Kirill Smelkov <kirr@xxxxxxxxxxxxxx> writes:
>> I do not necessarily buy your "so we HAVE TO, OR ELSE".
>>
>> Even though I can understand "We can sort the list of tests _if_ we do not
>> want them executed in seemingly random order when running 'make -j1'", I
>> tend to think that *if* is a big one. Aren't these tests designed not to
>> depend on each other anyway?
>
> Yes, they don't depend on each other, but what's the point in not
> sorting them? I usually watch test progress visually, and if tests are
> sorted, even with make -j4 they go more or less incrementally by their t
> number.
>
> On my netbook, adding $(sort ...) adds approximately 0.008s to make
> startup, so imho there is no performance penalty to adding that sort.
Heh, who said anything about performance?
I was pointing out that your justification "we HAVE TO" was wrong.
If you are doing this for perceived prettyness and not as a fix for any
correctness issue, I want to see the patch honestly described as such;
that's all.
By the way, if I recall correctly, $(sort) in GNU make not just sorts but
as a nice side effect removes duplicates. So if we used a(n fictional)
construct in our Makefile like this:
T = $(wildcard *.sh a.*)
that might produce duplicates (i.e. "a.sh" might appear twice), which
might leave us two identical pathnames in $T and cause us trouble. Even
if we do not have such a use currently, rewriting $(wildcard) like your
patch does using $(sort $(wildcard ...)) may be a good way to future-proof
our Makefile, and if you justify your patch that way, it would be a
possible correctness hardening, not just cosmetics, and phrasing it with
"HAVE TO" may be justifiable.
Care to try if $(wildcard *.sh a.*) give you duplicated output with newer
GNU make? I am lazy but am a bit curious ;-)
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html