OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
84,898 Commits 8,304 Branches 150 Tags 1.5 GiB
1cb4e2df65
Commit Graph

73 Commits

Author SHA1 Message Date
Xuehai Pan
1cb4e2df65 [BE][PYFMT] migrate PYFMT for torch._inductor to ruff format (#144550) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/144550
Approved by: https://github.com/jansel
 2025-02-28 13:33:19 +00:00
Tom Ritchford
44ee9ca593 [inductor] Add type annotations to _inductor/utils.py (#144108) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/144108
Approved by: https://github.com/eellison
 2025-02-15 23:13:41 +00:00
Jason Ansel
06604c4ec1 [inductor] Refactor op handlers part 5 (#146257) 
This makes OpHandler just a normal class using inheritance, and removes typing workarounds needed because it wasn't

Pull Request resolved: https://github.com/pytorch/pytorch/pull/146257
Approved by: https://github.com/shunting314
ghstack dependencies: #146252, #146254, #146255
 2025-02-08 18:00:30 +00:00
PyTorch MergeBot
49effa0deb Revert "[inductor] Refactor op handlers part 5 (#146257)" 
This reverts commit d3dd3eeb7f.

Reverted https://github.com/pytorch/pytorch/pull/146257 on behalf of https://github.com/atalman due to Sorry need to revert https://github.com/pytorch/pytorch/pull/146252 ([comment](https://github.com/pytorch/pytorch/pull/146257#issuecomment-2638251994))
 2025-02-05 23:20:38 +00:00
Jason Ansel
d3dd3eeb7f [inductor] Refactor op handlers part 5 (#146257) 
This makes OpHandler just a normal class using inheritance, and removes typing workarounds needed because it wasn't

Pull Request resolved: https://github.com/pytorch/pytorch/pull/146257
Approved by: https://github.com/shunting314
ghstack dependencies: #146225, #146226, #146235, #146252, #146254, #146255
 2025-02-04 23:36:25 +00:00
Jason Ansel
67be5953fe [inductor] Refactor op handlers part 1 (#146235) 
This enforces the invariant that every backend implements the same set of ops and removes a layer of indirection for BasicMathOps.

Interestingly this is a small compile time win:
```
...
WIN: benchmark ('add_loop_inductor', 'compile_time_instruction_count') failed, actual result 30151159301 is -6.13% lower than expected 32120000000 ±1.50% please update the expected results.

please update all results that changed significantly, and not only the failed ones
PASS: benchmark ('add_loop_inductor_dynamic_gpu', 'compile_time_instruction_count') pass, actual result 44447549162 -1.69% is within expected 45210000000 ±2.50%

WIN: benchmark ('add_loop_inductor_gpu', 'compile_time_instruction_count') failed, actual result 26743557195 is -2.25% lower than expected 27360000000 ±1.50% please update the expected results.

please update all results that changed significantly, and not only the failed ones
PASS: benchmark ('basic_modules_ListOfLinears_eager', 'compile_time_instruction_count') pass, actual result 945129734 +0.93% is within expected 936400000 ±1.50%

WIN: benchmark ('basic_modules_ListOfLinears_inductor', 'compile_time_instruction_count') failed, actual result 18984384503 is -3.19% lower than expected 19610000000 ±1.50% please update the expected results.

please update all results that changed significantly, and not only the failed ones
WIN: benchmark ('basic_modules_ListOfLinears_inductor_gpu_force_shape_pad', 'compile_time_instruction_count') failed, actual result 17258025389 is -1.94% lower than expected 17600000000 ±1.50% please update the expected results.
```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/146235
Approved by: https://github.com/shunting314
ghstack dependencies: #146225, #146226
 2025-02-04 23:35:53 +00:00
Jason Ansel
e9f6e273e7 [inductor] Add typing to common.CSE (#145993) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/145993
Approved by: https://github.com/yanboliang
ghstack dependencies: #145916
 2025-02-04 16:05:39 +00:00
PyTorch MergeBot
d3c7e4bb9c Revert "[inductor] Add typing to common.CSE (#145993)" 
This reverts commit 8c657ae4be.

Reverted https://github.com/pytorch/pytorch/pull/145993 on behalf of https://github.com/atalman due to Sorry need to revert https://github.com/pytorch/pytorch/pull/145916 ([comment](https://github.com/pytorch/pytorch/pull/145993#issuecomment-2632712384))
 2025-02-04 03:04:01 +00:00
PyTorch MergeBot
2f40f789da Revert "[inductor] Refactor op handlers part 1 (#146235)" 
This reverts commit 204be4e0a2.

Reverted https://github.com/pytorch/pytorch/pull/146235 on behalf of https://github.com/atalman due to Breaks lint, sorry: Definition of polygamma in base class MetalOverrides is incompatible with definition in base class OpsHandler. Please rebase fix lint and reland ([comment](https://github.com/pytorch/pytorch/pull/146235#issuecomment-2632444514))
 2025-02-04 00:00:08 +00:00
Jason Ansel
204be4e0a2 [inductor] Refactor op handlers part 1 (#146235) 
This enforces the invariant that every backend implements the same set of ops and removes a layer of indirection for BasicMathOps.

Interestingly this is a small compile time win:
```
...
WIN: benchmark ('add_loop_inductor', 'compile_time_instruction_count') failed, actual result 30151159301 is -6.13% lower than expected 32120000000 ±1.50% please update the expected results.

please update all results that changed significantly, and not only the failed ones
PASS: benchmark ('add_loop_inductor_dynamic_gpu', 'compile_time_instruction_count') pass, actual result 44447549162 -1.69% is within expected 45210000000 ±2.50%

WIN: benchmark ('add_loop_inductor_gpu', 'compile_time_instruction_count') failed, actual result 26743557195 is -2.25% lower than expected 27360000000 ±1.50% please update the expected results.

please update all results that changed significantly, and not only the failed ones
PASS: benchmark ('basic_modules_ListOfLinears_eager', 'compile_time_instruction_count') pass, actual result 945129734 +0.93% is within expected 936400000 ±1.50%

WIN: benchmark ('basic_modules_ListOfLinears_inductor', 'compile_time_instruction_count') failed, actual result 18984384503 is -3.19% lower than expected 19610000000 ±1.50% please update the expected results.

please update all results that changed significantly, and not only the failed ones
WIN: benchmark ('basic_modules_ListOfLinears_inductor_gpu_force_shape_pad', 'compile_time_instruction_count') failed, actual result 17258025389 is -1.94% lower than expected 17600000000 ±1.50% please update the expected results.
```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/146235
Approved by: https://github.com/shunting314
ghstack dependencies: #146225, #146226
 2025-02-03 23:15:13 +00:00
Jason Ansel
8c657ae4be [inductor] Add typing to common.CSE (#145993) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/145993
Approved by: https://github.com/yanboliang
ghstack dependencies: #145913, #145914, #145915, #145916
 2025-02-01 16:34:18 +00:00
Jason Ansel
2df2f9d895 [inductor] Change type of get_backend_features to OrderedSet (#145692) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/145692
Approved by: https://github.com/yanboliang
 2025-01-28 01:44:32 +00:00
Aaron Orenstein
2bf772d1ba PEP585 update - torch/_inductor/codegen (#145106) 
See #145101 for details.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/145106
Approved by: https://github.com/bobrenjc93
 2025-01-18 06:56:03 +00:00
bobrenjc93
a3ab27b8e0 Migrate from Tuple -> tuple in torch/_inductor (#144264) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/144264
Approved by: https://github.com/eellison
 2025-01-07 03:27:27 +00:00
Tom Ritchford
da67a6a7bb [inductor] Replace set by OrderedSet (#138466) 
Uses the set_linter from https://github.com/pytorch/pytorch/pull/138454
and considerable manual editing

Pull Request resolved: https://github.com/pytorch/pytorch/pull/138466
Approved by: https://github.com/eellison
 2024-12-13 16:08:45 +00:00
Tom Ritchford
dc23f1944a Remove unused Python variables in torch/[_-a]* (#133492) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/133492
Approved by: https://github.com/albanD
 2024-12-12 17:39:14 +00:00
PyTorch MergeBot
5c97ac9721 Revert "Remove unused Python variables in torch/[_-a]* (#133492)" 
This reverts commit fda975a7b3.

Reverted https://github.com/pytorch/pytorch/pull/133492 on behalf of https://github.com/clee2000 due to Sorry, I need to revert this in order to revert something else.  The only thing you need to do is rebase and remerge ([comment](https://github.com/pytorch/pytorch/pull/133492#issuecomment-2536635516))
 2024-12-11 17:29:12 +00:00
Tom Ritchford
fda975a7b3 Remove unused Python variables in torch/[_-a]* (#133492) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/133492
Approved by: https://github.com/albanD
 2024-12-10 21:48:44 +00:00
Blaine Burton Rister
f9af86de01 [Inductor] Represent tiling as a dict (#141751) 
# Summary

Preparatory refactor for https://github.com/pytorch/pytorch/pull/137243. This makes it easier to generalize to multi-dimensional reductions.

This diff refactors `self.numels` from a tuple like `(8,16)` to a dict like `{"x": 8, "r": 16}`.

Note: this is based off of https://github.com/pytorch/pytorch/pull/141738, which enables `tree.is_reduction`. That PR should land first.

# Test plan
The existing CI provides good coverage.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/141751
Approved by: https://github.com/jansel
 2024-12-05 02:28:16 +00:00
PyTorch MergeBot
38d10a1b17 Revert "[Inductor] Represent tiling as a dict (#141751)" 
This reverts commit 5deca07c0d.

Reverted https://github.com/pytorch/pytorch/pull/141751 on behalf of https://github.com/atalman due to Failing internal builds ([comment](https://github.com/pytorch/pytorch/pull/141751#issuecomment-2517815899))
 2024-12-04 15:43:16 +00:00
Blaine Burton Rister
5deca07c0d [Inductor] Represent tiling as a dict (#141751) 
# Summary

Preparatory refactor for https://github.com/pytorch/pytorch/pull/137243. This makes it easier to generalize to multi-dimensional reductions.

This diff refactors `self.numels` from a tuple like `(8,16)` to a dict like `{"x": 8, "r": 16}`.

Note: this is based off of https://github.com/pytorch/pytorch/pull/141738, which enables `tree.is_reduction`. That PR should land first.

# Test plan
The existing CI provides good coverage.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/141751
Approved by: https://github.com/jansel
 2024-12-01 09:54:34 +00:00
Blaine Burton Rister
49fde426ba [Inductor] Use a helper function to tell if a tree or prefix is a reduction (#141738) 
Preparatory refactor for https://github.com/pytorch/pytorch/pull/137243. Previously, we would typically check for reductions by `tree.prefix == "r"`. This PR moves the check into a helper function. This makes it easier to generalize the code to multi-dimensional reductions, which could have multiple prefixes like `("r0_", "r1_")`.

Tested by the existing CI.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/141738
Approved by: https://github.com/jansel
 2024-11-30 22:38:13 +00:00
Isuru Fernando
44186a0a4e Move Sympy printers to torch/utils/_sympy/printers.py (#140597) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/140597
Approved by: https://github.com/ezyang, https://github.com/anijain2305
 2024-11-26 18:11:00 +00:00
PyTorch MergeBot
f23621ec56 Revert "Move Sympy printers to torch/utils/_sympy/printers.py (#140597)" 
This reverts commit c25b201583.

Reverted https://github.com/pytorch/pytorch/pull/140597 on behalf of https://github.com/huydhn due to Trunk is sad again after this lands, this looks like a landrace this time, so please do a rebase ([comment](https://github.com/pytorch/pytorch/pull/140597#issuecomment-2494052978))
 2024-11-22 15:43:39 +00:00
Isuru Fernando
c25b201583 Move Sympy printers to torch/utils/_sympy/printers.py (#140597) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/140597
Approved by: https://github.com/ezyang, https://github.com/anijain2305
 2024-11-22 02:04:36 +00:00
PyTorch MergeBot
701e06b643 Revert "Move Sympy printers to torch/utils/_sympy/printers.py (#140597)" 
This reverts commit aefcdb3c9f.

Reverted https://github.com/pytorch/pytorch/pull/140597 on behalf of https://github.com/huydhn due to Sorry for reverting your change but I think it fails inductor/test_padding in trunk. This is a target determination miss and that failed test was not run in your PR ([comment](https://github.com/pytorch/pytorch/pull/140597#issuecomment-2489641453))
 2024-11-20 22:13:57 +00:00
Isuru Fernando
aefcdb3c9f Move Sympy printers to torch/utils/_sympy/printers.py (#140597) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/140597
Approved by: https://github.com/ezyang, https://github.com/anijain2305
 2024-11-20 20:26:49 +00:00
Aaron Gokaslan
12e95aa4ee [BE]: Apply PERF401 autofixes from ruff (#140980) 
* Automatically applies ruff rule 401. Turns loops into equivalent list comprehensions which are faster and do not leak the scope of the loop variables.
* list comprehensions not only often have better typing, but are 50+% faster than for loops on overhead. They also preserve length information etc and are better for the interpreter to optimize.
* Manually went back and made mypy happy after the change.
* Also fixed style lints in files covered by flake8 but not by pyfmt

Pull Request resolved: https://github.com/pytorch/pytorch/pull/140980
Approved by: https://github.com/justinchuby, https://github.com/malfet
 2024-11-20 17:52:07 +00:00
Jason Ansel
ed30fa74ab [inductor] sympy.Integer([01]) -> sympy.S.(Zero|One) (#139523) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/139523
Approved by: https://github.com/ezyang
ghstack dependencies: #139364, #139365, #139370, #139452
 2024-11-04 04:28:40 +00:00
PyTorch MergeBot
98e11b0021 Revert "[inductor] sympy.Integer([01]) -> sympy.S.(Zero|One) (#139523)" 
This reverts commit c53beab377.

Reverted https://github.com/pytorch/pytorch/pull/139523 on behalf of https://github.com/huydhn due to Sorry for reverting your change but it is failing lots of internal tests in D65345157 ([comment](https://github.com/pytorch/pytorch/pull/139364#issuecomment-2452897337))
 2024-11-02 06:49:10 +00:00
Jason Ansel
c53beab377 [inductor] sympy.Integer([01]) -> sympy.S.(Zero|One) (#139523) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/139523
Approved by: https://github.com/ezyang
ghstack dependencies: #139364, #139365, #139370, #139452
 2024-11-02 03:04:22 +00:00
Jason Ansel
f9ef880c0b [inductor] Refactor kernel args into SIMDKernelFeatures (#139327) 
This is a refactor PR to move stuff around.  I'm planning to use the SIMDKernelFeatures class (in a future PR) to host new heuristics for selecting kernel types and block sizes.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139327
Approved by: https://github.com/eellison, https://github.com/shunting314
 2024-11-01 00:30:14 +00:00
Jason Ansel
2b937e4e6d [inductor] Cooperative reductions (#137756) 
Example generated code for `(x+y).sum()`:
```py
@triton.jit
def triton_unk_fused_add_sum_0(in_ptr0, in_ptr1, out_ptr0, ws_ptr, semaphores_ptr, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr, RSPLIT : tl.constexpr):
    xnumel = 1
    rnumel = 1048576
    rsplit_id = tl.program_id(0)
    num_rblocks = (rnumel + RBLOCK - 1) // RBLOCK
    rsplit_chunk = (num_rblocks + RSPLIT - 1) // RSPLIT * RBLOCK
    rsplit_start = rsplit_chunk * rsplit_id
    rsplit_end = rsplit_chunk * (rsplit_id + 1)
    xoffset = tl.program_id(1) * XBLOCK
    xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
    xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
    rbase = tl.arange(0, RBLOCK)[None, :]
    _tmp4 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
    for roffset in range(rsplit_start, rsplit_end, RBLOCK):
        rindex = roffset + rbase
        rmask = rindex < rnumel
        r0 = rindex
        tmp0 = tl.load(in_ptr0 + (r0), rmask, eviction_policy='evict_first', other=0.0)
        tmp1 = tl.load(in_ptr1 + (r0), rmask, eviction_policy='evict_first', other=0.0)
        tmp2 = tmp0 + tmp1
        tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
        tmp5 = _tmp4 + tmp3
        _tmp4 = tl.where(rmask, tmp5, _tmp4)
    tmp4 = tl.sum(_tmp4, 1)[:, None]
    if RSPLIT > 1:
        tmp4_ws = (ws_ptr + 0).to(tl.pointer_type(tl.float32))
        tl.store(tmp4_ws + (xindex * RSPLIT + rsplit_id), tmp4, None)
    if RSPLIT > 1:
        triton_helpers.gpu_barrier(semaphores_ptr + (2 * tl.program_id(1) + 0), RSPLIT, True)
    if RSPLIT > 1:
        tmp4_peers = tl.load(tmp4_ws + (xindex * RSPLIT + tl.arange(0, RSPLIT)[None,:]), None, eviction_policy='evict_first')
        tmp4 = tl.sum(tmp4_peers, 1)[:, None]
    if rsplit_id == (0 % RSPLIT):
        tl.store(out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp4, None)
```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/137756
Approved by: https://github.com/eellison
 2024-10-29 00:45:53 +00:00
PyTorch MergeBot
60d1c7138d Revert "[inductor] Cooperative reductions (#137756)" 
This reverts commit fed37dbfbc.

Reverted https://github.com/pytorch/pytorch/pull/137756 on behalf of https://github.com/jeanschmidt due to ROCM tests are timing out :( ([comment](https://github.com/pytorch/pytorch/pull/137756#issuecomment-2441579322))
 2024-10-28 13:24:33 +00:00
Jason Ansel
fed37dbfbc [inductor] Cooperative reductions (#137756) 
Example generated code for `(x+y).sum()`:
```py
@triton.jit
def triton_unk_fused_add_sum_0(in_ptr0, in_ptr1, out_ptr0, ws_ptr, semaphores_ptr, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr, RSPLIT : tl.constexpr):
    xnumel = 1
    rnumel = 1048576
    rsplit_id = tl.program_id(0)
    num_rblocks = (rnumel + RBLOCK - 1) // RBLOCK
    rsplit_chunk = (num_rblocks + RSPLIT - 1) // RSPLIT * RBLOCK
    rsplit_start = rsplit_chunk * rsplit_id
    rsplit_end = rsplit_chunk * (rsplit_id + 1)
    xoffset = tl.program_id(1) * XBLOCK
    xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
    xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
    rbase = tl.arange(0, RBLOCK)[None, :]
    _tmp4 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
    for roffset in range(rsplit_start, rsplit_end, RBLOCK):
        rindex = roffset + rbase
        rmask = rindex < rnumel
        r0 = rindex
        tmp0 = tl.load(in_ptr0 + (r0), rmask, eviction_policy='evict_first', other=0.0)
        tmp1 = tl.load(in_ptr1 + (r0), rmask, eviction_policy='evict_first', other=0.0)
        tmp2 = tmp0 + tmp1
        tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
        tmp5 = _tmp4 + tmp3
        _tmp4 = tl.where(rmask, tmp5, _tmp4)
    tmp4 = tl.sum(_tmp4, 1)[:, None]
    if RSPLIT > 1:
        tmp4_ws = (ws_ptr + 0).to(tl.pointer_type(tl.float32))
        tl.store(tmp4_ws + (xindex * RSPLIT + rsplit_id), tmp4, None)
    if RSPLIT > 1:
        triton_helpers.gpu_barrier(semaphores_ptr + (2 * tl.program_id(1) + 0), RSPLIT, True)
    if RSPLIT > 1:
        tmp4_peers = tl.load(tmp4_ws + (xindex * RSPLIT + tl.arange(0, RSPLIT)[None,:]), None, eviction_policy='evict_first')
        tmp4 = tl.sum(tmp4_peers, 1)[:, None]
    if rsplit_id == (0 % RSPLIT):
        tl.store(out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp4, None)
```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/137756
Approved by: https://github.com/eellison
ghstack dependencies: #138970
 2024-10-27 16:31:38 +00:00
Xinran / Allan Rui
ba6526814a Add dtype attribute to CSEVariable (#136778) 
Summary:
- This diff introduces `dtype` attribute to `TritonCSEVariable` and a dtype propagation helper function to infer dtype from input to output for each op.

- There will be a follow-up diff that uses this `dtype` information in `TritonCSEVariable` to perform dtype-aware codegen.

Test Plan: CI

Differential Revision: D61815079

Pull Request resolved: https://github.com/pytorch/pytorch/pull/136778
Approved by: https://github.com/eellison, https://github.com/blaine-rister
 2024-10-25 18:00:30 +00:00
Jason Ansel
4632594546 [inductor] Move V.graph.scheduler.current_device to V.graph.current_device (#138252) 
There are some places where it would be nice to use this, but the scheduler hasn't yet been created.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/138252
Approved by: https://github.com/eellison
ghstack dependencies: #138170
 2024-10-18 23:05:54 +00:00
Jez Ng
71aac59e93 Add Triton CPU as an Inductor backend (#133408) 
The goal is to use Inductor-generated kernels to stress test the new Triton CPU backend.

Differential Revision: [D63298968](https://our.internmc.facebook.com/intern/diff/D63298968)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/133408
Approved by: https://github.com/jansel, https://github.com/blaine-rister, https://github.com/malfet
 2024-09-30 20:24:52 +00:00
PyTorch MergeBot
36428f91e9 Revert "Add Triton CPU as an Inductor backend (#133408)" 
This reverts commit 31c0467594.

Reverted https://github.com/pytorch/pytorch/pull/133408 on behalf of https://github.com/int3 due to internal tests failing ([comment](https://github.com/pytorch/pytorch/pull/133408#issuecomment-2379692517))
 2024-09-27 16:54:27 +00:00
Jez Ng
31c0467594 Add Triton CPU as an Inductor backend (#133408) 
The goal is to use Inductor-generated kernels to stress test the new Triton CPU backend.

Differential Revision: [D63298968](https://our.internmc.facebook.com/intern/diff/D63298968)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/133408
Approved by: https://github.com/jansel, https://github.com/blaine-rister, https://github.com/malfet
 2024-09-26 15:35:26 +00:00
PyTorch MergeBot
d0cebedb31 Revert "Add Triton CPU as an Inductor backend (#133408)" 
This reverts commit e498b02b47.

Reverted https://github.com/pytorch/pytorch/pull/133408 on behalf of https://github.com/jeanschmidt due to Broke internal signals, see D62737208 for more details ([comment](https://github.com/pytorch/pytorch/pull/133408#issuecomment-2353623816))
 2024-09-16 18:33:33 +00:00
Jez Ng
e498b02b47 Add Triton CPU as an Inductor backend (#133408) 
The goal is to use Inductor-generated kernels to stress test the new Triton CPU backend.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/133408
Approved by: https://github.com/jansel
 2024-09-14 21:45:19 +00:00
xinan.lin
13ee85ca5e [Inductor] Generalize cuda cpp wrapper as common triton based GPU cpp wrapper, will be reused by xpu in next PR. (#135312) 
[Inductor] Generalize cuda cpp wrapper as common triton based GPU cpp wrapper, will be reused by xpu in next PR.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/135312
Approved by: https://github.com/jansel, https://github.com/desertfire, https://github.com/eellison
 2024-09-11 23:59:54 +00:00
Isuru Fernando
1b10a5c652 Allow SymInts and SymFloats as other in div_softmax_pattern (#133989) 
Fixes https://github.com/pytorch/pytorch/issues/133759

Pull Request resolved: https://github.com/pytorch/pytorch/pull/133989
Approved by: https://github.com/ezyang
 2024-08-22 14:36:01 +00:00
Isuru Fernando
7470ae85e4 Fix triton codegen with math.trunc (#133354) 
Fixes https://github.com/pytorch/pytorch/issues/133172

Pull Request resolved: https://github.com/pytorch/pytorch/pull/133354
Approved by: https://github.com/ezyang, https://github.com/jansel
 2024-08-15 16:38:26 +00:00
Oguz Ulgen
09f9c256ad Add basic mypy annotations to inductor (#132416) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/132416
Approved by: https://github.com/XuehaiPan, https://github.com/jamesjwu
ghstack dependencies: #132415
 2024-08-04 18:43:37 +00:00
PyTorch MergeBot
f2ddd5e9e0 Revert "Add basic mypy annotations to inductor (#132416)" 
This reverts commit 78927d37f6.

Reverted https://github.com/pytorch/pytorch/pull/132416 on behalf of https://github.com/ZainRizvi due to Sorry, this PR has entered a weird state in the diff train. Trying to revert it to skip it, and then we can try relanding it ([comment](https://github.com/pytorch/pytorch/pull/132415#issuecomment-2267631785))
 2024-08-04 18:39:29 +00:00
Oguz Ulgen
78927d37f6 Add basic mypy annotations to inductor (#132416) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/132416
Approved by: https://github.com/XuehaiPan, https://github.com/jamesjwu
ghstack dependencies: #132415
 2024-08-01 20:14:25 +00:00
eellison
f32ab3b9e3 Migrate Inductor scheduler, dependencies, ir, and codegen/common to use OrderedSet (#130004) 
Python's set is non deterministic. There is an internal failure which we recently ran into which did not consistently fail.

See, repro here: P1453035092.

Now, with these changes, it does consistently fail. In follow ups we could also consider adding a lintrule for uses of either set() or set literals.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/130004
Approved by: https://github.com/oulgen
 2024-08-01 04:37:15 +00:00
PyTorch MergeBot
784a6ec5a3 Revert "Migrate Inductor scheduler, dependencies, ir, and codegen/common to use OrderedSet (#130004)" 
This reverts commit 13d744464f.

Reverted https://github.com/pytorch/pytorch/pull/130004 on behalf of https://github.com/clee2000 due to broke lint [GH job link](https://github.com/pytorch/pytorch/actions/runs/10183945999/job/28170099930) [HUD commit link](13d744464f) probably a landrace, the base is 21 hours old ([comment](https://github.com/pytorch/pytorch/pull/130004#issuecomment-2260946562))
 2024-07-31 16:49:21 +00:00