# BLAS library path

**URL:** <https://discuss.hail.is/t/blas-library-path/1720>\
**Category:** Hail Query & hailctl\
**Created:** [October 1, 2020, 5:43pm UTC](https://discuss.hail.is/t/blas-library-path/1720 "2020-10-01T17:43:45Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![EZXYZ](https://avatars.discourse-cdn.com/v4/letter/e/13edae/32.png) [@EZXYZ](https://discuss.hail.is/u/EZXYZ)\
**Post date:** [October 1, 2020, 5:43pm UTC](https://discuss.hail.is/t/blas-library-path/1720/1 "2020-10-01T17:43:45Z")

</div>

Hi there,

I have encountered an error that I think the solution would be pertinent to the suggestion posted here:

> [@Undefined symbol: cblas\_dgemm](https://discuss.hail.is/t/undefined-symbol-cblas-dgemm/1488):
>
> Hi! I get the following error when I execute pipelines including linear\_regression\_rows: /usr/java/default/bin/java: symbol lookup error: /.../...netlib-native\_system-linux-x86\_64.so: undefined symbol: cblas\_dgemv How do I avoid this error?

However, I’m not sure about how to find the path to my BLAS library. May I ask for some instructions? Thank you very much!

---

<div class="post-metadata">

**Author:** ![danking](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.hail.is/danking/32/43_2.png) [@danking](https://discuss.hail.is/u/danking)\
**Post date:** [October 1, 2020, 5:57pm UTC](https://discuss.hail.is/t/blas-library-path/1720/2 "2020-10-01T17:57:19Z")

</div>

If you’re on a GNU/Linux system, check `/lib`, `/usr/lib`, and any of the directories listed in `/etc/ld.so.conf`. In general, I’d ask whoever install BLAS where it is. Most package managers can tell you what files were installed by a given package.

---

<div class="post-metadata">

**Author:** ![EZXYZ](https://avatars.discourse-cdn.com/v4/letter/e/13edae/32.png) [@EZXYZ](https://discuss.hail.is/u/EZXYZ)\
**Post date:** [October 1, 2020, 6:08pm UTC](https://discuss.hail.is/t/blas-library-path/1720/3 "2020-10-01T18:08:20Z")

</div>

Thanks! So after I find the path to BLAS, should I just enter it after the “spark\_conf=” in hl.init(spark\_conf=…)?

---

<div class="post-metadata">

**Author:** ![danking](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.hail.is/danking/32/43_2.png) [@danking](https://discuss.hail.is/u/danking)\
**Post date:** [October 1, 2020, 6:11pm UTC](https://discuss.hail.is/t/blas-library-path/1720/4 "2020-10-01T18:11:59Z")

</div>

`spark_conf` takes a dictionary whose keys are spark parameters, like `"spark.executor.extraClassPath"`, and whose values are a setting for the parameter:

```auto
hl.init(spark_conf={"spark.executor.extraClassPath": "/foo/bar/baz.so"})

```