Source-Changes-D archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: CVS commit: src



On 11.09.2020 06:57, Rin Okuyama wrote:
> On 2020/09/10 18:28, Kamil Rytarowski wrote:
>> On 10.09.2020 10:50, Rin Okuyama wrote:
>>> On 2020/09/10 16:41, Kamil Rytarowski wrote:
>>>> On 10.09.2020 03:53, Rin Okuyama wrote:
>>>>> Module Name:    src
>>>>> Committed By:    rin
>>>>> Date:        Thu Sep 10 01:53:22 UTC 2020
>>>>>
>>>>> Modified Files:
>>>>>      src/distrib/sets/lists/base: mi
>>>>>      src/distrib/sets/lists/comp: mi
>>>>>      src/sys/dev: Makefile
>>>>>
>>>>> Log Message:
>>>>> Unconditionally install kernel headers for iSCSI as required by
>>>>> sanitizer shipped with GCC9.
>>>>>
>>>>> Fix build release with HAVE_GCC=9 for sun2, where MKISCSI=no by
>>>>> default.
>>>>>
>>>>
>>>> Please redo this commit with __has_include(), example:
>>>>
>>>> https://github.com/llvm/llvm-project/commit/7f6b25ad1bb3f8057655a9bad2a3b69621f4a543#diff-fa188a123646bb8c30d7fa22d61ef680
>>>>
>>>>
>>>>
>>>
>>> Hmm, I'm not sure whether it is worth the maintenance cost and
>>> complexities it adds... Is there any reason why we should not
>>> install these headers?
>>>
>>
>> The headers are optional and sanitizers shall not dictate what headers
>> are installed.
>>
>> Meanwhile, I'm going to do this in upstream LLVM.
> 
> Yes, I really appreciate your continuous efforts in syncing our
> codes with upstream.
> 
> However, still, there is ``the maintenance cost'' in our local
> repository. With __has_include(), we must maintain two copies of
> headers in the tree. This is dangerous potentially.
> 

There is no danger in terms of security.

> My question was whether it is worth the risk not to install few
> number of small headers.
> 

I've updated the LLVM interceptors. Please revert.

> Thanks,
> rin


Attachment: signature.asc
Description: OpenPGP digital signature



Home | Main Index | Thread Index | Old Index