On Wed, Apr 13, 2016 at 8:54 AM, Anna Henningsen wrote:
> Note how the first lines of your attached strace outputs are different;
> one time, you execute /usr/bin/xz, and one time /usr/local/bin/xz is run.
Gah! So obvious!
Sorry for wasting your time. I will now hang my head in shame.
Matt
On 12.04.2016 22:47, Matt Parlane wrote:
> On Tue, Apr 12, 2016 at 9:51 PM, Lasse Collin
> wrote:
>> If something like SELinux blocks reading such files, then the core
>> count detection won't work. Compare the strace output in the two
>> situations:
>>
>> strace -o /home/foo/strace.txt xz -T
On Tue, Apr 12, 2016 at 9:51 PM, Lasse Collin wrote:
> Check if the environment variables XZ_DEFAULTS and XZ_OPT differ
> between the two situations. I guess it's unlikely, but it's easy to
> check, so I mentioned this first.
I'm not seeing any difference in those vars between the two environmen
On 2016-04-12 Matt Parlane wrote:
> I have compiled xz from source (rev ac398c3), and I am trying to use
> multithread compression by using -T 0 in the command line.
>
> What I am finding is that when run straight from the command line, it
> uses multiple cores as expected. However, when running v