[aur-general] beecrypt build issues
beecrypt fails to build if I just run makepkg having an issue ( see below ) weird thing is if I cd to src/beecrypt-4.2.1 and run ./configure --prefix=/usr --sysconfdir=/etc --without-java --disable-openmp && make and then I change back to the beecrypt directory and run makepkg then mkpkg works... I don't get it (since this is essentially what the pkgbuild does). I suspect it's an environmental issue but I don't know what. libtool: link: g++ -shared -nostdlib /usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.3/../../../../lib/crti.o /usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.3/crtbeginS.o .libs/aes.o .libs/base64.o .libs/beecrypt.o .libs/blockmode.o .libs/blockpad.o .libs/blowfish.o .libs/dhies.o .libs/dldp.o .libs/dlkp.o .libs/dlpk.o .libs/dlsvdp-dh.o .libs/dsa.o .libs/elgamal.o .libs/endianness.o .libs/entropy.o .libs/fips186.o .libs/hmac.o .libs/hmacmd5.o .libs/hmacsha1.o .libs/hmacsha224.o .libs/hmacsha256.o .libs/md4.o .libs/md5.o .libs/hmacsha384.o .libs/hmacsha512.o .libs/memchunk.o .libs/mp.o .libs/mpbarrett.o .libs/mpnumber.o .libs/mpprime.o .libs/mtprng.o .libs/pkcs1.o .libs/pkcs12.o .libs/ripemd128.o .libs/ripemd160.o .libs/ripemd256.o .libs/ripemd320.o .libs/rsa.o .libs/rsakp.o .libs/rsapk.o .libs/sha1.o .libs/sha224.o .libs/sha256.o .libs/sha384.o .libs/sha512.o .libs/sha2k32.o .libs/sha2k64.o .libs/timestamp.o .libs/cppglue.o .libs/blowfishopt.o .libs/mpopt.o .libs/sha1opt.o -lrt -lpthread -ldl -L/usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.3 -L/usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.3/../../../../lib -L/lib/../lib -L/usr/lib/../lib -L/usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.3/../../.. -lstdc++ -lm -lc -lgcc_s /usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.3/crtendS.o /usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.3/../../../../lib/crtn.o -march=core2 -Wl,--hash-style=gnu -Wl,--as-needed -Wl,-soname -Wl,libbeecrypt.so.7 -o .libs/libbeecrypt.so.7.0.0 .libs/mpopt.o: In function `mpzero': (.text+0x0): multiple definition of `mpzero' .libs/mp.o:mp.c:(.text+0x0): first defined here .libs/mpopt.o: In function `mpfill': (.text+0x10): multiple definition of `mpfill' .libs/mp.o:mp.c:(.text+0x20): first defined here .libs/mpopt.o: In function `mpeven': (.text+0x20): multiple definition of `mpeven' .libs/mp.o:mp.c:(.text+0x50): first defined here .libs/mpopt.o: In function `mpodd': (.text+0x30): multiple definition of `mpodd' .libs/mp.o:mp.c:(.text+0x40): first defined here .libs/mpopt.o: In function `mpaddw': (.text+0x40): multiple definition of `mpaddw' .libs/mp.o:mp.c:(.text+0xb80): first defined here .libs/mpopt.o: In function `mpsubw': (.text+0x70): multiple definition of `mpsubw' .libs/mp.o:mp.c:(.text+0xcc0): first defined here .libs/mpopt.o: In function `mpadd': (.text+0xa0): multiple definition of `mpadd' .libs/mp.o:mp.c:(.text+0xbe0): first defined here .libs/mpopt.o: In function `mpsub': (.text+0xd0): multiple definition of `mpsub' .libs/mp.o:mp.c:(.text+0xd20): first defined here .libs/mpopt.o: In function `mpdivtwo': (.text+0x100): multiple definition of `mpdivtwo' .libs/mp.o:mp.c:(.text+0x11b0): first defined here .libs/mpopt.o: In function `mpmultwo': (.text+0x120): multiple definition of `mpmultwo' .libs/mp.o:mp.c:(.text+0x1260): first defined here .libs/mpopt.o: In function `mpsetmul': (.text+0x140): multiple definition of `mpsetmul' .libs/mp.o:mp.c:(.text+0xe40): first defined here .libs/mpopt.o: In function `mpaddmul': (.text+0x170): multiple definition of `mpaddmul' .libs/mp.o:mp.c:(.text+0xef0): first defined here .libs/mpopt.o: In function `mpaddsqrtrc': (.text+0x1b0): multiple definition of `mpaddsqrtrc' .libs/mp.o:mp.c:(.text+0x10c0): first defined here collect2: ld returned 1 exit status make[2]: *** [libbeecrypt.la] Error 1 make[2]: Leaving directory `/var/abs/local/beecrypt/src/beecrypt-4.2.1' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/var/abs/local/beecrypt/src/beecrypt-4.2.1' make: *** [all] Error 2 ==> ERROR: Build Failed. Aborting... -- Caleb Cushing http://xenoterracide.blogspot.com
On 04/15/2010 07:20 PM, Caleb Cushing wrote:
beecrypt fails to build if I just run makepkg
having an issue ( see below ) weird thing is if I cd to src/beecrypt-4.2.1 and run ./configure --prefix=/usr --sysconfdir=/etc --without-java --disable-openmp&& make and then I change back to the beecrypt directory and run makepkg then mkpkg works... I don't get it (since this is essentially what the pkgbuild does). I suspect it's an environmental issue but I don't know what.
<snip> add before ./configure: export LDFLAGS="${LDFLAGS//-Wl,--as-needed}" -- Ionut
On Thu, Apr 15, 2010 at 12:48 PM, Ionut Biru <biru.ionut@gmail.com> wrote:
add before ./configure: export LDFLAGS="${LDFLAGS//-Wl,--as-needed}"
tried changing build to the following but it still bails out with the same error... but that gave me a new idea... hmm... it's my FLAGS... CFLAGS="-march=core2 -O2 -pipe" CXXFLAGS="-march=core2 -O2 -pipe" dunno why it doesn't like core2 optimizations but it seems to bail for this package when using those... reverting them to CFLAGS="-march=x86-64 -mtune=generic -O2 -pipe" CXXFLAGS="-march=x86-64 -mtune=generic -O2 -pipe" works... problem solved... PEBCAK? -- Caleb Cushing http://xenoterracide.blogspot.com
On Thu, Apr 15, 2010 at 12:19 PM, Caleb Cushing <xenoterracide@gmail.com> wrote:
dunno why it doesn't like core2 optimizations but it seems to bail for this package when using those... reverting them to
CFLAGS="-march=x86-64 -mtune=generic -O2 -pipe" CXXFLAGS="-march=x86-64 -mtune=generic -O2 -pipe"
works...
problem solved... PEBCAK? -- Caleb Cushing
latest versions of gcc support the -march=native flag. Does that work?
On Thu, Apr 15, 2010 at 2:57 PM, Gary Wright <wriggary@gmail.com> wrote:
latest versions of gcc support the -march=native flag. Does that work?
That works -- Caleb Cushing http://xenoterracide.blogspot.com
On Fri, Apr 16, 2010 at 11:06 AM, Caleb Cushing <xenoterracide@gmail.com> wrote:
That works
I don't know much about -march=native so I wonder if it it's supposed to basically detect and make it compile the same way as -march=core2 on a core2? I'm thinking that maybe you had me do that because you think that I set my march wrong? -- Caleb Cushing http://xenoterracide.blogspot.com
On Fri, Apr 16, 2010 at 11:11 AM, Caleb Cushing <xenoterracide@gmail.com> wrote:
On Fri, Apr 16, 2010 at 11:06 AM, Caleb Cushing <xenoterracide@gmail.com> wrote:
That works
I don't know much about -march=native so I wonder if it it's supposed to basically detect and make it compile the same way as -march=core2 on a core2? I'm thinking that maybe you had me do that because you think that I set my march wrong?
-- Caleb Cushing
No, I didn't think you set your -march wrong, When I last reinstalled, I looked up the safe cflags wiki page for my own use, and it redirects to a line in the makepkg.conf wiki page that states: As of version 4.3.0, the gcc compiler offers the -march=native switch that enables CPU auto-detection and automatically selects optimizations supported by the local machine at gcc runtime. To use it, just modify the Arch Linux default settings by changing the CFLAGS line as follows: etc..... But, while going over to gentoo's safe cflags wiki entry, I did notice that it does list -march=core2 as a valid march entry. Perhaps you've stumbled into a bug?
On Fri, Apr 16, 2010 at 2:23 PM, Gary Wright <wriggary@gmail.com> wrote:
But, while going over to gentoo's safe cflags wiki entry, I did notice that it does list -march=core2 as a valid march entry. Perhaps you've stumbled into a bug?
Yeah I was a gentoo user for 5+ years. I know I had this machine set on core2 for at least a little bit on gentoo (before I left). But I would guess that this library might just be sensitive to that optimization. I suppose I could report it to beecrypt... but I don't think it's that big of a deal... honestly am only using it because I need to run oracle for school. -- Caleb Cushing http://xenoterracide.blogspot.com
participants (3)
-
Caleb Cushing
-
Gary Wright
-
Ionut Biru