Decode fix and drop unused triton kernel#18
Merged
Conversation
Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>
xintin
pushed a commit
that referenced
this pull request
Jun 10, 2025
Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>
xintin
pushed a commit
that referenced
this pull request
Jun 12, 2025
Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>
Hardcode84
added a commit
to Hardcode84/sglang
that referenced
this pull request
Jun 18, 2025
Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>
raikonenfnu
pushed a commit
that referenced
this pull request
Jun 27, 2025
Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>
nithinsubbiah
pushed a commit
that referenced
this pull request
Jul 8, 2025
Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>
xintin
pushed a commit
that referenced
this pull request
Jul 14, 2025
Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>
willghatch
pushed a commit
that referenced
this pull request
Jul 28, 2025
Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com>
xintin
pushed a commit
that referenced
this pull request
Aug 15, 2025
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
Hardcode84
pushed a commit
to Hardcode84/sglang
that referenced
this pull request
Aug 27, 2025
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (harsh-nod#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (harsh-nod#6)" (harsh-nod#7) This reverts commit eac4599. Wave Backend decode (harsh-nod#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (harsh-nod#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (harsh-nod#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (harsh-nod#14) Set unique cache dir for each worker (harsh-nod#16) update kernel (harsh-nod#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (harsh-nod#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (harsh-nod#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (harsh-nod#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (harsh-nod#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (harsh-nod#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
raikonenfnu
added a commit
that referenced
this pull request
Sep 8, 2025
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
willghatch
pushed a commit
that referenced
this pull request
Oct 17, 2025
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
xintin
pushed a commit
that referenced
this pull request
Nov 3, 2025
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
Hardcode84
pushed a commit
to Hardcode84/sglang
that referenced
this pull request
Nov 10, 2025
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (harsh-nod#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (harsh-nod#6)" (harsh-nod#7) This reverts commit eac4599. Wave Backend decode (harsh-nod#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (harsh-nod#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (harsh-nod#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (harsh-nod#14) Set unique cache dir for each worker (harsh-nod#16) update kernel (harsh-nod#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (harsh-nod#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (harsh-nod#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (harsh-nod#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (harsh-nod#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (harsh-nod#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
raikonenfnu
added a commit
that referenced
this pull request
Nov 17, 2025
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
panditsa
pushed a commit
that referenced
this pull request
Nov 20, 2025
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
willghatch
pushed a commit
that referenced
this pull request
Dec 15, 2025
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
willghatch
pushed a commit
that referenced
this pull request
Dec 19, 2025
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
xintin
pushed a commit
that referenced
this pull request
Jan 5, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
Hardcode84
pushed a commit
to Hardcode84/sglang
that referenced
this pull request
Jan 12, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (harsh-nod#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (harsh-nod#6)" (harsh-nod#7) This reverts commit eac4599. Wave Backend decode (harsh-nod#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (harsh-nod#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (harsh-nod#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (harsh-nod#14) Set unique cache dir for each worker (harsh-nod#16) update kernel (harsh-nod#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (harsh-nod#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (harsh-nod#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (harsh-nod#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (harsh-nod#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (harsh-nod#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
Hardcode84
pushed a commit
to Hardcode84/sglang
that referenced
this pull request
Jan 12, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (harsh-nod#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (harsh-nod#6)" (harsh-nod#7) This reverts commit eac4599. Wave Backend decode (harsh-nod#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (harsh-nod#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (harsh-nod#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (harsh-nod#14) Set unique cache dir for each worker (harsh-nod#16) update kernel (harsh-nod#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (harsh-nod#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (harsh-nod#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (harsh-nod#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (harsh-nod#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (harsh-nod#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
Hardcode84
pushed a commit
that referenced
this pull request
Jan 12, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
Hardcode84
pushed a commit
that referenced
this pull request
Jan 12, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
panditsa
pushed a commit
that referenced
this pull request
Jan 16, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
panditsa
pushed a commit
that referenced
this pull request
Jan 16, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
raikonenfnu
added a commit
that referenced
this pull request
Jan 26, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
raikonenfnu
added a commit
that referenced
this pull request
Jan 26, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
willghatch
pushed a commit
that referenced
this pull request
Feb 12, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
willghatch
pushed a commit
that referenced
this pull request
Feb 12, 2026
Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Add wave extend attention kernel Signed-off-by: Harsh Menon <harsh@nod-labs.com> [Wave] Adding logit_cap and layer scaling to API Also add support for the wave backend to the model runner. And use Triton decode kernels for now. [Wave] Run chunked prefill for perf comparison on Wave test Need to rename the non chunked/regular prefill version because otherwise rpd will treat it as the same kernel Signed-off-by: Stanley Winata <stanley.winata@amd.com> [Wave] Cache the function that loads the wave kernel Also maintain a global kernel hash to avoid recomputing the hash on every call. [Wave] Don't specify block size and enable buffer ops [Wave] Enable wave runtime and update scheduling API [Wave] Update API to use wave_compile & WaveCompileOptions [Wave] Update wave backend and extend attention to latest [Wave] Add speculative decode kernel Signed-off-by: nithinsubbiah <nithinsubbiah@gmail.com> cache kernels using lru_cache Update WaveBackend to use Wave Decode (#6) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Revert "Update WaveBackend to use Wave Decode (#6)" (#7) This reverts commit eac4599. Wave Backend decode (#8) * align shapes Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> * fix Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> --------- Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Wave backend fixes (#10) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> More fixes to Wave decode (#12) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> is_causal Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Enable the grok in3 model (#14) Set unique cache dir for each worker (#16) update kernel (#18) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> updated spec decode test as per wave Signed-off-by: xintin <gaurav.verma@amd.com> fix extend (#23) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Refactor paged decode intermediate arrays shapes (#24) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> remove dyn symbols (#26) Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> cleanup shapes (#27) Some fields were removed from `paged_decode_attention_shape`. Signed-off-by: Ivan Butygin <ivan.butygin@gmail.com> Remove `mha` param from Wave decode attention kernel (#28) Depends on iree-org/iree-turbine#1039 Signed-off-by: Paul Zhang <paul.zhang@amd.com> nfc: fix problems reported by linting update references from iree.turbine to wave_lang
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.