Alter how relocation packing cuts holes from libchrome.so.

Alter how relocation packing cuts holes from libchrome.so.

The current ELF file rewrite alters both offsets and addresses.
Altering addresses affects debugging tools such as breakpad.

This change leaves addresses unchanged, and instead splits a single
LOAD segment into two, leaving a mapping hole at the point where
the hole was created in the ELF file.  It works by repurposing the
PT_GNU_STACK segment, unused on Android.

The split into two LOAD segments no longer requires relocation
packing to adjust relocations or the symbol table, and in particular
means the the call frame information extracted by Dwarf DIE data
extractors for breakpad from an unstripped (and not packed) library
will match the stripped and packed one.

BUG=385553, 394703
Committed: https://crrev.com/1ce088922527f3ddfeb5618b871e05473de53c02
Cr-Commit-Position: refs/heads/master@{#294413}

[simonb (inactive), 2014-09-03 14:57:48]

simonb@chromium.org changed reviewers:
+ anton@chromium.org

[Anton, 2014-09-03 16:57:20]

https://codereview.chromium.org/535943002/diff/1/tools/relocation_packer/src/...
File tools/relocation_packer/src/elf_file.cc (right):

https://codereview.chromium.org/535943002/diff/1/tools/relocation_packer/src/...
tools/relocation_packer/src/elf_file.cc:20: 
The comment for SplitProgramHeadersForHole is pretty useful, but it is buried
half way down the file. I would like there to be something at the start which
gives a high-level about what this is all about and possibly just sends you to
the this comment for more detail.

https://codereview.chromium.org/535943002/diff/1/tools/relocation_packer/src/...
tools/relocation_packer/src/elf_file.cc:335: CHECK(first_loadable_segment);
Here and below I want a bit more text to come out with each of these crashing
errors.

https://codereview.chromium.org/535943002/diff/1/tools/relocation_packer/src/...
tools/relocation_packer/src/elf_file.cc:817: Elf_Scn* dynamic_section =
GetDynamicSection(elf);;
two semicolons, just to be sure.

[simonb (inactive), 2014-09-03 17:52:22]

Thanks.  Updated in patch set 2.

https://codereview.chromium.org/535943002/diff/1/tools/relocation_packer/src/...
File tools/relocation_packer/src/elf_file.cc (right):

https://codereview.chromium.org/535943002/diff/1/tools/relocation_packer/src/...
tools/relocation_packer/src/elf_file.cc:20: 
On 2014/09/03 16:57:20, Anton wrote:
> The comment for SplitProgramHeadersForHole is pretty useful, but it is buried
> half way down the file. I would like there to be something at the start which
> gives a high-level about what this is all about and possibly just sends you to
> the this comment for more detail.

Done.

https://codereview.chromium.org/535943002/diff/1/tools/relocation_packer/src/...
tools/relocation_packer/src/elf_file.cc:335: CHECK(first_loadable_segment);
On 2014/09/03 16:57:20, Anton wrote:
> Here and below I want a bit more text to come out with each of these crashing
> errors.

Done.

https://codereview.chromium.org/535943002/diff/1/tools/relocation_packer/src/...
tools/relocation_packer/src/elf_file.cc:817: Elf_Scn* dynamic_section =
GetDynamicSection(elf);;
On 2014/09/03 16:57:20, Anton wrote:
> two semicolons, just to be sure.

Done.

[Anton, 2014-09-04 10:51:53]

lgtm

[simonb (inactive), 2014-09-04 13:35:13]

The CQ bit was checked by simonb@chromium.org

[commit-bot: I haz the power, 2014-09-04 13:36:54]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/simonb@chromium.org/535943002/40001

[commit-bot: I haz the power, 2014-09-04 13:36:55]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-09-04 13:36:56]

No LGTM from a valid reviewer yet. Only full committers are accepted.
Even if an LGTM may have been provided, it was from a non-committer or
a provisional committer, _not_ a full super star committer.
See http://www.chromium.org/getting-involved/become-a-committer
Note that this has nothing to do with OWNERS files.

[simonb (inactive), 2014-09-04 14:48:25]

simonb@chromium.org changed reviewers:
+ skyostil@chromium.org

[simonb (inactive), 2014-09-04 14:48:26]

+Sami, for committer lgtm.  Thanks.

[Sami, 2014-09-04 14:55:37]

My eyes are indeed bleeding but lgtm.

[simonb (inactive), 2014-09-04 15:03:21]

The CQ bit was checked by simonb@chromium.org

[commit-bot: I haz the power, 2014-09-04 15:04:46]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/simonb@chromium.org/535943002/40001

[commit-bot: I haz the power, 2014-09-04 15:59:02]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-09-04 15:59:03]

Try jobs failed on following builders:
  linux_gpu_triggered_tests on tryserver.chromium.gpu
(http://build.chromium.org/p/tryserver.chromium.gpu/builders/linux_gpu_trigger...)

[simonb (inactive), 2014-09-11 16:51:03]

The CQ bit was checked by simonb@chromium.org

[commit-bot: I haz the power, 2014-09-11 16:51:38]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patchset/535943002/40001

[commit-bot: I haz the power, 2014-09-11 17:29:46]

Committed patchset #3 (id:40001) as 5112bfceccae1919568bceec245f98b8031f10a2

[commit-bot: I haz the power, 2014-09-11 18:42:48]

Patchset 3 (id:??) landed as
https://crrev.com/1ce088922527f3ddfeb5618b871e05473de53c02
Cr-Commit-Position: refs/heads/master@{#294413}