aboutsummaryrefslogtreecommitdiff
path: root/archival/libarchive
AgeCommit message (Collapse)Author
2018-02-07bzip2: expose tuning knob for faster/smaller codeDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-06ar: hopefully fix out-of-bounds read in get_header_ar()Denys Vlasenko
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=882175 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-05bzip2: work around bad compiler optimizationDenys Vlasenko
gc-6.1.1 x86_64: function old new delta generateMTFValues 380 367 -13 gcc-4.3.1 386: function old new delta inner_loop - 41 +41 generateMTFValues 357 294 -63 ------------------------------------------------------------------------------ (add/remove: 1/0 grow/shrink: 0/1 up/down: 41/-63) Total: -22 bytes gcc-6.3.0 386: function old new delta inner_loop - 36 +36 generateMTFValues 363 250 -113 ------------------------------------------------------------------------------ (add/remove: 1/0 grow/shrink: 0/1 up/down: 36/-113) Total: -77 bytes The last case, gcc-6.3.0, runs almost 3 times faster after this change. Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-04gunzip: fix from gzip-1.3.12 for gzip file with all zero length codesDenys Vlasenko
Corresponding changelog from gzip-1.3.12 reads: """ 2006-12-20 Paul Eggert <eggert@cs.ucla.edu> * inflate.c (huft_build): Fix regression that caused gzip to refuse to uncompress null input (all zero length codes). Problem reported by Yiorgos Adamopoulos. This regression was caused by the security patch installed 2006-11-20, which in turn came from Debian, which in turn apparently came from Thomas Biege of SuSe. """ function old new delta huft_build 1176 1216 +40 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: move runningOrder[] back to stack - 256 bytes is not muchDenys Vlasenko
function old new delta mainSort 1119 1108 -11 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: move ->origPtr out of struct EState, make a few members smallerDenys Vlasenko
function old new delta BZ2_compressBlock 223 228 +5 BZ2_blockSort 85 88 +3 generateMTFValues 356 357 +1 handle_compress 355 349 -6 compressStream 538 531 -7 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 3/2 up/down: 9/-13) Total: -4 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: pass sorting params through EState* pointerDenys Vlasenko
function old new delta mainGtU 499 515 +16 sendMTFValues 2085 2094 +9 mainSort 1116 1119 +3 generateMTFValues 357 356 -1 fallbackSort 1719 1705 -14 mainQSort3 1163 1141 -22 BZ2_blockSort 118 85 -33 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 3/4 up/down: 28/-70) Total: -42 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03gzip2: small simplification in mainSimpleSort()Denys Vlasenko
function old new delta mainQSort3 1165 1163 -2 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: a few more locals converted to generic typesDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: eliminate one parameter to mainQSort3()Denys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: small simplification in mainSort()Denys Vlasenko
function old new delta mainSort 1127 1118 -9 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: convert some locals to unsigned'sDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: runningOrder[] values are always 0..255, make it uint8Denys Vlasenko
function old new delta mainSort 1171 1124 -47 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: remove redundant clearing of an alredy unset bitDenys Vlasenko
function old new delta mainSort 1178 1171 -7 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: eliminate write-only local numQSortedDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: make locals in mainSort() saner, convert one of them from uint16 to ↵Denys Vlasenko
unsigned function old new delta mainSort 1192 1178 -14 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: remove redundant loop termination check in mainSort()Denys Vlasenko
function old new delta mainSort 1202 1192 -10 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: reduce indentation, no code changesDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: reuse zPend processing codeDenys Vlasenko
function old new delta generateMTFValues 378 357 -21 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: shrink makeMaps_e()Denys Vlasenko
function old new delta generateMTFValues 378 368 -10 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: optimize zPend variable codeDenys Vlasenko
function old new delta generateMTFValues 433 378 -55 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: have two separate "store bit 0" and "store bit 1" functionsDenys Vlasenko
function old new delta sendMTFValues 2051 2085 +34 bsW1_0 - 33 +33 BZ2_compressBlock 225 223 -2 bsW1 52 - -52 ------------------------------------------------------------------------------ (add/remove: 1/1 grow/shrink: 1/1 up/down: 67/-54) Total: 13 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: ~1% speedup by special-casing "store 1 bit" functionDenys Vlasenko
function old new delta bsW1 - 52 +52 BZ2_compressBlock 230 225 -5 BZ2_blockSort 125 118 -7 sendMTFValues 2070 2051 -19 ------------------------------------------------------------------------------ (add/remove: 1/0 grow/shrink: 0/3 up/down: 52/-31) Total: 21 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: rewrite bit of code which depends on integer overflowDenys Vlasenko
function old new delta sendMTFValues 2093 2070 -23 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: delete write-only fave[] arrayDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bgip2: fewer specifically-sized [u]int32_t's locals in sendMTFValuesDenys Vlasenko
Generic ints/unsigneds are usually fine. Yes, really. function old new delta sendMTFValues 2100 2093 -7 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: code shrinkDenys Vlasenko
function old new delta sendMTFValues 2111 2100 -11 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-03bzip2: code shrinkDenys Vlasenko
function old new delta BZ2_compressBlock 225 230 +5 handle_compress 356 355 -1 bsW16 59 56 -3 bsW 64 61 -3 bsFinishWrite 37 32 -5 prepare_new_block 48 34 -14 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 1/5 up/down: 5/-26) Total: -21 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-02bzip2: code shrinkDenys Vlasenko
function old new delta bsW16 - 59 +59 sendMTFValues 2116 2111 -5 bsPutU16 36 - -36 bsPutU32 76 31 -45 ------------------------------------------------------------------------------ (add/remove: 1/1 grow/shrink: 0/2 up/down: 59/-86) Total: -27 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-01libbb: commonalize a bit of little-endian CRC32 table generation codeDenys Vlasenko
function old new delta global_crc32_new_table_le - 11 +11 crc32_new_table_le - 9 +9 inflate_unzip_internal 560 556 -4 flash_eraseall_main 823 819 -4 unpack_xz_stream 2403 2394 -9 lzop_main 121 112 -9 gzip_main 187 178 -9 ------------------------------------------------------------------------------ (add/remove: 2/0 grow/shrink: 0/5 up/down: 20/-35) Total: -15 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-02-01lzop: code shrinkDenys Vlasenko
function old new delta lzo_decompress 526 524 -2 lzo_compress 473 470 -3 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2018-01-27tar: accomodate non-terminated tar.chksum fields as seen from github.comDenys Vlasenko
function old new delta get_header_tar 1783 1696 -87 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-10-27unlzma: fix SEGV, closes 10436Denys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-10-22bunzip2: fix runCnt overflow from bug 10431Denys Vlasenko
This particular corrupted file can be dealth with by using "unsigned". If there will be cases where it genuinely overflows, there is a disabled code to deal with that too. function old new delta get_next_block 1678 1667 -11 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-10-05whitespace and comment format fixes, no code changesDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-10-05whitespace and comment format fixes, no code changesDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-08-22libarchive: fix build failure on NOMMU systemsRon Yorston
In the old code fd was an argument, now we need to get the file descriptor from the xstate structure. Signed-off-by: Ron Yorston <rmy@pobox.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-08-10libarchive: do not extract unsafe symlinks unless $EXTRACT_UNSAFE_SYMLINKS=1Denys Vlasenko
function old new delta unsafe_symlink_target - 147 +147 unzip_main 2711 2732 +21 copy_file 1657 1678 +21 tar_main 999 971 -28 data_extract_all 1038 984 -54 ------------------------------------------------------------------------------ (add/remove: 2/0 grow/shrink: 2/2 up/down: 189/-82) Total: 107 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-08-10rpm2cpio: handle LZMA compressed rpms. closes 10166Denys Vlasenko
function old new delta rpm2cpio_main 78 120 +42 setup_lzma_on_fd - 29 +29 fork_transformer_and_free - 28 +28 ... setup_unzip_on_fd 56 32 -24 ------------------------------------------------------------------------------ (add/remove: 2/0 grow/shrink: 3/5 up/down: 104/-67) Total: 37 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-08-07unxz: get_le32 macro is obviously wrongDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-07-24tar: postpone creation of symlinks with "suspicious" targets. Closes 8411Denys Vlasenko
function old new delta data_extract_all 968 1038 +70 tar_main 952 986 +34 scan_tree 258 262 +4 ------------------------------------------------------------------------------ (add/remove: 0/0 grow/shrink: 3/0 up/down: 108/0) Total: 108 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-07-15Move get_unaligned_le32() macros to platform.hDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-04-17Spelling fixes in comments, documentation, tests and examplesDenys Vlasenko
By klemens <ka7@github.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-02-17Fix FEATURE_{GZIP,BZIP2}_DECOMPRESS link errorKang-Che Sung
... when gzip is selected but not gunzip nor zcat, or when bzip2 is selected but not bunzip2 nor bzcat. This regression is introduced in b130f9f758b6404c6d0911a1c120937ae6ab47f8 ("Allow 'gzip -d' and 'bzip2 -d' without gunzip or bunzip2") Signed-off-by: Kang-Che Sung <explorer09@gmail.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-02-04Replace int -> uint to avoid signed integer overflowRostislav Skudnov
An example of such an error (should be compiled with DEBUG_SANITIZE): runtime error: left shift of 1 by 31 places cannot be represented in type 'int' Signed-off-by: Rostislav Skudnov <rostislav@tuxera.com> Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-01-26link: new appletDenys Vlasenko
coreutils grew itself a tiny simplistic alternative to ln: Usage: link FILE LINK Create hard LINK to FILE function old new delta link_main - 75 +75 packed_usage 31114 31131 +17 applet_names 2564 2569 +5 applet_main 1480 1484 +4 applet_install_loc 185 186 +1 ------------------------------------------------------------------------------ (add/remove: 1/0 grow/shrink: 4/0 up/down: 102/0) Total: 102 bytes Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-01-09unlzma: expand comments, no code changesDenys Vlasenko
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-01-09unlzma: fix erroneous "while" instead of "if". Closes 4682Denys Vlasenko
These parts of the code essentially check whether stepping back by rep0 goes negative or not. LZMA SDK from lzma1604.7z has the following in the corresponding places: ... = dic[dicPos - rep0 + (dicPos < rep0 ? dicBufSize : 0)] Clearly, not loop here. Technically, "while" here works: if condition is false (because pos underflowed), it iterates once, adds header.dict_size (a.k.a. dicBufSize), this makes pos positive but smaller than header.dict_size, and loop exits. Now we'll just check for negative result of subtraction, which is less code: function old new delta unpack_lzma_stream 2659 2641 -18 (I hope 2 Gbyte+ dictionaries won't be in use soon). Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-01-09unzip: optional support for xzDenys Vlasenko
function old new delta unzip_main 2476 2491 +15 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2017-01-09unzip: optional support for bzip2 and lzmaDenys Vlasenko
function old new delta unzip_main 2376 2476 +100 bbunpack 750 745 -5 Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>