Fix VK WritePixels with offset rect

Fix VK WritePixels with offset rect

VkUploadPixelsTest was failing because it was reallocating 
the image when it didn't need to, because it was not 
resetting the rowBytes when it did need to, and because the
origin was incorrect when using bottom-up images.

GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1940963002

Committed: https://skia.googlesource.com/skia/+/c578b06319f271ac46ac78e3b86f74fbceed360e

[jvanverth1, 2016-05-02 16:53:17]

Description was changed from

==========
Fix VK WritePixels with offset rect
==========

to

==========
Fix VK WritePixels with offset rect
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[jvanverth1, 2016-05-02 16:55:01]

Description was changed from

==========
Fix VK WritePixels with offset rect
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Fix VK WritePixels with offset rect

VkUploadPixelsTest was failing because it was reallocating the image when it
didn't need to, and because it was not resetting the rowBytes when it did need
to.

GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[jvanverth1, 2016-05-02 16:59:46]

Description was changed from

==========
Fix VK WritePixels with offset rect

VkUploadPixelsTest was failing because it was reallocating the image when it
didn't need to, and because it was not resetting the rowBytes when it did need
to.

GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Fix VK WritePixels with offset rect

VkUploadPixelsTest was failing because it was reallocating 
the image when it didn't need to, because it was not 
resetting the rowBytes when it did need to, and because the
origin was incorrect when using bottom-up images.

GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[jvanverth1, 2016-05-02 17:00:01]

jvanverth@google.com changed reviewers:
+ egdaniel@google.com

[jvanverth1, 2016-05-02 17:30:24]

Updated to fix VkClearTests.

[egdaniel, 2016-05-02 17:38:24]

lgtm with the one question

https://codereview.chromium.org/1940963002/diff/1/src/gpu/vk/GrVkGpu.cpp
File src/gpu/vk/GrVkGpu.cpp (right):

https://codereview.chromium.org/1940963002/diff/1/src/gpu/vk/GrVkGpu.cpp#newc...
src/gpu/vk/GrVkGpu.cpp:422: for (int currentMipLevel = 1; currentMipLevel <
texelsShallowCopy.count(); currentMipLevel++) {
In this loop I notice you removed the SkTMax to keep width/height at least 1. Is
that something the mip level calculation already guarantees us, and thus why you
removed it? Is there any value in an assert that it is >= 1?

[jvanverth1, 2016-05-02 17:46:08]

https://codereview.chromium.org/1940963002/diff/1/src/gpu/vk/GrVkGpu.cpp
File src/gpu/vk/GrVkGpu.cpp (right):

https://codereview.chromium.org/1940963002/diff/1/src/gpu/vk/GrVkGpu.cpp#newc...
src/gpu/vk/GrVkGpu.cpp:422: for (int currentMipLevel = 1; currentMipLevel <
texelsShallowCopy.count(); currentMipLevel++) {
On 2016/05/02 17:38:23, egdaniel wrote:
> In this loop I notice you removed the SkTMax to keep width/height at least 1.
Is
> that something the mip level calculation already guarantees us, and thus why
you
> removed it? Is there any value in an assert that it is >= 1?

Whoops, fixed.

[jvanverth1, 2016-05-02 17:46:44]

The CQ bit was checked by jvanverth@google.com

[jvanverth1, 2016-05-02 17:46:44]

The patchset sent to the CQ was uploaded after l-g-t-m from egdaniel@google.com
Link to the patchset: https://codereview.chromium.org/1940963002/#ps30001
(title: "Clamp miplevel dimensions to be at least 1"")

[commit-bot: I haz the power, 2016-05-02 17:46:49]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1940963002/30001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1940963002/30001

[commit-bot: I haz the power, 2016-05-02 17:58:15]

Description was changed from

==========
Fix VK WritePixels with offset rect

VkUploadPixelsTest was failing because it was reallocating 
the image when it didn't need to, because it was not 
resetting the rowBytes when it did need to, and because the
origin was incorrect when using bottom-up images.

GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Fix VK WritePixels with offset rect

VkUploadPixelsTest was failing because it was reallocating 
the image when it didn't need to, because it was not 
resetting the rowBytes when it did need to, and because the
origin was incorrect when using bottom-up images.

GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/c578b06319f271ac46ac78e3b86f74fbceed360e
==========

[commit-bot: I haz the power, 2016-05-02 17:58:16]

Committed patchset #3 (id:30001) as
https://skia.googlesource.com/skia/+/c578b06319f271ac46ac78e3b86f74fbceed360e