[Slackbuilds-users] "make -j4" [was: linuxwacom slackbuild]

[Murat D. Kadirov]

19:37 Fri 08 Aug     , Jim Diamond написал:
> On Fri, Aug  8, 2008 at 12:10 (-0500), Robby Workman wrote:
> 
> > Jim Diamond wrote:
> 
> >>P.S. For everyone who makes slack packages...  Maybe this has been
> >>hashed out 100 times and you don't do it for some good reason, but if
> >>not, would you consider using "-j 4" or something like than on your
> >>"make" commands?  Even on a single core, single CPU machine this helps
> >>noticeably, and on a multi-core machine it makes a big difference.
> 
> 
> > No, that's not the best (IMHO) way to do it.  While "make -j4" might
> > be appropriate for you, "make -j8" might be appropriate for me, and
> > "make -j2" might be best for someone else.  As I see it, the best
> > thing to do is use plain "make" -- that way, I (and everyone else)
> > can set MAKEFLAGS in my shell environment.
> 
> > Example:
> 
> > MAKEFLAGS="-j8" make
> 
> Well, as you say, some people might like different choices.  However,
> I suggest that extremely few people will benefit with "-j 1", and so
> making the default choice even 2 would speed things up for (I suggest)
> virtually everyone.  The mileage of those people still running 486's
> with <= 32 Mbytes memory may vary :-)
> 
> One data point: my venerable Dell I8000 with its measly 192 MB of
> memory and sloth-like P3-700 compiles things much more quickly with "-j 4".
> 
> Cheers.
> 
> 				Jim

My example :) My laptop with Celeron 1.4Ghz and 256Mb RAM with "-j 2"
compile corrupted binary file djview3 (djvulibre package).

-- 
Murat D. Kadirov
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://lists.slackbuilds.org/pipermail/slackbuilds-users/attachments/20080809/10f7ab4a/attachment.sig>