Use offsetof in ZR_HELPER_ for Clang.

Use offsetof in ZR_HELPER_ for Clang.

As an immediate reason for the change is that making an invalid
reinterpret_cast(16) causes Control Flow Integrity and UBSan vptr
sanitizers to crash on reading the vtable pointer. Ignoring this
problem (by suppressing a check) would mask real failures, some
of which have security consequences.

Theoretically, a similar change could be made for GCC, but there's
no immediate pressure to do that, as GCC does not have any type-checking
subsystem for reinterpret_casts.

NOTE: this patch will need to be reverted, when the github repo gets the official fix submitted by me to the internal Google repo.

BUG=605933
Committed: https://crrev.com/7809c2d9948352fb73d50d3b88b45e42bcfed5cc
Cr-Commit-Position: refs/heads/master@{#390568}

[krasin, 2016-04-28 23:20:28]

The CQ bit was checked by krasin@google.com to run a CQ dry run

[commit-bot: I haz the power, 2016-04-28 23:20:45]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1929193002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1929193002/1

[krasin, 2016-04-28 23:26:35]

krasin@google.com changed reviewers:
+ xyzzyz@chromium.org

[krasin, 2016-04-28 23:27:23]

Description was changed from

==========
Use offsetof in ZR_HELPER_ for Clang.

As an immediate reason for the change is that making an invalid
reinterpret_cast(16) causes Control Flow Integrity and UBSan vptr
sanitizers to crash on reading the vtable pointer. Ignoring this
problem (by suppressing a check) would mask real failures, some
of which have security consequences.

Theoretically, a similar change could be made for GCC, but there's
no immediate pressure to do that, as GCC does not have any type-checking
subsystem for reinterpret_casts.

BUG=605933
==========

to

==========
Use offsetof in ZR_HELPER_ for Clang.

As an immediate reason for the change is that making an invalid
reinterpret_cast(16) causes Control Flow Integrity and UBSan vptr
sanitizers to crash on reading the vtable pointer. Ignoring this
problem (by suppressing a check) would mask real failures, some
of which have security consequences.

Theoretically, a similar change could be made for GCC, but there's
no immediate pressure to do that, as GCC does not have any type-checking
subsystem for reinterpret_casts.

NOTE: this patch will need to be reverted, when the github repo gets the
official fix submitted by me to the internal Google repo.

BUG=605933
==========

[xyzzyz, 2016-04-28 23:42:56]

lgtm

https://codereview.chromium.org/1929193002/diff/1/third_party/protobuf/README...
File third_party/protobuf/README.chromium (right):

https://codereview.chromium.org/1929193002/diff/1/third_party/protobuf/README...
third_party/protobuf/README.chromium:95: the internal Google repo.
Please link to the issue on github, so that we know what to track.

[krasin, 2016-04-28 23:50:41]

The CQ bit was unchecked by krasin@google.com

[krasin, 2016-04-28 23:50:45]

The CQ bit was checked by krasin@google.com

[commit-bot: I haz the power, 2016-04-28 23:50:51]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1929193002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1929193002/1

[krasin, 2016-04-28 23:50:51]

The CQ bit was unchecked by krasin@google.com

[commit-bot: I haz the power, 2016-04-28 23:50:52]

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

[commit-bot: I haz the power, 2016-04-28 23:50:53]

No L-G-T-M from a valid reviewer yet. 
CQ run can only be started by full committers or once the patch has
received an L-G-T-M from a full committer.
Even if an L-G-T-M may have been provided, it was from a non-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.

[krasin1, 2016-04-28 23:57:02]

I almost missed the comment to link to the github issue. Added it to
README.chromium. Also added a crbug issue and updated the github issue with the
current status.

Also, it appears that I've created this CL from the wrong non-committer account,
so I have to say LGTM to myself to bless it.

[krasin1, 2016-04-28 23:57:10]

The CQ bit was checked by krasin@chromium.org

[krasin1, 2016-04-28 23:57:10]

The patchset sent to the CQ was uploaded after l-g-t-m from xyzzyz@chromium.org
Link to the patchset: https://codereview.chromium.org/1929193002/#ps20001
(title: "link github and crbug issues")

[commit-bot: I haz the power, 2016-04-28 23:57:29]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1929193002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1929193002/20001

[commit-bot: I haz the power, 2016-04-29 01:51:08]

Description was changed from

==========
Use offsetof in ZR_HELPER_ for Clang.

As an immediate reason for the change is that making an invalid
reinterpret_cast(16) causes Control Flow Integrity and UBSan vptr
sanitizers to crash on reading the vtable pointer. Ignoring this
problem (by suppressing a check) would mask real failures, some
of which have security consequences.

Theoretically, a similar change could be made for GCC, but there's
no immediate pressure to do that, as GCC does not have any type-checking
subsystem for reinterpret_casts.

NOTE: this patch will need to be reverted, when the github repo gets the
official fix submitted by me to the internal Google repo.

BUG=605933
==========

to

==========
Use offsetof in ZR_HELPER_ for Clang.

As an immediate reason for the change is that making an invalid
reinterpret_cast(16) causes Control Flow Integrity and UBSan vptr
sanitizers to crash on reading the vtable pointer. Ignoring this
problem (by suppressing a check) would mask real failures, some
of which have security consequences.

Theoretically, a similar change could be made for GCC, but there's
no immediate pressure to do that, as GCC does not have any type-checking
subsystem for reinterpret_casts.

NOTE: this patch will need to be reverted, when the github repo gets the
official fix submitted by me to the internal Google repo.

BUG=605933
==========

[commit-bot: I haz the power, 2016-04-29 01:51:09]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-04-30 17:23:37]

Patchset 2 (id:??) landed as
https://crrev.com/7809c2d9948352fb73d50d3b88b45e42bcfed5cc
Cr-Commit-Position: refs/heads/master@{#390568}

[commit-bot: I haz the power, 2016-04-30 17:58:58]

Description was changed from

==========
Use offsetof in ZR_HELPER_ for Clang.

As an immediate reason for the change is that making an invalid
reinterpret_cast(16) causes Control Flow Integrity and UBSan vptr
sanitizers to crash on reading the vtable pointer. Ignoring this
problem (by suppressing a check) would mask real failures, some
of which have security consequences.

Theoretically, a similar change could be made for GCC, but there's
no immediate pressure to do that, as GCC does not have any type-checking
subsystem for reinterpret_casts.

NOTE: this patch will need to be reverted, when the github repo gets the
official fix submitted by me to the internal Google repo.

BUG=605933
==========

to

==========
Use offsetof in ZR_HELPER_ for Clang.

As an immediate reason for the change is that making an invalid
reinterpret_cast(16) causes Control Flow Integrity and UBSan vptr
sanitizers to crash on reading the vtable pointer. Ignoring this
problem (by suppressing a check) would mask real failures, some
of which have security consequences.

Theoretically, a similar change could be made for GCC, but there's
no immediate pressure to do that, as GCC does not have any type-checking
subsystem for reinterpret_casts.

NOTE: this patch will need to be reverted, when the github repo gets the
official fix submitted by me to the internal Google repo.

BUG=605933

Committed: https://crrev.com/7809c2d9948352fb73d50d3b88b45e42bcfed5cc
Cr-Commit-Position: refs/heads/master@{#390568}
==========