|
From: | Antonio Diaz Diaz |
Subject: | Re: [Lzip-bug] bug report: 'plzip -vf9' fails to compress files over a certain size |
Date: | Sat, 16 Dec 2017 01:21:49 +0100 |
User-agent: | Mozilla/5.0 (X11; U; Linux i586; en-US; rv:1.9.1.19) Gecko/20110420 SeaMonkey/2.0.14 |
Hi Erik, Erik Jan Tromp wrote:
I expect to release new versions of lzlib and plzip in a couple days. Stay tuned.Looking forward to it!
I announced them just hours ago: http://lists.nongnu.org/archive/html/lzip-bug/2017-12/msg00011.html http://lists.nongnu.org/archive/html/lzip-bug/2017-12/msg00012.html
Continuing with some custom scripts related to slackware package conversion (nothing official, just personal stuff), plzip 1.6 with the latest patch (I think of it as the '3GB patch') was throwing errors on my i7 about running out of memory. Oddly enough running identical tests on an identical data set would sometimes succeed, sometimes fail, or sometimes say it succeeded but silently fail. Very weird. Reverting to unpatched plzip 1.6& restoring the thread limiting loop in my scripts gave me predictable/desirable results. Essentially, I *think* the patch originally worked by accident on my somewhat odd hardware/software combination. Only by having a larger (very time consuming) data set did I spot this.
It is possible that the 3 GiB limit is too high for your machine[1]. This is why I first tried the more conservative 2 GiB limit.
[1] http://en.wikipedia.org/wiki/3_GB_barrierIt specially worries me the "sometimes say it succeeded but silently fail". May you describe what happened? Does plzip exit with zero status but leaves the file uncompressed, for example?
Maybe you could try the new plzip-1.7-rc1 and reduce progressively the 0xBFFFFFFFLL limit in line 714 of main.cc until it always works? Thanks.
Best regards, Antonio.
[Prev in Thread] | Current Thread | [Next in Thread] |