diff mbox

[PATCHv2,04/21] toolchain: move helpers.mk into toolchain-external/

Message ID 1381069171-29748-5-git-send-email-thomas.petazzoni@free-electrons.com
State Superseded
Headers show

Commit Message

Thomas Petazzoni Oct. 6, 2013, 2:19 p.m. UTC
The helpers.mk file is no longer used by the Crosstool-NG backend, so
it is moved into the toolchain/toolchain-external/ directory.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
 toolchain/toolchain-buildroot.mk              | 2 --
 toolchain/toolchain-external.mk               | 2 +-
 toolchain/{ => toolchain-external}/helpers.mk | 0
 3 files changed, 1 insertion(+), 3 deletions(-)
 rename toolchain/{ => toolchain-external}/helpers.mk (100%)

diff --git a/toolchain/helpers.mk b/toolchain/toolchain-external/helpers.mk
similarity index 100%
rename from toolchain/helpers.mk
rename to toolchain/toolchain-external/helpers.mk

Comments

Peter Korsgaard Oct. 6, 2013, 7:31 p.m. UTC | #1
>>>>> "Thomas" == Thomas Petazzoni <thomas.petazzoni@free-electrons.com> writes:

 Thomas> The helpers.mk file is no longer used by the Crosstool-NG backend, so
 Thomas> it is moved into the toolchain/toolchain-external/ directory.

 Thomas> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
 Thomas> ---
 Thomas>  toolchain/toolchain-buildroot.mk              | 2 --
 Thomas>  toolchain/toolchain-external.mk               | 2 +-
 Thomas>  toolchain/{ => toolchain-external}/helpers.mk | 0
 Thomas>  3 files changed, 1 insertion(+), 3 deletions(-)
 Thomas>  rename toolchain/{ => toolchain-external}/helpers.mk (100%)

 Thomas> diff --git a/toolchain/toolchain-buildroot.mk b/toolchain/toolchain-buildroot.mk
 Thomas> index 3a05800..ff89125 100644
 Thomas> --- a/toolchain/toolchain-buildroot.mk
 Thomas> +++ b/toolchain/toolchain-buildroot.mk
 Thomas> @@ -2,8 +2,6 @@
 Thomas>  # build of binutils, uClibc, kernel headers and all the intermediate
 Thomas>  # gcc steps.
 
 Thomas> -include toolchain/helpers.mk
 Thomas> -

That doesn't seem right as we have:

git grep copy_toolchain package/
package/glibc/glibc.mk:         $(call copy_toolchain_lib_root,$(STAGING_DIR)/,,

And indeed:

>>> glibc 2.17-svnr22064 Installing to target
for libs in ld*.so libc.so libcrypt.so libdl.so libgcc_s.so libm.so libnsl.so libpthread.so libresolv.so librt.so libutil.so libnss_files.so libnss_dns.so; do  ; done
/bin/sh: -c: line 0: syntax error near unexpected token `;'
/bin/sh: -c: line 0: `for libs in ld*.so libc.so libcrypt.so libdl.so libgcc_s.so libm.so libnsl.so libpthread.so libresolv.so librt.so libutil.so libnss_files.so libnss_dns.so; do  ; done'
make: *** [/home/peko/source/buildroot/output/build/glibc-2.17-svnr22064/.stamp_target_installed] Error 1
Thomas Petazzoni Oct. 6, 2013, 9:26 p.m. UTC | #2
Dear Peter Korsgaard,

On Sun, 06 Oct 2013 21:31:46 +0200, Peter Korsgaard wrote:

>  Thomas> -include toolchain/helpers.mk
>  Thomas> -
> 
> That doesn't seem right as we have:
> 
> git grep copy_toolchain package/
> package/glibc/glibc.mk:         $(call copy_toolchain_lib_root,$(STAGING_DIR)/,,

Oh, I'm impressed how well you know the code, even code that I have
written. We really need some kind of "test suite" to check patches or
something like that: a patch set such as this toolchain-related patch
set is really a pain to test in all the various possible conditions. I
did test the internal backend, but only with uClibc toolchain.

I'll rework my patch series.

Thanks!

Thomas
Peter Korsgaard Oct. 6, 2013, 9:40 p.m. UTC | #3
>>>>> "Thomas" == Thomas Petazzoni <thomas.petazzoni@free-electrons.com> writes:

Hi,

 >> git grep copy_toolchain package/
 >> package/glibc/glibc.mk:         $(call copy_toolchain_lib_root,$(STAGING_DIR)/,,

 Thomas> Oh, I'm impressed how well you know the code, even code that I have
 Thomas> written.

git grep is your friend ;)

 Thomas> We really need some kind of "test suite" to check patches or
 Thomas> something like that: a patch set such as this toolchain-related patch
 Thomas> set is really a pain to test in all the various possible conditions. I
 Thomas> did test the internal backend, but only with uClibc toolchain.

Yeah, but I'm afraid the only real way of testing things is to build the
various affected variants (internal uclibc/glibc + external).

 Thomas> I'll rework my patch series.

Great, thanks!
diff mbox

Patch

diff --git a/toolchain/toolchain-buildroot.mk b/toolchain/toolchain-buildroot.mk
index 3a05800..ff89125 100644
--- a/toolchain/toolchain-buildroot.mk
+++ b/toolchain/toolchain-buildroot.mk
@@ -2,8 +2,6 @@ 
 # build of binutils, uClibc, kernel headers and all the intermediate
 # gcc steps.
 
-include toolchain/helpers.mk
-
 BUILDROOT_LIBC = $(call qstrip,$(BR2_TOOLCHAIN_BUILDROOT_LIBC))
 
 toolchain-buildroot: host-gcc-final
diff --git a/toolchain/toolchain-external.mk b/toolchain/toolchain-external.mk
index 2f43db7..2644689 100644
--- a/toolchain/toolchain-external.mk
+++ b/toolchain/toolchain-external.mk
@@ -1,4 +1,4 @@ 
 # Required includes for the external toolchain backend
 
-include toolchain/helpers.mk
+include toolchain/toolchain-external/helpers.mk
 include toolchain/toolchain-external/ext-tool.mk