Replace matmul(v2) with fused_matmul during oneDNN fuse passes

[Silv3S]

PR types

Others

PR changes

OPs

Describe

• introduce fused_matmul to handle fusion logic from oneDNN matmul and matmul_v2 fuse passes ,
• remove fusion logic from base matmul_v2 op and kernel,
• adjust existing fuse passes and op compat checks.

[paddle-bot]

你的PR提交成功，感谢你对开源项目的贡献!
请关注后续CI自动化测试结果，详情请参考Paddle-CI手册。
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

[sfraczek]

LGTM

[tsocha]

TODOs

[qingqing01]

LGTM

[qingqing01]

2023-01-18 23:38:58 ****************
2023-01-18 23:38:58 0. You must have Dianhai or XiaoguangHu01 approval for change 20+ files or add than 1000+ lines of content.
2023-01-18 23:38:58 There are 1 approved errors.
2023-01-18 23:38:58 ****************

@raindrops2sea @XiaoguangHu01 please help to review~

[wawltor]

There have some broadcast computational logic difference between matmul_v1 and matmul_v2， the code as follows

import paddle 
data1 = paddle.rand((3, 1, 10, 10))
data2 = paddle.rand((1, 2, 10, 10))


# The paddle.matmul (matmul_v2)
paddle.matmul(data1, data2).shape # [3, 2, 10, 10]


# The fluid.layers.matmul (matmul_v1)
paddle.fluid.layers.matmul(data1, data2).shape # 直接报错
# ValueError: (InvalidArgument) The batch size of the two matrices should be equal, or at least one is zero.
# But received X's shape: [3, 10, 10], Y's shape: [2, 10, 10]

Is there some problems when replacing directly ?

[Silv3S]

Hi, I don't think that it could be problem. For some time oneDNN version of both matmul and matmul_v2 are already using same code #44640, #48162. Declared attributes of these two are mostly the same. Only difference is alpha not supported in matmul_v2, and transpose is renamed to trans).

Goal of this PR is just to extract fusion logic from base op and base kernel. Fused op is superset, which has all extra attributes declared in OPMaker and fused kernel has implemented support for handling fusion logic.

I've adjusted all fuse pass unit test to work with fused_matmul. Also this PR has been checked in our internal validation and it didn't report any accuracy drop.

[wawltor]

LGTM

[XiaoguangHu01]

LGTM

[Aurelius84]

LGTM for data type Registeration