You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by "cbalint13 (via GitHub)" <gi...@apache.org> on 2025/08/22 19:14:59 UTC
[I] [Bug][RELAX][PIPELINE] Metaschedule failure due to pipeline passes [tvm]
cbalint13 opened a new issue, #18224:
URL: https://github.com/apache/tvm/issues/18224
Failures are encontered about tuning a small example with tensorization.
This also affects our work in [PR#18182](https://github.com/apache/tvm/pull/18182) where this issue was discovered.
----
#### Reproducer
- Here is a standalone minimal program reproducing the issue: [x86-dense-relax-metaschedule.py](https://github.com/user-attachments/files/21943488/x86-dense-relax-metaschedule.py)
- Swapping out L146 (enabling) and L151(disabling) make the program running, schedules are tensorized as expected.
On **success**, the program will find and print the tensorized IR from the tuning database.
On **failure** (with many obvious errors inside tuner) the program finds zero tensorized schedules.
#### Descriptions
* It seems to be a subtle difference in IR originating from ```bb.emit_te(tvm.topi.nn.dense)``` vs ```bb.emit(relax.op.matmul```.
#### Investigation
* Investigatiated and found a very subtle difference, betweeen the good/bad relax IR:
```
$ diff -Nru relax-ir-te.txt relax-ir-pp.txt
--- relax-ir-te.txt 2025-08-22 21:42:34.567064474 +0300
+++ relax-ir-pp.txt 2025-08-22 21:46:34.309096420 +0300
@@ -5,22 +5,22 @@
@I.ir_module
class Module:
@T.prim_func(private=True)
- def dense1(data: T.Buffer((T.int64(4), T.int64(4)), "uint8"), weight: T.Buffer((T.int64(4), T.int64(4)), "int8"), T_matmul_NT: T.Buffer((T.int64(4), T.int64(4)), "int32")):
+ def matmul1(data: T.Buffer((T.int64(4), T.int64(4)), "uint8"), weight: T.Buffer((T.int64(4), T.int64(4)), "int8"), matmul: T.Buffer((T.int64(4), T.int64(4)), "int32")):
T.func_attr({"layout_free_buffers": [], "op_pattern": 4, "tir.noalias": True})
# with T.block("root"):
for i0, i1, k in T.grid(T.int64(4), T.int64(4), T.int64(4)):
- with T.block("T_matmul_NT"):
+ with T.block("matmul"):
v_i0, v_i1, v_k = T.axis.remap("SSR", [i0, i1, k])
- T.reads(data[v_i0, v_k], weight[v_i1, v_k])
- T.writes(T_matmul_NT[v_i0, v_i1])
+ T.reads(data[v_i0, v_k], weight[v_k, v_i1])
+ T.writes(matmul[v_i0, v_i1])
with T.init():
- T_matmul_NT[v_i0, v_i1] = 0
- T_matmul_NT[v_i0, v_i1] = T_matmul_NT[v_i0, v_i1] + T.Cast("int32", data[v_i0, v_k]) * T.Cast("int32", weight[v_i1, v_k])
+ matmul[v_i0, v_i1] = 0
+ matmul[v_i0, v_i1] = matmul[v_i0, v_i1] + T.Cast("int32", data[v_i0, v_k]) * T.Cast("int32", weight[v_k, v_i1])
@R.function
def main(data: R.Tensor((4, 4), dtype="uint8"), weight: R.Tensor((4, 4), dtype="int8")) -> R.Tensor((4, 4), dtype="int32"):
cls = Module
with R.dataflow():
- gv = R.call_tir(cls.dense1, (data, weight), out_sinfo=R.Tensor((4, 4), dtype="int32"))
+ gv = R.call_tir(cls.matmul1, (data, weight), out_sinfo=R.Tensor((4, 4), dtype="int32"))
R.output(gv)
return gv
```
I am seeing only naming differences, no structural** or other differences of IR
Looking at the differences [(-)good, (+)bad] I cannot tell what is the issue.
I incline that there is a missing pass, but which one ? Or is there a bug in TIR processing ?
Cc @MasterJH5574 @Hzfengsy @mshr-h @tqchen
Cc @fzi-peccia
Thank you !
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] [Bug][RELAX][PIPELINE] Metaschedule failure due to pipeline passes [tvm]
Posted by "cbalint13 (via GitHub)" <gi...@apache.org>.
cbalint13 commented on issue #18224:
URL: https://github.com/apache/tvm/issues/18224#issuecomment-3215871334
It seems that the difference is the access pattern.
For ```relax.op.matmul``` (being transposed by default) adapting the template for tensorizer instrinsic works.
```
--- dot_product_4x4_u8i8i32_desc_old.py
+++ dot_product_4x4_u8i8i32_desc_new.py
@@ -8,4 +8,4 @@
for i in T.serial(0, 4):
for k in T.serial(0, 4):
with T.block("update"):
- vi, vk = T.axis.remap("SR", [i, k])
- C[vi] = C[vi] + T.cast(A[vk], "int32") * T.cast(B[vi, vk], "int32")
+ vi, vk = T.axis.remap("SR", [i, k])
+ C[vi] = C[vi] + T.cast(A[vk], "int32") * T.cast(B[vk, vi], "int32")
```
Beyond R.matmul, one must be also careful migrating relay->relax about layout transforms, NHWC -> HCHW (relax default).
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [I] [Bug][RELAX][PIPELINE] Metaschedule failure due to pipeline passes [tvm]
Posted by "cbalint13 (via GitHub)" <gi...@apache.org>.
cbalint13 closed issue #18224: [Bug][RELAX][PIPELINE] Metaschedule failure due to pipeline passes
URL: https://github.com/apache/tvm/issues/18224
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org