Bobby Wang

Q: 不能调试其它进程么？

13

今晚上试着调试下inputsurfaceflinger 但是怎么都不成功！ 猜想： inputsurfaceflinger 运行在system_process, 而博文的是调试一个app进程的 ，垃圾回收机制运行于JVM中，所以能调试？？ 而如果想调试一些其它的比较 Media/SurfaceFlinger这些module 貌似不行了？？

[QST] Is there a way to re-construct cudf DataFrame by __cuda_array_interface__

4

Hi there, I'm asking if there is a way to re-construct cudf dataframe or Series according to __cuda_array_interface__? the __cuda_array_interface__ is like ``` json {'shape': [5], 'data': ['xxxxxx', False], 'typestr':...

[BUG] "import cudf" has changed the device ID

3

Looks like "import cudf" will change the device ID on 22.06 release ## rapids 22.06 ``` bash conda create -n rapids-22.06 -c rapidsai -c nvidia -c conda-forge \ rapids=22.06 python=3.9...

[jvm-packages] bridge the gaps between jvm package and native xgboost

10

JVM-packages is far behind the native XGBoost. I would like to file this issue to track some missing features or bugs that should be fixed in the incoming 2.0.0 release....

This is the follow up of https://github.com/dmlc/xgboost/pull/7836, which has introduced the barrier execution mode, and the killing SparkContext is not needed anymore, it's time to re-org the testing framework

Spark 3.3.0 has introduced a new API **mapInArrow** in PySpark DataFrame, see [SPARK-37228](https://issues.apache.org/jira/browse/SPARK-37228?page=com.atlassian.jira.plugin.system.issuetabpanels%3Aall-tabpanel) and PR https://github.com/apache/spark/pull/34505. mapInArrow is quite similar with mapInPandas, the only difference is the input is Iterable[pa.RecordBatch]...

[SPARK-40407][SQL] Fix the potential data skew caused by df.repartition

8

### What changes were proposed in this pull request? ``` scala val df = spark.range(0, 100, 1, 50).repartition(4) val v = df.rdd.mapPartitions { iter => { Iterator.single(iter.length) }.collect() println(v.mkString(",")) ```...

[FEA] [JSON reader] to support column prune

8

This is part of FEA of https://github.com/NVIDIA/spark-rapids/issues/9 We have a JSON file with below lines ``` json {"name":"Michael"} {"name":"Andy", "age":30} {"name":"Justin", "age":19} ``` when specifying the reading column name only...

[jvm-packages] fix spark-rapids compatibility issue

1

spark-rapids (from 22.10) has shimmed GpuColumnVector, which means we can't call it directly. So this PR call the UnshimmedGpuColumnVector