1 |
2016-06-21 23:07 GMT+08:00 Peter Smith <peter.smith@××××××.org>: |
2 |
> Hello Lei, |
3 |
> |
4 |
> The changes to llvm and clang look ok to me. I've got some suggestions |
5 |
> for testing. |
6 |
> |
7 |
> For the clang patch, it looks like there isn't a test to check that |
8 |
> musleabihf implies hard floating point. It looks like |
9 |
> Driver/arm-mfpu.c CHECK-HF might be a good candidate to add a test. |
10 |
> |
11 |
> For the llvm patch |
12 |
> |
13 |
> I think you should be able to find a test that checks the behaviour of |
14 |
> GNUEABI and GNUEABIHF for each of the properties that you've added |
15 |
> Subtarget->isTargetMuslAEABI() to or equivalent. It would be useful to |
16 |
> add a test case for MUSLEABI and/or MUSLEABIHF. For example in the |
17 |
> RTLIB case there are a large number of tests that check whether the |
18 |
> correct __aeabi_ function is called. |
19 |
> |
20 |
> Some files I came across (there are many more) that might be a good |
21 |
> place to check that musleabi and musleabihf behaves like gnueabi and |
22 |
> gnueabihf: |
23 |
> CodeGen/ARM/memfunc.ll |
24 |
> CodeGen/Thumb2/float-ops.ll |
25 |
> CodeGen/ARM/divmod-eabi.ll |
26 |
> CodeGen/ARM/fp16.ll (hard-float for HF) |
27 |
> MC/ARM/eh-directives-personalityindex.s |
28 |
|
29 |
Thanks for the pointers! Please see the refined (again) patches. |
30 |
|
31 |
As a side note, there's no "gnueabi" in float-ops.ll or |
32 |
eh-directive-personalityindex.s, so I skipped them. In addition, I |
33 |
found a few other relevant test files to patch thanks to your advice. |
34 |
|
35 |
|
36 |
Lei |