* [Buildroot] [PATCH] poco: avoid build failures on multicore hosts
@ 2017-01-20 18:18 Carlos Santos
2017-01-20 20:19 ` Peter Korsgaard
0 siblings, 1 reply; 2+ messages in thread
From: Carlos Santos @ 2017-01-20 18:18 UTC (permalink / raw)
To: buildroot
Parallel build still fails on heavilly multicore machines (e.g. -j25)
and hacks likecommit 32f4957b153bdabe7af60d529942aca7d1a4783d do not
seem to be effective.
Let's simply use MAKE1 for the build step, instead.
Fixes:
http://autobuild.buildroot.net/results/388/38834ad3013fe79e5e4f75997133f1bd827be6dc
Signed-off-by: Carlos Santos <casantos@datacom.ind.br>
---
Extra information: here at DATACOM we have a build farm with several
multicore blades that usually build with "make -j25". With this setup we
have been observing compilation failures each two days.
Change-Id: If80cd0b01ceb8b0aee886ea96a266fa864fa8eb0
---
package/poco/poco.mk | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/package/poco/poco.mk b/package/poco/poco.mk
index 79452c0..54ec701 100644
--- a/package/poco/poco.mk
+++ b/package/poco/poco.mk
@@ -57,8 +57,9 @@ define POCO_CONFIGURE_CMDS
--no-samples)
endef
+# Use $(MAKE1) to avoid failures on heavilly parallel machines (e.g. -j25)
define POCO_BUILD_CMDS
- $(MAKE) POCO_TARGET_OSARCH=$(ARCH) CROSS_COMPILE=$(TARGET_CROSS) \
+ $(MAKE1) POCO_TARGET_OSARCH=$(ARCH) CROSS_COMPILE=$(TARGET_CROSS) \
MYSQL_LIBDIR=$(STAGING_DIR)/usr/lib/mysql \
MYSQL_INCDIR=$(STAGING_DIR)/usr/include/mysql \
DEFAULT_TARGET=$(POCO_MAKE_TARGET) -C $(@D)
--
2.7.4
^ permalink raw reply related [flat|nested] 2+ messages in thread
* [Buildroot] [PATCH] poco: avoid build failures on multicore hosts
2017-01-20 18:18 [Buildroot] [PATCH] poco: avoid build failures on multicore hosts Carlos Santos
@ 2017-01-20 20:19 ` Peter Korsgaard
0 siblings, 0 replies; 2+ messages in thread
From: Peter Korsgaard @ 2017-01-20 20:19 UTC (permalink / raw)
To: buildroot
>>>>> "Carlos" == Carlos Santos <casantos@datacom.ind.br> writes:
> Parallel build still fails on heavilly multicore machines (e.g. -j25)
> and hacks likecommit 32f4957b153bdabe7af60d529942aca7d1a4783d do not
> seem to be effective.
> Let's simply use MAKE1 for the build step, instead.
> Fixes:
> http://autobuild.buildroot.net/results/388/38834ad3013fe79e5e4f75997133f1bd827be6dc
> Signed-off-by: Carlos Santos <casantos@datacom.ind.br>
> ---
> Extra information: here at DATACOM we have a build farm with several
> multicore blades that usually build with "make -j25". With this setup we
> have been observing compilation failures each two days.
> Change-Id: If80cd0b01ceb8b0aee886ea96a266fa864fa8eb0
Committed, thanks.
--
Bye, Peter Korsgaard
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2017-01-20 20:19 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-01-20 18:18 [Buildroot] [PATCH] poco: avoid build failures on multicore hosts Carlos Santos
2017-01-20 20:19 ` Peter Korsgaard
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox