On Sat, Dec 4, 2010 at 2:09 AM, Raphael Geissert geiss...@debian.org wrote:
Hi Niels!
[OT but using the opportunity: congrats! :)]
Thanks :)
Niels Thykier wrote:
Interestingly enough this does not appear to be the current bottle
neck of lintian (or I am doing something wrong in my patch).
On Sat, Dec 4, 2010 at 1:57 AM, Russ Allbery r...@debian.org wrote:
Niels Thykier nthyk...@gmail.com writes:
Interestingly enough this does not appear to be the current bottle neck
of lintian (or I am doing something wrong in my patch).
It's probably a good change regardless, but sorting is
On Tue, Dec 7, 2010 at 10:02 PM, Niels Thykier nthyk...@gmail.com wrote:
Hmm... could be interesting to see if we could do something about
that. If time permits I will have a look.
~Niels
I guess I had time! Trivial testing on the eclipse-platform-data shows
that neither objdump-info[1]
Package: lintian
Version: 2.4.4
Hi
I examined the source of lintian (pulled from git) and noticed that
lintian could probably benefit from looking at common access patterns.
In the checks, there are 13 accesses to 'keys %{$info-index}'; 11 of
these are 'sort keys %{$info-index}'. This means
Interestingly enough this does not appear to be the current bottle
neck of lintian (or I am doing something wrong in my patch).
~Niels
$ dpkg --contents eclipse-platform-data_3.5.2-8_all.deb | wc -l
1480
lintian 2.4.3
$ for i in {1..5} ; do time lintian -EI --pedantic
Hi Niels!
[OT but using the opportunity: congrats! :)]
Niels Thykier wrote:
Interestingly enough this does not appear to be the current bottle
neck of lintian (or I am doing something wrong in my patch).
Thanks for the analysis and the suggestion (don't know if Russ is going to
beat me at
Niels Thykier nthyk...@gmail.com writes:
Interestingly enough this does not appear to be the current bottle neck
of lintian (or I am doing something wrong in my patch).
It's probably a good change regardless, but sorting is generally very
fast.
--
Russ Allbery (r...@debian.org)
7 matches
Mail list logo
