[XPU] support xpu pin memory copy

[DongBaiYue]

PR Category

Performance Optimization

PR Types

New features

Description

支持xpu上的pin_memory相关 API。
使用示例：

# one case
tensor_device = paddle.rand((624, 1024, 1024))
tensor_pin = test_tensor.pin_memory(blocking=True)

# another case
tensor_device = paddle.rand((624, 1024, 1024))
tensor_pin = paddle.empty_like(tensor_device, pin_memory=True)
tensor_pin.copy_(test_tensor, non_blocking=False)

pcard-67164

[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.

[DongBaiYue]

/re-run all-failed

[codecov-commenter]

Codecov Report

❌ Patch coverage is 90.00000% with 1 line in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@f07eb72). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
paddle/fluid/imperative/tracer.cc	50.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop   #76092   +/-   ##
==========================================
  Coverage           ?   90.00%           
==========================================
  Files              ?        2           
  Lines              ?       10           
  Branches           ?        0           
==========================================
  Hits               ?        9           
  Misses             ?        1           
  Partials           ?        0           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[DongBaiYue]

/re-run all-failed

[yongqiangma]

为啥把CustomDevice的逻辑也删除了

[DongBaiYue]

这部分是无效代码，核心逻辑

#ifdef PADDLE_WITH_CUDA || defined(PADDLE_WITH_HIP)
template <> PADDLE_API void Copy<phi::Place, phi::Place>{
    ……
#ifdef PADDLE_WITH_XPU
    ……
#endif
#ifdef PADDLE_WITH_CUSTOM_DEVICE
    ……
#endif
}
#endif

分支执行不到

[yongqiangma]

LGTM