Please mark dzip-2.9-r2.ebuild stable.
this is a regression... $ dzip test.txt -o test.dz *** glibc detected *** dzip: malloc(): memory corruption: 0x083aa3b8 *** ======= Backtrace: ========= /lib/libc.so.6[0xf7631fc4] /lib/libc.so.6[0xf7634896] /lib/libc.so.6(__libc_malloc+0x5a)[0xf763677a] dzip[0x8050ee2] dzip[0x8050f12] /lib/libz.so.1(deflateInit2_+0x1fe)[0xf771459e] /lib/libz.so.1(deflateInit_+0x52)[0xf77146c2] dzip[0x8049a98] dzip[0x8049d3f] dzip[0x8050c0f] dzip[0x8051736] /lib/libc.so.6(__libc_start_main+0xe5)[0xf75dca65] dzip[0x8049061] ======= Memory map: ======== 08048000-08053000 r-xp 00000000 fe:07 132453 /usr/games/bin/dzip 08053000-08054000 r--p 0000a000 fe:07 132453 /usr/games/bin/dzip 08054000-08055000 rw-p 0000b000 fe:07 132453 /usr/games/bin/dzip 08055000-08077000 rw-p 00000000 00:00 0 08392000-083cb000 rw-p 00000000 00:00 0 [heap] f7400000-f7421000 rw-p 00000000 00:00 0 f7421000-f7500000 ---p 00000000 00:00 0 f75c5000-f75c6000 rw-p 00000000 00:00 0 f75c6000-f7708000 r-xp 00000000 fe:07 465741 /lib/libc-2.10.1.so f7708000-f770a000 r--p 00142000 fe:07 465741 /lib/libc-2.10.1.so f770a000-f770b000 rw-p 00144000 fe:07 465741 /lib/libc-2.10.1.so f770b000-f770e000 rw-p 00000000 00:00 0 f770e000-f7720000 r-xp 00000000 fe:07 55963 /lib/libz.so.1.2.3 f7720000-f7721000 r--p 00011000 fe:07 55963 /lib/libz.so.1.2.3 f7721000-f7722000 rw-p 00012000 fe:07 55963 /lib/libz.so.1.2.3 f7722000-f7723000 rw-p 00000000 00:00 0 f7758000-f7764000 r-xp 00000000 fe:07 914816 /usr/lib/gcc/i686-pc-linux-gnu/4.3.4/libgcc_s.so.1 f7764000-f7765000 r--p 0000b000 fe:07 914816 /usr/lib/gcc/i686-pc-linux-gnu/4.3.4/libgcc_s.so.1 f7765000-f7766000 rw-p 0000c000 fe:07 914816 /usr/lib/gcc/i686-pc-linux-gnu/4.3.4/libgcc_s.so.1 f7766000-f7767000 rw-p 00000000 00:00 0 f7767000-f7783000 r-xp 00000000 fe:07 464836 /lib/ld-2.10.1.so f7783000-f7784000 r--p 0001c000 fe:07 464836 /lib/ld-2.10.1.so f7784000-f7785000 rw-p 0001d000 fe:07 464836 /lib/ld-2.10.1.so ffc31000-ffc47000 rw-p 00000000 00:00 0 [stack] ffffe000-fffff000 r-xp 00000000 00:00 0 [vdso] Aborted
works fine for me on stable x86.
Which Glibc version are you two using? I can reproduce the failure and I have Glibc 2.10 (for testing).
Created attachment 216738 [details, diff] fixing segfault I don't think it has anything to do with glibc, rather zlib version. Package had bundled version of zlib which defined: typedef voidpf (*alloc_func) OF((uInt size)); typedef void (*free_func) OF((voidpf address)); while on system zlib we have: typedef voidpf (*alloc_func) OF((voidpf opaque, uInt items, uInt size)); typedef void (*free_func) OF((voidpf opaque, voidpf address)); There two options: 1) Use default functions from zlib lib i.e. pass Z_NULL to both functions (see patch) 2) Rewrite Dzip_calloc function in conmain.c.
I masked it for removal.
I'll take it
Added -r3, also fixed crc32 on amd64. Will request x86/amd64 stabilization after 30 days period so we can kill the old broken versions. I also cleaned the ebuild up a bit and didn't use the games eclass any more. It's just an archiver, nothing really related to games (the only reason it probably landed there is that it was developed from a gaming-related website).
