GN: Strip NaCl IRT and save its .debug file

GN: Strip NaCl IRT and save its .debug file

With a simple tweak, the toolchain already knows how to strip the
nexe and leave the unstripped nexe where it can be found.  After
that, the unstripped nexe can just be copied as the .debug file and
a single objcopy command suffices to annotate the stripped file with
a pointer to the .debug file.

Also factor out computations about toolchain directory names into
//build/config/nacl/config.gni where they can be reused.

BUG=546352
R=bbudge@chromium.org, dpranke@chromium.org

Committed: https://crrev.com/7a379aa7f9ff6fe49bd441869887c7b517a1a0a0
Cr-Commit-Position: refs/heads/master@{#356340}

[bbudge, 2015-10-22 21:00:53]

rubber stamp LGTM

[Dirk Pranke, 2015-10-23 02:39:34]

dpranke@chromium.org changed reviewers:
+ brettw@chromium.org

[Dirk Pranke, 2015-10-23 02:39:35]

This mostly looks fine, but the call out to the generic run script is probably
out (at least as I understand things).

Brett, do I have this right? Any suggestion for something other than a custom
single-purpose script?

https://codereview.chromium.org/1415293005/diff/1/ppapi/native_client/BUILD.gn
File ppapi/native_client/BUILD.gn (right):

https://codereview.chromium.org/1415293005/diff/1/ppapi/native_client/BUILD.g...
ppapi/native_client/BUILD.gn:85: script = "//build/run.py"
As I understand it, we very intentionally do not have a generic run command like
this, because it would be too prone to abuse as an escape hatch.

You might have to just write a script that wraps objcopy only.

Perhaps Brett has other thoughts on this ...

[Petr Hosek, 2015-10-23 03:05:01]

On 2015/10/23 02:39:35, Dirk Pranke wrote:
> This mostly looks fine, but the call out to the generic run script is probably
> out (at least as I understand things).
> 
> Brett, do I have this right? Any suggestion for something other than a custom
> single-purpose script?
> 
> https://codereview.chromium.org/1415293005/diff/1/ppapi/native_client/BUILD.gn
> File ppapi/native_client/BUILD.gn (right):
> 
>
https://codereview.chromium.org/1415293005/diff/1/ppapi/native_client/BUILD.g...
> ppapi/native_client/BUILD.gn:85: script = "//build/run.py"
> As I understand it, we very intentionally do not have a generic run command
like
> this, because it would be too prone to abuse as an escape hatch.
> 
> You might have to just write a script that wraps objcopy only.
> 
> Perhaps Brett has other thoughts on this ...

Regarding the toolchain directory names refactoring, I had already done that in
issue 1386733002 which everyone seem to be ignoring. Shall I delete it then?

[Roland McGrath, 2015-10-23 18:12:56]

On 2015/10/23 03:05:01, Petr Hosek wrote:
> Regarding the toolchain directory names refactoring, I had already done that
in
> issue 1386733002 which everyone seem to be ignoring. Shall I delete it then?

I'd forgotten about that one and it wasn't clear to me if it was ever going to
land.
Your change and mine have much the same effect, but go about it in very
different ways.
Yours seems like it might be cleaner, but it's really a matter of GN style more
than anything else.
So I'd like Dirk and/or Brett to say which approach is preferred.

Since my change makes more use of the variables than yours does and I'd have to
re-check everything to see if yours covers my new uses fully, perhaps it makes
sense just to roll them together even if we use your approach rather than mine.


Thanks,
Roland

[Roland McGrath, 2015-10-26 19:06:32]

PTAL

[Dirk Pranke, 2015-10-27 00:28:46]

lgtm

[Roland McGrath, 2015-10-27 18:05:31]

The CQ bit was checked by mcgrathr@chromium.org

[Roland McGrath, 2015-10-27 18:05:31]

The patchset sent to the CQ was uploaded after l-g-t-m from bbudge@chromium.org
Link to the patchset: https://codereview.chromium.org/1415293005/#ps20001
(title: "use bespoke action script")

[commit-bot: I haz the power, 2015-10-27 18:06:29]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1415293005/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1415293005/20001

[commit-bot: I haz the power, 2015-10-27 18:11:36]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-10-27 18:14:27]

Patchset 2 (id:??) landed as
https://crrev.com/7a379aa7f9ff6fe49bd441869887c7b517a1a0a0
Cr-Commit-Position: refs/heads/master@{#356340}