[qcms] Allow negative XYZ for display profiles on the APPLE port

[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the layout tests.

BUG=562951
Committed: https://crrev.com/38a801c16ed58e8dde9e4fb98b8bf99edfdad3e2
Cr-Commit-Position: refs/heads/master@{#362410}

[Noel Gordon, 2015-11-30 04:39:55]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace.txt: use it to convince
the bots that this CL should run the webkit layout tests.

BUG=562951
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace.txt: use it to convince
the bots that this CL should run the webkit layout tests. At least
I was going to do this until the presubmit checked failed.

BUG=562951
==========

[Noel Gordon, 2015-11-30 04:40:19]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace.txt: use it to convince
the bots that this CL should run the webkit layout tests. At least
I was going to do this until the presubmit checked failed.

BUG=562951
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace.txt: use it to convince
the bots that this CL should run the webkit layout tests. [1]

BUG=562951

At least I was going to do this until the presubmit checked failed.
==========

[Noel Gordon, 2015-11-30 04:40:45]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace.txt: use it to convince
the bots that this CL should run the webkit layout tests. [1]

BUG=562951

At least I was going to do this until the presubmit checked failed.
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace.txt: use it to convince
the bots that this CL should run the webkit layout tests. [1]

BUG=562951

[1] Was going to do this until the presubmit checked failed.
==========

[Noel Gordon, 2015-11-30 05:31:17]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace.txt: use it to convince
the bots that this CL should run the webkit layout tests. [1]

BUG=562951

[1] Was going to do this until the presubmit checked failed.
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to convince the the
bots that this CL should run the webkit layout
tests [1].

BUG=562951

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ lat week, not sure why.
==========

[Noel Gordon, 2015-11-30 05:34:48]

+tkent@ was using Source/Tools/whitespace-file.txt to force webkit tests run on
the bots.  The file was removed.

Is there some other recommended way to convince the bots that this CL should run
the webkit tests?

[Noel Gordon, 2015-11-30 05:35:31]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to convince the the
bots that this CL should run the webkit layout
tests [1].

BUG=562951

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ lat week, not sure why.
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
convince the the bots that this CL should run the webkit layout
tests [1].

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ lat week, not sure why.

BUG=562951
==========

[tkent, 2015-11-30 06:08:10]

On 2015/11/30 at 05:34:48, noel wrote:
> +tkent@ was using Source/Tools/whitespace-file.txt to force webkit tests run
on the bots.  The file was removed.
> 
> Is there some other recommended way to convince the bots that this CL should
run the webkit tests?

Can you use CQ_INCLUDE_TRYBOTS=linux_blink_rel or something?

Fixing dependency like crbug.com/546901 is the right approach.  However, I think
it's ok to re-add whitespace-file.txt.

[Noel Gordon, 2015-11-30 06:24:22]

On 2015/11/30 06:08:10, tkent wrote:

> Can you use CQ_INCLUDE_TRYBOTS=linux_blink_rel or something?

Nope: no longer works, I believe.

> Fixing dependency like crbug.com/546901 is the right approach.

We're gonna create a very long list if we go that way :/

> However, I think it's ok to re-add whitespace-file.txt.

Lemme try on this CL.  I got a pre-submit failure on upload when I tried to
re-add it as mentioned in the change log.

[Noel Gordon, 2015-11-30 06:28:16]

On 2015/11/30 06:24:22, noel gordon wrote:
> On 2015/11/30 06:08:10, tkent wrote:
> 
> > Can you use CQ_INCLUDE_TRYBOTS=linux_blink_rel or something?
> 
> Nope: no longer works, I believe.

Meanwhile, I will manually add the blink try bots.

[Noel Gordon, 2015-11-30 06:36:58]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
convince the the bots that this CL should run the webkit layout
tests [1].

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ lat week, not sure why.

BUG=562951
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the webkit layout
tests [1].

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ lat week, not sure why.

BUG=562951
==========

[Noel Gordon, 2015-11-30 07:15:10]

On 2015/11/30 06:24:22, noel gordon wrote:
> >
> > However, I think it's ok to re-add whitespace-file.txt.
> 
> Lemme try on this CL.  I got a pre-submit failure on upload when I tried to
> re-add it as mentioned in the change log.

I "re-added" it as WebKit/LayoutTests/whitespace.txt in patch set #2, and
  - no presubmit warning on patch upload (good)
  - the bots are running the layout tests (good)

~noel

[Noel Gordon, 2015-11-30 08:11:56]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the webkit layout
tests [1].

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ lat week, not sure why.

BUG=562951
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the webkit layout
tests [1].

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ lat week, not sure why.
Fixed in patch set #2.

BUG=562951
==========

[Noel Gordon, 2015-11-30 14:06:54]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the webkit layout
tests [1].

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ lat week, not sure why.
Fixed in patch set #2.

BUG=562951
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the webkit layout
tests [1].

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ lat week, not sure why.
Fixed the presubmit fu with patch set #2.

BUG=562951
==========

[Noel Gordon, 2015-11-30 14:08:10]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the webkit layout
tests [1].

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ lat week, not sure why.
Fixed the presubmit fu with patch set #2.

BUG=562951
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the webkit layout
tests [1].

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ last week, not sure why.
Fixed the presubmit fu with patch set #2.

BUG=562951
==========

[Justin Novosad, 2015-11-30 15:17:32]

This seems fine to me, but don't you want to land this in the upstream qcms?

[Noel Gordon, 2015-11-30 15:26:30]

On 2015/11/30 15:17:32, Justin Novosad wrote:
> This seems fine to me, but don't you want to land this in the upstream qcms?

Nope, refer the README.

[Noel Gordon, 2015-11-30 15:49:52]

noel@chromium.org changed reviewers:
+ junov@chromium.org

[Noel Gordon, 2015-12-01 00:29:23]

+robert@  Doesn't affect your work, but if you have read the bug (sorry it's
long) you should also review please.

[Noel Gordon, 2015-12-01 00:29:56]

noel@chromium.org changed reviewers:
+ robert.bradford@intel.com

[Justin Novosad, 2015-12-01 03:29:55]

lgtm with nit

https://codereview.chromium.org/1485583002/diff/20001/third_party/qcms/src/ic...
File third_party/qcms/src/iccread.c (right):

https://codereview.chromium.org/1485583002/diff/20001/third_party/qcms/src/ic...
third_party/qcms/src/iccread.c:336: // FIXME: allow this relaxation on all
ports?
FIXME is deprecated, use TODO(noel)

[Noel Gordon, 2015-12-01 03:58:37]

On 2015/12/01 03:29:55, Justin Novosad wrote:

> FIXME is deprecated, use TODO(noel)

QCMS is FIXME.  Lots of bugs [1], and anyone can pitch in to fix them. 
Appreciate any help.

[1]
https://bugzilla.mozilla.org/buglist.cgi?component=GFX%3A%20Color%20Managemen...

[Noel Gordon, 2015-12-01 04:00:16]

https://codereview.chromium.org/1485583002/diff/20001/third_party/qcms/src/ic...
File third_party/qcms/src/iccread.c (right):

https://codereview.chromium.org/1485583002/diff/20001/third_party/qcms/src/ic...
third_party/qcms/src/iccread.c:301: // Sum the XYZ values: they should add up
D50 white, within tolerance.
Note to self: change to "should add to" on submit.

[Noel Gordon, 2015-12-01 12:48:58]

The CQ bit was checked by noel@chromium.org

[Noel Gordon, 2015-12-01 12:48:58]

The patchset sent to the CQ was uploaded after l-g-t-m from junov@chromium.org
Link to the patchset: https://codereview.chromium.org/1485583002/#ps40001
(title: "Patch for landing")

[commit-bot: I haz the power, 2015-12-01 12:49:23]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1485583002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1485583002/40001

[Noel Gordon, 2015-12-01 12:49:36]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the webkit layout
tests [1].

[1] Was going to do this until the pre-submit check failed. The
white space file was removed by tkent@ last week, not sure why.
Fixed the presubmit fu with patch set #2.

BUG=562951
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the layout tests.

BUG=562951
==========

[robert.bradford, 2015-12-01 12:56:04]

lgtm. qcms whitespace improvements one CL at a time :-)

[commit-bot: I haz the power, 2015-12-01 14:20:51]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the layout tests.

BUG=562951
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the layout tests.

BUG=562951
==========

[commit-bot: I haz the power, 2015-12-01 14:20:51]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-12-01 14:21:34]

Description was changed from

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the layout tests.

BUG=562951
==========

to

==========
[qcms] Allow negative XYZ for display profiles on the APPLE port

Allow negative XYZ values (only on the Apple port for now) in a
display profile to support iMac P3 displays. The bogus checking
using negative XYZ values is questionable and may be removed in
future. Retain the D50 white-point tolerance testing. Add notes
about it, add FIXME.

Also: remove MSVC win32 warning (use a cast) and fix formatting
of the code (yes, the story of qcms and tabs is not good).

Recover the recently removed whitespace change file - use it to
try convince the bots that this CL should run the layout tests.

BUG=562951

Committed: https://crrev.com/38a801c16ed58e8dde9e4fb98b8bf99edfdad3e2
Cr-Commit-Position: refs/heads/master@{#362410}
==========

[commit-bot: I haz the power, 2015-12-01 14:21:34]

Patchset 3 (id:??) landed as
https://crrev.com/38a801c16ed58e8dde9e4fb98b8bf99edfdad3e2
Cr-Commit-Position: refs/heads/master@{#362410}

[Noel Gordon, 2015-12-10 00:54:02]

On 2015/12/01 12:56:04, robert.bradford wrote:
> lgtm. qcms whitespace improvements one CL at a time :-)

I'm gonna buy you a beer on day, bro :)

[Noel Gordon, 2015-12-10 00:54:28]

On 2015/12/10 00:54:02, noel gordon wrote:
> On 2015/12/01 12:56:04, robert.bradford wrote:
> > lgtm. qcms whitespace improvements one CL at a time :-)
> 
> I'm gonna buy you a beer on day, bro :)

... one day when I'm in the UK.