Fix the tools and processor build for Android

Fix the tools and processor build for Android

This small patch allows the build of the tools and processor
when targetting Android with the Automake/Autconf build.

Not that these necessarily work correctly at the moment,
but there is no need for --disable-tools --disable-processor
now when using --host=arm-linux-androideabi or
--host=i686-linux-android.

+ Modify android/run-checks.sh to build all binaries with
  the Automake build.

+ Tiny fix for --abi=x86 in android/run-checks.sh
Committed: https://code.google.com/p/google-breakpad/source/detail?r=1018

[digit]

I tested that this didn't change the results of the Linux/x86_64 and Linux/x86
test suites.

'make check' doesn't work when targeting Android, yet.

[Mark Mentovai]

http://breakpad.appspot.com/438002/diff/4001/android/run-checks.sh
File android/run-checks.sh (right):

http://breakpad.appspot.com/438002/diff/4001/android/run-checks.sh#newcode490
android/run-checks.sh:490: --host="$GNU_CONFIG" &&
Alignment (existing problem, but as long as you’re touching the line…)

http://breakpad.appspot.com/438002/diff/4001/src/tools/linux/md2core/minidump...
File src/tools/linux/md2core/minidump-2-core.cc (right):

http://breakpad.appspot.com/438002/diff/4001/src/tools/linux/md2core/minidump...
src/tools/linux/md2core/minidump-2-core.cc:205: #elif defined(__x86_64__)
Share this with the __i386__ branch by using an || on the #if defined(…)?

http://breakpad.appspot.com/438002/diff/4001/src/tools/linux/md2core/minidump...
src/tools/linux/md2core/minidump-2-core.cc:209: user_regs regs;
This is getting annoying, to have these #ifdefs spilling all over the file like
this. Would it make more sense to have a (file-local, perhaps) |typedef
user_regs user_regs_struct;| for __arm__ in just one spot to make up for glibc’s
lack of cross-architecture compatibility? Looks like a lot of the rest of the
code can just be shared among architectures then.

[digit]

http://breakpad.appspot.com/438002/diff/4001/android/run-checks.sh
File android/run-checks.sh (right):

http://breakpad.appspot.com/438002/diff/4001/android/run-checks.sh#newcode490
android/run-checks.sh:490: --host="$GNU_CONFIG" &&
Done.

http://breakpad.appspot.com/438002/diff/4001/src/tools/linux/md2core/minidump...
File src/tools/linux/md2core/minidump-2-core.cc (right):

http://breakpad.appspot.com/438002/diff/4001/src/tools/linux/md2core/minidump...
src/tools/linux/md2core/minidump-2-core.cc:205: #elif defined(__x86_64__)
Sure.

http://breakpad.appspot.com/438002/diff/4001/src/tools/linux/md2core/minidump...
src/tools/linux/md2core/minidump-2-core.cc:209: user_regs regs;
That works too :)

[Mark Mentovai]

http://breakpad.appspot.com/438002/diff/11001/src/tools/linux/md2core/minidum...
File src/tools/linux/md2core/minidump-2-core.cc (right):

http://breakpad.appspot.com/438002/diff/11001/src/tools/linux/md2core/minidum...
src/tools/linux/md2core/minidump-2-core.cc:80: // platforms. This file-local
typedef simplifies the source code.
“architectures” instead of “platforms” to make it clearer.

http://breakpad.appspot.com/438002/diff/11001/src/tools/linux/md2core/minidum...
src/tools/linux/md2core/minidump-2-core.cc:205: user_regs_struct regs;
Why don’t you just have this user_regs_struct thing be unconditional for all
CPUs, and then have the fpregs field be i386/x86-64 and fpxregs be i386-only? As
it is now, you’re still duplicating the regs field.

[digit]

http://breakpad.appspot.com/438002/diff/11001/src/tools/linux/md2core/minidum...
File src/tools/linux/md2core/minidump-2-core.cc (right):

http://breakpad.appspot.com/438002/diff/11001/src/tools/linux/md2core/minidum...
src/tools/linux/md2core/minidump-2-core.cc:80: // platforms. This file-local
typedef simplifies the source code.
On 2012/08/21 17:00:54, Mark Mentovai wrote:
> “architectures” instead of “platforms” to make it clearer.
Right. Done.

http://breakpad.appspot.com/438002/diff/11001/src/tools/linux/md2core/minidum...
src/tools/linux/md2core/minidump-2-core.cc:205: user_regs_struct regs;
The main idea was to keep the #error to more easily identify porting effort in
the future (e.g. for MIPS). I've removed it to simplify the code.

[Mark Mentovai]

LGTM