pkgsrc-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: pkgsrc scanning performance benchmarks



On 12/2/2016 01:31, Martin Husemann wrote:
On Thu, Dec 01, 2016 at 07:54:27PM -0600, John Marino wrote:
parallel.   For example, it splits the tree into 12 equal batches.  In
sequential order, it uses bmake to return the value of 10 variables.

Could you please name the 10 variables queried?

I don't think it matters. The same work is required for 1 or 100 and the name isn't important. But to answer your question:

.MAKE.EXPAND_VARIABLES=yes -VPKGVERSION -VPKGFILE:T -V_MAKE_JOBS:C/^-j//" -V_CBBH_MSGS -VTOOL_DEPENDS -VBUILD_DEPENDS -VDEPENDS" -VPKG_OPTIONS -VPKG_DISABLED_OPTIONS -VEMUL_PLATFORMS

The Port version has more variables than this, mainly because it has several more flavors of "DEPENDS".


I do not believe the pkgsrc framework is 28 times more complex than the
Ports Collection framework.  It's just much more inefficient.  I know
such statements rankle some pkgsrc devs, but numbers don't lie.

Indeed, interesting numbers. Rumours say most of it is spent in make's
string operations, but hard numbers (and reproducable ways to get them)
are good, thanks for sharing!

This is reproducible. If you proposed a fix, you could measure exactly the effect using the same machine with synth.





---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus


Home | Main Index | Thread Index | Old Index