[Buildroot] [PATCH 0/2 v3] arch: allow passing compile flags (branch yem/arc-atomic)

[Buildroot] [PATCH 0/2 v3] arch: allow passing compile flags (branch yem/arc-atomic)

[Yann E. MORIN]

Hello All!

This series makes it so that architectures can set compilation flags,
passed down via the wrapper.

This is then used by arc to enforce -matomic for those cores that need
it, namely arc7x0, as reported by Alexey:

    https://patchwork.ozlabs.org/patch/1087480/
    https://patchwork.ozlabs.org/patch/1087471/

Changes v2 -> v3:
  - fix 'y' -> 'yy' in condition

Changes v1 -> v2:
  - move condition that was previously in package/Makefile.in  (Alexey)


Regards,
Yann E. MORIN.


The following changes since commit ed3eef6bb49270de619e5a86a249fdd8724c40cd

  package/glib-networking: bump to version 2.61.1 (2019-07-14 16:11:45 +0200)


are available in the git repository at:

  git://git.buildroot.org/~ymorin/git/buildroot.git

for you to fetch changes up to 7fe54408b9c04d8d5799a167aded7c0b0157c572

  arch/arc: always needs -matomic with atomix extensions (2019-07-18 22:51:24 +0200)


----------------------------------------------------------------
Yann E. MORIN (2):
      toolchain: allow architectures to enforce compilation flags
      arch/arc: always needs -matomic with atomix extensions

 arch/arch.mk.arc               | 4 ++++
 package/Makefile.in            | 4 ----
 toolchain/toolchain-wrapper.mk | 1 +
 3 files changed, 5 insertions(+), 4 deletions(-)
 create mode 100644 arch/arch.mk.arc

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 561 099 427 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

[Buildroot] [PATCH 1/2 v3] toolchain: allow architectures to enforce compilation flags

[Yann E. MORIN]

Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Arnout Vandecappelle <arnout@mind.be>
Acked-by: Alexey Brodkin <abrodkin@synopsys.com>
---
 toolchain/toolchain-wrapper.mk | 1 +
 1 file changed, 1 insertion(+)

diff --git a/toolchain/toolchain-wrapper.mk b/toolchain/toolchain-wrapper.mk
index ca66fa7ba4..8f872506ec 100644
--- a/toolchain/toolchain-wrapper.mk
+++ b/toolchain/toolchain-wrapper.mk
@@ -17,6 +17,7 @@ TOOLCHAIN_WRAPPER_ARGS = $($(PKG)_TOOLCHAIN_WRAPPER_ARGS)
 TOOLCHAIN_WRAPPER_ARGS += -DBR_SYSROOT='"$(STAGING_SUBDIR)"'
 
 TOOLCHAIN_WRAPPER_OPTS = \
+	$(ARCH_TOOLCHAIN_WRAPPER_OPTS) \
 	$(call qstrip,$(BR2_SSP_OPTION)) \
 	$(call qstrip,$(BR2_TARGET_OPTIMIZATION))
 
-- 
2.20.1

[Buildroot] [PATCH 2/2 v3] arch/arc: always needs -matomic with atomix extensions

[Yann E. MORIN]

As reported by Alexey:
    https://patchwork.ozlabs.org/patch/1087480/
    https://patchwork.ozlabs.org/patch/1087471/

Reported-by: Alexey Brodkin <Alexey.Brodkin@synopsys.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Alexey Brodkin <Alexey.Brodkin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Arnout Vandecappelle <arnout@mind.be>
Acked-by: Alexey Brodkin <abrodkin@synopsys.com>

---
Changes v2 -> v3:
  - fix the condition test ('yy', not 'y')

Changes v1 -> v2:
  - move the condition out of package/Makefile.in  (Alexey)
---
 arch/arch.mk.arc    | 4 ++++
 package/Makefile.in | 4 ----
 2 files changed, 4 insertions(+), 4 deletions(-)
 create mode 100644 arch/arch.mk.arc

diff --git a/arch/arch.mk.arc b/arch/arch.mk.arc
new file mode 100644
index 0000000000..5fcffb7f4b
--- /dev/null
+++ b/arch/arch.mk.arc
@@ -0,0 +1,4 @@
+# -matomic is always required when the ARC core has the atomic extensions
+ifeq ($(BR2_arc)$(BR2_ARC_ATOMIC_EXT),yy)
+ARCH_TOOLCHAIN_WRAPPER_OPTS = -matomic
+endif
diff --git a/package/Makefile.in b/package/Makefile.in
index f05b9cc3aa..0a7899c852 100644
--- a/package/Makefile.in
+++ b/package/Makefile.in
@@ -112,10 +112,6 @@ ifeq ($(BR2_xtensa),y)
 TARGET_ABI += -mlongcalls -mauto-litpools
 endif
 
-ifeq ($(BR2_arc)$(BR2_ARC_ATOMIC_EXT),yy)
-TARGET_ABI += -matomic
-endif
-
 STAGING_SUBDIR = $(GNU_TARGET_NAME)/sysroot
 STAGING_DIR    = $(HOST_DIR)/$(STAGING_SUBDIR)
 
-- 
2.20.1

[Buildroot] [PATCH 0/2 v3] arch: allow passing compile flags (branch yem/arc-atomic)

[Thomas Petazzoni]

Hello,

On Thu, 18 Jul 2019 22:52:46 +0200
"Yann E. MORIN" <yann.morin.1998@free.fr> wrote:

> Yann E. MORIN (2):
>       toolchain: allow architectures to enforce compilation flags
>       arch/arc: always needs -matomic with atomix extensions

Thanks, I've applied the series. I extended the commit log of the
second patch, because it was really terse.

Following this, I believe it would make sense to move the two remaining
uses of TARGET_ABI in package/Makefile.in to the appropriate arch/
file, and drop TARGET_ABI entirely. Indeed, those ABI options are
really mandatory, if they don't get passed to the cross-compiler,
things are going to be clearly wrong.

Best regards,

Thomas
-- 
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

[Buildroot] [PATCH 0/2 v3] arch: allow passing compile flags (branch yem/arc-atomic)

[Thomas Petazzoni]

On Thu, 18 Jul 2019 22:57:40 +0200
Thomas Petazzoni <thomas.petazzoni@bootlin.com> wrote:

> On Thu, 18 Jul 2019 22:52:46 +0200
> "Yann E. MORIN" <yann.morin.1998@free.fr> wrote:
> 
> > Yann E. MORIN (2):
> >       toolchain: allow architectures to enforce compilation flags
> >       arch/arc: always needs -matomic with atomix extensions  
> 
> Thanks, I've applied the series. I extended the commit log of the
> second patch, because it was really terse.
> 
> Following this, I believe it would make sense to move the two remaining
> uses of TARGET_ABI in package/Makefile.in to the appropriate arch/
> file, and drop TARGET_ABI entirely. Indeed, those ABI options are
> really mandatory, if they don't get passed to the cross-compiler,
> things are going to be clearly wrong.

Also, in relation to this, I think we could probably revert commit
18855d79e2e86dc646182bd6f4861c491a05c624 one the TARGET_ABI flags are
encoded into the compiler wrapper.

Thomas
-- 
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com