alsa makefiles don't work with -jN where N > 1

alsa makefiles don't work with -jN where N > 1

[Lee Revell]

Subject says it all.  Here is the relevant section from the GNU make documentation:

`warning: jobserver unavailable: using -j1. Add `+' to parent make rule.'
        In order for make processes to communicate, the parent will pass
        information to the child. Since this could result in problems if
        the child process isn't actually a make, the parent will only do
        this if it thinks the child is a make. The parent uses the
        normal algorithms to determine this (see section 5.6.1 How the
        MAKEVariable Works). If the makefile is constructed such that
        the parent doesn't know the child is a make process, then the
        child will receive only part of the information necessary. In
        this case, the child will generate this warning message and
        proceed with its build in a sequential manner.

I believe this is because some of the rules in the makefile aren't make
commands, but rm -f's.  This always seemed dangerous and sloppy to me -
why does make need to rm -f anything?  Isn't this the whole point of
makefiles, so you don't have to rm -f your tree and rebuild every time
you change something? 

Lee



-------------------------------------------------------
SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank Media
100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for only $33
Save 50% off Retail on Ink & Toner - Free Shipping and Free Gift.
http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285

Re: alsa makefiles don't work with -jN where N > 1

[Takashi Iwai]

[-- Attachment #1: Type: text/plain, Size: 1302 bytes --]

At Tue, 24 Aug 2004 15:44:40 -0400,
Lee Revell wrote:
> 
> Subject says it all.  Here is the relevant section from the GNU make documentation:
> 
> `warning: jobserver unavailable: using -j1. Add `+' to parent make rule.'
>         In order for make processes to communicate, the parent will pass
>         information to the child. Since this could result in problems if
>         the child process isn't actually a make, the parent will only do
>         this if it thinks the child is a make. The parent uses the
>         normal algorithms to determine this (see section 5.6.1 How the
>         MAKEVariable Works). If the makefile is constructed such that
>         the parent doesn't know the child is a make process, then the
>         child will receive only part of the information necessary. In
>         this case, the child will generate this warning message and
>         proceed with its build in a sequential manner.
> 
> I believe this is because some of the rules in the makefile aren't make
> commands, but rm -f's.  This always seemed dangerous and sloppy to me -
> why does make need to rm -f anything?  Isn't this the whole point of
> makefiles, so you don't have to rm -f your tree and rebuild every time
> you change something? 

The attached oneliner should fix this.


Takashi

[-- Attachment #2: Type: text/plain, Size: 515 bytes --]

Index: alsa-driver/Makefile
===================================================================
RCS file: /suse/tiwai/cvs/alsa/alsa-driver/Makefile,v
retrieving revision 1.73
diff -u -r1.73 Makefile
--- alsa-driver/Makefile	3 Jun 2004 17:04:36 -0000	1.73
+++ alsa-driver/Makefile	25 Aug 2004 13:56:14 -0000
@@ -98,7 +98,7 @@
 all-deps: toplevel.config.in acinclude.m4 include/config1.h.in
 
 include/sndversions.h:
-	make dep
+	$(MAKE) dep
 
 .PHONY: compile
 compile: include/sound/version.h include/sndversions.h