Re: Windows: core.useBuiltinFSMonitor without core.untrackedcache - performance hazard?

From: Jeff Hostetler <git@jeffhostetler.com>
Date: Jun 24, 2021, 13:10 UTC
Message-ID: <596386e9-f9bf-c465-ec5e-dcf4356b6194@jeffhostetler.com>
In-Reply-To: <CAPMMpojKMZmorKPgMASkHsvKcQTaaGyWURsgDS+KTGVeuBh1ww@mail.gmail.com>

On 6/24/21 1:25 AM, Tao Klerks wrote:

Show 26 quoted lines

> Hi Jeff,
> 
> On Mon, Jun 21, 2021 at 8:41 PM Jeff Hostetler <git@jeffhostetler.com> wrote:
>> We're currently looking at a problem that we believe is in the
>> untracked-cache code.  This is causing some of our Scalar tests
>> to fail on Windows when the untracked-cache is turned on.  This
>> is independent of whether FSMonitor or FSCache is turned on.
>> We're still tracking this down.
>>
>> And yes, the best possible solution is to turn on FSMonitor *and*
>> the untracked-cache, so that the "untracked" status code doesn't
>> have to do anything.  So I want to look at tracking down the above
>> problem before doing anything else.
> 
> I got a bit excited about a possible clean path forward to getting
> -uall to work well with untracked cache, and submitted a patch along
> those lines, but rereading the above I should probably have been a
> little more patient.
> 
> Is there anything "we" can do to see/understand the
> scalar-test-suite-error you describe above, or is this
> microsoft-internal?
> 
> Thanks
> Tao
>

Thanks for looking into this. The untracked-cache code is pretty dense and having another set of eyes is good. I apologize that I'm still working thru the backlog from my vacation and haven't gotten to spend any "quality" time with the untracked-cache code yet, so I need to do some homework and study the questions/issues that you've found so far. (Thanks again)

All of our work is done in the open, so yes you should be able to see what we're doing and the errors that we're getting.

The source for our Scalar functional tests is in: https://github.com/microsoft/scalar

My test branch 'test-no-fscache' is in my personal development fork: https://github.com/jeffhostetler/git

I have a PR against microsoft/git (a fork of git-for-windows which is a fork of core git) which turns off a bunch of things to try to isolate the failures: https://github.com/microsoft/git/pull/383

Output for a recent run can be seen here: https://github.com/microsoft/git/pull/383/checks?check_run_id=2884524819

Let us know if you have questions. Jeff