Displaying 4 results from an estimated 4 matches for "r270942".
2016 May 28
2
[LibFuzzer] Recent performance regression due to r270942
Hi,
This started as an off hand comment in [1] but this appears to be a
real issue so I'm moving the discussion to the mailing list.
In r270942 the time taken to run LibFuzzer's test became noticeably
longer. I am building on
* Arch Linux (4.5.4-1-ARCH #1 SMP PREEMPT Wed May 11 22:21:28 CEST
2016 x86_64 GNU/Linux)
* I am building libFuzzer and running its tests like so
```
CC=<new_clang> CXX=<new_clang++> cmake -DLLVM_USE...
2016 May 28
0
[LibFuzzer] Recent performance regression due to r270942
...ecause it sees everything it wants to see.
--kcc
On Fri, May 27, 2016 at 6:18 PM, Dan Liew <dan at su-root.co.uk> wrote:
> Hi,
>
> This started as an off hand comment in [1] but this appears to be a
> real issue so I'm moving the discussion to the mailing list.
>
> In r270942 the time taken to run LibFuzzer's test became noticeably
> longer. I am building on
>
> * Arch Linux (4.5.4-1-ARCH #1 SMP PREEMPT Wed May 11 22:21:28 CEST
> 2016 x86_64 GNU/Linux)
> * I am building libFuzzer and running its tests like so
>
> ```
> CC=<new_clang> CX...
2016 May 28
2
[LibFuzzer] Recent performance regression due to r270942
...;
> --kcc
>
> On Fri, May 27, 2016 at 6:18 PM, Dan Liew <dan at su-root.co.uk> wrote:
>
>> Hi,
>>
>> This started as an off hand comment in [1] but this appears to be a
>> real issue so I'm moving the discussion to the mailing list.
>>
>> In r270942 the time taken to run LibFuzzer's test became noticeably
>> longer. I am building on
>>
>> * Arch Linux (4.5.4-1-ARCH #1 SMP PREEMPT Wed May 11 22:21:28 CEST
>> 2016 x86_64 GNU/Linux)
>> * I am building libFuzzer and running its tests like so
>>
>> ```...
2016 May 28
0
[LibFuzzer] Recent performance regression due to r270942
On 27 May 2016 at 21:26, Kostya Serebryany <kcc at google.com> wrote:
> Done. r271095
Thanks that fixed the issue for me. But now ``fuzzer.test`` is failing
for me. Specifically
```
not LLVMFuzzer-NullDerefTest -close_fd_mask=3 2>&1 | FileCheck %s
--check-prefix=NullDerefTest
```
However it looks like this is to be expected because this test is
relying on the symbol