[Buildroot] [PATCH] uclibc: disable parallel build for uClibc 0.9.31

[Buildroot] [PATCH] uclibc: disable parallel build for uClibc 0.9.31

[Alexey Brodkin]

This is only required for uClibc 0.9.31 and this veriosn is only used by
AVR32 so could be safely removed together with AVR32.

Even 0.9.32 builds well in parallel (at least on x86) so IMHO there's no need
in >= 0.9.33, just disable for a particular broken version.

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>

Cc: Anton Kolesov <akolesov@synopsys.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Gustavo Zacarias <gustavo@zacarias.com.ar>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
---
 package/uclibc/uclibc.mk | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/package/uclibc/uclibc.mk b/package/uclibc/uclibc.mk
index 28a8a08..af3a4d4 100644
--- a/package/uclibc/uclibc.mk
+++ b/package/uclibc/uclibc.mk
@@ -493,8 +493,20 @@ define UCLIBC_BUILD_TEST_SUITE
 endef
 endif
 
+# In uClibc 0.9.31 parallel building is broken so we have to disable it
+# Fortunately uClibc 0.9.31 is only used by AVR32 and in its turn AVR32 is
+# about to be removed from buildroot.
+#
+# So as soon as AVR32 is removed please revert this patch so instead of
+# MAKE_UCLIBC normal "MAKE" is used in UCLIBC_BUILD_CMDS
+ifeq ($(BR2_UCLIBC_VERSION_0_9_31),y)
+	MAKE_UCLIBC = $(MAKE1)
+else
+	MAKE_UCLIBC = $(MAKE)
+endif
+
 define UCLIBC_BUILD_CMDS
-	$(MAKE) -C $(@D) \
+	$(MAKE_UCLIBC) -C $(@D) \
 		$(UCLIBC_MAKE_FLAGS) \
 		PREFIX= \
 		DEVEL_PREFIX=/ \
-- 
1.9.3

[Buildroot] [PATCH] uclibc: disable parallel build for uClibc 0.9.31

[Thomas Petazzoni]

Dear Alexey Brodkin,

On Mon, 30 Jun 2014 18:30:20 +0400, Alexey Brodkin wrote:

> +# In uClibc 0.9.31 parallel building is broken so we have to disable it
> +# Fortunately uClibc 0.9.31 is only used by AVR32 and in its turn AVR32 is
> +# about to be removed from buildroot.
> +#
> +# So as soon as AVR32 is removed please revert this patch so instead of
> +# MAKE_UCLIBC normal "MAKE" is used in UCLIBC_BUILD_CMDS
> +ifeq ($(BR2_UCLIBC_VERSION_0_9_31),y)
> +	MAKE_UCLIBC = $(MAKE1)
> +else
> +	MAKE_UCLIBC = $(MAKE)
> +endif

Sorry to nitpick, but I'd prefer to use UCLIBC_MAKE as the variable
name. We use <PKG>_<SOMETHING> for all variables, so it would make more
sense to do the same.

>  define UCLIBC_BUILD_CMDS
> -	$(MAKE) -C $(@D) \
> +	$(MAKE_UCLIBC) -C $(@D) \

Your original patch changing MAKE1 -> MAKE changed two places, and
you're changing only one place here. To be sure, we should probably
change back both places, no?

Thanks,

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

[Buildroot] [PATCH] uclibc: disable parallel build for uClibc 0.9.31

[Alexey Brodkin]

Hi Thomas,

On Mon, 2014-06-30 at 16:42 +0200, Thomas Petazzoni wrote:
> Sorry to nitpick, but I'd prefer to use UCLIBC_MAKE as the variable
> name. We use <PKG>_<SOMETHING> for all variables, so it would make more
> sense to do the same.

Ok I didn't think about this concept - just used whatever first appeared
in my mind. So will use UCLIBC_MAKE - makes sense.

> >  define UCLIBC_BUILD_CMDS
> > -	$(MAKE) -C $(@D) \
> > +	$(MAKE_UCLIBC) -C $(@D) \
> 
> Your original patch changing MAKE1 -> MAKE changed two places, and
> you're changing only one place here. To be sure, we should probably
> change back both places, no?

I do see that host utilities were built perfectly fine in parallel mode
then why do more changes than needed?

-Alexey

[Buildroot] [PATCH] uclibc: disable parallel build for uClibc 0.9.31

[Thomas Petazzoni]

Dear Alexey Brodkin,

On Mon, 30 Jun 2014 16:01:19 +0000, Alexey Brodkin wrote:

> On Mon, 2014-06-30 at 16:42 +0200, Thomas Petazzoni wrote:
> > Sorry to nitpick, but I'd prefer to use UCLIBC_MAKE as the variable
> > name. We use <PKG>_<SOMETHING> for all variables, so it would make more
> > sense to do the same.
> 
> Ok I didn't think about this concept - just used whatever first appeared
> in my mind. So will use UCLIBC_MAKE - makes sense.
> 
> > >  define UCLIBC_BUILD_CMDS
> > > -	$(MAKE) -C $(@D) \
> > > +	$(MAKE_UCLIBC) -C $(@D) \
> > 
> > Your original patch changing MAKE1 -> MAKE changed two places, and
> > you're changing only one place here. To be sure, we should probably
> > change back both places, no?
> 
> I do see that host utilities were built perfectly fine in parallel mode
> then why do more changes than needed?

Ok, fine with me.

Thanks,

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com