On 22.11.25 09:54, Thomas Munro wrote:
> On Fri, Nov 21, 2025 at 11:48 PM Peter Eisentraut <peter@eisentraut.org> wrote:
>> Currently, headerscheck and cpluspluscheck are very slow, and they
>> defeat use of ccache. I have fixed that, and now they are much faster. :-)
>>
>> The problem was (I think) that the test files are created in a
>> randomly-named directory (`mktemp -d /tmp/$me.XXXXXX`), and this
>> directory is named on the compiler command line, which is part of the
>> cache key.
>>
>> My solution is to create the test files in the build directory. For
>> example, for src/include/storage/ipc.h I generate
>>
>> headerscheck_src_include_storage_ipc_h.c (or .cpp)
>>
>> Now ccache works. (And it's also a bit easier to debug everything with
>> this naming.)
>>
>> The observed speedup on Cirrus CI for headerscheck plus cpluspluscheck
>> is from about 1min 20s to only 20s. In local use, the speedups are similar.
>
> +1
>
> I wrote an almost identical patch[1] and then lost it down the back of
> the sofa. I was wondering about parallelising it next...
>
> [1] https://www.postgresql.org/message-id/CA%2BhUKGJjQyZUvcu6udk5OKz5rnaF4a_hm5nb_VtZHYMH%2BvsN0g%40mail.gmail.com
Ah yes, that's about the same idea. The difference is that yours
requires specifying TMPDIR on the make invocation. So it wouldn't
happen by default for local (non-CI) use. I think I would like an
implementation that also worked out of the box locally.
