| Message ID | AM6PR10MB2566B6AD3CAA9BEF6CA99C9DE4B90@AM6PR10MB2566.EURPRD10.PROD.OUTLOOK.COM |
|---|---|
| State | New |
| Headers | show |
| Series | [LIBPHOBOS] Fix multi-lib RUNTESTFLAGS handling | expand |
On Tue, 3 Sep 2019 at 08:10, Bernd Edlinger <bernd.edlinger@hotmail.de> wrote: > > Hi, > > > I've noticed that testing libphobos fails for multi-lib configs: > > $ make check-target-libphobos RUNTESTFLAGS="--target_board=unix\{-m32,\}" > > fails for every 32bit execution, because the host libgcc_s.so is used which > is not the correct version: > > spawn [open ...] > ./test_aa.exe: /lib/i386-linux-gnu/libgcc_s.so.1: version `GCC_7.0.0' not found (required by ./test_aa.exe) > FAIL: libphobos.aa/test_aa.d execution test > > This can be fixed by adding a few lines from libstdc++/testsuite/lib/libstdc++.exp > to libphobos/testsuite/lib/libphobos.exp, see attached patch. > > > Bootstrapped and reg-tested on x86_64-pc-linux-gnu. > Is it OK for trunk? > OK.
Hi! On 2019-09-03T10:04:14+0200, Iain Buclaw <ibuclaw@gdcproject.org> wrote: > On Tue, 3 Sep 2019 at 08:10, Bernd Edlinger <bernd.edlinger@hotmail.de> wrote: >> I've noticed that testing libphobos fails for multi-lib configs: >> >> $ make check-target-libphobos RUNTESTFLAGS="--target_board=unix\{-m32,\}" >> >> fails for every 32bit execution, because the host libgcc_s.so is used which >> is not the correct version: >> >> spawn [open ...] >> ./test_aa.exe: /lib/i386-linux-gnu/libgcc_s.so.1: version `GCC_7.0.0' not found (required by ./test_aa.exe) >> FAIL: libphobos.aa/test_aa.d execution test >> >> This can be fixed by adding a few lines from libstdc++/testsuite/lib/libstdc++.exp >> to libphobos/testsuite/lib/libphobos.exp, see attached patch. >> >> >> Bootstrapped and reg-tested on x86_64-pc-linux-gnu. >> Is it OK for trunk? > > OK. The very same problem existed since the beginning of D language support in GCC, so I backported this to gcc-9-branch in r277611 "[LIBPHOBOS] Fix multi-lib RUNTESTFLAGS handling", see attached. Grüße Thomas
Hi! Just for posterity. On 2019-04-20T23:22:31+0200, Iain Buclaw <ibuclaw@gdcproject.org> wrote: > On Sat, 20 Apr 2019 at 22:30, Thomas Schwinge <thomas@codesourcery.com> wrote: >> On Tue, 18 Sep 2018 02:39:46 +0200, Iain Buclaw <ibuclaw@gdcproject.org> wrote: >> > This patch adds the configure and make files used for building D >> > runtime and Phobos. As well as running all unittests and the >> > testsuite. >> >> With a x86_64-pc-linux-gnu build, I've noticed breakage in '-m32' >> multilib testing, made apparent by message: "[...]: >> /lib/i386-linux-gnu/libgcc_s.so.1: version `GCC_7.0.0' not found >> (required by [...])". (That is, the system 'libgcc_s.so.1' being >> dynamically linked instead of the just built one.) This is because of >> incomplete 'gccdir' setup in the '*.exp' file. In such a multilibbed >> configuration, there are 'build-gcc/gcc/libgcc.*' and >> 'build-gcc/gcc/32/libgcc.*' (for example); for '-m32' multilib testing, >> paths need to be set up to point to the latter instead of the former. It >> seems as if some of this '*.exp' stuff has been copied from libffi (?); >> the attached patch copies the missing pieces from there, too. I've been >> tempted to commit this "as obvious", but then thought I'll still get some >> review/approval first. If approving this patch, please respond with >> "Reviewed-by: NAME <EMAIL>" so that your effort will be recorded in the >> commit log, see <https://gcc.gnu.org/wiki/Reviewed-by>. > > Seems reasonable to me. I didn't get this committed promptly -- and in the mean time (what's half a year, eh?), Bernd addressed the same issue: <http://mid.mail-archive.com/AM6PR10MB2566B6AD3CAA9BEF6CA99C9DE4B90@AM6PR10MB2566.EURPRD10.PROD.OUTLOOK.COM>. There remain a few incremental changes from his committed to my originally proposed version (see attached), but I'm not committing that now, as it's not affecting test results, and refactoring all the '*.exp' files, unifying what has been copied from elsewhere, is a (big) task for another day (weeks...). Grüße Thomas
2019-09-02 Bernd Edlinger <bernd.edlinger@hotmail.de> * testsuite/lib/libphobos.exp (libphobos_init): Add multi-lib libgcc dirs to the ld_library_path var. Index: libphobos/testsuite/lib/libphobos.exp =================================================================== --- libphobos/testsuite/lib/libphobos.exp (revision 275320) +++ libphobos/testsuite/lib/libphobos.exp (working copy) @@ -170,6 +170,25 @@ proc libphobos_init { args } { append ld_library_path ":${blddir}/src/.libs" } + # Compute what needs to be added to the existing LD_LIBRARY_PATH. + if {$gccdir != ""} { + set compiler ${gccdir}/gdc + + if { [is_remote host] == 0 && [which $compiler] != 0 } { + foreach i "[exec $compiler --print-multi-lib]" { + set mldir "" + regexp -- "\[a-z0-9=_/\.-\]*;" $i mldir + set mldir [string trimright $mldir "\;@"] + if { "$mldir" == "." } { + continue + } + if { [llength [glob -nocomplain ${gccdir}/${mldir}/libgcc_s*.so.*]] >= 1 } { + append ld_library_path ":${gccdir}/${mldir}" + } + } + } + } + set_ld_library_path_env_vars libphobos_maybe_build_wrapper "${objdir}/testglue.o"
Hi, I've noticed that testing libphobos fails for multi-lib configs: $ make check-target-libphobos RUNTESTFLAGS="--target_board=unix\{-m32,\}" fails for every 32bit execution, because the host libgcc_s.so is used which is not the correct version: spawn [open ...] ./test_aa.exe: /lib/i386-linux-gnu/libgcc_s.so.1: version `GCC_7.0.0' not found (required by ./test_aa.exe) FAIL: libphobos.aa/test_aa.d execution test This can be fixed by adding a few lines from libstdc++/testsuite/lib/libstdc++.exp to libphobos/testsuite/lib/libphobos.exp, see attached patch. Bootstrapped and reg-tested on x86_64-pc-linux-gnu. Is it OK for trunk? Thanks Bernd.