Lvm2-2.02.70 fails to build with dmsetup-related errors. I tried also with MAKEOPTS="-j1" and distcc disabled, without any luck.
Created attachment 240583 [details] build.log
Created attachment 240585 [details] emerge --info
i686-pc-linux-gnu-gcc -O2 -march=pentium3 -pipe -O2 -march=pentium3 -pipe -fPIC -Wall -Wundef -Wshadow -Wcast-align -Wwrite-strings -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -Winline -Wmissing-noreturn -Wformat-security -O2 -O2 -march=pentium3 -pipe -fPIC -Wall -Wundef -Wshadow -Wcast-align -Wwrite-strings -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -Winline -Wmissing-noreturn -Wformat-security -O2 -Wl,-O1 -Wl,--as-needed -Wl,-O1 -Wl,--as-needed -Wl,--export-dynamic -L./libdm -L./lib -L./daemons/dmeventd -Wl,-O1 -Wl,--as-needed -Wl,--export-dynamic -L../libdm -L../lib -L../daemons/dmeventd -L../libdm \ -o dmsetup dmsetup.o -ldevmapper -lm -ludev dmsetup.o: In function `_get_split_name': dmsetup.c:(.text+0x2b6): undefined reference to `dm_split_lvm_name'
My current upstream build gives just: gcc -fPIC -Wall -Wundef -Wshadow -Wcast-align -Wwrite-strings -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -Winline -Wmissing-noreturn -Wformat-security -Wredundant-decls -g -fno-omit-frame-pointer -Wl,--export-dynamic -L../libdm -L../lib -L../daemons/dmeventd -L../libdm \ -o dmsetup dmsetup.o -ldevmapper -lm -ludev - Are output lines getting conflated or are all those options really getting duplicated? Can you test moving the 'dmsetup.o' on the last line right to the end of it, after the -ludev ?
(an easy way to test is to keep the failed build directory around, cd into it and cut-and-paste the failing command - see that it fails the same way - then tweak it to see if you can find a 'right' version of it.)
l33tmmx: please repeat with MAKEOPTS=-j1 and post the build.log from that run, as I think some of the output in your log is mixed up.
Created attachment 240965 [details] build.log using -j1
(In reply to comment #7) > Created an attachment (id=240965) [details] > build.log using -j1 Can you please test 2.02.70-r1 that has some as-needed fixes, and if that solves it for you, I'll upstream that patch to agk.
(In reply to comment #8) > (In reply to comment #7) > > Created an attachment (id=240965) [details] [details] > > build.log using -j1 > Can you please test 2.02.70-r1 that has some as-needed fixes, and if that > solves it for you, I'll upstream that patch to agk. > Both lvm2-2.02.70-r1 and lvm2-2.02.72 still fail :(
Same here, unfortunately :( . lvm2-2.02.70-r1 fails regardless of which way I build it
(In reply to comment #10) > Same here, unfortunately :( . lvm2-2.02.70-r1 fails regardless of which way I > build it If you drop that --as-needed, does it start to compile?
(In reply to comment #11) > (In reply to comment #10) > > Same here, unfortunately :( . lvm2-2.02.70-r1 fails regardless of which way I > > build it > If you drop that --as-needed, does it start to compile? Still fails.
build.log without the --as-needed please?
Created attachment 241375 [details, diff] lvm2-locale-killer.patch Gah... should've first to build under LC_ALL=C :S This is locale issue.
Regression was caused by this commit: http://sources.redhat.com/git/gitweb.cgi?p=lvm2.git;a=commit;f=make.tmpl.in;h=1c78aa19ff8945362ad6635c5f064c8590853dd3
fun. merging into tree now. agk: want to merge that LC_ALL=C bit upstream too?
This was already fixed upstream.
