Update faq #3596
Update faq #3596
Conversation
parbenc
requested review from
gandryey,
mangupta and
rakesroy
as code owners
neon60
force-pushed
the
docs/develop
branch
from
98f9599
to
73814d1
Compare
MKKnorr
force-pushed
the
update-faq
branch
4 times, most recently
from
fdb0996
to
0d53bfd
Compare
| @@ -5,6 +5,9 @@ defaults: | |||
| maxdepth: 6 | |||
| root: index | |||
| subtrees: | |||
| - entries: | |||
| - file: faq | |||
| title: Frequently asked questions | |||
There was a problem hiding this comment.
Please put this to the bottom.
docs/faq.rst
Outdated
| ------------------ | ||
|
|
||
| HIP-Clang is a Clang/LLVM based compiler to compile HIP programs for AMD | ||
| platforms. It's executable is ``amdclang++``. |
There was a problem hiding this comment.
At windows, they just call it clang++. Should we mention here?
| This means you can compile host code with the compiler of your choice and link the | ||
| generated host object code with device code. | ||
|
|
||
| Can HIP applications be compiled with a C compiler? |
There was a problem hiding this comment.
Probably we should add these here:
https://advanced-micro-devices-demo--3622.com.readthedocs.build/projects/HIP/en/3622/understand/compilers.html
There was a problem hiding this comment.
I would have to rebase on top of that. Let's wait until either is merged in and add it to the other one
| Miscellaneous | ||
| ============= | ||
|
|
||
| How to create a guard for code that is specific to the host or the GPU? |
There was a problem hiding this comment.
This should be in the porting guide and we should link here.
There was a problem hiding this comment.
I think this does not (only) belong to the porting guide, this is also relevant for other users. So either the programming guide, the compilers. I'd vote for the compilers or somewhere where we explain the compilation process.
| Compiler related questions | ||
| ========================== | ||
|
|
||
| How to use HIP-Clang to build HIP programs? |
There was a problem hiding this comment.
Please link this tutorial: https://rocm.docs.amd.com/projects/HIP/en/latest/tutorial/saxpy.html#compiling-on-the-command-line
It is written to represent the compiler usage on different OS or platform.
There was a problem hiding this comment.
I was thinking of whether to even keep this question in the FAQ. That tutorial is a bit overkill for this question. Maybe something similar can also be added to the HIP compilers page of #3622 and we could link to that
| * HIP provides device-level control over memory allocation and placement. | ||
| * HIP offers an offline compilation model. | ||
|
|
||
| How does porting CUDA to HIP compare to porting CUDA to OpenCL? |
There was a problem hiding this comment.
|
|
||
| Yes. Beware, that you still need a compatible GPU to actually run the compiled code. | ||
|
|
||
| HIP detected my platform (``amd`` vs ``nvidia``) incorrectly. What should I do? |
There was a problem hiding this comment.
Should we add this to porting guide and link here?
https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html#hip-porting-guide
MKKnorr
force-pushed
the
update-faq
branch
2 times, most recently
from
49a908d
to
958ca19
Compare
No description provided.