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib, display_rq_regs_
 	dml_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param);
 	extract_rq_regs(mode_lib, rq_regs, rq_param);
 
-	print__rq_regs_st(mode_lib, *rq_regs);
+	print__rq_regs_st(mode_lib, rq_regs);
 }
 
 static void calculate_ttu_cursor(
@@ -1669,8 +1669,8 @@ static void dml_rq_dlg_get_dlg_params(
 	disp_ttu_regs->min_ttu_vblank = min_ttu_vblank * refclk_freq_in_mhz;
 	ASSERT(disp_ttu_regs->min_ttu_vblank < dml_pow(2, 24));
 
-	print__ttu_regs_st(mode_lib, *disp_ttu_regs);
-	print__dlg_regs_st(mode_lib, *disp_dlg_regs);
+	print__ttu_regs_st(mode_lib, disp_ttu_regs);
+	print__dlg_regs_st(mode_lib, disp_dlg_regs);
 }
 
 void dml31_rq_dlg_get_dlg_reg(
@@ -1699,7 +1699,7 @@ void dml31_rq_dlg_get_dlg_reg(
 	dlg_sys_param.total_flip_bw = get_total_immediate_flip_bw(mode_lib, e2e_pipe_param, num_pipes);
 	dlg_sys_param.total_flip_bytes = get_total_immediate_flip_bytes(mode_lib, e2e_pipe_param, num_pipes);
 
-	print__dlg_sys_params_st(mode_lib, dlg_sys_param);
+	print__dlg_sys_params_st(mode_lib, &dlg_sys_param);
 
 	// system parameter calculation done
 
drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.c +128 -128

499 lines not on this page · raw message

drivers/gpu/drm/amd/display/dc/dml/display_rq_dlg_helpers.h +10 -10
index 2555ef0358c2..ebcd717744e5 100644
@@ -31,16 +31,16 @@
 /* Function: Printer functions
  *  Print various struct
  */
-void print__rq_params_st(struct display_mode_lib *mode_lib, display_rq_params_st rq_param);
-void print__data_rq_sizing_params_st(struct display_mode_lib *mode_lib, display_data_rq_sizing_params_st rq_sizing);
-void print__data_rq_dlg_params_st(struct display_mode_lib *mode_lib, display_data_rq_dlg_params_st rq_dlg_param);
-void print__data_rq_misc_params_st(struct display_mode_lib *mode_lib, display_data_rq_misc_params_st rq_misc_param);
-void print__rq_dlg_params_st(struct display_mode_lib *mode_lib, display_rq_dlg_params_st rq_dlg_param);
-void print__dlg_sys_params_st(struct display_mode_lib *mode_lib, display_dlg_sys_params_st dlg_sys_param);
+void print__rq_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_rq_params_st *rq_param);
+void print__data_rq_sizing_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_sizing_params_st *rq_sizing);
+void print__data_rq_dlg_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_dlg_params_st *rq_dlg_param);
+void print__data_rq_misc_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_misc_params_st *rq_misc_param);
+void print__rq_dlg_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_rq_dlg_params_st *rq_dlg_param);
+void print__dlg_sys_params_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_dlg_sys_params_st *dlg_sys_param);
 
-void print__data_rq_regs_st(struct display_mode_lib *mode_lib, display_data_rq_regs_st data_rq_regs);
-void print__rq_regs_st(struct display_mode_lib *mode_lib, display_rq_regs_st rq_regs);
-void print__dlg_regs_st(struct display_mode_lib *mode_lib, display_dlg_regs_st dlg_regs);
-void print__ttu_regs_st(struct display_mode_lib *mode_lib, display_ttu_regs_st ttu_regs);
+void print__data_rq_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_data_rq_regs_st *rq_regs);
+void print__rq_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_rq_regs_st *rq_regs);
+void print__dlg_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_dlg_regs_st *dlg_regs);
+void print__ttu_regs_st(struct display_mode_lib *mode_lib, const struct _vcs_dpi_display_ttu_regs_st *ttu_regs);
 
 #endif
drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c +14 -14
index 8f2b1684c231..9bde3f961409 100644
@@ -206,26 +206,26 @@ static unsigned int get_blk_size_bytes(const enum source_macro_tile_size tile_si
 static void extract_rq_sizing_regs(
 		struct display_mode_lib *mode_lib,
 		struct _vcs_dpi_display_data_rq_regs_st *rq_regs,
-		const struct _vcs_dpi_display_data_rq_sizing_params_st rq_sizing)
+		const struct _vcs_dpi_display_data_rq_sizing_params_st *rq_sizing)
 {
 	DTRACE("DLG: %s: rq_sizing param", __func__);
 	print__data_rq_sizing_params_st(mode_lib, rq_sizing);
 
-	rq_regs->chunk_size = dml_log2(rq_sizing.chunk_bytes) - 10;
+	rq_regs->chunk_size = dml_log2(rq_sizing->chunk_bytes) - 10;
 
-	if (rq_sizing.min_chunk_bytes == 0)
+	if (rq_sizing->min_chunk_bytes == 0)
 		rq_regs->min_chunk_size = 0;
 	else
-		rq_regs->min_chunk_size = dml_log2(rq_sizing.min_chunk_bytes) - 8 + 1;
+		rq_regs->min_chunk_size = dml_log2(rq_sizing->min_chunk_bytes) - 8 + 1;
 
-	rq_regs->meta_chunk_size = dml_log2(rq_sizing.meta_chunk_bytes) - 10;
-	if (rq_sizing.min_meta_chunk_bytes == 0)
+	rq_regs->meta_chunk_size = dml_log2(rq_sizing->meta_chunk_bytes) - 10;
+	if (rq_sizing->min_meta_chunk_bytes == 0)
 		rq_regs->min_meta_chunk_size = 0;
 	else
-		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing.min_meta_chunk_bytes) - 6 + 1;
+		rq_regs->min_meta_chunk_size = dml_log2(rq_sizing->min_meta_chunk_bytes) - 6 + 1;
 
-	rq_regs->dpte_group_size = dml_log2(rq_sizing.dpte_group_bytes) - 6;
-	rq_regs->mpte_group_size = dml_log2(rq_sizing.mpte_group_bytes) - 6;
+	rq_regs->dpte_group_size = dml_log2(rq_sizing->dpte_group_bytes) - 6;
+	rq_regs->mpte_group_size = dml_log2(rq_sizing->mpte_group_bytes) - 6;
 }
 
 void dml1_extract_rq_regs(
@@ -236,9 +236,9 @@ void dml1_extract_rq_regs(
 	unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024;
 	unsigned int detile_buf_plane1_addr = 0;
 
-	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), rq_param.sizing.rq_l);
+	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), &rq_param.sizing.rq_l);
 	if (rq_param.yuv420)
-		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), rq_param.sizing.rq_c);
+		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), &rq_param.sizing.rq_c);
 
 	rq_regs->rq_regs_l.swath_height = dml_log2(rq_param.dlg.rq_l.swath_height);
 	rq_regs->rq_regs_c.swath_height = dml_log2(rq_param.dlg.rq_c.swath_height);
@@ -988,7 +988,7 @@ void dml1_rq_dlg_get_rq_params(
 
 	/* calculate how to split the det buffer space between luma and chroma */
 	handle_det_buf_split(mode_lib, rq_param, pipe_src_param);
-	print__rq_params_st(mode_lib, *rq_param);
+	print__rq_params_st(mode_lib, rq_param);
 }
 
 /* Note: currently taken in as is.
@@ -1927,6 +1927,6 @@ void dml1_rq_dlg_get_dlg_params(
 	disp_ttu_regs->min_ttu_vblank = min_ttu_vblank * refclk_freq_in_mhz;
 	ASSERT(disp_ttu_regs->min_ttu_vblank < dml_pow(2, 24));
 
-	print__ttu_regs_st(mode_lib, *disp_ttu_regs);
-	print__dlg_regs_st(mode_lib, *disp_dlg_regs);
+	print__ttu_regs_st(mode_lib, disp_ttu_regs);
+	print__dlg_regs_st(mode_lib, disp_dlg_regs);
 }
Signature · 1 line
2.33.0
#4 lore raw

Re: [2/4] drm/amd/display: Pass all structs in display_rq_dlg_helpers by pointer

Hi Harry,

This patch fixes the following CID's. Thanks.

Addresses-Coverity-ID: 1423868: ("Big parameter passed by value")
Addresses-Coverity-ID: 1423870: ("Big parameter passed by value")

-- Anson

author maintainer
#5 lore raw

PATCH 3/4 drm/amd/display: Fix rest of pass-by-value structs in DML

In reply to Harry Wentland

Passing structs adds a lot of overhead. We don't ever want to pass anything bigger than primitives by value.

Cc Nick Desaulniers <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Linux Kernel Mailing List <[email protected]>
Cc Arnd Bergmann <[email protected]>
Cc Leo Li <[email protected]>
Cc Alex Deucher <[email protected]>
Cc Christian König <[email protected]>
Cc Xinhui Pan <[email protected]>
Cc Nathan Chancellor <[email protected]>
Cc Guenter Roeck <[email protected]>
---
 .../gpu/drm/amd/display/dc/calcs/dcn_calcs.c  |  10 +-
 .../dc/dml/dcn20/display_rq_dlg_calc_20.c     | 126 +++++-----
 .../dc/dml/dcn20/display_rq_dlg_calc_20v2.c   | 124 +++++-----
 .../dc/dml/dcn21/display_rq_dlg_calc_21.c     |  68 +++---
 .../dc/dml/dcn30/display_rq_dlg_calc_30.c     |  34 +--
 .../dc/dml/dcn31/display_rq_dlg_calc_31.c     |  72 +++---
 .../display/dc/dml/dml1_display_rq_dlg_calc.c | 222 +++++++++---------
 .../display/dc/dml/dml1_display_rq_dlg_calc.h |  10 +-
 8 files changed, 333 insertions(+), 333 deletions(-)
drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +5 -5
index 0e18df1283b6..8e3a9294be3a 100644
@@ -502,15 +502,15 @@ static void dcn_bw_calc_rq_dlg_ttu(
 	/*todo: soc->sr_enter_plus_exit_time??*/
 	dlg_sys_param.t_srx_delay_us = dc->dcn_ip->dcfclk_cstate_latency / v->dcf_clk_deep_sleep;
 
-	dml1_rq_dlg_get_rq_params(dml, &rq_param, input.pipe.src);
-	dml1_extract_rq_regs(dml, rq_regs, rq_param);
+	dml1_rq_dlg_get_rq_params(dml, &rq_param, &input.pipe.src);
+	dml1_extract_rq_regs(dml, rq_regs, &rq_param);
 	dml1_rq_dlg_get_dlg_params(
 			dml,
 			dlg_regs,
 			ttu_regs,
-			rq_param.dlg,
-			dlg_sys_param,
-			input,
+			&rq_param.dlg,
+			&dlg_sys_param,
+			&input,
 			true,
 			true,
 			v->pte_enable == dcn_bw_yes,
drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c +63 -63

255 lines not on this page · raw message

drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c +62 -62

253 lines not on this page · raw message

drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c +34 -34

154 lines not on this page · raw message

drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c +17 -17
index 1535b01a8c5d..aef854270054 100644
@@ -113,28 +113,28 @@ static void extract_rq_sizing_regs(struct display_mode_lib *mode_lib,
 
 static void extract_rq_regs(struct display_mode_lib *mode_lib,
 	display_rq_regs_st *rq_regs,
-	const display_rq_params_st rq_param)
+	const display_rq_params_st *rq_param)
 {
 	unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024;
 	unsigned int detile_buf_plane1_addr = 0;
 
-	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), &rq_param.sizing.rq_l);
+	extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_l), &rq_param->sizing.rq_l);
 
-	rq_regs->rq_regs_l.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_l.dpte_row_height),
+	rq_regs->rq_regs_l.pte_row_height_linear = dml_floor(dml_log2(rq_param->dlg.rq_l.dpte_row_height),
 		1) - 3;
 
-	if (rq_param.yuv420) {
-		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), &rq_param.sizing.rq_c);
-		rq_regs->rq_regs_c.pte_row_height_linear = dml_floor(dml_log2(rq_param.dlg.rq_c.dpte_row_height),
+	if (rq_param->yuv420) {
+		extract_rq_sizing_regs(mode_lib, &(rq_regs->rq_regs_c), &rq_param->sizing.rq_c);
+		rq_regs->rq_regs_c.pte_row_height_linear = dml_floor(dml_log2(rq_param->dlg.rq_c.dpte_row_height),
 			1) - 3;
 	}
 
-	rq_regs->rq_regs_l.swath_height = dml_log2(rq_param.dlg.rq_l.swath_height);
-	rq_regs->rq_regs_c.swath_height = dml_log2(rq_param.dlg.rq_c.swath_height);
+	rq_regs->rq_regs_l.swath_height = dml_log2(rq_param->dlg.rq_l.swath_height);
+	rq_regs->rq_regs_c.swath_height = dml_log2(rq_param->dlg.rq_c.swath_height);
 
 	// FIXME: take the max between luma, chroma chunk size?
 	// okay for now, as we are setting chunk_bytes to 8kb anyways
-	if (rq_param.sizing.rq_l.chunk_bytes >= 32 * 1024 || (rq_param.yuv420 && rq_param.sizing.rq_c.chunk_bytes >= 32 * 1024)) { //32kb
+	if (rq_param->sizing.rq_l.chunk_bytes >= 32 * 1024 || (rq_param->yuv420 && rq_param->sizing.rq_c.chunk_bytes >= 32 * 1024)) { //32kb
 		rq_regs->drq_expansion_mode = 0;
 	} else {
 		rq_regs->drq_expansion_mode = 2;
@@ -143,9 +143,9 @@ static void extract_rq_regs(struct display_mode_lib *mode_lib,
 	rq_regs->mrq_expansion_mode = 1;
 	rq_regs->crq_expansion_mode = 1;
 
-	if (rq_param.yuv420) {
-	if ((double)rq_param.misc.rq_l.stored_swath_bytes
-			/ (double)rq_param.misc.rq_c.stored_swath_bytes <= 1.5) {
+	if (rq_param->yuv420) {
+	if ((double)rq_param->misc.rq_l.stored_swath_bytes
+			/ (double)rq_param->misc.rq_c.stored_swath_bytes <= 1.5) {
 			detile_buf_plane1_addr = (detile_buf_size_in_bytes / 2.0 / 64.0); // half to chroma
 		} else {
 			detile_buf_plane1_addr = dml_round_to_multiple((unsigned int)((2.0 * detile_buf_size_in_bytes) / 3.0),
@@ -158,7 +158,7 @@ static void extract_rq_regs(struct display_mode_lib *mode_lib,
 
 static void handle_det_buf_split(struct display_mode_lib *mode_lib,
 	display_rq_params_st *rq_param,
-	const display_pipe_source_params_st pipe_src_param)
+	const display_pipe_source_params_st *pipe_src_param)
 {
 	unsigned int total_swath_bytes = 0;
 	unsigned int swath_bytes_l = 0;
@@ -167,8 +167,8 @@ static void handle_det_buf_split(struct display_mode_lib *mode_lib,
 	unsigned int full_swath_bytes_packed_c = 0;
 	bool req128_l = false;
 	bool req128_c = false;
-	bool surf_linear = (pipe_src_param.sw_mode == dm_sw_linear);
-	bool surf_vert = (pipe_src_param.source_scan == dm_vert);
+	bool surf_linear = (pipe_src_param->sw_mode == dm_sw_linear);
+	bool surf_vert = (pipe_src_param->source_scan == dm_vert);
 	unsigned int log2_swath_height_l = 0;
 	unsigned int log2_swath_height_c = 0;
 	unsigned int detile_buf_size_in_bytes = mode_lib->ip.det_buffer_size_kbytes * 1024;
@@ -871,7 +871,7 @@ static void dml_rq_dlg_get_rq_params(struct display_mode_lib *mode_lib,
 	}
 
 	// calculate how to split the det buffer space between luma and chroma
-	handle_det_buf_split(mode_lib, rq_param, pipe_param->src);
+	handle_det_buf_split(mode_lib, rq_param, &pipe_param->src);
 	print__rq_params_st(mode_lib, rq_param);
 }
 
@@ -883,7 +883,7 @@ void dml30_rq_dlg_get_rq_reg(struct display_mode_lib *mode_lib,
 
 	memset(rq_regs, 0, sizeof(*rq_regs));
 	dml_rq_dlg_get_rq_params(mode_lib, &rq_param, pipe_param);
-	extract_rq_regs(mode_lib, rq_regs, rq_param);
+	extract_rq_regs(mode_lib, rq_regs, &rq_param);
 
 	print__rq_regs_st(mode_lib, rq_regs);
 }
drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c +36 -36

151 lines not on this page · raw message

drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c +111 -111

413 lines not on this page · raw message

drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.h +5 -5
index 9c06913ad767..e19ee3bde45f 100644
@@ -33,7 +33,7 @@ struct display_mode_lib;
 void dml1_extract_rq_regs(
 		struct display_mode_lib *mode_lib,
 		struct _vcs_dpi_display_rq_regs_st *rq_regs,
-		const struct _vcs_dpi_display_rq_params_st rq_param);
+		const struct _vcs_dpi_display_rq_params_st *rq_param);
 /* Function: dml_rq_dlg_get_rq_params
  *  Calculate requestor related parameters that register definition agnostic
  *  (i.e. this layer does try to separate real values from register definition)
@@ -45,7 +45,7 @@ void dml1_extract_rq_regs(
 void dml1_rq_dlg_get_rq_params(
 		struct display_mode_lib *mode_lib,
 		struct _vcs_dpi_display_rq_params_st *rq_param,
-		const struct _vcs_dpi_display_pipe_source_params_st pipe_src_param);
+		const struct _vcs_dpi_display_pipe_source_params_st *pipe_src_param);
 
 
 /* Function: dml_rq_dlg_get_dlg_params
@@ -55,9 +55,9 @@ void dml1_rq_dlg_get_dlg_params(
 		struct display_mode_lib *mode_lib,
 		struct _vcs_dpi_display_dlg_regs_st *dlg_regs,
 		struct _vcs_dpi_display_ttu_regs_st *ttu_regs,
-		const struct _vcs_dpi_display_rq_dlg_params_st rq_dlg_param,
-		const struct _vcs_dpi_display_dlg_sys_params_st dlg_sys_param,
-		const struct _vcs_dpi_display_e2e_pipe_params_st e2e_pipe_param,
+		const struct _vcs_dpi_display_rq_dlg_params_st *rq_dlg_param,
+		const struct _vcs_dpi_display_dlg_sys_params_st *dlg_sys_param,
+		const struct _vcs_dpi_display_e2e_pipe_params_st *e2e_pipe_param,
 		const bool cstate_en,
 		const bool pstate_en,
 		const bool vm_en,
Signature · 1 line
2.33.0
#6 lore raw

Re: [3/4] drm/amd/display: Fix rest of pass-by-value structs in DML

Hi Harry,

This patch fixes the following CID's. Thanks.

Addresses-Coverity-ID: 1424031: ("Big parameter passed by value")
Addresses-Coverity-ID: 1424055: ("Big parameter passed by value")
Addresses-Coverity-ID: 1424072: ("Big parameter passed by value")
Addresses-Coverity-ID: 1423779: ("Big parameter passed by value")

-- Anson

author maintainer
#7 lore raw

PATCH 4/4 drm/amd/display: Allocate structs needed by dcn_bw_calc_rq_dlg_ttu in pipe_ctx

In reply to Harry Wentland

[Why & How]
dcn_bw_calc_rq_dlg_ttu uses a stack frame great than 1024. To solve this we could allocate the rq_param, dlg_sys_param, and input structs dynamically. Since this function is inside a kernel_fpu_begin()/end() call we want to avoid memory allocation. Instead it's much safer to pre-allocate these on the pipe_ctx.

Fixes 3fe617ccafd6 ("Enable '-Werror' by default for all kernel builds")
Cc Nick Desaulniers <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Linux Kernel Mailing List <[email protected]>
Cc Arnd Bergmann <[email protected]>
Cc Leo Li <[email protected]>
Cc Alex Deucher <[email protected]>
Cc Christian König <[email protected]>
Cc Xinhui Pan <[email protected]>
Cc Nathan Chancellor <[email protected]>
Cc Guenter Roeck <[email protected]>
---
 .../gpu/drm/amd/display/dc/calcs/dcn_calcs.c  | 61 ++++++++++---------
 .../gpu/drm/amd/display/dc/inc/core_types.h   |  3 +
 2 files changed, 35 insertions(+), 29 deletions(-)
drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +32 -29
index 8e3a9294be3a..6b248cd2a461 100644
@@ -459,9 +459,9 @@ static void dcn_bw_calc_rq_dlg_ttu(
 	struct _vcs_dpi_display_dlg_regs_st *dlg_regs = &pipe->dlg_regs;
 	struct _vcs_dpi_display_ttu_regs_st *ttu_regs = &pipe->ttu_regs;
 	struct _vcs_dpi_display_rq_regs_st *rq_regs = &pipe->rq_regs;
-	struct _vcs_dpi_display_rq_params_st rq_param = {0};
-	struct _vcs_dpi_display_dlg_sys_params_st dlg_sys_param = {0};
-	struct _vcs_dpi_display_e2e_pipe_params_st input = { { { 0 } } };
+	struct _vcs_dpi_display_rq_params_st *rq_param = &pipe->dml_rq_param;
+	struct _vcs_dpi_display_dlg_sys_params_st *dlg_sys_param = &pipe->dml_dlg_sys_param;
+	struct _vcs_dpi_display_e2e_pipe_params_st *input = &pipe->dml_input;
 	float total_active_bw = 0;
 	float total_prefetch_bw = 0;
 	int total_flip_bytes = 0;
@@ -470,47 +470,50 @@ static void dcn_bw_calc_rq_dlg_ttu(
 	memset(dlg_regs, 0, sizeof(*dlg_regs));
 	memset(ttu_regs, 0, sizeof(*ttu_regs));
 	memset(rq_regs, 0, sizeof(*rq_regs));
+	memset(rq_param, 0, sizeof(*rq_param));
+	memset(dlg_sys_param, 0, sizeof(*dlg_sys_param));
+	memset(input, 0, sizeof(*input));
 
 	for (i = 0; i < number_of_planes; i++) {
 		total_active_bw += v->read_bandwidth[i];
 		total_prefetch_bw += v->prefetch_bandwidth[i];
 		total_flip_bytes += v->total_immediate_flip_bytes[i];
 	}
-	dlg_sys_param.total_flip_bw = v->return_bw - dcn_bw_max2(total_active_bw, total_prefetch_bw);
-	if (dlg_sys_param.total_flip_bw < 0.0)
-		dlg_sys_param.total_flip_bw = 0;
-
-	dlg_sys_param.t_mclk_wm_us = v->dram_clock_change_watermark;
-	dlg_sys_param.t_sr_wm_us = v->stutter_enter_plus_exit_watermark;
-	dlg_sys_param.t_urg_wm_us = v->urgent_watermark;
-	dlg_sys_param.t_extra_us = v->urgent_extra_latency;
-	dlg_sys_param.deepsleep_dcfclk_mhz = v->dcf_clk_deep_sleep;
-	dlg_sys_param.total_flip_bytes = total_flip_bytes;
-
-	pipe_ctx_to_e2e_pipe_params(pipe, &input.pipe);
-	input.clks_cfg.dcfclk_mhz = v->dcfclk;
-	input.clks_cfg.dispclk_mhz = v->dispclk;
-	input.clks_cfg.dppclk_mhz = v->dppclk;
-	input.clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0;
-	input.clks_cfg.socclk_mhz = v->socclk;
-	input.clks_cfg.voltage = v->voltage_level;
+	dlg_sys_param->total_flip_bw = v->return_bw - dcn_bw_max2(total_active_bw, total_prefetch_bw);
+	if (dlg_sys_param->total_flip_bw < 0.0)
+		dlg_sys_param->total_flip_bw = 0;
+
+	dlg_sys_param->t_mclk_wm_us = v->dram_clock_change_watermark;
+	dlg_sys_param->t_sr_wm_us = v->stutter_enter_plus_exit_watermark;
+	dlg_sys_param->t_urg_wm_us = v->urgent_watermark;
+	dlg_sys_param->t_extra_us = v->urgent_extra_latency;
+	dlg_sys_param->deepsleep_dcfclk_mhz = v->dcf_clk_deep_sleep;
+	dlg_sys_param->total_flip_bytes = total_flip_bytes;
+
+	pipe_ctx_to_e2e_pipe_params(pipe, &input->pipe);
+	input->clks_cfg.dcfclk_mhz = v->dcfclk;
+	input->clks_cfg.dispclk_mhz = v->dispclk;
+	input->clks_cfg.dppclk_mhz = v->dppclk;
+	input->clks_cfg.refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0;
+	input->clks_cfg.socclk_mhz = v->socclk;
+	input->clks_cfg.voltage = v->voltage_level;
 //	dc->dml.logger = pool->base.logger;
-	input.dout.output_format = (v->output_format[in_idx] == dcn_bw_420) ? dm_420 : dm_444;
-	input.dout.output_type  = (v->output[in_idx] == dcn_bw_hdmi) ? dm_hdmi : dm_dp;
+	input->dout.output_format = (v->output_format[in_idx] == dcn_bw_420) ? dm_420 : dm_444;
+	input->dout.output_type  = (v->output[in_idx] == dcn_bw_hdmi) ? dm_hdmi : dm_dp;
 	//input[in_idx].dout.output_standard;
 
 	/*todo: soc->sr_enter_plus_exit_time??*/
-	dlg_sys_param.t_srx_delay_us = dc->dcn_ip->dcfclk_cstate_latency / v->dcf_clk_deep_sleep;
+	dlg_sys_param->t_srx_delay_us = dc->dcn_ip->dcfclk_cstate_latency / v->dcf_clk_deep_sleep;
 
-	dml1_rq_dlg_get_rq_params(dml, &rq_param, &input.pipe.src);
-	dml1_extract_rq_regs(dml, rq_regs, &rq_param);
+	dml1_rq_dlg_get_rq_params(dml, rq_param, &input->pipe.src);
+	dml1_extract_rq_regs(dml, rq_regs, rq_param);
 	dml1_rq_dlg_get_dlg_params(
 			dml,
 			dlg_regs,
 			ttu_regs,
-			&rq_param.dlg,
-			&dlg_sys_param,
-			&input,
+			&rq_param->dlg,
+			dlg_sys_param,
+			input,
 			true,
 			true,
 			v->pte_enable == dcn_bw_yes,
drivers/gpu/drm/amd/display/dc/inc/core_types.h +3 -0
index 45a6216dfa2a..4dca14b598dd 100644
@@ -366,6 +366,9 @@ struct pipe_ctx {
 	struct _vcs_dpi_display_ttu_regs_st ttu_regs;
 	struct _vcs_dpi_display_rq_regs_st rq_regs;
 	struct _vcs_dpi_display_pipe_dest_params_st pipe_dlg_param;
+	struct _vcs_dpi_display_rq_params_st dml_rq_param;
+	struct _vcs_dpi_display_dlg_sys_params_st dml_dlg_sys_param;
+	struct _vcs_dpi_display_e2e_pipe_params_st dml_input;
 	int det_buffer_size_kb;
 	bool unbounded_req;
 #endif
Signature · 1 line
2.33.0
#8 lore raw
In reply to Harry Wentland

It's nice to see at least some of them addressed, feel free to add an Acked-by: Christian König <[email protected]>

Regards,
Christian.

Am 09.09.21 um 03:00 schrieb Harry Wentland:

Quoted 50 lines

With the '-Werror' enablement patch the amdgpu build was failing on clang builds because a bunch of functions were blowing past the 1024 byte stack frame default. Due to this we also noticed that a lot of functions were passing large structs by value instead of by pointer.

This series attempts to fix this.

There is still one remaining function that blows the 1024 limit by 40 bytes:

drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn21/display_mode_vba_21.c:3397:6:

error: stack frame size of 1064 bytes in function 'dml21_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than=]

This will be a slightly more challenging fix but I'll see if we can get it below 1024 by breaking it into smaller functions.

With this series I can build amdgpu with CC=clang and a stack frame limit of 1064.

This series boots on a Radeon RX 5500 XT.

Harry Wentland (4):
drm/amd/display: Pass display_pipe_params_st as const in DML drm/amd/display: Pass all structs in display_rq_dlg_helpers by pointer drm/amd/display: Fix rest of pass-by-value structs in DML drm/amd/display: Allocate structs needed by dcn_bw_calc_rq_dlg_ttu in
pipe_ctx

  .../gpu/drm/amd/display/dc/calcs/dcn_calcs.c  |  55 ++--
  .../drm/amd/display/dc/dcn20/dcn20_resource.c |   2 +-
  .../dc/dml/dcn20/display_rq_dlg_calc_20.c     | 158 +++++------
  .../dc/dml/dcn20/display_rq_dlg_calc_20.h     |   4 +-
  .../dc/dml/dcn20/display_rq_dlg_calc_20v2.c   | 156 +++++------
  .../dc/dml/dcn20/display_rq_dlg_calc_20v2.h   |   4 +-
  .../dc/dml/dcn21/display_rq_dlg_calc_21.c     | 156 +++++------
  .../dc/dml/dcn21/display_rq_dlg_calc_21.h     |   4 +-
  .../dc/dml/dcn30/display_rq_dlg_calc_30.c     | 132 ++++-----
  .../dc/dml/dcn30/display_rq_dlg_calc_30.h     |   4 +-
  .../dc/dml/dcn31/display_rq_dlg_calc_31.c     | 166 ++++++------
  .../dc/dml/dcn31/display_rq_dlg_calc_31.h     |   4 +-
  .../drm/amd/display/dc/dml/display_mode_lib.h |   4 +-
  .../display/dc/dml/display_rq_dlg_helpers.c   | 256 +++++++++---------
  .../display/dc/dml/display_rq_dlg_helpers.h   |  20 +-
  .../display/dc/dml/dml1_display_rq_dlg_calc.c | 246 ++++++++---------
  .../display/dc/dml/dml1_display_rq_dlg_calc.h |  10 +-
  .../gpu/drm/amd/display/dc/inc/core_types.h   |   3 +
  18 files changed, 695 insertions(+), 689 deletions(-)
#9 lore raw
In reply to Harry Wentland
Quoted 21 lines · Harry Wentland

With the '-Werror' enablement patch the amdgpu build was failing on clang builds because a bunch of functions were blowing past the 1024 byte stack frame default. Due to this we also noticed that a lot of functions were passing large structs by value instead of by pointer.

This series attempts to fix this.

There is still one remaining function that blows the 1024 limit by 40 bytes:

drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn21/display_mode_vba_21.c:3397:6:

error: stack frame size of 1064 bytes in function 'dml21_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than=]

This will be a slightly more challenging fix but I'll see if we can get it below 1024 by breaking it into smaller functions.

With this series I can build amdgpu with CC=clang and a stack frame limit of 1064.

Quoted 8 lines

This series boots on a Radeon RX 5500 XT.

Harry Wentland (4):
drm/amd/display: Pass display_pipe_params_st as const in DML drm/amd/display: Pass all structs in display_rq_dlg_helpers by pointer drm/amd/display: Fix rest of pass-by-value structs in DML drm/amd/display: Allocate structs needed by dcn_bw_calc_rq_dlg_ttu in
pipe_ctx

This series resolves some warnings that were reported on our issue tracker for 32-bit x86. I do see some other warnings in amdgpu with clang in various configurations but this is a great start. Thank you for taking a look at them. For the series:

Quoted 22 lines
 .../gpu/drm/amd/display/dc/calcs/dcn_calcs.c  |  55 ++--
 .../drm/amd/display/dc/dcn20/dcn20_resource.c |   2 +-
 .../dc/dml/dcn20/display_rq_dlg_calc_20.c     | 158 +++++------
 .../dc/dml/dcn20/display_rq_dlg_calc_20.h     |   4 +-
 .../dc/dml/dcn20/display_rq_dlg_calc_20v2.c   | 156 +++++------
 .../dc/dml/dcn20/display_rq_dlg_calc_20v2.h   |   4 +-
 .../dc/dml/dcn21/display_rq_dlg_calc_21.c     | 156 +++++------
 .../dc/dml/dcn21/display_rq_dlg_calc_21.h     |   4 +-
 .../dc/dml/dcn30/display_rq_dlg_calc_30.c     | 132 ++++-----
 .../dc/dml/dcn30/display_rq_dlg_calc_30.h     |   4 +-
 .../dc/dml/dcn31/display_rq_dlg_calc_31.c     | 166 ++++++------
 .../dc/dml/dcn31/display_rq_dlg_calc_31.h     |   4 +-
 .../drm/amd/display/dc/dml/display_mode_lib.h |   4 +-
 .../display/dc/dml/display_rq_dlg_helpers.c   | 256 +++++++++---------
 .../display/dc/dml/display_rq_dlg_helpers.h   |  20 +-
 .../display/dc/dml/dml1_display_rq_dlg_calc.c | 246 ++++++++---------
 .../display/dc/dml/dml1_display_rq_dlg_calc.h |  10 +-
 .../gpu/drm/amd/display/dc/inc/core_types.h   |   3 +
 18 files changed, 695 insertions(+), 689 deletions(-)
Signature · 1 line
2.33.0
maintainer Reviewed-by on the v1 series
#10 lore raw
In reply to Harry Wentland
Quoted 21 lines · Harry Wentland

With the '-Werror' enablement patch the amdgpu build was failing on clang builds because a bunch of functions were blowing past the 1024 byte stack frame default. Due to this we also noticed that a lot of functions were passing large structs by value instead of by pointer.

This series attempts to fix this.

There is still one remaining function that blows the 1024 limit by 40 bytes:

drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn21/display_mode_vba_21.c:3397:6:

error: stack frame size of 1064 bytes in function 'dml21_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than=]

This will be a slightly more challenging fix but I'll see if we can get it below 1024 by breaking it into smaller functions.

With this series I can build amdgpu with CC=clang and a stack frame limit of 1064.

Quoted 8 lines

This series boots on a Radeon RX 5500 XT.

Harry Wentland (4):
drm/amd/display: Pass display_pipe_params_st as const in DML drm/amd/display: Pass all structs in display_rq_dlg_helpers by pointer drm/amd/display: Fix rest of pass-by-value structs in DML drm/amd/display: Allocate structs needed by dcn_bw_calc_rq_dlg_ttu in
pipe_ctx

Series LGTM,

Thanks!

Quoted 21 lines
  .../gpu/drm/amd/display/dc/calcs/dcn_calcs.c  |  55 ++--
  .../drm/amd/display/dc/dcn20/dcn20_resource.c |   2 +-
  .../dc/dml/dcn20/display_rq_dlg_calc_20.c     | 158 +++++------
  .../dc/dml/dcn20/display_rq_dlg_calc_20.h     |   4 +-
  .../dc/dml/dcn20/display_rq_dlg_calc_20v2.c   | 156 +++++------
  .../dc/dml/dcn20/display_rq_dlg_calc_20v2.h   |   4 +-
  .../dc/dml/dcn21/display_rq_dlg_calc_21.c     | 156 +++++------
  .../dc/dml/dcn21/display_rq_dlg_calc_21.h     |   4 +-
  .../dc/dml/dcn30/display_rq_dlg_calc_30.c     | 132 ++++-----
  .../dc/dml/dcn30/display_rq_dlg_calc_30.h     |   4 +-
  .../dc/dml/dcn31/display_rq_dlg_calc_31.c     | 166 ++++++------
  .../dc/dml/dcn31/display_rq_dlg_calc_31.h     |   4 +-
  .../drm/amd/display/dc/dml/display_mode_lib.h |   4 +-
  .../display/dc/dml/display_rq_dlg_helpers.c   | 256 +++++++++---------
  .../display/dc/dml/display_rq_dlg_helpers.h   |  20 +-
  .../display/dc/dml/dml1_display_rq_dlg_calc.c | 246 ++++++++---------
  .../display/dc/dml/dml1_display_rq_dlg_calc.h |  10 +-
  .../gpu/drm/amd/display/dc/inc/core_types.h   |   3 +
  18 files changed, 695 insertions(+), 689 deletions(-)