| To: | Mark Goodwin <mgoodwin@xxxxxxxxxx> |
|---|---|
| Subject: | Re: [pcp] pcp update: more make -j support |
| From: | "Frank Ch. Eigler" <fche@xxxxxxxxxx> |
| Date: | Fri, 15 May 2015 08:12:18 -0400 |
| Cc: | pcp developers <pcp@xxxxxxxxxxx> |
| Delivered-to: | pcp@xxxxxxxxxxx |
| In-reply-to: | <555589D1.2090100@xxxxxxxxxx> |
| References: | <20150515010809.GC19346@xxxxxxxxxx> <555589D1.2090100@xxxxxxxxxx> |
| User-agent: | Mutt/1.4.2.2i |
Hi, Mark -
Thanks for taking a look!
> [...]
> Hi Frank, on my 4-core RHEL7.1 box, your fche/make-j tree builds :
> real 8m16.531s
> [...]
> Compared to current master branch :
> real 10m9.608s
> [...]
Yeah; the %{_smp_mflags} underestimates the available parallelism; the
load average here rarely gets close to the number-of-cpus. 2 x #cpus
that makes things faster on my 4/8-cpu boxes.
> So the changes look good - reviewing the code the need for the '+'
> prefix on certain rules wasn't obvious (I know about '@' to suppress
> echo but not '+'). [...]
The '+' seems to be a hint to gnu-make that nested $(MAKE)s should be
specially treated; see [1].
[1] https://www.gnu.org/software/make/manual/html_node/MAKE-Variable.html
- FChE
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Mac OS X builds, Nathan Scott |
|---|---|
| Next by Date: | Re: PMIE : Using raw value for a counter, Frank Ch. Eigler |
| Previous by Thread: | Re: [pcp] pcp update: more make -j support, Mark Goodwin |
| Next by Thread: | [Bug 1111] New: --container=NO_SUCH_CONTAINER succeeds, bugzilla-daemon |
| Indexes: | [Date] [Thread] [Top] [All Lists] |