Patchwork gcc: arm: linux-eabi: fix handling of armv4 bx fixups when linking

login
register
mail settings
Submitter Mike Frysinger
Date April 19, 2013, 9:40 p.m.
Message ID <1366407647-1956-1-git-send-email-vapier@gentoo.org>
Download mbox | patch
Permalink /patch/238124/
State New
Headers show

Comments

Mike Frysinger - April 19, 2013, 9:40 p.m.
The bpabi.h header already sets up defines to automatically use the
--fix-v4bx flag with the assembler & linker as needed, and creates a
default assembly & linker spec which uses those.  Unfortunately, the
linux-eabi.h header clobbers the LINK_SPEC define and doesn't include
the v4bx define when setting up its own.  So while the assembler spec
is retained and works fine to generate the right relocs, building for
armv4 targets doesn't invoke the linker correctly so all the relocs
get processed as if we had an armv4t target.

You can see this with -dumpspecs when configuring gcc for an armv4
target and using --with-arch=armv4:
$ armv4l-unknown-linux-gnueabi-gcc -dumpspecs |& grep -B 1 fix-v4bx
*subtarget_extra_asm_spec:
.... %{mcpu=arm8|mcpu=arm810|mcpu=strongarm*|march=armv4|mcpu=fa526|mcpu=fa626:--fix-v4bx} ...

With this fix in place, we also get the link spec:
$ armv4l-unknown-linux-gnueabi-gcc -dumpspecs |& grep -B 1 fix-v4bx
*link:
...  %{mcpu=arm8|mcpu=arm810|mcpu=strongarm*|march=armv4|mcpu=fa526|mcpu=fa626:--fix-v4bx} ...

And all my hello world tests / glibc builds automatically turn the
bx insn into the 'mov pc, lr' insn and all is right in the world.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>

2013-04-19  Mike Frysinger  <vapier@gentoo.org>

	* config/arm/linux-eabi.h (LINK_SPEC): Add TARGET_FIX_V4BX_SPEC.
---
Note: This issue seems to exist since the code was first introduced.
      At least, I've tested gcc-4.5.x and gcc-4.8.x and they both fail,
      and the code looks the same in gcc-4.[467].x.  That means it's
      not technically a regression, so I guess policy dictates that it
      can't be merged into older branches?

 gcc/config/arm/linux-eabi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Richard Earnshaw - April 24, 2013, 3:45 p.m.
On 19/04/13 22:40, Mike Frysinger wrote:
> The bpabi.h header already sets up defines to automatically use the
> --fix-v4bx flag with the assembler & linker as needed, and creates a
> default assembly & linker spec which uses those.  Unfortunately, the
> linux-eabi.h header clobbers the LINK_SPEC define and doesn't include
> the v4bx define when setting up its own.  So while the assembler spec
> is retained and works fine to generate the right relocs, building for
> armv4 targets doesn't invoke the linker correctly so all the relocs
> get processed as if we had an armv4t target.
>
> You can see this with -dumpspecs when configuring gcc for an armv4
> target and using --with-arch=armv4:
> $ armv4l-unknown-linux-gnueabi-gcc -dumpspecs |& grep -B 1 fix-v4bx
> *subtarget_extra_asm_spec:
> .... %{mcpu=arm8|mcpu=arm810|mcpu=strongarm*|march=armv4|mcpu=fa526|mcpu=fa626:--fix-v4bx} ...
>
> With this fix in place, we also get the link spec:
> $ armv4l-unknown-linux-gnueabi-gcc -dumpspecs |& grep -B 1 fix-v4bx
> *link:
> ...  %{mcpu=arm8|mcpu=arm810|mcpu=strongarm*|march=armv4|mcpu=fa526|mcpu=fa626:--fix-v4bx} ...
>
> And all my hello world tests / glibc builds automatically turn the
> bx insn into the 'mov pc, lr' insn and all is right in the world.
>
> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
>
> 2013-04-19  Mike Frysinger  <vapier@gentoo.org>
>
> 	* config/arm/linux-eabi.h (LINK_SPEC): Add TARGET_FIX_V4BX_SPEC.
> ---
> Note: This issue seems to exist since the code was first introduced.
>        At least, I've tested gcc-4.5.x and gcc-4.8.x and they both fail,
>        and the code looks the same in gcc-4.[467].x.  That means it's
>        not technically a regression, so I guess policy dictates that it
>        can't be merged into older branches?
>
>   gcc/config/arm/linux-eabi.h | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/gcc/config/arm/linux-eabi.h b/gcc/config/arm/linux-eabi.h
> index 4a425c8..8b7ebb2 100644
> --- a/gcc/config/arm/linux-eabi.h
> +++ b/gcc/config/arm/linux-eabi.h
> @@ -80,7 +80,7 @@
>   /* At this point, bpabi.h will have clobbered LINK_SPEC.  We want to
>      use the GNU/Linux version, not the generic BPABI version.  */
>   #undef  LINK_SPEC
> -#define LINK_SPEC BE8_LINK_SPEC						\
> +#define LINK_SPEC TARGET_FIX_V4BX_SPEC BE8_LINK_SPEC			\
>     LINUX_OR_ANDROID_LD (LINUX_TARGET_LINK_SPEC,				\
>   		       LINUX_TARGET_LINK_SPEC " " ANDROID_LINK_SPEC)
>
>


Mike,

Thanks for the patch.  The concept is right, but I think we need a 
better approach to avoid problems like this cropping up again.

What I'd like to see is BPABI_LINK_SPEC in bpabi.h split into 
EABI_LINK_SPEC (and containing TARGET_FIX_V4BX_SPEC & BE8_LINK_SPEC) and 
then all EABI-based targets using EABI_LINK_SPEC as part of their rule set.

R.

Patch

diff --git a/gcc/config/arm/linux-eabi.h b/gcc/config/arm/linux-eabi.h
index 4a425c8..8b7ebb2 100644
--- a/gcc/config/arm/linux-eabi.h
+++ b/gcc/config/arm/linux-eabi.h
@@ -80,7 +80,7 @@ 
 /* At this point, bpabi.h will have clobbered LINK_SPEC.  We want to
    use the GNU/Linux version, not the generic BPABI version.  */
 #undef  LINK_SPEC
-#define LINK_SPEC BE8_LINK_SPEC						\
+#define LINK_SPEC TARGET_FIX_V4BX_SPEC BE8_LINK_SPEC			\
   LINUX_OR_ANDROID_LD (LINUX_TARGET_LINK_SPEC,				\
 		       LINUX_TARGET_LINK_SPEC " " ANDROID_LINK_SPEC)