ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-20 06:49:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

[CK_TILE] FMHA BWD Remove Unnecessary Padding (#2550)

Browse Source 
* Remove unnecessary pssk

* Add BlockFmhaBwdDQDKDVPipeline wrapper

* Resolve copilot comments & Remove kpad & fix

* Remove spad
This commit is contained in:
Yi DING
2025-08-07 21:24:43 +08:00
committed by GitHub
parent ffdee5e774
commit b0a97498b0
11 changed files with 158 additions and 189 deletions
Download Patch File Download Diff File Expand all files Collapse all files

54
include/ck_tile/ops/fmha/kernel/fmha_bwd_kernel.hpp
View File

@@ -52,8 +52,6 @@ struct FmhaBwdDQDKDVKernel
using BiasGradDataType = ck_tile::remove_cvref_t<typename FmhaPipeline::BiasGradDataType>;
static constexpr bool kIsGroupMode = FmhaPipeline::kIsGroupMode;
static constexpr bool kPadSeqLenQ = FmhaPipeline::kPadSeqLenQ;
static constexpr bool kPadSeqLenK = FmhaPipeline::kPadSeqLenK;
static constexpr bool kPadHeadDimQ = FmhaPipeline::kPadHeadDimQ;
static constexpr bool kPadHeadDimV = FmhaPipeline::kPadHeadDimV;
static constexpr auto BiasEnum = FmhaPipeline::BiasEnum;
@@ -85,8 +83,6 @@ struct FmhaBwdDQDKDVKernel
#define _TS_ std::to_string
auto pn = [&] () {
std::string n;
if (kPadSeqLenQ) n += "s";
if (kPadSeqLenK) n += "sk";
if (kPadHeadDimQ) n += "d";
if (kPadHeadDimV) n += "dv";
return n.empty() ? n : std::string("p") + n; }();
@@ -100,7 +96,7 @@ struct FmhaBwdDQDKDVKernel
"r" + _TS_(gbr4::at(ck_tile::number<0>{})) + "x" + _TS_(gbr4::at(ck_tile::number<1>{})) + "x" + _TS_(gbr4::at(ck_tile::number<2>{})) + "_" +
"w" + _TS_(gwt0::at(ck_tile::number<0>{})) + "x" + _TS_(gwt0::at(ck_tile::number<1>{})) + "x" + _TS_(gwt0::at(ck_tile::number<2>{})) + "_" +
"w" + _TS_(gwt1::at(ck_tile::number<0>{})) + "x" + _TS_(gwt1::at(ck_tile::number<1>{})) + "x" + _TS_(gwt1::at(ck_tile::number<2>{})) + "_" +
("o" + _TS_(kBlockPerCu) + "_") + _SS_(FmhaPipeline::name) + (pn.empty() ? "_npad" : "_" + pn) +
("o" + _TS_(kBlockPerCu)) + (pn.empty() ? "_npad" : "_" + pn) +
(BiasEnum == BlockAttentionBiasEnum::NO_BIAS ? _SS_("_nbias") : (_SS_("_") + BlockAttentionBiasEnumToStr<BiasEnum>::name)) +
(kHasBiasGrad ? "_dbias" : "_ndbias") + (kHasMask ? "_" + _SS_(FmhaMask::name) : "_nmask") + (kHasDropout ? "_dropout" : "_ndropout" ) +
(kIsStoreRandval ? "_storerandval" : "" ) + (kIsDeterministic ? "_deterministic" : "_ndeterministic" );
@@ -1221,7 +1217,7 @@ struct FmhaBwdDQDKDVKernel
const auto q_dram = pad_tensor_view(
q_dram_naive,
make_tuple(number<FmhaPipeline::kM0>{}, number<FmhaPipeline::kQKHeaddim>{}),
sequence<kPadSeqLenQ, kPadHeadDimQ>{});
sequence<false, kPadHeadDimQ>{});
const auto k_dram_naive = make_naive_tensor_view<address_space_enum::global>(
k_ptr,
@@ -1232,7 +1228,7 @@ struct FmhaBwdDQDKDVKernel
const auto k_dram = pad_tensor_view(
k_dram_naive,
make_tuple(number<FmhaPipeline::kN0>{}, number<FmhaPipeline::kQKHeaddim>{}),
sequence<kPadSeqLenK, kPadHeadDimQ>{});
sequence<false, kPadHeadDimQ>{});
const auto v_dram = [&]() {
const auto v_dram_naive = make_naive_tensor_view<address_space_enum::global>(
@@ -1244,22 +1240,15 @@ struct FmhaBwdDQDKDVKernel
return pad_tensor_view(
v_dram_naive,
make_tuple(number<FmhaPipeline::kN0>{}, number<FmhaPipeline::kVHeaddim>{}),
sequence<kPadSeqLenK, kPadHeadDimV>{});
sequence<false, kPadHeadDimV>{});
}();
const auto lse_dram = [&]() {
const auto lse_dram_naive = make_naive_tensor_view_packed<address_space_enum::global>(
lse_ptr, make_tuple(kargs.seqlen_q), number<1>{});
return pad_tensor_view(
lse_dram_naive, make_tuple(number<FmhaPipeline::kM0>{}), sequence<kPadSeqLenQ>{});
}();
// lse and d should be fine to read unpaded data as they are not on the reduction dimension
const auto lse_dram = make_naive_tensor_view_packed<address_space_enum::global>(
lse_ptr, make_tuple(kargs.seqlen_q), number<FmhaPipeline::kM0>{});
const auto d_dram = [&]() {
const auto d_dram_naive = make_naive_tensor_view_packed<address_space_enum::global>(
d_ptr, make_tuple(kargs.seqlen_q), number<1>{});
return pad_tensor_view(
d_dram_naive, make_tuple(number<FmhaPipeline::kM0>{}), sequence<kPadSeqLenQ>{});
}();
const auto d_dram = make_naive_tensor_view_packed<address_space_enum::global>(
d_ptr, make_tuple(kargs.seqlen_q), number<FmhaPipeline::kM0>{});
const auto do_dram_naive = make_naive_tensor_view<address_space_enum::global>(
do_ptr,
@@ -1270,7 +1259,7 @@ struct FmhaBwdDQDKDVKernel
const auto do_dram = pad_tensor_view(
do_dram_naive,
make_tuple(number<FmhaPipeline::kM0>{}, number<FmhaPipeline::kVHeaddim>{}),
sequence<kPadSeqLenQ, kPadHeadDimV>{});
sequence<false, kPadHeadDimV>{});
auto q_dram_window = make_tile_window(
q_dram,
@@ -1313,7 +1302,7 @@ struct FmhaBwdDQDKDVKernel
return pad_tensor_view(
dq_acc_dram_naive,
make_tuple(number<FmhaPipeline::kM0>{}, number<FmhaPipeline::kQKHeaddim>{}),
sequence<kPadSeqLenQ, kPadHeadDimQ>{});
sequence<false, kPadHeadDimQ>{});
}();
return make_tile_window(
@@ -1341,7 +1330,7 @@ struct FmhaBwdDQDKDVKernel
return pad_tensor_view(
dq_acc_dram_naive,
make_tuple(number<FmhaPipeline::kM0>{}, number<FmhaPipeline::kQKHeaddim>{}),
sequence<kPadSeqLenQ, kPadHeadDimQ>{});
sequence<false, kPadHeadDimQ>{});
}();
return make_tile_window(
@@ -1376,9 +1365,8 @@ struct FmhaBwdDQDKDVKernel
number<FmhaPipeline::kAlignmentBias>{},
number<1>{});
return pad_tensor_view(bias_dram_naive,
bias_dram_window_lengths,
sequence<kPadSeqLenQ, kPadSeqLenK>{});
return pad_tensor_view(
bias_dram_naive, bias_dram_window_lengths, sequence<false, true>{});
}();
return make_tile_window(bias_dram, bias_dram_window_lengths, {0, i_n0});
@@ -1406,9 +1394,8 @@ struct FmhaBwdDQDKDVKernel
number<FmhaPipeline::kAlignmentBias>{},
number<1>{});
return pad_tensor_view(dbias_dram_naive,
bias_dram_window_lengths,
sequence<kPadSeqLenQ, kPadSeqLenK>{});
return pad_tensor_view(
dbias_dram_naive, bias_dram_window_lengths, sequence<false, true>{});
}();
return make_tile_window(dbias_dram, bias_dram_window_lengths, {0, i_n0});
@@ -1495,9 +1482,8 @@ struct FmhaBwdDQDKDVKernel
number<1>{},
number<1>{});
return pad_tensor_view(randval_dram_naive,
randval_dram_window_lengths,
sequence<kPadSeqLenQ, kPadSeqLenK>{});
return pad_tensor_view(
randval_dram_naive, randval_dram_window_lengths, sequence<false, true>{});
}();
return make_tile_window(randval_dram, randval_dram_window_lengths, {0, i_n0});
@@ -1550,7 +1536,7 @@ struct FmhaBwdDQDKDVKernel
return pad_tensor_view(
dk_dram_naive,
make_tuple(number<FmhaPipeline::kN0>{}, number<FmhaPipeline::kQKHeaddim>{}),
sequence<kPadSeqLenK, kPadHeadDimQ>{});
sequence<false, kPadHeadDimQ>{});
}();
auto dv_dram = [&]() {
@@ -1564,7 +1550,7 @@ struct FmhaBwdDQDKDVKernel
return pad_tensor_view(
dv_dram_naive,
make_tuple(number<FmhaPipeline::kN0>{}, number<FmhaPipeline::kVHeaddim>{}),
sequence<kPadSeqLenK, kPadHeadDimV>{});
sequence<false, kPadHeadDimV>{});
}();
auto dk_dram_window = make_tile_window(