Broken in 48.0.2538.0
Bisect info: 354613 (good) - 354624 (bad)
https://chromium.googlesource.com/chromium/src/+log/e9db8061..6b1707c8?pretty=fuller
Suspecting r354615 "Revalidate using the same Resource, attempt #3"

Repro note: open http://aspectusaurei.nl/test.html and refresh it five times in 1 second intervals.

Able to reproduce this issue on Windows-10 and Ubuntu 14.04 using chrome latest stable #59.0.3071.115.

As per the bisect information provided in the comment #1 assigning to the concerned person.

japhet@ - Could you please check whether this is caused with respect to your change, if not please help us in assigning it to the right owner. 

Thanks!

Missing content justifies release blocking. Marking it so.

Given that this bug has been present on stable for well over a year, I'm inclined to mark this as needed for the next branch, rather than trying to block an already-branched released on it. Therefore, RB-Beta for M62.

Cc'ing  mkwst@ as well for inputs.

Tested on latest Canary #62.0.3168.0 on Windows 7, Mac 10.12.6 and Ubuntu 14.04. Issue is still reproducible 

@japhet -- Is there any latest update on this. Thanks.

Has anyone been able to reproduce this locally? As far as I can tell, I've got my local server sending the exact same headers and body as http://aspectusaurei.nl is, but I can't get the bug to reproduce. I don't immediately see what I'm missing.

Hmm, this seems to be caused by the combination of SVG + revalidation + style. If I removed the <style> element entirely, then the SVG image is shown as expected...minimizing.

Minimized test case (both "position: absolute" and "width: 100%" are required).

<div style="position: absolute">
<img style="width: 100%" src="http://aspectusaurei.nl/beelden/mefecit.svg" onload="onLoad()">
</div>
<script>
function onLoad() {
  // Always outputs 454.
  console.log(document.getElementsByTagName('img')[0].naturalWidth);

  // 454 when loaded for the first time or reloaded with cache disabled.
  // 0 when reloaded normally.
  console.log(document.getElementsByTagName('img')[0].width);
}
</script>

When I added "width: 100%" or "width: 1000px" etc. to <div>, then the image was shown and the img's width seen from JavaScript became non-zero:

<div style="position: absolute; width: 100%">
<img style="width: 100%" src="http://aspectusaurei.nl/beelden/mefecit.svg" onload="onLoad()">
</div>

Friendly ping to get an update on this issue marked as Beta blocker. I am still able to reproduce the issue on the latest canary(62.0.3178.0) on Windows-10 using http://aspectusaurei.nl/test.html.

We are branching in few days. Please have a fix ASAP.

Just to Update,
Able to reproduce the issue on Windows 10 with stable #60.0.3112.101, beta #61.0.3163.49, 
dev #62.0.3192.0, canary #62.0.3193.0

Since this issue marked as Beta blocker, japhet@ could you please look into this issue.

I've got a fix: https://chromium-review.googlesource.com/c/chromium/src/+/634131, haven't finished the regression test yet, though.

Just to update, M-62 would be branched on 08/31. Friendly ping for an update to land the CL before branch point. 

Thank you!

The fix is slowed down in review, I may miss the branch point.

This bug is fairly painful when it triggers, but it requires a fairly unusual set of circumstance to trigger (which is probably how it was on stable for over a year before being reported).

Bumping to RB-Stable, will make sure I get a fix merged. Sorry for the delay :/

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/379d0b2a6c3a03aa61f2eb7c15392ca6d59b4cc2

commit 379d0b2a6c3a03aa61f2eb7c15392ca6d59b4cc2
Author: Nate Chapin <japhet@chromium.org>
Date: Thu Aug 31 19:04:37 2017

Don't use an SVGImage's size info if it's being revalidated

The sizing info isn't expected to be available yet, and it might get
used if we layout during revalidation. In that case, when the 
revalidation completes, the intrinisic size won't change, so a
relayout won't happen, which can result in the SVGImage being in a 0x0
parent.

Bug:  747672 
Test: http/tests/svg/svgimage-revalidation.html
Change-Id: I1649b67f8f03eae7999f2c461ab1bb64a7cb2d88
Reviewed-on: https://chromium-review.googlesource.com/634131
Commit-Queue: Nate Chapin <japhet@chromium.org>
Reviewed-by: Hiroshige Hayashizaki <hiroshige@chromium.org>
Cr-Commit-Position: refs/heads/master@{#498942}
[add] https://crrev.com/379d0b2a6c3a03aa61f2eb7c15392ca6d59b4cc2/third_party/WebKit/LayoutTests/http/tests/svg/resources/rect-100x100-delayed-304.php
[add] https://crrev.com/379d0b2a6c3a03aa61f2eb7c15392ca6d59b4cc2/third_party/WebKit/LayoutTests/http/tests/svg/resources/rect-100x100.svg
[add] https://crrev.com/379d0b2a6c3a03aa61f2eb7c15392ca6d59b4cc2/third_party/WebKit/LayoutTests/http/tests/svg/svgimage-revalidation-expected.html
[add] https://crrev.com/379d0b2a6c3a03aa61f2eb7c15392ca6d59b4cc2/third_party/WebKit/LayoutTests/http/tests/svg/svgimage-revalidation.html
[modify] https://crrev.com/379d0b2a6c3a03aa61f2eb7c15392ca6d59b4cc2/third_party/WebKit/Source/core/layout/LayoutImage.cpp
[modify] https://crrev.com/379d0b2a6c3a03aa61f2eb7c15392ca6d59b4cc2/third_party/WebKit/Source/core/loader/resource/ImageResourceContent.cpp
[modify] https://crrev.com/379d0b2a6c3a03aa61f2eb7c15392ca6d59b4cc2/third_party/WebKit/Source/core/loader/resource/ImageResourceContent.h

This landed before the branch, so no merge should be required.

I opened https://bugs.chromium.org/p/chromium/issues/detail?id=761026 to explore a better design for this fix.

[Auto-generated comment by a script] We noticed that this issue is targeted for M-62; it appears the fix may have landed after branch point, meaning a merge might be required. Please confirm if a merge is required here - if so add Merge-Request-62 label, otherwise remove Merge-TBD label. Thanks.