[Mesa-dev] [PATCH 4/4] OPTIONAL: anv/gen9: Optimize slice and subslice load balancing behavior.
Francisco Jerez
currojerez at riseup.net
Sat Aug 10 21:55:55 UTC 2019
Jason Ekstrand <jason at jlekstrand.net> writes:
> On Sat, Aug 10, 2019 at 2:22 PM Francisco Jerez <currojerez at riseup.net>
> wrote:
>
>> Jason Ekstrand <jason at jlekstrand.net> writes:
>>
>> > On Fri, Aug 9, 2019 at 7:22 PM Francisco Jerez <currojerez at riseup.net>
>> > wrote:
>> >
>> >> See "i965/gen9: Optimize slice and subslice load balancing behavior."
>> >> for the rationale. Marked optional because no performance evaluation
>> >> has been done on this commit, it is provided to match the hashing
>> >> settings of the Iris driver. Test reports welcome.
>> >>
>> >
>> > Improves Aztec Ruins by 2.7% (tested 5 times in each configuration). I
>> > also tested Batman: Arkham City but it only reports integer FPS numbers
>> and
>> > no change was noticable.
>> >
>> >
>> >> ---
>> >> src/intel/vulkan/anv_genX.h | 4 ++
>> >> src/intel/vulkan/anv_private.h | 6 ++
>> >> src/intel/vulkan/genX_blorp_exec.c | 6 ++
>> >> src/intel/vulkan/genX_cmd_buffer.c | 96 ++++++++++++++++++++++++++++++
>> >> 4 files changed, 112 insertions(+)
>> >>
>> >> diff --git a/src/intel/vulkan/anv_genX.h b/src/intel/vulkan/anv_genX.h
>> >> index a5435e566a3..06c6b467acf 100644
>> >> --- a/src/intel/vulkan/anv_genX.h
>> >> +++ b/src/intel/vulkan/anv_genX.h
>> >> @@ -44,6 +44,10 @@ void genX(cmd_buffer_apply_pipe_flushes)(struct
>> >> anv_cmd_buffer *cmd_buffer);
>> >>
>> >> void genX(cmd_buffer_emit_gen7_depth_flush)(struct anv_cmd_buffer
>> >> *cmd_buffer);
>> >>
>> >> +void genX(cmd_buffer_emit_hashing_mode)(struct anv_cmd_buffer
>> *cmd_buffer,
>> >> + unsigned width, unsigned
>> height,
>> >> + unsigned scale);
>> >> +
>> >> void genX(flush_pipeline_select_3d)(struct anv_cmd_buffer *cmd_buffer);
>> >> void genX(flush_pipeline_select_gpgpu)(struct anv_cmd_buffer
>> *cmd_buffer);
>> >>
>> >> diff --git a/src/intel/vulkan/anv_private.h
>> >> b/src/intel/vulkan/anv_private.h
>> >> index 2465f264354..b381386a716 100644
>> >> --- a/src/intel/vulkan/anv_private.h
>> >> +++ b/src/intel/vulkan/anv_private.h
>> >> @@ -2421,6 +2421,12 @@ struct anv_cmd_state {
>> >>
>> >> bool
>> >> conditional_render_enabled;
>> >>
>> >> + /**
>> >> + * Last rendering scale argument provided to
>> >> + * genX(cmd_buffer_emit_hashing_mode)().
>> >> + */
>> >> + unsigned current_hash_scale;
>> >> +
>> >> /**
>> >> * Array length is anv_cmd_state::pass::attachment_count. Array
>> >> content is
>> >> * valid only when recording a render pass instance.
>> >> diff --git a/src/intel/vulkan/genX_blorp_exec.c
>> >> b/src/intel/vulkan/genX_blorp_exec.c
>> >> index 1592e7f7e3d..e9eedc06696 100644
>> >> --- a/src/intel/vulkan/genX_blorp_exec.c
>> >> +++ b/src/intel/vulkan/genX_blorp_exec.c
>> >> @@ -223,6 +223,12 @@ genX(blorp_exec)(struct blorp_batch *batch,
>> >> genX(cmd_buffer_config_l3)(cmd_buffer, cfg);
>> >> }
>> >>
>> >> + const unsigned scale = params->fast_clear_op ? UINT_MAX : 1;
>> >> + if (cmd_buffer->state.current_hash_scale != scale) {
>> >> + genX(cmd_buffer_emit_hashing_mode)(cmd_buffer, params->x1 -
>> >> params->x0,
>> >> + params->y1 - params->y0,
>> scale);
>> >> + }
>> >> +
>> >> #if GEN_GEN >= 11
>> >> /* The PIPE_CONTROL command description says:
>> >> *
>> >> diff --git a/src/intel/vulkan/genX_cmd_buffer.c
>> >> b/src/intel/vulkan/genX_cmd_buffer.c
>> >> index 86ef1663ac4..e9e5570d49f 100644
>> >> --- a/src/intel/vulkan/genX_cmd_buffer.c
>> >> +++ b/src/intel/vulkan/genX_cmd_buffer.c
>> >> @@ -1595,6 +1595,7 @@ genX(CmdExecuteCommands)(
>> >> */
>> >> primary->state.current_pipeline = UINT32_MAX;
>> >> primary->state.current_l3_config = NULL;
>> >> + primary->state.current_hash_scale = 0;
>> >>
>> >> /* Each of the secondary command buffers will use its own state base
>> >> * address. We need to re-emit state base address for the primary
>> >> after
>> >> @@ -2663,6 +2664,9 @@ genX(cmd_buffer_flush_state)(struct anv_cmd_buffer
>> >> *cmd_buffer)
>> >>
>> >> genX(cmd_buffer_config_l3)(cmd_buffer, pipeline->urb.l3_config);
>> >>
>> >> + if (cmd_buffer->state.current_hash_scale != 1)
>> >> + genX(cmd_buffer_emit_hashing_mode)(cmd_buffer, UINT_MAX,
>> UINT_MAX,
>> >> 1);
>> >>
>> >
>> > Can we do the check against current_hash_scale as an early return inside
>> > the emit function since that's already where the "current" value is
>> > updated? Maybe call it cmd_buffer_config_hashing_mode if "emit" no
>> longer
>> > seems right with that change? I don't really care about the rename.
>> >
>>
>> That's what my first revision did. I moved the current_hash_scale check
>> one level up because the function call overhead was visible in
>> brw_upload_pipeline_state(), which is quite a hot path in i965.
>
>
> That's a bit surprising but believable if the two are in different files
> and there's a real function call going on there.
>
>
>> That
>> said I don't have any data justifying the change in the Vulkan driver --
>> Because caller and callee are defined in the same compilation unit it
>> *might* be that the function is inlined into
>> genX(cmd_buffer_flush_state), though it's large enough that it may not
>>
>
> I highly doubt it would be a problem. It's a lot of lines of code but once
> you optimize things a bit, it's actually quite a small function.
>
>
>> be, either way it seemed like a good idea to do things consistently
>> across drivers.
>>
>
> The functions right next to it for l3$ configuration and PIPELINE_SELECT
> are both called unconditionally. IMO, consistency within the Vulkan driver
> is more important than consistency between iris and ANV given that they
> have completely different state tracking models.
>
I don't really care one way or another since I have no data justifying
the micro-optimization in Vulkan (I do have data showing a statistically
significant benefit in CPU-bound tests from the corresponding change on
i965). Will gladly undo the change in exchange for an R-b.
> --Jason
>
>
>> >
>> >> +
>> >> genX(flush_pipeline_select_3d)(cmd_buffer);
>> >>
>> >> if (vb_emit) {
>> >> @@ -3925,6 +3929,98 @@ genX(cmd_buffer_emit_gen7_depth_flush)(struct
>> >> anv_cmd_buffer *cmd_buffer)
>> >> }
>> >> }
>> >>
>> >> +/**
>> >> + * Update the pixel hashing modes that determine the balancing of PS
>> >> threads
>> >> + * across subslices and slices.
>> >> + *
>> >> + * \param width Width bound of the rendering area (already scaled down
>> if
>> >> \p
>> >> + * scale is greater than 1).
>> >> + * \param height Height bound of the rendering area (already scaled
>> down
>> >> if \p
>> >> + * scale is greater than 1).
>> >> + * \param scale The number of framebuffer samples that could
>> potentially
>> >> be
>> >> + * affected by an individual channel of the PS thread.
>> This
>> >> is
>> >> + * typically one for single-sampled rendering, but for
>> >> operations
>> >> + * like CCS resolves and fast clears a single PS
>> invocation
>> >> may
>> >> + * update a huge number of pixels, in which case a finer
>> >> + * balancing is desirable in order to maximally utilize
>> the
>> >> + * bandwidth available. UINT_MAX can be used as shorthand
>> >> for
>> >> + * "finest hashing mode available".
>> >> + */
>> >> +void
>> >> +genX(cmd_buffer_emit_hashing_mode)(struct anv_cmd_buffer *cmd_buffer,
>> >> + unsigned width, unsigned height,
>> >> + unsigned scale)
>> >> +{
>> >> +#if GEN_GEN == 9
>> >> + const struct gen_device_info *devinfo = &cmd_buffer->device->info;
>> >> + const unsigned slice_hashing[] = {
>> >> + /* Because all Gen9 platforms with more than one slice require
>> >> + * three-way subslice hashing, a single "normal" 16x16 slice
>> hashing
>> >> + * block is guaranteed to suffer from substantial imbalance, with
>> >> one
>> >> + * subslice receiving twice as much work as the other two in the
>> >> + * slice.
>> >> + *
>> >> + * The performance impact of that would be particularly severe
>> when
>> >> + * three-way hashing is also in use for slice balancing (which is
>> >> the
>> >> + * case for all Gen9 GT4 platforms), because one of the slices
>> >> + * receives one every three 16x16 blocks in either direction,
>> which
>> >> + * is roughly the periodicity of the underlying subslice
>> imbalance
>> >> + * pattern ("roughly" because in reality the hardware's
>> >> + * implementation of three-way hashing doesn't do exact modulo 3
>> >> + * arithmetic, which somewhat decreases the magnitude of this
>> effect
>> >> + * in practice). This leads to a systematic subslice imbalance
>> >> + * within that slice regardless of the size of the primitive.
>> The
>> >> + * 32x32 hashing mode guarantees that the subslice imbalance
>> within
>> >> a
>> >> + * single slice hashing block is minimal, largely eliminating
>> this
>> >> + * effect.
>> >> + */
>> >> + _32x32,
>> >> + /* Finest slice hashing mode available. */
>> >> + NORMAL
>> >> + };
>> >> + const unsigned subslice_hashing[] = {
>> >> + /* 16x16 would provide a slight cache locality benefit especially
>> >> + * visible in the sampler L1 cache efficiency of low-bandwidth
>> >> + * non-LLC platforms, but it comes at the cost of greater
>> subslice
>> >> + * imbalance for primitives of dimensions approximately
>> intermediate
>> >> + * between 16x4 and 16x16.
>> >> + */
>> >> + _16x4,
>> >> + /* Finest subslice hashing mode available. */
>> >> + _8x4
>> >> + };
>> >> + /* Dimensions of the smallest hashing block of a given hashing mode.
>> >> If
>> >> + * the rendering area is smaller than this there can't possibly be
>> any
>> >> + * benefit from switching to this mode, so we optimize out the
>> >> + * transition.
>> >> + */
>> >> + const unsigned min_size[][2] = {
>> >> + { 16, 4 },
>> >> + { 8, 4 }
>> >> + };
>> >> + const unsigned idx = scale > 1;
>> >> +
>> >> + if (width > min_size[idx][0] || height > min_size[idx][1]) {
>> >> + uint32_t gt_mode;
>> >> +
>> >> + anv_pack_struct(>_mode, GENX(GT_MODE),
>> >> + .SliceHashing = (devinfo->num_slices > 1 ?
>> >> slice_hashing[idx] : 0),
>> >> + .SliceHashingMask = (devinfo->num_slices > 1 ?
>> -1 :
>> >> 0),
>> >> + .SubsliceHashing = subslice_hashing[idx],
>> >> + .SubsliceHashingMask = -1);
>> >> +
>> >> + anv_batch_emit(&cmd_buffer->batch, GENX(PIPE_CONTROL), pc) {
>> >> + pc.StallAtPixelScoreboard = true;
>> >> + pc.CommandStreamerStallEnable = true;
>> >> + }
>> >>
>> >
>> > This would be better as
>> >
>> > cmd_buffer->state.pending_pipe_bits |= ANV_PIPE_CS_STALL_BIT;
>> > genX(cmd_buffer_apply_pipe_flushes)(cmd_buffer);
>> >
>> > That way workarounds get applied more properly and we can avoid redundant
>> > PIPE_CONTROLs. apply_pipe_flushes will insert the StallAtPixelScoreboard
>> > for you if needed. That said, maybe it's best to also OR in
>> > ANV_PIPE_STALL_AT_SCOREBOARD_BIT just to be sure.
>> >
>>
>> OK, changed locally.
>>
>> > Other than that, the patch looks good to me. Happy to see the perf
>> bump. :)
>> >
>>
>> Thanks for testing it out!
>>
>> > I did consider maybe trying to move things around so that the change is
>> > done more as a subpass thing but there are some subtle interactions to
>> > think through there and I'm convinced what you have here is correct so
>> > let's go with it for now.
>> >
>> > --Jason
>> >
>> >
>> >> +
>> >> + emit_lri(&cmd_buffer->batch, GENX(GT_MODE_num), gt_mode);
>> >> +
>> >> + cmd_buffer->state.current_hash_scale = scale;
>> >> + }
>> >> +#endif
>> >> +}
>> >> +
>> >> static void
>> >> cmd_buffer_emit_depth_stencil(struct anv_cmd_buffer *cmd_buffer)
>> >> {
>> >> --
>> >> 2.22.0
>> >>
>> >> _______________________________________________
>> >> mesa-dev mailing list
>> >> mesa-dev at lists.freedesktop.org
>> >> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
>>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 227 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20190810/1a63da87/attachment.sig>
More information about the mesa-dev
mailing list