[Mesa-dev] [RFC] i965: Factor out descriptor building for indirect send messages
Francisco Jerez
currojerez at riseup.net
Sat Mar 7 06:15:08 PST 2015
Topi Pohjolainen <topi.pohjolainen at intel.com> writes:
> The original patch from Curro was based on something that is not
> present in the master yet. This patch tries to mimick the logic on
> top master.
> I wanted to see if could separate the building of the descriptor
> instruction from building of the send instruction. This logic now
> allows the caller to construct any kind of sequence of instructions
> filling in the descriptor before giving it to the send instruction
> builder.
>
> This is only compile tested. Curro, how would you feel about this
> sort of approach? I apologise for muddying the waters again but I
> wasn't entirely comfortable with the logic and wanted to try to
> something else.
>
> I believe patch number five should go nicely on top of this as
> the descriptor instruction could be followed by (or preceeeded by)
> any additional instructions modifying the descriptor register
> before the actual send instruction.
>
Topi, the goal I had in mind with PATCH 01 was to refactor a commonly
recurring pattern. In terms of the functions defined in this patch my
example from yesterday [1] would now look like:
| if (index.file == BRW_IMMEDIATE_VALUE) {
|
| uint32_t surf_index = index.dw1.ud;
|
| brw_inst *send = brw_next_insn(p, BRW_OPCODE_SEND);
| brw_set_dest(p, send, retype(dst, BRW_REGISTER_TYPE_UW));
| brw_set_src0(p, send, offset);
| brw_set_sampler_message(p, send,
| surf_index,
| 0, /* LD message ignores sampler unit */
| GEN5_SAMPLER_MESSAGE_SAMPLE_LD,
| rlen,
| mlen,
| false, /* no header */
| simd_mode,
| 0);
|
| brw_mark_surface_used(prog_data, surf_index);
|
| } else {
|
| struct brw_reg addr = vec1(retype(brw_address_reg(0), BRW_REGISTER_TYPE_UD));
|
| brw_push_insn_state(p);
| brw_set_default_mask_control(p, BRW_MASK_DISABLE);
| brw_set_default_access_mode(p, BRW_ALIGN_1);
|
| /* a0.0 = surf_index & 0xff */
| brw_inst *insn_and = brw_next_insn(p, BRW_OPCODE_AND);
| brw_inst_set_exec_size(p->brw, insn_and, BRW_EXECUTE_1);
| brw_set_dest(p, insn_and, addr);
| brw_set_src0(p, insn_and, vec1(retype(index, BRW_REGISTER_TYPE_UD)));
| brw_set_src1(p, insn_and, brw_imm_ud(0x0ff));
|
|
| /* a0.0 |= <descriptor> */
| brw_inst *descr_inst = brw_build_indirect_message_descr(p, addr, addr);
| brw_set_sampler_message(p, descr_inst,
| 0 /* surface */,
| 0 /* sampler */,
| GEN5_SAMPLER_MESSAGE_SAMPLE_LD,
| rlen /* rlen */,
| mlen /* mlen */,
| false /* header */,
| simd_mode,
| 0);
|
| /* dst = send(offset, a0.0) */
| brw_send_indirect_message(p, BRW_SFID_SAMPLER, dst, offset, addr);
|
| brw_pop_insn_state(p);
|
| /* visitor knows more than we do about the surface limit required,
| * so has already done marking.
| */
| }
I agree that this would be a modest improvement, but it still requires
the caller to handle the immediate and indirect descriptor cases
explicitly, which is what I wanted to avoid because the exact same
mechanism would still have to be repeated in each of the 11 indirect
message sends (counting the new surface opcodes). If you look at my
PATCH 11-12, you won't find any explicit handling for indirect surface
indices, the code for emitting each of the typed or untyped surface
opcodes looks like:
| void
| brw_typed_surface_read(/* ... */)
| {
| const struct brw_context *brw = p->brw;
| const unsigned sfid = /* ... */;
| struct brw_inst *insn = brw_send_indirect_surface_message(
| p, sfid, dst, payload, surface, msg_length,
| brw_surface_payload_size(p, num_channels,
| brw->gen >= 8 || brw->is_haswell, false),
| true);
|
| brw_set_dp_typed_surface_read_message(
| p, insn, num_channels);
| }
Which works regardless of whether surface is an immediate or a register
thanks to brw_send_indirect_message().
[1] http://lists.freedesktop.org/archives/mesa-dev/2015-March/078746.html
> Signed-off-by: Topi Pohjolainen <topi.pohjolainen at intel.com>
> ---
> src/mesa/drivers/dri/i965/brw_eu.h | 17 +++++--
> src/mesa/drivers/dri/i965/brw_eu_emit.c | 53 ++++++++++++++++------
> src/mesa/drivers/dri/i965/brw_fs_generator.cpp | 58 ++++++------------------
> src/mesa/drivers/dri/i965/brw_vec4_generator.cpp | 38 ++++------------
> 4 files changed, 75 insertions(+), 91 deletions(-)
>
> diff --git a/src/mesa/drivers/dri/i965/brw_eu.h b/src/mesa/drivers/dri/i965/brw_eu.h
> index 736c54b..82570f5 100644
> --- a/src/mesa/drivers/dri/i965/brw_eu.h
> +++ b/src/mesa/drivers/dri/i965/brw_eu.h
> @@ -205,11 +205,6 @@ void brw_set_sampler_message(struct brw_compile *p,
> unsigned simd_mode,
> unsigned return_format);
>
> -void brw_set_indirect_send_descriptor(struct brw_compile *p,
> - brw_inst *insn,
> - unsigned sfid,
> - struct brw_reg descriptor);
> -
> void brw_set_dp_read_message(struct brw_compile *p,
> brw_inst *insn,
> unsigned binding_table_index,
> @@ -242,6 +237,18 @@ void brw_urb_WRITE(struct brw_compile *p,
> unsigned offset,
> unsigned swizzle);
>
> +struct brw_inst *
> +brw_build_indirect_message_descr(struct brw_compile *p,
> + struct brw_reg dst,
> + struct brw_reg src);
> +
> +struct brw_inst *
> +brw_send_indirect_message(struct brw_compile *p,
> + unsigned sfid,
> + struct brw_reg dst,
> + struct brw_reg payload,
> + struct brw_reg desc);
> +
> void brw_ff_sync(struct brw_compile *p,
> struct brw_reg dest,
> unsigned msg_reg_nr,
> diff --git a/src/mesa/drivers/dri/i965/brw_eu_emit.c b/src/mesa/drivers/dri/i965/brw_eu_emit.c
> index 6f29468..cd0d199 100644
> --- a/src/mesa/drivers/dri/i965/brw_eu_emit.c
> +++ b/src/mesa/drivers/dri/i965/brw_eu_emit.c
> @@ -772,21 +772,6 @@ brw_set_sampler_message(struct brw_compile *p,
> }
> }
>
> -void brw_set_indirect_send_descriptor(struct brw_compile *p,
> - brw_inst *insn,
> - unsigned sfid,
> - struct brw_reg descriptor)
> -{
> - /* Only a0.0 may be used as SEND's descriptor operand. */
> - assert(descriptor.file == BRW_ARCHITECTURE_REGISTER_FILE);
> - assert(descriptor.type == BRW_REGISTER_TYPE_UD);
> - assert(descriptor.nr == BRW_ARF_ADDRESS);
> - assert(descriptor.subnr == 0);
> -
> - brw_set_message_descriptor(p, insn, sfid, 0, 0, false, false);
> - brw_set_src1(p, insn, descriptor);
> -}
> -
> static void
> gen7_set_dp_scratch_message(struct brw_compile *p,
> brw_inst *inst,
> @@ -2495,6 +2480,44 @@ void brw_urb_WRITE(struct brw_compile *p,
> swizzle);
> }
>
> +struct brw_inst *
> +brw_build_indirect_message_descr(struct brw_compile *p,
> + struct brw_reg dst,
> + struct brw_reg src)
> +{
> + assert(dst.type == BRW_REGISTER_TYPE_UD);
> + assert(src.type == BRW_REGISTER_TYPE_UD);
> +
> + brw_push_insn_state(p);
> + brw_set_default_access_mode(p, BRW_ALIGN_1);
> + brw_set_default_mask_control(p, BRW_MASK_DISABLE);
> + brw_set_default_predicate_control(p, BRW_PREDICATE_NONE);
> +
> + struct brw_inst *setup = brw_OR(p, dst, src, brw_imm_ud(0));
> +
> + brw_pop_insn_state(p);
> +
> + return setup;
> +}
> +
> +struct brw_inst *
> +brw_send_indirect_message(struct brw_compile *p,
> + unsigned sfid,
> + struct brw_reg dst,
> + struct brw_reg payload,
> + struct brw_reg desc)
> +{
> + const struct brw_context *brw = p->brw;
> + struct brw_inst *send = next_insn(p, BRW_OPCODE_SEND);
> +
> + brw_set_src1(p, send, desc);
> + brw_set_dest(p, send, retype(dst, BRW_REGISTER_TYPE_UD));
> + brw_set_src0(p, send, retype(payload, BRW_REGISTER_TYPE_UD));
> + brw_inst_set_sfid(brw, send, sfid);
> +
> + return send;
> +}
> +
> static int
> brw_find_next_block_end(struct brw_compile *p, int start_offset)
> {
> diff --git a/src/mesa/drivers/dri/i965/brw_fs_generator.cpp b/src/mesa/drivers/dri/i965/brw_fs_generator.cpp
> index 02ea3b6..38419cb 100644
> --- a/src/mesa/drivers/dri/i965/brw_fs_generator.cpp
> +++ b/src/mesa/drivers/dri/i965/brw_fs_generator.cpp
> @@ -751,9 +751,10 @@ fs_generator::generate_tex(fs_inst *inst, struct brw_reg dst, struct brw_reg src
> brw_AND(p, addr, addr, brw_imm_ud(0x0ff));
> brw_OR(p, addr, addr, temp);
>
> - /* a0.0 |= <descriptor> */
> - brw_inst *insn_or = brw_next_insn(p, BRW_OPCODE_OR);
> - brw_set_sampler_message(p, insn_or,
> + /* dst = send(offset, a0.0 | <descriptor>) */
> + struct brw_reg descr = retype(brw_address_reg(0), BRW_REGISTER_TYPE_UD);
> + brw_inst *descr_inst = brw_build_indirect_message_descr(p, descr, addr);
> + brw_set_sampler_message(p, descr_inst,
> 0 /* surface */,
> 0 /* sampler */,
> msg_type,
> @@ -762,17 +763,7 @@ fs_generator::generate_tex(fs_inst *inst, struct brw_reg dst, struct brw_reg src
> inst->header_present /* header */,
> simd_mode,
> return_format);
> - brw_inst_set_exec_size(p->brw, insn_or, BRW_EXECUTE_1);
> - brw_inst_set_src1_reg_type(p->brw, insn_or, BRW_REGISTER_TYPE_UD);
> - brw_set_src0(p, insn_or, addr);
> - brw_set_dest(p, insn_or, addr);
> -
> -
> - /* dst = send(offset, a0.0) */
> - brw_inst *insn_send = brw_next_insn(p, BRW_OPCODE_SEND);
> - brw_set_dest(p, insn_send, dst);
> - brw_set_src0(p, insn_send, src);
> - brw_set_indirect_send_descriptor(p, insn_send, BRW_SFID_SAMPLER, addr);
> + brw_send_indirect_message(p, BRW_SFID_SAMPLER, dst, src, descr);
>
> brw_pop_insn_state(p);
>
> @@ -1080,10 +1071,10 @@ fs_generator::generate_uniform_pull_constant_load_gen7(fs_inst *inst,
> brw_set_src0(p, insn_and, vec1(retype(index, BRW_REGISTER_TYPE_UD)));
> brw_set_src1(p, insn_and, brw_imm_ud(0x0ff));
>
> -
> - /* a0.0 |= <descriptor> */
> - brw_inst *insn_or = brw_next_insn(p, BRW_OPCODE_OR);
> - brw_set_sampler_message(p, insn_or,
> + /* dst = send(payload, a0.0 | <descriptor>) */
> + struct brw_reg descr = retype(brw_address_reg(0), BRW_REGISTER_TYPE_UD);
> + brw_inst *descr_inst = brw_build_indirect_message_descr(p, descr, addr);
> + brw_set_sampler_message(p, descr_inst,
> 0 /* surface */,
> 0 /* sampler */,
> GEN5_SAMPLER_MESSAGE_SAMPLE_LD,
> @@ -1092,17 +1083,7 @@ fs_generator::generate_uniform_pull_constant_load_gen7(fs_inst *inst,
> header_present,
> BRW_SAMPLER_SIMD_MODE_SIMD4X2,
> 0);
> - brw_inst_set_exec_size(p->brw, insn_or, BRW_EXECUTE_1);
> - brw_inst_set_src1_reg_type(p->brw, insn_or, BRW_REGISTER_TYPE_UD);
> - brw_set_src0(p, insn_or, addr);
> - brw_set_dest(p, insn_or, addr);
> -
> -
> - /* dst = send(offset, a0.0) */
> - brw_inst *insn_send = brw_next_insn(p, BRW_OPCODE_SEND);
> - brw_set_dest(p, insn_send, dst);
> - brw_set_src0(p, insn_send, src);
> - brw_set_indirect_send_descriptor(p, insn_send, BRW_SFID_SAMPLER, addr);
> + brw_send_indirect_message(p, BRW_SFID_SAMPLER, dst, src, descr);
>
> brw_pop_insn_state(p);
>
> @@ -1240,9 +1221,10 @@ fs_generator::generate_varying_pull_constant_load_gen7(fs_inst *inst,
> brw_set_src1(p, insn_and, brw_imm_ud(0x0ff));
>
>
> - /* a0.0 |= <descriptor> */
> - brw_inst *insn_or = brw_next_insn(p, BRW_OPCODE_OR);
> - brw_set_sampler_message(p, insn_or,
> + /* dst = send(offset, a0.0 | <descriptor>) */
> + struct brw_reg descr = retype(brw_address_reg(0), BRW_REGISTER_TYPE_UD);
> + brw_inst *descr_inst = brw_build_indirect_message_descr(p, descr, addr);
> + brw_set_sampler_message(p, descr_inst,
> 0 /* surface */,
> 0 /* sampler */,
> GEN5_SAMPLER_MESSAGE_SAMPLE_LD,
> @@ -1251,17 +1233,7 @@ fs_generator::generate_varying_pull_constant_load_gen7(fs_inst *inst,
> false /* header */,
> simd_mode,
> 0);
> - brw_inst_set_exec_size(p->brw, insn_or, BRW_EXECUTE_1);
> - brw_inst_set_src1_reg_type(p->brw, insn_or, BRW_REGISTER_TYPE_UD);
> - brw_set_src0(p, insn_or, addr);
> - brw_set_dest(p, insn_or, addr);
> -
> -
> - /* dst = send(offset, a0.0) */
> - brw_inst *insn_send = brw_next_insn(p, BRW_OPCODE_SEND);
> - brw_set_dest(p, insn_send, retype(dst, BRW_REGISTER_TYPE_UW));
> - brw_set_src0(p, insn_send, offset);
> - brw_set_indirect_send_descriptor(p, insn_send, BRW_SFID_SAMPLER, addr);
> + brw_send_indirect_message(p, BRW_SFID_SAMPLER, dst, offset, descr);
>
> brw_pop_insn_state(p);
>
> diff --git a/src/mesa/drivers/dri/i965/brw_vec4_generator.cpp b/src/mesa/drivers/dri/i965/brw_vec4_generator.cpp
> index 010a5c4..6cdd611 100644
> --- a/src/mesa/drivers/dri/i965/brw_vec4_generator.cpp
> +++ b/src/mesa/drivers/dri/i965/brw_vec4_generator.cpp
> @@ -419,9 +419,10 @@ vec4_generator::generate_tex(vec4_instruction *inst,
> brw_AND(p, addr, addr, brw_imm_ud(0x0ff));
> brw_OR(p, addr, addr, temp);
>
> - /* a0.0 |= <descriptor> */
> - brw_inst *insn_or = brw_next_insn(p, BRW_OPCODE_OR);
> - brw_set_sampler_message(p, insn_or,
> + /* dst = send(offset, a0.0 | <descriptor>) */
> + struct brw_reg descr = retype(brw_address_reg(0), BRW_REGISTER_TYPE_UD);
> + brw_inst *descr_inst = brw_build_indirect_message_descr(p, descr, addr);
> + brw_set_sampler_message(p, descr_inst,
> 0 /* surface */,
> 0 /* sampler */,
> msg_type,
> @@ -430,17 +431,7 @@ vec4_generator::generate_tex(vec4_instruction *inst,
> inst->header_present /* header */,
> BRW_SAMPLER_SIMD_MODE_SIMD4X2,
> return_format);
> - brw_inst_set_exec_size(p->brw, insn_or, BRW_EXECUTE_1);
> - brw_inst_set_src1_reg_type(p->brw, insn_or, BRW_REGISTER_TYPE_UD);
> - brw_set_src0(p, insn_or, addr);
> - brw_set_dest(p, insn_or, addr);
> -
> -
> - /* dst = send(offset, a0.0) */
> - brw_inst *insn_send = brw_next_insn(p, BRW_OPCODE_SEND);
> - brw_set_dest(p, insn_send, dst);
> - brw_set_src0(p, insn_send, src);
> - brw_set_indirect_send_descriptor(p, insn_send, BRW_SFID_SAMPLER, addr);
> + brw_send_indirect_message(p, BRW_SFID_SAMPLER, dst, src, descr);
>
> brw_pop_insn_state(p);
>
> @@ -1082,9 +1073,10 @@ vec4_generator::generate_pull_constant_load_gen7(vec4_instruction *inst,
> brw_set_src1(p, insn_and, brw_imm_ud(0x0ff));
>
>
> - /* a0.0 |= <descriptor> */
> - brw_inst *insn_or = brw_next_insn(p, BRW_OPCODE_OR);
> - brw_set_sampler_message(p, insn_or,
> + /* dst = send(offset, a0.0 | <descriptor>) */
> + struct brw_reg descr = retype(brw_address_reg(0), BRW_REGISTER_TYPE_UD);
> + brw_inst *descr_inst = brw_build_indirect_message_descr(p, descr, addr);
> + brw_set_sampler_message(p, descr_inst,
> 0 /* surface */,
> 0 /* sampler */,
> GEN5_SAMPLER_MESSAGE_SAMPLE_LD,
> @@ -1093,17 +1085,7 @@ vec4_generator::generate_pull_constant_load_gen7(vec4_instruction *inst,
> false /* header */,
> BRW_SAMPLER_SIMD_MODE_SIMD4X2,
> 0);
> - brw_inst_set_exec_size(p->brw, insn_or, BRW_EXECUTE_1);
> - brw_inst_set_src1_reg_type(p->brw, insn_or, BRW_REGISTER_TYPE_UD);
> - brw_set_src0(p, insn_or, addr);
> - brw_set_dest(p, insn_or, addr);
> -
> -
> - /* dst = send(offset, a0.0) */
> - brw_inst *insn_send = brw_next_insn(p, BRW_OPCODE_SEND);
> - brw_set_dest(p, insn_send, dst);
> - brw_set_src0(p, insn_send, offset);
> - brw_set_indirect_send_descriptor(p, insn_send, BRW_SFID_SAMPLER, addr);
> + brw_send_indirect_message(p, BRW_SFID_SAMPLER, dst, offset, descr);
>
> brw_pop_insn_state(p);
>
> --
> 1.9.3
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/mesa-dev
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 212 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/mesa-dev/attachments/20150307/49cf1648/attachment-0001.sig>
More information about the mesa-dev
mailing list