Re: [PATCH v4 03/14] trace2: collect platform-specific process information
- From
- Jeff Hostetler <git@jeffhostetler.com>
- Date
- Feb 1, 2019, 11:12 UTC
- Message-ID
- <a1bf3e53-3f79-32f4-f349-a3165c02f595@jeffhostetler.com>
- In-Reply-To
- <20190131231528.GC10587@szeder.dev>
On 1/31/2019 6:15 PM, SZEDER Gábor wrote:
Show 46 quoted lines
> On Wed, Jan 30, 2019 at 12:56:24PM -0800, Jeff Hostetler via GitGitGadget wrote:
>> Add optional platform-specific code to log information about
>> the current process.
>>
>> On Windows, this includes whether git.exe is running under a
>> debugger and information about the ancestors of the process.
>>
>> The purpose of this information is to help indicate if the
>> process was launched interactively or in the background by
>> an IDE, for example.
>
>> diff --git a/trace2.h b/trace2.h
>> index a0e99d9c26..cb11a46366 100644
>> --- a/trace2.h
>> +++ b/trace2.h
>> @@ -363,4 +363,18 @@ __attribute__((format (printf, 1, 2)))
>> void trace2_printf(const char *fmt, ...);
>> #endif
>>
>> +/*
>> + * Optional platform-specific code to dump information about the
>> + * current and any parent process(es). This is intended to allow
>> + * post-processors to know who spawned this git instance and anything
>> + * else the platform may be able to tell us about the current process.
>> + */
>> +#if defined(GIT_WINDOWS_NATIVE)
>> +void trace2_collect_process_info(void);
>> +#else
>> +#define trace2_collect_process_info() \
>> + do { \
>> + } while (0)
>> +#endif
>
> Please consider mentioning in the commit message that on other
> platforms this is a noop. I was scrolling through the whole patch,
> skipping over the Windows-specific parts, to see how you did it on
> Linux, only to find the above do-nothing loop. It was anticlimactic :)
>
> Why is it a noop on other platforms? I suspect that (since your main
> focus is supporting Windows devs using Git on Windows) it's along the
> lines of "I just didn't want to bother, and left it as future work for
> anyone interested", which a perfectly valid reason in my book.
> However, if you did look into it, and found some major difficulties or
> downright showstoppers, then that might be worth mentioning.
> (Portability? I would expect that it would need a bunch of '#elif
> defined(...)')Right, my primary focus was Windows at the time and I only added the Windows version. I'll document that.
Thanks Jeff