| Message ID | 20201110120923.21759-1-pvorel@suse.cz |
|---|---|
| State | Superseded |
| Headers | show |
| Series | [1/1] m4: Remove workaround for old pkg-config | expand |
Hi! Looks good to me, let's apply this unless anybody complains.
Hi, On Wed, Nov 11, 2020 at 5:38 PM Cyril Hrubis <chrubis@suse.cz> wrote: > Hi! > Looks good to me, let's apply this unless anybody complains. > Hold on, please. Seems we have to replace TIRPC_LIBS/TIRPC_CFLAGS with LIBTIRPC_* in all respective Makefiles? Otherwise, it will fail like the Travis CI job I guess.
Hi, > On Wed, Nov 11, 2020 at 5:38 PM Cyril Hrubis <chrubis@suse.cz> wrote: > > Hi! > > Looks good to me, let's apply this unless anybody complains. > Hold on, please. > Seems we have to replace TIRPC_LIBS/TIRPC_CFLAGS with LIBTIRPC_* in all > respective Makefiles? Otherwise, it will fail like the Travis CI job I > guess. Correct, currently would fail without it. https://travis-ci.org/github/pevik/ltp/builds/742685737 I'll send v2 where I implement this. Kind regards, Petr
On Wed, Nov 11, 2020 at 5:56 PM Petr Vorel <pvorel@suse.cz> wrote: > Hi, > > > On Wed, Nov 11, 2020 at 5:38 PM Cyril Hrubis <chrubis@suse.cz> wrote: > > > > Hi! > > > Looks good to me, let's apply this unless anybody complains. > > > > Hold on, please. > > > Seems we have to replace TIRPC_LIBS/TIRPC_CFLAGS with LIBTIRPC_* in all > > respective Makefiles? Otherwise, it will fail like the Travis CI job I > > guess. > > Correct, currently would fail without it. > https://travis-ci.org/github/pevik/ltp/builds/742685737 > > I'll send v2 where I implement this. > Thanks, I modified base on your patch, test result FYI https://github.com/wangli5665/ltp/commit/dad10021771101e7fdd416af894be7484fcb7a52 https://travis-ci.org/github/wangli5665/ltp/builds/742909686
Hi Li, > On Wed, Nov 11, 2020 at 5:56 PM Petr Vorel <pvorel@suse.cz> wrote: > > Hi, > > > On Wed, Nov 11, 2020 at 5:38 PM Cyril Hrubis <chrubis@suse.cz> wrote: > > > > Hi! > > > > Looks good to me, let's apply this unless anybody complains. > > > Hold on, please. > > > Seems we have to replace TIRPC_LIBS/TIRPC_CFLAGS with LIBTIRPC_* in all > > > respective Makefiles? Otherwise, it will fail like the Travis CI job I > > > guess. > > Correct, currently would fail without it. > > https://travis-ci.org/github/pevik/ltp/builds/742685737 > > I'll send v2 where I implement this. > Thanks, I modified base on your patch, test result FYI > https://github.com/wangli5665/ltp/commit/dad10021771101e7fdd416af894be7484fcb7a52 > https://travis-ci.org/github/wangli5665/ltp/builds/742909686 You implemented it, thank you. Please push it after Travis finishes. Just an explanation, why I introduced TIRPC_{CFLAGS,LIBS}: there was Sun-RPC implementation in glibc, which I wanted to test. Thus variable name without "LIB" to say, that it could be also glibc. But Sun-RPC is broken, was deprecated and my patchset which removes it from glibc was accepted and released in glibc 2.32. Thus I'm going to send patch which reverts the commit 60b20c428060b20c428 ("rpc: Enable and fix build basic RPC tests with glibc SunRPC"). My long term goal: fix and cleanup RPC tests (some are failing also with libtirpc), move these tests to libtirpc sources (they needs to be cleaned. And also I'd prefer to not put there whole LTP library, but just needed subset). Kind regards, Petr
Hi Petr, On Wed, Nov 11, 2020 at 7:48 PM Petr Vorel <pvorel@suse.cz> wrote: > Hi Li, > > > On Wed, Nov 11, 2020 at 5:56 PM Petr Vorel <pvorel@suse.cz> wrote: > > > > Hi, > > > > > On Wed, Nov 11, 2020 at 5:38 PM Cyril Hrubis <chrubis@suse.cz> > wrote: > > > > > > Hi! > > > > > Looks good to me, let's apply this unless anybody complains. > > > > > > Hold on, please. > > > > > Seems we have to replace TIRPC_LIBS/TIRPC_CFLAGS with LIBTIRPC_* in > all > > > > respective Makefiles? Otherwise, it will fail like the Travis CI job > I > > > > guess. > > > > Correct, currently would fail without it. > > > https://travis-ci.org/github/pevik/ltp/builds/742685737 > > > > I'll send v2 where I implement this. > > > > Thanks, I modified base on your patch, test result FYI > > > https://github.com/wangli5665/ltp/commit/dad10021771101e7fdd416af894be7484fcb7a52 > > https://travis-ci.org/github/wangli5665/ltp/builds/742909686 > You implemented it, thank you. Please push it after Travis finishes. Pushed, and thanks for the explanation. > > Just an explanation, why I introduced TIRPC_{CFLAGS,LIBS}: there was > Sun-RPC > implementation in glibc, which I wanted to test. Thus variable name without > "LIB" to say, that it could be also glibc. > > But Sun-RPC is broken, was deprecated and my patchset which removes it from > glibc was accepted and released in glibc 2.32. Thus I'm going to send patch > which reverts the commit 60b20c428060b20c428 ("rpc: Enable and fix build > basic > RPC tests with glibc SunRPC"). > > My long term goal: fix and cleanup RPC tests (some are failing also with > libtirpc), move these tests to libtirpc sources (they needs to be cleaned. > And > also I'd prefer to not put there whole LTP library, but just needed > subset). > > Kind regards, > Petr > >
diff --git a/m4/ltp-tirpc.m4 b/m4/ltp-tirpc.m4 index 4d9701469..f06fed3d9 100644 --- a/m4/ltp-tirpc.m4 +++ b/m4/ltp-tirpc.m4 @@ -22,9 +22,4 @@ AC_DEFUN([LTP_CHECK_TIRPC], [ if test "x$have_libtirpc" = "xyes" -o "x$have_rpc_glibc" = "xyes"; then AC_SUBST(HAVE_RPC, 1) fi - - dnl fix for old pkg-config (< 0.24) - dnl https://autotools.io/pkgconfig/pkg_check_modules.html - AC_SUBST(TIRPC_CFLAGS) - AC_SUBST(TIRPC_LIBS) ])
fix was for 0.24 which is probably old enough (released in 2010; problematic 0.23 was released 2008). Signed-off-by: Petr Vorel <pvorel@suse.cz> --- Travis CI: https://travis-ci.org/github/pevik/ltp/builds/742685737 m4/ltp-tirpc.m4 | 5 ----- 1 file changed, 5 deletions(-)