| 1 |
On Mon, 21 Mar 2011 13:07:33 +0100 |
| 2 |
"Paweł Hajdan, Jr." <phajdan.jr@g.o> wrote: |
| 3 |
|
| 4 |
> sys-devel/gcc runs tests, but the results are ignored and I remember the |
| 5 |
> tests fail most of the time. |
| 6 |
|
| 7 |
s/most/all |
| 8 |
|
| 9 |
> Because the tests take long time to run and fail anyway (I understand |
| 10 |
> it's non-trivial to fix those on Gentoo side), I wonder whether it makes |
| 11 |
> sense to run them at all: |
| 12 |
|
| 13 |
It does to me, I use them all the time. ;) The important part is that we |
| 14 |
install the test results, which can then be used for regression testing when |
| 15 |
rolling patchsets. |
| 16 |
|
| 17 |
> toolchain.eclass: |
| 18 |
> |
| 19 |
> gcc_src_test() { |
| 20 |
> cd "${WORKDIR}"/build |
| 21 |
> emake -j1 -k check || ewarn "check failed and that sucks :(" |
| 22 |
> } |
| 23 |
> |
| 24 |
> My suggestion is to make the src_test empty (I think the default one |
| 25 |
> still calls make). I can produce a patch if needed. |
| 26 |
> |
| 27 |
> What do you think? |
| 28 |
|
| 29 |
I think that glibc and gcc tests and other testsuites that nearly always |
| 30 |
fail shouldn't be run for the average user but should still be easily |
| 31 |
accessible in a standard way. I think we need a more finely grained test |
| 32 |
setup, where we can say tests are "expensive" or "interesting only to |
| 33 |
developers" or "known to fail", and let people opt-in to these on a |
| 34 |
per-package basis. Right now you always have to opt-out using |
| 35 |
package.use.mask which "works" but is unintuitive. |
| 36 |
|
| 37 |
|
| 38 |
-- |
| 39 |
fonts, gcc-porting, it makes no sense how it makes no sense |
| 40 |
toolchain, wxwidgets but i'll take it free anytime |
| 41 |
@ gentoo.org EFFD 380E 047A 4B51 D2BD C64F 8AA8 8346 F9A4 0662 |
Archives