1247034 - Black and white bitmap renders upside down

Status: RESOLVED FIXED

(Core :: Graphics: ImageLib, defect)

Description

[balage78]

Attachment: QVHighResImage.bmp

User Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; Trident/7.0; Touch; .NET4.0C; .NET4.0E; .NET CLR 2.0.50727; .NET CLR 3.0.30729; .NET CLR 3.5.30729; Tablet PC 2.0; McAfee; rv:11.0) like Gecko

Steps to reproduce:

Load black and white bitmap in FireFox (file drag an drop or when served over HTTP)

(tested on 44 and 44.0.1 on Win 10; did not happen on Chrome, IE11, Opera, Safari or FireFox on OSX)


Actual results:

Bitmap loads and displays, but the image is flipped upside down 


Expected results:

Bitmap load, display right side up

Comment 1

[balage78]

Attachment: QVHighResImage.z_p

my uploaded bitmap was not exhibiting the same behavior so I'm attaching the original bitmap image as a zip file

Comment 2

[u279076]

I can reproduce on Mac OS 10.11 with the (zip) testcase.

Comment 3

[YF (Yang)]

https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=f088e6dbb1ba6f27e0fb6bd42ba090b464d3a5df&tochange=c9a6d9c234181121b667aea4087680fdbd254ba8

Comment 4

[u279076]

[Tracking Requested - why for this release]: Regression in BMP rendering starting with Firefox 43.

mozilla-central
===============
Last good revision: 01ae99b53561a2c3b40533d8c1c92bd3efc42d00 (2015-09-09)
First bad revision: dd2a1d737a64d9a3f23714ec5cc623ec8933b51f (2015-09-10)
Pushlog: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=01ae99b53561a2c3b40533d8c1c92bd3efc42d00&tochange=dd2a1d737a64d9a3f23714ec5cc623ec8933b51f

mozilla-inbound
===============
Last good revision: f088e6dbb1ba6f27e0fb6bd42ba090b464d3a5df
First bad revision: c9a6d9c234181121b667aea4087680fdbd254ba8
Pushlog: https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=f088e6dbb1ba6f27e0fb6bd42ba090b464d3a5df&tochange=c9a6d9c234181121b667aea4087680fdbd254ba8

Blame: Bug 1062066 -  Add downscale-during-decode support for the BMP decoder (Seth Fowler)

Comment 5

[u279076]

Seth, can you please look in to this?

Comment 6

[balage78]

I dug around a bit and it turns out this test image has a negative height (of the correct magnitude) specified in the header.

Comment 7

[Ritu Kothari (:ritu) (Inactive, please n-i to RyanVM, jcristau, or pascal)]

While this is a valid issue, does not seem critical enough to warrant (inclusion in) a dot release. Wontfix for Fx44.

Comment 8

[Timothy Nikkel (:tnikkel)]

Fixed by bug 1249550 ?

Comment 9

[Seth Fowler [:seth] [:s2h]]

I cannot reproduce this on Nightly. I strongly suspect this was indeed fixed by bug 1249550. I'm going to resolve this; please reopen if someone can reproduce again in Firefox 47 or later.