Any plans to add cmake support?

[StephanDollberg]

See title.

bazel is not really widely used.

[ckennelly]

Not at this time

[JonKohler]

Would you accept PRs for someone adding that? or just "hard no"? :)

[jedisct1]

Autotools, please.

[EricWF]

It's not a hard no to CMake. We've discussed it, and would like to support it in the future. However, time is a finite resource, and nobody is scheduled to work on it. I've told @ckennelly that I'll look into it in the coming weeks, but I have a lot on my plate already.

We likely want to model the behavior of Abseil in it's handling of CMake and Bazel, so PR's that implement something different would not be accepted at this time.

Supporting autotools is a non-goal of the project.

[tru]

When you say:

We likely want to model the behavior of Abseil in it's handling of CMake and Bazel, so PR's that implement something different would not be accepted at this time.

What are you referring to specifically then?

I will probably write some CMake for tcmalloc soon in any case - so if I can model it "right" from the beginning I am happy to do so.

[AsteriskZuo]

Compile failed. All day. I want to use Cmake.

[alexey-milovidov]

Some example of integration of tcmalloc to a project with CMake build system can be found here: ClickHouse/ClickHouse#11590

[ipizhenko]

+1 to this. Please publish build instructions without Bazel, using cmake or just GNU make.

[ObiWahn]

Hi, I have started to play here:

https://github.com/extcpp/tcmalloc/tree/cmake

Anybody is welcome to use the files make suggestions and create PRs. My hope is that the branch will eventually be merged into this repository. My aim is not to replace Bazel but to make tcmalloc usable for projects that use CMake and want to build tcmalloc from source. Therefore I personally neglect (tests - in the beginning) installation and especially packaging.

[Dan-Do]

When you say:

We likely want to model the behavior of Abseil in it's handling of CMake and Bazel, so PR's that implement something different would not be accepted at this time.

What are you referring to specifically then?

I will probably write some CMake for tcmalloc soon in any case - so if I can model it "right" from the beginning I am happy to do so.

Hello @tru
How is the status of making cmake build?

[tru]

We dropped it and went with rpmalloc instead.

[todd-richmond]

a cmake PR has been made that looks promising. would be great to have a tcmalloc official release once that gets in and the core code is at a known stable point. gperftools follows this project too slowly

[matta]

a cmake PR has been made that looks promising.

I assume this refers to #104, just to note it to make it easier to find. Cmake support would be in line with https://opensource.google/documentation/policies/cplusplus-support.

[patty641]

Hi @tru
Are there any updates regarding the status of Cmake Build?

[tru]

Hi @tru Are there any updates regarding the status of Cmake Build?

#4 (comment)

[ObiWahn]

cryptopp has a nice solution. I could create something similar to https://github.com/noloader/cryptopp-cmake. It would be nice if there was somebody to test and and give feedback.

UPDATE:
I have started here: https://github.com/ObiWahn/tcmalloc-cmake