| Message ID | 20260319132904.1308436-1-joaomarcos.costa@bootlin.com |
|---|---|
| State | Changes Requested |
| Headers | show |
| Series | ltp: fix warning in remove_broken_musl_sources | expand |
On Thu Mar 19, 2026 at 2:29 PM CET, Joao Marcos Costa via lists.openembedded.org wrote: > If this was originally intended to be an actual warning, it was not > behaving as such: echo only prints to log.do_patch, so the message ends > up hidden there. > > Replace 'echo' by bbwarn to display the warning message correctly. > > Signed-off-by: João Marcos Costa <joaomarcos.costa@bootlin.com> > --- Hi João, Thanks for your patch. > meta/recipes-extended/ltp/ltp_20260130.bb | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/meta/recipes-extended/ltp/ltp_20260130.bb b/meta/recipes-extended/ltp/ltp_20260130.bb > index 75c5b8b7bd..36432c15b2 100644 > --- a/meta/recipes-extended/ltp/ltp_20260130.bb > +++ b/meta/recipes-extended/ltp/ltp_20260130.bb > @@ -126,7 +126,7 @@ remove_broken_musl_sources() { > [ "${TCLIBC}" = "musl" ] || return 0 > > cd ${S} > - echo "WARNING: remove unsupported tests (until they're fixed)" > + bbwarn "remove unsupported tests (until they're fixed)" So I believe this will add a WARNING on all musl builds, right? I'm not sure we can really accept that. Thanks, Mathieu
Hello, On 3/19/26 18:38, Mathieu Dubois-Briand wrote: > On Thu Mar 19, 2026 at 2:29 PM CET, Joao Marcos Costa via lists.openembedded.org wrote: >> If this was originally intended to be an actual warning, it was not >> behaving as such: echo only prints to log.do_patch, so the message ends >> up hidden there. >> >> Replace 'echo' by bbwarn to display the warning message correctly. >> >> Signed-off-by: João Marcos Costa <joaomarcos.costa@bootlin.com> >> --- > > Hi João, > > Thanks for your patch. > >> meta/recipes-extended/ltp/ltp_20260130.bb | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/meta/recipes-extended/ltp/ltp_20260130.bb b/meta/recipes-extended/ltp/ltp_20260130.bb >> index 75c5b8b7bd..36432c15b2 100644 >> --- a/meta/recipes-extended/ltp/ltp_20260130.bb >> +++ b/meta/recipes-extended/ltp/ltp_20260130.bb >> @@ -126,7 +126,7 @@ remove_broken_musl_sources() { >> [ "${TCLIBC}" = "musl" ] || return 0 >> >> cd ${S} >> - echo "WARNING: remove unsupported tests (until they're fixed)" >> + bbwarn "remove unsupported tests (until they're fixed)" > > So I believe this will add a WARNING on all musl builds, right? I'm not > sure we can really accept that. > Yep, it will. Would an INFO/NOTE be acceptable instead? Thanks!
On Thu Mar 19, 2026 at 6:55 PM CET, Joao Marcos Costa wrote: > Hello, > > On 3/19/26 18:38, Mathieu Dubois-Briand wrote: >> On Thu Mar 19, 2026 at 2:29 PM CET, Joao Marcos Costa via lists.openembedded.org wrote: >>> If this was originally intended to be an actual warning, it was not >>> behaving as such: echo only prints to log.do_patch, so the message ends >>> up hidden there. >>> >>> Replace 'echo' by bbwarn to display the warning message correctly. >>> >>> Signed-off-by: João Marcos Costa <joaomarcos.costa@bootlin.com> >>> --- >> >> Hi João, >> >> Thanks for your patch. >> >>> meta/recipes-extended/ltp/ltp_20260130.bb | 2 +- >>> 1 file changed, 1 insertion(+), 1 deletion(-) >>> >>> diff --git a/meta/recipes-extended/ltp/ltp_20260130.bb b/meta/recipes-extended/ltp/ltp_20260130.bb >>> index 75c5b8b7bd..36432c15b2 100644 >>> --- a/meta/recipes-extended/ltp/ltp_20260130.bb >>> +++ b/meta/recipes-extended/ltp/ltp_20260130.bb >>> @@ -126,7 +126,7 @@ remove_broken_musl_sources() { >>> [ "${TCLIBC}" = "musl" ] || return 0 >>> >>> cd ${S} >>> - echo "WARNING: remove unsupported tests (until they're fixed)" >>> + bbwarn "remove unsupported tests (until they're fixed)" >> >> So I believe this will add a WARNING on all musl builds, right? I'm not >> sure we can really accept that. >> > Yep, it will. Would an INFO/NOTE be acceptable instead? > > Thanks! Yes, I believe info or note are acceptable. Thanks, Mathieu
Hello, On 3/19/26 19:17, Mathieu Dubois-Briand wrote: > On Thu Mar 19, 2026 at 6:55 PM CET, Joao Marcos Costa wrote: >> Hello, >> >> On 3/19/26 18:38, Mathieu Dubois-Briand wrote: >>> On Thu Mar 19, 2026 at 2:29 PM CET, Joao Marcos Costa via lists.openembedded.org wrote: >>>> If this was originally intended to be an actual warning, it was not >>>> behaving as such: echo only prints to log.do_patch, so the message ends >>>> up hidden there. >>>> >>>> Replace 'echo' by bbwarn to display the warning message correctly. >>>> >>>> Signed-off-by: João Marcos Costa <joaomarcos.costa@bootlin.com> >>>> --- >>> (...)>>> >> Yep, it will. Would an INFO/NOTE be acceptable instead? >> >> Thanks! > > Yes, I believe info or note are acceptable. > > Thanks, > Mathieu I sent a v2 to fix that. One more thing, though: this same function uses an 'rm -rfv' and the "-f" option will not raise an error if the file is not there. Do you think this should be addressed? I would send this in a different patch, of course. Thanks!
On Fri, 2026-03-20 at 10:16 +0100, Joao Marcos Costa via lists.openembedded.org wrote: > Hello, > > On 3/19/26 19:17, Mathieu Dubois-Briand wrote: > > On Thu Mar 19, 2026 at 6:55 PM CET, Joao Marcos Costa wrote: > > > Hello, > > > > > > On 3/19/26 18:38, Mathieu Dubois-Briand wrote: > > > > On Thu Mar 19, 2026 at 2:29 PM CET, Joao Marcos Costa via lists.openembedded.org wrote: > > > > > If this was originally intended to be an actual warning, it was not > > > > > behaving as such: echo only prints to log.do_patch, so the message ends > > > > > up hidden there. > > > > > > > > > > Replace 'echo' by bbwarn to display the warning message correctly. > > > > > > > > > > Signed-off-by: João Marcos Costa <joaomarcos.costa@bootlin.com> > > > > > --- > > > > > (...)>>> > > > Yep, it will. Would an INFO/NOTE be acceptable instead? > > > > > > Thanks! > > > > Yes, I believe info or note are acceptable. > > > > Thanks, > > Mathieu > > I sent a v2 to fix that. > > One more thing, though: this same function uses an 'rm -rfv' and the > "-f" option will not raise an error if the file is not there. > > Do you think this should be addressed? I would send this in a different > patch, of course. Tasks need to be idempotent, we can run then multiple times with the same result. I didn't check where this is being called but the -f is probably fine. Cheers, Richard
diff --git a/meta/recipes-extended/ltp/ltp_20260130.bb b/meta/recipes-extended/ltp/ltp_20260130.bb index 75c5b8b7bd..36432c15b2 100644 --- a/meta/recipes-extended/ltp/ltp_20260130.bb +++ b/meta/recipes-extended/ltp/ltp_20260130.bb @@ -126,7 +126,7 @@ remove_broken_musl_sources() { [ "${TCLIBC}" = "musl" ] || return 0 cd ${S} - echo "WARNING: remove unsupported tests (until they're fixed)" + bbwarn "remove unsupported tests (until they're fixed)" # sync with upstream # https://github.com/linux-test-project/ltp/blob/master/ci/alpine.sh#L33
If this was originally intended to be an actual warning, it was not behaving as such: echo only prints to log.do_patch, so the message ends up hidden there. Replace 'echo' by bbwarn to display the warning message correctly. Signed-off-by: João Marcos Costa <joaomarcos.costa@bootlin.com> --- meta/recipes-extended/ltp/ltp_20260130.bb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)