[Intel-gfx] [Mesa-dev] [PATCH 3/5] intel gen4-5: Correctly setup the parameters in the sf.
Paul Berry
stereotype441 at gmail.com
Tue Jul 17 14:50:39 CEST 2012
On 30 June 2012 11:50, Olivier Galibert <galibert at pobox.com> wrote:
> This patch also correct a couple of problems with noperspective
> interpolation.
>
> At that point all the glsl 1.1/1.3 interpolation tests that do not
> clip pass (the -none ones).
>
> The fs code does not use the pre-resolved interpolation modes in order
> not to mess with gen6+. Sharing the resolution would require putting
> brw_wm_prog before brw_clip_prog and brw_sf_prog. This may be a good
> thing, but it could have unexpected consequences, so it's better be
> done independently in any case.
>
> Signed-off-by: Olivier Galibert <galibert at pobox.com>
> ---
> src/mesa/drivers/dri/i965/brw_fs.cpp | 2 +-
> src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 5 +
> src/mesa/drivers/dri/i965/brw_sf.c | 9 +-
> src/mesa/drivers/dri/i965/brw_sf.h | 2 +-
> src/mesa/drivers/dri/i965/brw_sf_emit.c | 164
> +++++++++++++-------------
> 5 files changed, 95 insertions(+), 87 deletions(-)
>
> diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp
> b/src/mesa/drivers/dri/i965/brw_fs.cpp
> index 710f2ff..b142f2b 100644
> --- a/src/mesa/drivers/dri/i965/brw_fs.cpp
> +++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
> @@ -506,7 +506,7 @@ fs_visitor::emit_general_interpolation(ir_variable *ir)
> struct brw_reg interp = interp_reg(location, k);
> emit_linterp(attr, fs_reg(interp), interpolation_mode,
> ir->centroid);
> - if (intel->gen < 6) {
> + if (intel->gen < 6 && interpolation_mode ==
> INTERP_QUALIFIER_SMOOTH) {
> emit(BRW_OPCODE_MUL, attr, attr, this->pixel_w);
> }
> }
> diff --git a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
> b/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
> index 9bd1e67..ab83a95 100644
> --- a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
> +++ b/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
> @@ -1924,6 +1924,11 @@ fs_visitor::emit_interpolation_setup_gen4()
> emit(BRW_OPCODE_ADD,
> this->delta_y[BRW_WM_PERSPECTIVE_PIXEL_BARYCENTRIC],
> this->pixel_y, fs_reg(negate(brw_vec1_grf(1, 1))));
>
> + this->delta_x[BRW_WM_NONPERSPECTIVE_PIXEL_BARYCENTRIC] =
> + this->delta_x[BRW_WM_PERSPECTIVE_PIXEL_BARYCENTRIC];
> + this->delta_y[BRW_WM_NONPERSPECTIVE_PIXEL_BARYCENTRIC] =
> + this->delta_y[BRW_WM_PERSPECTIVE_PIXEL_BARYCENTRIC];
> +
>
Can we add a comment explaining why this is correct? Something like this
maybe:
"On Gen4-5, we accomplish perspective-correct interpolation by dividing the
attribute values by w in the vertex shader, interpolating the result
linearly in screen space, and then multiplying by w in the fragment
shader. So the interpolation step is always linear in screen space,
regardless of whether the attribute is perspective or non-perspective.
Accordingly, we use the same delta_x and delta_y values for both kinds of
interpolation."
> this->current_annotation = "compute pos.w and 1/pos.w";
> /* Compute wpos.w. It's always in our setup, since it's needed to
> * interpolate the other attributes.
> diff --git a/src/mesa/drivers/dri/i965/brw_sf.c
> b/src/mesa/drivers/dri/i965/brw_sf.c
> index 0cc4fc7..85f5f51 100644
> --- a/src/mesa/drivers/dri/i965/brw_sf.c
> +++ b/src/mesa/drivers/dri/i965/brw_sf.c
> @@ -139,6 +139,7 @@ brw_upload_sf_prog(struct brw_context *brw)
> struct brw_sf_prog_key key;
> /* _NEW_BUFFERS */
> bool render_to_fbo = _mesa_is_user_fbo(ctx->DrawBuffer);
> + int i;
>
> memset(&key, 0, sizeof(key));
>
> @@ -191,7 +192,13 @@ brw_upload_sf_prog(struct brw_context *brw)
> key.sprite_origin_lower_left = true;
>
> /* _NEW_LIGHT */
> - key.do_flat_shading = (ctx->Light.ShadeModel == GL_FLAT);
> + key.has_flat_shading = 0;
> + for (i = 0; i < BRW_VERT_RESULT_MAX; i++) {
> + if (brw_get_interpolation_mode(brw, i) == INTERP_QUALIFIER_FLAT) {
> + key.has_flat_shading = 1;
> + break;
> + }
> + }
> key.do_twoside_color = (ctx->Light.Enabled &&
> ctx->Light.Model.TwoSide) ||
> ctx->VertexProgram._TwoSideEnabled;
> brw_copy_interpolation_modes(brw, key.interpolation_mode);
> diff --git a/src/mesa/drivers/dri/i965/brw_sf.h
> b/src/mesa/drivers/dri/i965/brw_sf.h
> index 0a8135c..c718072 100644
> --- a/src/mesa/drivers/dri/i965/brw_sf.h
> +++ b/src/mesa/drivers/dri/i965/brw_sf.h
> @@ -50,7 +50,7 @@ struct brw_sf_prog_key {
> uint8_t point_sprite_coord_replace;
> GLuint primitive:2;
> GLuint do_twoside_color:1;
> - GLuint do_flat_shading:1;
> + GLuint has_flat_shading:1;
> GLuint frontface_ccw:1;
> GLuint do_point_sprite:1;
> GLuint do_point_coord:1;
> diff --git a/src/mesa/drivers/dri/i965/brw_sf_emit.c
> b/src/mesa/drivers/dri/i965/brw_sf_emit.c
> index 9d8aa38..387685a 100644
> --- a/src/mesa/drivers/dri/i965/brw_sf_emit.c
> +++ b/src/mesa/drivers/dri/i965/brw_sf_emit.c
> @@ -44,6 +44,17 @@
>
>
> /**
> + * Determine the vue slot corresponding to the given half of the given
> + * register. half=0 means the first half of a register, half=1 means the
> + * second half.
> + */
> +static inline int vert_reg_to_vue_slot(struct brw_sf_compile *c, GLuint
> reg,
> + int half)
> +{
> + return (reg + c->urb_entry_read_offset) * 2 + half;
> +}
> +
> +/**
> * Determine the vert_result corresponding to the given half of the given
> * register. half=0 means the first half of a register, half=1 means the
> * second half.
> @@ -51,11 +62,24 @@
> static inline int vert_reg_to_vert_result(struct brw_sf_compile *c,
> GLuint reg,
> int half)
> {
> - int vue_slot = (reg + c->urb_entry_read_offset) * 2 + half;
> + int vue_slot = vert_reg_to_vue_slot(c, reg, half);
> return c->vue_map.slot_to_vert_result[vue_slot];
> }
>
> /**
> + * Determine the register corresponding to the given vue slot.
> + */
> +static struct brw_reg get_vue_slot(struct brw_sf_compile *c,
> + struct brw_reg vert,
> + int vue_slot)
> +{
> + GLuint off = vue_slot / 2 - c->urb_entry_read_offset;
> + GLuint sub = vue_slot % 2;
> +
> + return brw_vec4_grf(vert.nr + off, sub * 4);
>
+}
> +
> +/**
> * Determine the register corresponding to the given vert_result.
> */
> static struct brw_reg get_vert_result(struct brw_sf_compile *c,
> @@ -64,10 +88,7 @@ static struct brw_reg get_vert_result(struct
> brw_sf_compile *c,
> {
> int vue_slot = c->vue_map.vert_result_to_slot[vert_result];
> assert (vue_slot >= c->urb_entry_read_offset);
> - GLuint off = vue_slot / 2 - c->urb_entry_read_offset;
> - GLuint sub = vue_slot % 2;
> -
> - return brw_vec4_grf(vert.nr + off, sub * 4);
> + return get_vue_slot(c, vert, vue_slot);
> }
>
> static bool
> @@ -128,31 +149,37 @@ static void do_twoside_color( struct brw_sf_compile
> *c )
> * Flat shading
> */
>
> -#define VERT_RESULT_COLOR_BITS (BITFIELD64_BIT(VERT_RESULT_COL0) | \
> - BITFIELD64_BIT(VERT_RESULT_COL1))
> -
> -static void copy_colors( struct brw_sf_compile *c,
> - struct brw_reg dst,
> - struct brw_reg src,
> - int allow_twoside)
> +static void copy_flatshaded_attributes( struct brw_sf_compile *c,
> + struct brw_reg dst,
> + struct brw_reg src)
> {
> struct brw_compile *p = &c->func;
> + struct brw_context *brw = p->brw;
> GLuint i;
>
> - for (i = VERT_RESULT_COL0; i <= VERT_RESULT_COL1; i++) {
> - if (have_attr(c,i)) {
> + for (i = 0; i < BRW_VERT_RESULT_MAX; i++) {
> + if (brw_get_interpolation_mode(brw, i) == INTERP_QUALIFIER_FLAT) {
> brw_MOV(p,
> - get_vert_result(c, dst, i),
> - get_vert_result(c, src, i));
> + get_vue_slot(c, dst, i),
> + get_vue_slot(c, src, i));
>
> - } else if(allow_twoside && have_attr(c, i - VERT_RESULT_COL0 +
> VERT_RESULT_BFC0)) {
> - brw_MOV(p,
> - get_vert_result(c, dst, i - VERT_RESULT_COL0 +
> VERT_RESULT_BFC0),
> - get_vert_result(c, src, i - VERT_RESULT_COL0 +
> VERT_RESULT_BFC0));
> }
> }
> }
>
> +static GLuint count_flatshaded_attributes(struct brw_sf_compile *c )
> +{
> + struct brw_compile *p = &c->func;
> + struct brw_context *brw = p->brw;
> + GLuint count = 0;
> + GLuint i;
> + for (i = 0; i < BRW_VERT_RESULT_MAX; i++) {
> + if (brw_get_interpolation_mode(brw, i) == INTERP_QUALIFIER_FLAT)
> + count++;
> + }
> + return count;
> +}
> +
>
>
> /* Need to use a computed jump to copy flatshaded attributes as the
> @@ -168,18 +195,6 @@ static void do_flatshade_triangle( struct
> brw_sf_compile *c )
>
> GLuint nr;
>
> - if (c->key.do_twoside_color) {
> - nr = ((c->key.attrs & (BITFIELD64_BIT(VERT_RESULT_COL0) |
> BITFIELD64_BIT(VERT_RESULT_BFC0))) != 0) +
> - ((c->key.attrs & (BITFIELD64_BIT(VERT_RESULT_COL1) |
> BITFIELD64_BIT(VERT_RESULT_BFC1))) != 0);
> -
> - } else {
> - nr = ((c->key.attrs & BITFIELD64_BIT(VERT_RESULT_COL0)) != 0) +
> - ((c->key.attrs & BITFIELD64_BIT(VERT_RESULT_COL1)) != 0);
> - }
> -
> - if (!nr)
> - return;
> -
> /* Already done in clip program:
> */
> if (c->key.primitive == SF_UNFILLED_TRIS)
> @@ -188,21 +203,23 @@ static void do_flatshade_triangle( struct
> brw_sf_compile *c )
> if (intel->gen == 5)
> jmpi = 2;
>
> + nr = count_flatshaded_attributes(c);
> +
>
We used to have this optimization:
if (!nr)
return;
I understand why you removed it: because now this code should only be
called if c->key.has_flat_shading is true. However, for the sake of safety
(and documentation), can we add something like this:
/* This code should only be called if c->key.has_flat_shading is true,
* meaning there is at least one flatshaded attribute.
*/
assert(nr != 0);
If you would prefer to put this assertion inside
count_flatshaded_attributes(), I would be happy with that too.
> brw_push_insn_state(p);
>
> brw_MUL(p, c->pv, c->pv, brw_imm_d(jmpi*(nr*2+1)));
> brw_JMPI(p, ip, ip, c->pv);
>
> - copy_colors(c, c->vert[1], c->vert[0], c->key.do_twoside_color);
> - copy_colors(c, c->vert[2], c->vert[0], c->key.do_twoside_color);
> + copy_flatshaded_attributes(c, c->vert[1], c->vert[0]);
> + copy_flatshaded_attributes(c, c->vert[2], c->vert[0]);
> brw_JMPI(p, ip, ip, brw_imm_d(jmpi*(nr*4+1)));
>
> - copy_colors(c, c->vert[0], c->vert[1], c->key.do_twoside_color);
> - copy_colors(c, c->vert[2], c->vert[1], c->key.do_twoside_color);
> + copy_flatshaded_attributes(c, c->vert[0], c->vert[1]);
> + copy_flatshaded_attributes(c, c->vert[2], c->vert[1]);
> brw_JMPI(p, ip, ip, brw_imm_d(jmpi*nr*2));
>
> - copy_colors(c, c->vert[0], c->vert[2], c->key.do_twoside_color);
> - copy_colors(c, c->vert[1], c->vert[2], c->key.do_twoside_color);
> + copy_flatshaded_attributes(c, c->vert[0], c->vert[2]);
> + copy_flatshaded_attributes(c, c->vert[1], c->vert[2]);
>
> brw_pop_insn_state(p);
> }
> @@ -213,12 +230,9 @@ static void do_flatshade_line( struct brw_sf_compile
> *c )
> struct brw_compile *p = &c->func;
> struct intel_context *intel = &p->brw->intel;
> struct brw_reg ip = brw_ip_reg();
> - GLuint nr = _mesa_bitcount_64(c->key.attrs & VERT_RESULT_COLOR_BITS);
> + GLuint nr;
> GLuint jmpi = 1;
>
> - if (!nr)
> - return;
> -
> /* Already done in clip program:
> */
> if (c->key.primitive == SF_UNFILLED_TRIS)
> @@ -227,14 +241,16 @@ static void do_flatshade_line( struct brw_sf_compile
> *c )
> if (intel->gen == 5)
> jmpi = 2;
>
> + nr = count_flatshaded_attributes(c);
> +
>
A similar assertion should go here.
> brw_push_insn_state(p);
>
> brw_MUL(p, c->pv, c->pv, brw_imm_d(jmpi*(nr+1)));
> brw_JMPI(p, ip, ip, c->pv);
> - copy_colors(c, c->vert[1], c->vert[0], 0);
> + copy_flatshaded_attributes(c, c->vert[1], c->vert[0]);
>
> brw_JMPI(p, ip, ip, brw_imm_ud(jmpi*nr));
> - copy_colors(c, c->vert[0], c->vert[1], 0);
> + copy_flatshaded_attributes(c, c->vert[0], c->vert[1]);
>
> brw_pop_insn_state(p);
> }
> @@ -332,40 +348,25 @@ static void invert_det( struct brw_sf_compile *c)
>
> static bool
> calculate_masks(struct brw_sf_compile *c,
> - GLuint reg,
> - GLushort *pc,
> - GLushort *pc_persp,
> - GLushort *pc_linear)
> + GLuint reg,
> + GLushort *pc,
> + GLushort *pc_persp,
> + GLushort *pc_linear)
>
I like the way you've rewritten this function. Nicely done.
> {
> + struct brw_compile *p = &c->func;
> + struct brw_context *brw = p->brw;
> + enum glsl_interp_qualifier interp;
> bool is_last_attr = (reg == c->nr_setup_regs - 1);
> - GLbitfield64 persp_mask;
> - GLbitfield64 linear_mask;
> -
> - if (c->key.do_flat_shading)
> - persp_mask = c->key.attrs & ~(BITFIELD64_BIT(VERT_RESULT_HPOS) |
> - BITFIELD64_BIT(VERT_RESULT_COL0) |
> - BITFIELD64_BIT(VERT_RESULT_COL1) |
> - BITFIELD64_BIT(VERT_RESULT_BFC0) |
> - BITFIELD64_BIT(VERT_RESULT_BFC1));
> - else
> - persp_mask = c->key.attrs & ~(BITFIELD64_BIT(VERT_RESULT_HPOS));
> -
> - if (c->key.do_flat_shading)
> - linear_mask = c->key.attrs & ~(BITFIELD64_BIT(VERT_RESULT_COL0) |
> - BITFIELD64_BIT(VERT_RESULT_COL1) |
> - BITFIELD64_BIT(VERT_RESULT_BFC0) |
> - BITFIELD64_BIT(VERT_RESULT_BFC1));
> - else
> - linear_mask = c->key.attrs;
>
> *pc_persp = 0;
> *pc_linear = 0;
> *pc = 0xf;
> -
> - if (persp_mask & BITFIELD64_BIT(vert_reg_to_vert_result(c, reg, 0)))
> - *pc_persp = 0xf;
>
> - if (linear_mask & BITFIELD64_BIT(vert_reg_to_vert_result(c, reg, 0)))
> + interp = brw_get_interpolation_mode(brw, vert_reg_to_vue_slot(c, reg,
> 0));
> + if (interp == INTERP_QUALIFIER_SMOOTH) {
> + *pc_linear = 0xf;
> + *pc_persp = 0xf;
> + } else if(interp == INTERP_QUALIFIER_NOPERSPECTIVE)
> *pc_linear = 0xf;
>
> /* Maybe only processs one attribute on the final round:
> @@ -373,11 +374,12 @@ calculate_masks(struct brw_sf_compile *c,
> if (vert_reg_to_vert_result(c, reg, 1) != BRW_VERT_RESULT_MAX) {
> *pc |= 0xf0;
>
> - if (persp_mask & BITFIELD64_BIT(vert_reg_to_vert_result(c, reg, 1)))
> - *pc_persp |= 0xf0;
> -
> - if (linear_mask & BITFIELD64_BIT(vert_reg_to_vert_result(c, reg,
> 1)))
> - *pc_linear |= 0xf0;
> + interp = brw_get_interpolation_mode(brw, vert_reg_to_vue_slot(c,
> reg, 1));
> + if (interp == INTERP_QUALIFIER_SMOOTH) {
> + *pc_linear |= 0xf0;
> + *pc_persp |= 0xf0;
> + } else if(interp == INTERP_QUALIFIER_NOPERSPECTIVE)
> + *pc_linear |= 0xf0;
> }
>
> return is_last_attr;
> @@ -430,7 +432,7 @@ void brw_emit_tri_setup(struct brw_sf_compile *c, bool
> allocate)
> if (c->key.do_twoside_color)
> do_twoside_color(c);
>
> - if (c->key.do_flat_shading)
> + if (c->key.has_flat_shading)
> do_flatshade_triangle(c);
>
>
> @@ -443,7 +445,6 @@ void brw_emit_tri_setup(struct brw_sf_compile *c, bool
> allocate)
> struct brw_reg a2 = offset(c->vert[2], i);
> GLushort pc, pc_persp, pc_linear;
> bool last = calculate_masks(c, i, &pc, &pc_persp, &pc_linear);
> -
> if (pc_persp)
> {
> brw_set_predicate_control_flag_value(p, pc_persp);
> @@ -507,7 +508,6 @@ void brw_emit_line_setup(struct brw_sf_compile *c,
> bool allocate)
> struct brw_compile *p = &c->func;
> GLuint i;
>
> -
> c->nr_verts = 2;
>
> if (allocate)
> @@ -516,7 +516,7 @@ void brw_emit_line_setup(struct brw_sf_compile *c,
> bool allocate)
> invert_det(c);
> copy_z_inv_w(c);
>
> - if (c->key.do_flat_shading)
> + if (c->key.has_flat_shading)
> do_flatshade_line(c);
>
> for (i = 0; i < c->nr_setup_regs; i++)
> @@ -799,7 +799,3 @@ void brw_emit_anyprim_setup( struct brw_sf_compile *c )
>
> brw_emit_point_setup( c, false );
> }
> -
> -
> -
> -
> --
> 1.7.10.rc3.1.gb306
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/mesa-dev
>
With the minor changes noted above, this patch is:
Reviewed-by: Paul Berry <stereotype441 at gmail.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-gfx/attachments/20120717/790b428e/attachment.html>
More information about the Intel-gfx
mailing list