chiark / gitweb /
~ianmdlvl / elogind.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: f697185) | patch
journal/compress: use LZ4_compress_continue()
authorEvangelos Foutras <evangelos@foutrelis.com>
Sat, 30 Aug 2014 07:13:43 +0000 (10:13 +0300)
committerZbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
Sat, 30 Aug 2014 21:41:15 +0000 (17:41 -0400)
commitb4232628f3d4b00c967310d56c0e95715c9d05cd
tree0080a4a98ca571d48cb8d7ad7a5f1605a8f1409btree | snapshot
parentf697185e5b45287b6a62592129e726d8a636d3ffcommit | diff
journal/compress: use LZ4_compress_continue()

We can't use LZ4_compress_limitedOutput_continue() because in the
worst-case scenario the compressed output can be slightly bigger than
the input block. This generally affects very few blocks and is no reason
to abort the compression process.

I ran into this when I noticed that Chromium core dumps weren't being
compressed. After switching to LZ4_compress_continue() a ~330MB Chromium
core dump gets compressed to ~17M.
src/journal/compress.c diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.