| 1 |
2016-06-30 15:48 GMT+08:00 Luca Barbato <lu_zero@g.o>: |
| 2 |
> On 29/06/16 17:52, Lei Zhang wrote: |
| 3 |
>> Plan for today: |
| 4 |
>> - investigate clang's poor performance |
| 5 |
>> - study catalyst and stuff |
| 6 |
>> |
| 7 |
>> Progress: |
| 8 |
>> - I find out that optimization flags are not correctly passed to |
| 9 |
>> compiler when building the freestanding clang. The cause is a bit |
| 10 |
>> complicated: |
| 11 |
>> |
| 12 |
>> clang has to be built in-tree (i.e. along with LLVM); compiler-rt can |
| 13 |
>> be built out-of-tree, but it's a little inconvenient so I also built |
| 14 |
>> it in-tree. This means clang, compiler-rt and LLVM are built |
| 15 |
>> all-together. compiler-rt's sanitizer has some issue with musl, so I |
| 16 |
>> disabled building the sanitizer via some option. It turns out this |
| 17 |
>> option somehow leads cmake to forget passing optimization flags to |
| 18 |
>> compiler. The solution is simple: build compiler-rt out-of-tree. Or I |
| 19 |
>> can resolve the issue between compiler-rt's sanitizer and musl, when |
| 20 |
>> I've got time... |
| 21 |
> |
| 22 |
> Sounds a good plan =) |
| 23 |
|
| 24 |
I just found that compiler-rt is built as part of clang/LLVM in the |
| 25 |
default ebuilds. Perhaps I should leave it that way: not to create a |
| 26 |
new package for compiler-rt. That means compiler-rt still needs to be |
| 27 |
built in-tree, and I'll have to deal with the optimization flag issue. |
| 28 |
|
| 29 |
I also noticed that gcc's sanitizer doesn't work on the musl based |
| 30 |
system, so is it forgivable to not support clang's sanitizer either? |
| 31 |
(clang's sanitizer is implemented in compiler-rt) |
| 32 |
|
| 33 |
|
| 34 |
Lei |
Archives