85808 2012-05-07 10:34:07 -0700 [chromium] Non-preserves-3d requires explicit orthographic transform when computing screen-space transform 2012-08-06 10:43:22 -0700 1 1 1 Unclassified WebKit Layout and Rendering 528+ (Nightly build) Unspecified Unspecified RESOLVED FIXED P2 Normal --- 84195 1 shawnsingh shawnsingh cc-bugs danakj dtu enne jamesr reveman vangelis vollick webkit.review.bot oldest_to_newest 617703 0 shawnsingh 2012-05-07 10:34:07 -0700 Consider the following case: parent layer A has a perspective projection descendant surface B does not preserve 3d, thus flattening its children children of B have 3-d transforms In this case, children of B should be flattened to surface B, before the perspective projection is applied. However, in the current code, this flattening is accidentally skipped when computing the screen-space transform, and therefore it computes incorrectly. Thanks to danakj@ for finding this problem. =) The fix is simple, when there is a surface that does not preserve 3d, and it does not have a perspective projection already, then it should be using an orthographic projection instead. Because transforms accumulate over the entire hierarchy, this orthographic projection actually needs to be made explicit, so that children are correctly flattened before ancestor transforms do anything else. 618100 1 shawnsingh 2012-05-07 16:52:23 -0700 It turns out that adding the orthographic portion of the screen-space transform (which I'm pretty sure is correct) breaks one layout test that seems to be malformed in the first place. The fix for that is in bug 84195. 685538 2 shawnsingh 2012-08-02 10:18:39 -0700 So I finally have a working fix. Patch coming in a moment, no obvious regressions from unit tests and layout tests on mac. Instead of using orthographic projection, it works more correctly to use the "flattening" operation, which is similar to an orthographic projection, but has some blatant differences, too: - A canonical orthographic matrix post-multiplied to a transform would effectively set the matrix's 3rd column to (0, 0, 0, 0). -- yes, 3rd element is zero here. - The flatten operation effectively sets the matrix's 3rd row AND 3rd colum to (0, 0, 1, 0). I tried for a while to swallow why exactly the flatten operation is more correct, but I can't justify it to myself 100% yet. The best I could come up with is that x and y values behave exactly like an orthographic projection, while the z component of the matrix is overridden so that it preserves the z value that is input to this transform. I experimented using an orthographic projection instead of flatten operation, andin that case layer sorting seems to break, for example, on http://www.webkit.org/blog-files/3d-transforms/transform-style.html I narrowed it down to the difference in the value of m33() for these two transforms. when m33() is zero for the orthographic projection, the layer sorting seems to break. When it's 1 (flattening operation), the layer sorting works. So my best guess is that the flatten operation is necessary for our layer sorter, because it preserves the z-values of points on the subtree that is being flattened. At the same time, the transform remains "correct" for x and y, only because we know that inputs and outputs are 2d, even though 3d things can happen to them. Any thoughts? 685547 3 156117 shawnsingh 2012-08-02 10:30:29 -0700 Created attachment 156117 Patch 685706 4 vangelis 2012-08-02 13:07:38 -0700 (In reply to comment #2) > > I experimented using an orthographic projection instead of flatten operation, andin that case layer sorting seems to break, for example, on http://www.webkit.org/blog-files/3d-transforms/transform-style.html > > I narrowed it down to the difference in the value of m33() for these two transforms. when m33() is zero for the orthographic projection, the layer sorting seems to break. When it's 1 (flattening operation), the layer sorting works. So my best guess is that the flatten operation is necessary for our layer sorter, because it preserves the z-values of points on the subtree that is being flattened. At the same time, the transform remains "correct" for x and y, only because we know that inputs and outputs are 2d, even though 3d things can happen to them. > > Any thoughts? Interesting. If preserves 3D isn't set shouldn't we be skipping sorting of sublayers altogether? 685811 5 156117 enne 2012-08-02 14:14:23 -0700 Comment on attachment 156117 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=156117&action=review > Source/WebCore/platform/graphics/chromium/cc/CCMathUtil.cpp:371 > + // - For z values, the new transform overrides any effect that the transform had on > + // z, and instead it preserves the z value for any points that are transformed. Can you explain why z values are preserved? Shouldn't everything flatten to z=0 on the surface? 685860 6 shawnsingh 2012-08-02 14:54:54 -0700 (In reply to comment #5) > (From update of attachment 156117 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=156117&action=review > > > Source/WebCore/platform/graphics/chromium/cc/CCMathUtil.cpp:371 > > + // - For z values, the new transform overrides any effect that the transform had on > > + // z, and instead it preserves the z value for any points that are transformed. > > Can you explain why z values are preserved? Shouldn't everything flatten to z=0 on the surface? Are you asking why they should be preserved? Ian and I believe that it is an intentional hack that allows us to use the transform for two purposes simultaneously: x,y,w would be used for rendering, and they are correctly orthographically projected, and z would be used for sorting; whatever z-value the subtree would have produced is preserved by having this flattening here. Or, if you're asking how this transform actually preserves z, it's because of the 3rd row being set to 0, 0, 1, 0: So if we say p' = flattenedA * p then p'.z = 0*p.x + 0*p.y + 1*p.z + 0*p.w so p'.z = p.z. Suppose we have additional matrices that were post-multiplied (i.e. the subtree has more transforms applied after the sublayer flattened).... like p' = flattenedA * B * C * D * p then p'.z = z-value of (B * C * D * p). i.e. it preserves whatever z-value is given to flattenedA. It seems to me that is (almost) exactly what we want for layer sorting. 685867 7 enne 2012-08-02 14:59:09 -0700 (In reply to comment #6) > (In reply to comment #5) > > (From update of attachment 156117 [details] [details]) > > View in context: https://bugs.webkit.org/attachment.cgi?id=156117&action=review > > > > > Source/WebCore/platform/graphics/chromium/cc/CCMathUtil.cpp:371 > > > + // - For z values, the new transform overrides any effect that the transform had on > > > + // z, and instead it preserves the z value for any points that are transformed. > > > > Can you explain why z values are preserved? Shouldn't everything flatten to z=0 on the surface? > > Are you asking why they should be preserved? Ian and I believe that it is an intentional hack that allows us to use the transform for two purposes simultaneously: x,y,w would be used for rendering, and they are correctly orthographically projected, and z would be used for sorting; whatever z-value the subtree would have produced is preserved by having this flattening here. Yeah, I'm asking why. When do layers in non-preserves-3d contexts need to be sorted? 685884 8 shawnsingh 2012-08-02 15:12:50 -0700 (In reply to comment #7) > (In reply to comment #6) > > (In reply to comment #5) > > > (From update of attachment 156117 [details] [details] [details]) > > > View in context: https://bugs.webkit.org/attachment.cgi?id=156117&action=review > > > > > > > Source/WebCore/platform/graphics/chromium/cc/CCMathUtil.cpp:371 > > > > + // - For z values, the new transform overrides any effect that the transform had on > > > > + // z, and instead it preserves the z value for any points that are transformed. > > > > > > Can you explain why z values are preserved? Shouldn't everything flatten to z=0 on the surface? > > > > Are you asking why they should be preserved? Ian and I believe that it is an intentional hack that allows us to use the transform for two purposes simultaneously: x,y,w would be used for rendering, and they are correctly orthographically projected, and z would be used for sorting; whatever z-value the subtree would have produced is preserved by having this flattening here. > > Yeah, I'm asking why. > > When do layers in non-preserves-3d contexts need to be sorted? Yeah, good question... this is also to answer Comment #4. Actually the situation you guys are thinking of is not the situation we would be dealing with here. The situation we're considering here is: (1) an ancestor layer has preserve-3d==false. Then, a bunch of layers in the subtree are in a 3d-rendering context, and need to be sorted with each other. The transform these layers use to compute their z stuff for sorting is the drawTransform... and the drawTransform would include the ancestor's transform as part of the drawTransform. That is where z-values would get lost if we used a pure orthographic projection -- at the ancestor. By doing this flattening operation instead, we don't loose the z-values that we needed for correct sorting. Does that answer your question? =) 685953 9 enne 2012-08-02 16:11:12 -0700 (In reply to comment #8) > (In reply to comment #7) > > > > When do layers in non-preserves-3d contexts need to be sorted? > > Yeah, good question... this is also to answer Comment #4. > > Actually the situation you guys are thinking of is not the situation we would be dealing with here. > > The situation we're considering here is: (1) an ancestor layer has preserve-3d==false. Then, a bunch of layers in the subtree are in a 3d-rendering context, and need to be sorted with each other. > > The transform these layers use to compute their z stuff for sorting is the drawTransform... and the drawTransform would include the ancestor's transform as part of the drawTransform. That is where z-values would get lost if we used a pure orthographic projection -- at the ancestor. Ok, so keeping m33=1 and not forcing z=0 is only for the sublayer matrix flattening. The screen space transform doesn't need this, but it's not problematic to have it. 686028 10 shawnsingh 2012-08-02 17:25:42 -0700 (In reply to comment #9) > (In reply to comment #8) > > (In reply to comment #7) > > > > > > When do layers in non-preserves-3d contexts need to be sorted? > > > > Yeah, good question... this is also to answer Comment #4. > > > > Actually the situation you guys are thinking of is not the situation we would be dealing with here. > > > > The situation we're considering here is: (1) an ancestor layer has preserve-3d==false. Then, a bunch of layers in the subtree are in a 3d-rendering context, and need to be sorted with each other. > > > > The transform these layers use to compute their z stuff for sorting is the drawTransform... and the drawTransform would include the ancestor's transform as part of the drawTransform. That is where z-values would get lost if we used a pure orthographic projection -- at the ancestor. > > Ok, so keeping m33=1 and not forcing z=0 is only for the sublayer matrix flattening. The screen space transform doesn't need this, but it's not problematic to have it. The screen space transform does need it. That was the whole point of this bug. Dana had a lovely repro for this, I have it locally on my machine. How about I convert it into a layout test for one more "layer" of coverage (pun definitely intended)? This was related to the bug you reported, where the android apps page was clipping the 3d layer when resizing the window. http://code.google.com/p/chromium/issues/detail?id=121743 686046 11 enne 2012-08-02 17:48:29 -0700 (In reply to comment #10) > (In reply to comment #9) > > (In reply to comment #8) > > > (In reply to comment #7) > > > > > > > > When do layers in non-preserves-3d contexts need to be sorted? > > > > > > Yeah, good question... this is also to answer Comment #4. > > > > > > Actually the situation you guys are thinking of is not the situation we would be dealing with here. > > > > > > The situation we're considering here is: (1) an ancestor layer has preserve-3d==false. Then, a bunch of layers in the subtree are in a 3d-rendering context, and need to be sorted with each other. > > > > > > The transform these layers use to compute their z stuff for sorting is the drawTransform... and the drawTransform would include the ancestor's transform as part of the drawTransform. That is where z-values would get lost if we used a pure orthographic projection -- at the ancestor. > > > > Ok, so keeping m33=1 and not forcing z=0 is only for the sublayer matrix flattening. The screen space transform doesn't need this, but it's not problematic to have it. > > The screen space transform does need it. That was the whole point of this bug. I was asking about m33=1 vs. m33=0, not about whether flattening needs to be applied. Just trying to understand the math. 686227 12 shawnsingh 2012-08-02 22:15:57 -0700 (In reply to comment #11) > (In reply to comment #10) > > (In reply to comment #9) > > > (In reply to comment #8) > > > > (In reply to comment #7) > > > > > > > > > > When do layers in non-preserves-3d contexts need to be sorted? > > > > > > > > Yeah, good question... this is also to answer Comment #4. > > > > > > > > Actually the situation you guys are thinking of is not the situation we would be dealing with here. > > > > > > > > The situation we're considering here is: (1) an ancestor layer has preserve-3d==false. Then, a bunch of layers in the subtree are in a 3d-rendering context, and need to be sorted with each other. > > > > > > > > The transform these layers use to compute their z stuff for sorting is the drawTransform... and the drawTransform would include the ancestor's transform as part of the drawTransform. That is where z-values would get lost if we used a pure orthographic projection -- at the ancestor. > > > > > > Ok, so keeping m33=1 and not forcing z=0 is only for the sublayer matrix flattening. The screen space transform doesn't need this, but it's not problematic to have it. > > > > The screen space transform does need it. That was the whole point of this bug. > > I was asking about m33=1 vs. m33=0, not about whether flattening needs to be applied. Just trying to understand the math. Oh I see what you're asking - I think you're right, the screen space transform doesn't need it. but I'll try it out and make sure. 686823 13 156117 enne 2012-08-03 09:21:22 -0700 Comment on attachment 156117 Patch R=me. Ok, I think I'm convinced by this. Thanks for the explanation about why the sublayer matrix can't have m33=0. 688208 14 156710 shawnsingh 2012-08-06 09:40:03 -0700 Created attachment 156710 Patch for landing 688213 15 156710 webkit.review.bot 2012-08-06 09:41:14 -0700 Comment on attachment 156710 Patch for landing Rejecting attachment 156710 from commit-queue. Failed to run "['/mnt/git/webkit-commit-queue/Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '-..." exit_code: 2 Last 500 characters of output: 1 cwd: /mnt/git/webkit-commit-queue/ Parsed 2 diffs from patch file(s). patching file Source/WebCore/ChangeLog Hunk #1 FAILED at 3. 1 out of 1 hunk FAILED -- saving rejects to file Source/WebCore/ChangeLog.rej patching file Source/WebKit/chromium/ChangeLog Hunk #1 FAILED at 3. 1 out of 1 hunk FAILED -- saving rejects to file Source/WebKit/chromium/ChangeLog.rej Failed to run "[u'/mnt/git/webkit-commit-queue/Tools/Scripts/svn-apply', u'--force']" exit_code: 1 cwd: /mnt/git/webkit-commit-queue/ Full output: http://queues.webkit.org/results/13443446 688216 16 shawnsingh 2012-08-06 09:43:44 -0700 > > I was asking about m33=1 vs. m33=0, not about whether flattening needs to be applied. Just trying to understand the math. > > > Oh I see what you're asking - I think you're right, the screen space transform doesn't need it. but I'll try it out and make sure. So, it looks like the screenSpaceTransform does actually require the m33==1 part, too. The reason is that the inverse screen-space transform is being used by CCOcclusionTracker, with a project operation. The projection would be incorrect if the z-value is not correctly preserved. So I guess that's additional (and probably more important) explanation about why we need to do this special flatten operation, rather than the pure orthographic projection. Nothing changes in the patch based on this new point, it's just about our conceptual understanding. Thanks for reviewing! 688220 17 156713 shawnsingh 2012-08-06 09:44:23 -0700 Created attachment 156713 Patch for landing 688263 18 156713 webkit.review.bot 2012-08-06 10:43:18 -0700 Comment on attachment 156713 Patch for landing Clearing flags on attachment: 156713 Committed r124782: <http://trac.webkit.org/changeset/124782> 688264 19 webkit.review.bot 2012-08-06 10:43:22 -0700 All reviewed patches have been landed. Closing bug. 156117 2012-08-02 10:30:29 -0700 2012-08-06 09:39:59 -0700 Patch bug-85808-20120802103010.patch text/plain 9575 shawnsingh U3VidmVyc2lvbiBSZXZpc2lvbjogMTI0NDc1CmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViQ29yZS9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViQ29yZS9DaGFuZ2VMb2cKaW5kZXggZGMzMWYwZDRkZTk4ZDM2 OGUxNGExMzExZDM2NjljNjBjYTUzODFlMC4uYTk5MmQ3OWFiOTg3OGFiNjI0YmM2YzNjYjIzOGQ3 YWY4NzE5ZDRkNiAxMDA2NDQKLS0tIGEvU291cmNlL1dlYkNvcmUvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJDb3JlL0NoYW5nZUxvZwpAQCAtMSwzICsxLDMxIEBACisyMDEyLTA4LTAyICBTaGF3 biBTaW5naCAgPHNoYXduc2luZ2hAY2hyb21pdW0ub3JnPgorCisgICAgICAgIFtjaHJvbWl1bV0g Tm9uLXByZXNlcnZlcy0zZCByZXF1aXJlcyBleHBsaWNpdCBmbGF0dGVuaW5nIG9mIHNjcmVlbi1z cGFjZSB0cmFuc2Zvcm0uCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVn LmNnaT9pZD04NTgwOAorCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisg ICAgICAgIFdoZW4gY29tcHV0aW5nIHRoZSBzY3JlZW4tc3BhY2UgdHJhbnNmb3JtcywgeiB2YWx1 ZXMgd2VyZSBub3QKKyAgICAgICAgYmVpbmcgZmxhdHRlbmVkIHdoZW4gdGhleSBzaG91bGQgYmUu IFRoaXMgY2F1c2VkIGluY29ycmVjdAorICAgICAgICBjbGlwcGluZyBkdWUgdG8gb2NjbHVzaW9u IHRyYWNraW5nIHRoYXQgcmVsaWVkIG9uIHRoZSBzY3JlZW4tc3BhY2UKKyAgICAgICAgdHJhbnNm b3JtLiBUaGUgZml4IGlzIHRvIGZsYXR0ZW4gdGhlIHNjcmVlbi1zcGFjZSB0cmFuc2Zvcm0ganVz dAorICAgICAgICBsaWtlIHRoZSAic3VibGF5ZXJNYXRyaXgiIGlzIGZsYXR0ZW5lZCwgd2hlbiB0 aGUgbGF5ZXIgZG9lcyBub3QKKyAgICAgICAgcHJlc2VydmUtM2QuCisKKyAgICAgICAgSW4gYWRk aXRpb24gdG8gbWFraW5nIHRoZSBzaW1wbGUgZml4LCBpdCB3YXMgY29udmVuaWVudCB0byBtYWtl IGEKKyAgICAgICAgaGVscGVyIGZ1bmN0aW9uIGZvciB0aGUgZmxhdHRlbmluZyBjb2RlLgorCisg ICAgICAgIEFkZGl0aW9uYWwgdW5pdCB0ZXN0IGFkZGVkIHRvIGV4ZXJjaXNlIHRoYXQgZmxhdHRl bmluZyBjb2RlOgorICAgICAgICAgIENDTGF5ZXJUcmVlSG9zdENvbW1vblRlc3QudmVyaWZ5VHJh bnNmb3Jtc0ZvckZsYXR0ZW5pbmdMYXllcgorCisgICAgICAgICogcGxhdGZvcm0vZ3JhcGhpY3Mv Y2hyb21pdW0vY2MvQ0NMYXllclRyZWVIb3N0Q29tbW9uLmNwcDoKKyAgICAgICAgKFdlYkNvcmU6 OmNhbGN1bGF0ZURyYXdUcmFuc2Zvcm1zSW50ZXJuYWwpOgorICAgICAgICAqIHBsYXRmb3JtL2dy YXBoaWNzL2Nocm9taXVtL2NjL0NDTWF0aFV0aWwuY3BwOgorICAgICAgICAoV2ViQ29yZTo6Q0NN YXRoVXRpbDo6ZmxhdHRlblRyYW5zZm9ybVRvMmQpOgorICAgICAgICAoV2ViQ29yZSk6CisgICAg ICAgICogcGxhdGZvcm0vZ3JhcGhpY3MvY2hyb21pdW0vY2MvQ0NNYXRoVXRpbC5oOgorICAgICAg ICAoQ0NNYXRoVXRpbCk6CisKIDIwMTItMDgtMDIgIEVyaWMgU2VpZGVsICA8ZXJpY0B3ZWJraXQu b3JnPgogCiAgICAgICAgIEFkZCBiYWNrIEFTU0VSVCghbmVlZHNMYXlvdXQpIHRvIFJlbmRlclRh YmxlU2VjdGlvbiB3aGljaCBpcyBub3cgdmFsaWQKZGlmZiAtLWdpdCBhL1NvdXJjZS9XZWJLaXQv Y2hyb21pdW0vQ2hhbmdlTG9nIGIvU291cmNlL1dlYktpdC9jaHJvbWl1bS9DaGFuZ2VMb2cKaW5k ZXggODRhOWZhODEyMGFmOTI4MmVhZDdjZTMyMjY3YTlmOTFlNjEzZmQ0OC4uMzQ3YWU1N2YzNzhj MzhkOGQ0MTRjNGYxNDkzY2RkZmQyYmNkMjg2NCAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdC9j aHJvbWl1bS9DaGFuZ2VMb2cKKysrIGIvU291cmNlL1dlYktpdC9jaHJvbWl1bS9DaGFuZ2VMb2cK QEAgLTEsMyArMSwxMiBAQAorMjAxMi0wOC0wMiAgU2hhd24gU2luZ2ggIDxzaGF3bnNpbmdoQGNo cm9taXVtLm9yZz4KKworICAgICAgICBbY2hyb21pdW1dIE5vbi1wcmVzZXJ2ZXMtM2QgcmVxdWly ZXMgZXhwbGljaXQgZmxhdHRlbmluZyBvZiBzY3JlZW4tc3BhY2UgdHJhbnNmb3JtLgorICAgICAg ICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9ODU4MDgKKworICAgICAg ICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICAqIHRlc3RzL0NDTGF5ZXJU cmVlSG9zdENvbW1vblRlc3QuY3BwOgorCiAyMDEyLTA4LTAyICBIYW5zIFdlbm5ib3JnICA8aGFu c0BjaHJvbWl1bS5vcmc+CiAKICAgICAgICAgU3BlZWNoIEphdmFTY3JpcHQgQVBJOiBGaXJlIHNw ZWVjaCBzdGFydCBldmVudCBhdCB0aGUgc2FtZSB0aW1lIGFzIHNvdW5kIHN0YXJ0IGV2ZW50CmRp ZmYgLS1naXQgYS9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFwaGljcy9jaHJvbWl1bS9jYy9D Q0xheWVyVHJlZUhvc3RDb21tb24uY3BwIGIvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3JhcGhp Y3MvY2hyb21pdW0vY2MvQ0NMYXllclRyZWVIb3N0Q29tbW9uLmNwcAppbmRleCAwMDk0NmY0MGY1 NjIyZWY1MzFiYTFlMjQ4NDY4MmI0YTZlODIwZmI5Li4yYTc5YWVhYWNlYTU1NmIwYWEzYzAxZWJj Y2Y2MTdiNTVlOTFlMGE0IDEwMDY0NAotLS0gYS9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFw aGljcy9jaHJvbWl1bS9jYy9DQ0xheWVyVHJlZUhvc3RDb21tb24uY3BwCisrKyBiL1NvdXJjZS9X ZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNzL2Nocm9taXVtL2NjL0NDTGF5ZXJUcmVlSG9zdENvbW1v bi5jcHAKQEAgLTU1Niw2ICs1NTYsOCBAQCBzdGF0aWMgdm9pZCBjYWxjdWxhdGVEcmF3VHJhbnNm b3Jtc0ludGVybmFsKExheWVyVHlwZSogbGF5ZXIsIExheWVyVHlwZSogcm9vdExheQogCiAgICAg Ly8gbGF5ZXJTY3JlZW5TcGFjZVRyYW5zZm9ybSByZXByZXNlbnRzIHRoZSB0cmFuc2Zvcm0gYmV0 d2VlbiByb290IGxheWVyJ3MgInNjcmVlbiBzcGFjZSIgYW5kIGxvY2FsIGNvbnRlbnQgc3BhY2Uu CiAgICAgV2ViVHJhbnNmb3JtYXRpb25NYXRyaXggbGF5ZXJTY3JlZW5TcGFjZVRyYW5zZm9ybSA9 IGZ1bGxIaWVyYXJjaHlNYXRyaXg7CisgICAgaWYgKCFsYXllci0+cHJlc2VydmVzM0QoKSkKKyAg ICAgICAgQ0NNYXRoVXRpbDo6ZmxhdHRlblRyYW5zZm9ybVRvMmQobGF5ZXJTY3JlZW5TcGFjZVRy YW5zZm9ybSk7CiAgICAgbGF5ZXJTY3JlZW5TcGFjZVRyYW5zZm9ybS5tdWx0aXBseShkcmF3VHJh bnNmb3JtKTsKICAgICBsYXllci0+c2V0U2NyZWVuU3BhY2VUcmFuc2Zvcm0obGF5ZXJTY3JlZW5T cGFjZVRyYW5zZm9ybSk7CiAKQEAgLTY3OSwxNSArNjgxLDggQEAgc3RhdGljIHZvaWQgY2FsY3Vs YXRlRHJhd1RyYW5zZm9ybXNJbnRlcm5hbChMYXllclR5cGUqIGxheWVyLCBMYXllclR5cGUqIHJv b3RMYXkKICAgICB9CiAKICAgICAvLyBGbGF0dGVuIHRvIDJEIGlmIHRoZSBsYXllciBkb2Vzbid0 IHByZXNlcnZlIDNELgotICAgIGlmICghbGF5ZXItPnByZXNlcnZlczNEKCkpIHsKLSAgICAgICAg c3VibGF5ZXJNYXRyaXguc2V0TTEzKDApOwotICAgICAgICBzdWJsYXllck1hdHJpeC5zZXRNMjMo MCk7Ci0gICAgICAgIHN1YmxheWVyTWF0cml4LnNldE0zMSgwKTsKLSAgICAgICAgc3VibGF5ZXJN YXRyaXguc2V0TTMyKDApOwotICAgICAgICBzdWJsYXllck1hdHJpeC5zZXRNMzMoMSk7Ci0gICAg ICAgIHN1YmxheWVyTWF0cml4LnNldE0zNCgwKTsKLSAgICAgICAgc3VibGF5ZXJNYXRyaXguc2V0 TTQzKDApOwotICAgIH0KKyAgICBpZiAoIWxheWVyLT5wcmVzZXJ2ZXMzRCgpKQorICAgICAgICBD Q01hdGhVdGlsOjpmbGF0dGVuVHJhbnNmb3JtVG8yZChzdWJsYXllck1hdHJpeCk7CiAKICAgICAv LyBBcHBseSB0aGUgc3VibGF5ZXIgdHJhbnNmb3JtIGF0IHRoZSBjZW50ZXIgb2YgdGhlIGxheWVy LgogICAgIHN1YmxheWVyTWF0cml4Lm11bHRpcGx5KGxheWVyLT5zdWJsYXllclRyYW5zZm9ybSgp KTsKZGlmZiAtLWdpdCBhL1NvdXJjZS9XZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNzL2Nocm9taXVt L2NjL0NDTWF0aFV0aWwuY3BwIGIvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3JhcGhpY3MvY2hy b21pdW0vY2MvQ0NNYXRoVXRpbC5jcHAKaW5kZXggZjliZTc2N2E4MTY3MWZiNDk1YTc1YjdkN2Fk NTIyNjU1ZjU5NGYxYy4uNGQ1ZWM5MzNkN2IxMDBjZDQ5ZWMzOWE2OGQyNWUzNjU0MzlkMmQ2MCAx MDA2NDQKLS0tIGEvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3JhcGhpY3MvY2hyb21pdW0vY2Mv Q0NNYXRoVXRpbC5jcHAKKysrIGIvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3JhcGhpY3MvY2hy b21pdW0vY2MvQ0NNYXRoVXRpbC5jcHAKQEAgLTM2MCw0ICszNjAsMjUgQEAgRmxvYXRQb2ludCBD Q01hdGhVdGlsOjpwcm9qZWN0UG9pbnQoY29uc3QgV2ViVHJhbnNmb3JtYXRpb25NYXRyaXgmIHRy YW5zZm9ybSwgY28KICAgICByZXR1cm4gaC5jYXJ0ZXNpYW5Qb2ludDJkKCk7CiB9CiAKK3ZvaWQg Q0NNYXRoVXRpbDo6ZmxhdHRlblRyYW5zZm9ybVRvMmQoV2ViVHJhbnNmb3JtYXRpb25NYXRyaXgm IHRyYW5zZm9ybSkKK3sKKyAgICAvLyBTZXQgYm90aCB0aGUgM3JkIHJvdyBhbmQgM3JkIGNvbHVt biB0byAoMCwgMCwgMSwgMCkuCisgICAgLy8KKyAgICAvLyBPbmUgdXNlZnVsIGludGVycHJldGF0 aW9uIG9mIGRvaW5nIHRoaXMgb3BlcmF0aW9uOgorICAgIC8vICAtIEZvciB4IGFuZCB5IHZhbHVl cywgdGhlIG5ldyB0cmFuc2Zvcm0gYmVoYXZlcyBlZmZlY3RpdmVseSBsaWtlIGFuIG9ydGhvZ3Jh cGhpYworICAgIC8vICAgIHByb2plY3Rpb24gd2FzIGFkZGVkIHRvIHRoZSBtYXRyaXggc2VxdWVu Y2UuCisgICAgLy8gIC0gRm9yIHogdmFsdWVzLCB0aGUgbmV3IHRyYW5zZm9ybSBvdmVycmlkZXMg YW55IGVmZmVjdCB0aGF0IHRoZSB0cmFuc2Zvcm0gaGFkIG9uCisgICAgLy8gICAgeiwgYW5kIGlu c3RlYWQgaXQgcHJlc2VydmVzIHRoZSB6IHZhbHVlIGZvciBhbnkgcG9pbnRzIHRoYXQgYXJlIHRy YW5zZm9ybWVkLgorICAgIC8vICAtIEJlY2F1c2Ugb2YgbGluZWFyaXR5IG9mIHRyYW5zZm9ybXMs IHRoaXMgZmxhdHRlbmVkIHRyYW5zZm9ybSBhbHNvIHByZXNlcnZlcyB0aGUKKyAgICAvLyAgICBl ZmZlY3QgdGhhdCBhbnkgc3Vic2VxdWVudCAocG9zdC1tdWx0aXBsaWVkKSB0cmFuc2Zvcm1zIHdv dWxkIGhhdmUgb24geiB2YWx1ZXMuCisgICAgLy8KKyAgICB0cmFuc2Zvcm0uc2V0TTEzKDApOwor ICAgIHRyYW5zZm9ybS5zZXRNMjMoMCk7CisgICAgdHJhbnNmb3JtLnNldE0zMSgwKTsKKyAgICB0 cmFuc2Zvcm0uc2V0TTMyKDApOworICAgIHRyYW5zZm9ybS5zZXRNMzMoMSk7CisgICAgdHJhbnNm b3JtLnNldE0zNCgwKTsKKyAgICB0cmFuc2Zvcm0uc2V0TTQzKDApOworfQorCiB9IC8vIG5hbWVz cGFjZSBXZWJDb3JlCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFwaGlj cy9jaHJvbWl1bS9jYy9DQ01hdGhVdGlsLmggYi9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFw aGljcy9jaHJvbWl1bS9jYy9DQ01hdGhVdGlsLmgKaW5kZXggMDRlODUwZWJkNmI0MjNjNTMzMzAw YWQ1YWRiZDUwMTExZDVlNTNjOC4uM2ZiNTkxNDI5MGEyYTc1YmQ1MjVmM2U0ODQzNjU5NGNhOGU4 NDUwYSAxMDA2NDQKLS0tIGEvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3JhcGhpY3MvY2hyb21p dW0vY2MvQ0NNYXRoVXRpbC5oCisrKyBiL1NvdXJjZS9XZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNz L2Nocm9taXVtL2NjL0NDTWF0aFV0aWwuaApAQCAtMTEzLDYgKzExMyw4IEBAIHB1YmxpYzoKICAg ICBzdGF0aWMgRmxvYXRQb2ludDNEIG1hcFBvaW50KGNvbnN0IFdlYktpdDo6V2ViVHJhbnNmb3Jt YXRpb25NYXRyaXgmLCBjb25zdCBGbG9hdFBvaW50M0QmLCBib29sJiBjbGlwcGVkKTsKICAgICBz dGF0aWMgRmxvYXRRdWFkIHByb2plY3RRdWFkKGNvbnN0IFdlYktpdDo6V2ViVHJhbnNmb3JtYXRp b25NYXRyaXgmLCBjb25zdCBGbG9hdFF1YWQmLCBib29sJiBjbGlwcGVkKTsKICAgICBzdGF0aWMg RmxvYXRQb2ludCBwcm9qZWN0UG9pbnQoY29uc3QgV2ViS2l0OjpXZWJUcmFuc2Zvcm1hdGlvbk1h dHJpeCYsIGNvbnN0IEZsb2F0UG9pbnQmLCBib29sJiBjbGlwcGVkKTsKKworICAgIHN0YXRpYyB2 b2lkIGZsYXR0ZW5UcmFuc2Zvcm1UbzJkKFdlYktpdDo6V2ViVHJhbnNmb3JtYXRpb25NYXRyaXgm KTsKIH07CiAKIH0gLy8gbmFtZXNwYWNlIFdlYkNvcmUKZGlmZiAtLWdpdCBhL1NvdXJjZS9XZWJL aXQvY2hyb21pdW0vdGVzdHMvQ0NMYXllclRyZWVIb3N0Q29tbW9uVGVzdC5jcHAgYi9Tb3VyY2Uv V2ViS2l0L2Nocm9taXVtL3Rlc3RzL0NDTGF5ZXJUcmVlSG9zdENvbW1vblRlc3QuY3BwCmluZGV4 IDNlZDJkNjBlMmM1OTc1NTM2YmZjNGI0NmQ1NDkxM2I2YmE2NTFjZjEuLjA4ZGMxOGZjNGUzYTc1 MGFiZDg0YTE3ODBmNjRhNDhkOGI5NTIyMjkgMTAwNjQ0Ci0tLSBhL1NvdXJjZS9XZWJLaXQvY2hy b21pdW0vdGVzdHMvQ0NMYXllclRyZWVIb3N0Q29tbW9uVGVzdC5jcHAKKysrIGIvU291cmNlL1dl YktpdC9jaHJvbWl1bS90ZXN0cy9DQ0xheWVyVHJlZUhvc3RDb21tb25UZXN0LmNwcApAQCAtOTAz LDYgKzkwMyw1MSBAQCBURVNUKENDTGF5ZXJUcmVlSG9zdENvbW1vblRlc3QsIHZlcmlmeVRyYW5z Zm9ybXNGb3JSZW5kZXJTdXJmYWNlSGllcmFyY2h5KQogICAgIEVYUEVDVF9GTE9BVF9FUSg1LCBn cmFuZENoaWxkT2ZSUzItPnNjcmVlblNwYWNlVHJhbnNmb3JtKCkubTQyKCkpOwogfQogCitURVNU KENDTGF5ZXJUcmVlSG9zdENvbW1vblRlc3QsIHZlcmlmeVRyYW5zZm9ybXNGb3JGbGF0dGVuaW5n TGF5ZXIpCit7CisgICAgLy8gRm9yIGxheWVycyB0aGF0IGZsYXR0ZW4gdGhlaXIgc3VidHJlZSwg dGhlcmUgc2hvdWxkIGJlIGFuIG9ydGhvZ3JhcGhpYyBwcm9qZWN0aW9uCisgICAgLy8gKGZvciB4 IGFuZCB5IHZhbHVlcykgaW4gdGhlIG1pZGRsZSBvZiB0aGUgdHJhbnNmb3JtIHNlcXVlbmNlLiBO b3RlIHRoYXQgdGhlIHdheSB0aGUKKyAgICAvLyBjb2RlIGlzIGN1cnJlbnRseSBpbXBsZW1lbnRl ZCwgaXQgaXMgbm90IGV4cGVjdGVkIHRvIHVzZSBhIGNhbm9uaWNhbCBvcnRob2dyYXBoaWMKKyAg ICAvLyBwcm9qZWN0aW9uLgorCisgICAgUmVmUHRyPExheWVyQ2hyb21pdW0+IHJvb3QgPSBMYXll ckNocm9taXVtOjpjcmVhdGUoKTsKKyAgICBSZWZQdHI8TGF5ZXJDaHJvbWl1bT4gY2hpbGQgPSBM YXllckNocm9taXVtOjpjcmVhdGUoKTsKKyAgICBSZWZQdHI8TGF5ZXJDaHJvbWl1bVdpdGhGb3Jj ZWREcmF3c0NvbnRlbnQ+IGdyYW5kQ2hpbGQgPSBhZG9wdFJlZihuZXcgTGF5ZXJDaHJvbWl1bVdp dGhGb3JjZWREcmF3c0NvbnRlbnQoKSk7CisKKyAgICBXZWJUcmFuc2Zvcm1hdGlvbk1hdHJpeCBy b3RhdGlvbkFib3V0WUF4aXM7CisgICAgcm90YXRpb25BYm91dFlBeGlzLnJvdGF0ZTNkKDAsIDMw LCAwKTsKKworICAgIGNvbnN0IFdlYlRyYW5zZm9ybWF0aW9uTWF0cml4IGlkZW50aXR5TWF0cml4 OworICAgIHNldExheWVyUHJvcGVydGllc0ZvclRlc3Rpbmcocm9vdC5nZXQoKSwgaWRlbnRpdHlN YXRyaXgsIGlkZW50aXR5TWF0cml4LCBGbG9hdFBvaW50Ojp6ZXJvKCksIEZsb2F0UG9pbnQ6Onpl cm8oKSwgSW50U2l6ZSgxMDAsIDEwMCksIGZhbHNlKTsKKyAgICBzZXRMYXllclByb3BlcnRpZXNG b3JUZXN0aW5nKGNoaWxkLmdldCgpLCByb3RhdGlvbkFib3V0WUF4aXMsIGlkZW50aXR5TWF0cml4 LCBGbG9hdFBvaW50Ojp6ZXJvKCksIEZsb2F0UG9pbnQ6Onplcm8oKSwgSW50U2l6ZSgxMCwgMTAp LCBmYWxzZSk7CisgICAgc2V0TGF5ZXJQcm9wZXJ0aWVzRm9yVGVzdGluZyhncmFuZENoaWxkLmdl dCgpLCByb3RhdGlvbkFib3V0WUF4aXMsIGlkZW50aXR5TWF0cml4LCBGbG9hdFBvaW50Ojp6ZXJv KCksIEZsb2F0UG9pbnQ6Onplcm8oKSwgSW50U2l6ZSgxMCwgMTApLCBmYWxzZSk7CisKKyAgICBy b290LT5hZGRDaGlsZChjaGlsZCk7CisgICAgY2hpbGQtPmFkZENoaWxkKGdyYW5kQ2hpbGQpOwor ICAgIGNoaWxkLT5zZXRGb3JjZVJlbmRlclN1cmZhY2UodHJ1ZSk7CisKKyAgICAvLyBObyBsYXll cnMgaW4gdGhpcyB0ZXN0IHNob3VsZCBwcmVzZXJ2ZSAzZC4KKyAgICBBU1NFUlRfRkFMU0Uocm9v dC0+cHJlc2VydmVzM0QoKSk7CisgICAgQVNTRVJUX0ZBTFNFKGNoaWxkLT5wcmVzZXJ2ZXMzRCgp KTsKKyAgICBBU1NFUlRfRkFMU0UoZ3JhbmRDaGlsZC0+cHJlc2VydmVzM0QoKSk7CisKKyAgICBX ZWJUcmFuc2Zvcm1hdGlvbk1hdHJpeCBleHBlY3RlZENoaWxkRHJhd1RyYW5zZm9ybSA9IHJvdGF0 aW9uQWJvdXRZQXhpczsKKyAgICBXZWJUcmFuc2Zvcm1hdGlvbk1hdHJpeCBleHBlY3RlZENoaWxk U2NyZWVuU3BhY2VUcmFuc2Zvcm0gPSByb3RhdGlvbkFib3V0WUF4aXM7CisgICAgV2ViVHJhbnNm b3JtYXRpb25NYXRyaXggZXhwZWN0ZWRHcmFuZENoaWxkRHJhd1RyYW5zZm9ybSA9IHJvdGF0aW9u QWJvdXRZQXhpczsgLy8gZHJhd3Mgb250byBjaGlsZCdzIHJlbmRlclN1cmZhY2UKKyAgICBXZWJU cmFuc2Zvcm1hdGlvbk1hdHJpeCBleHBlY3RlZEdyYW5kQ2hpbGRTY3JlZW5TcGFjZVRyYW5zZm9y bSA9IHJvdGF0aW9uQWJvdXRZQXhpcy50bzJkVHJhbnNmb3JtKCkgKiByb3RhdGlvbkFib3V0WUF4 aXM7CisKKyAgICBleGVjdXRlQ2FsY3VsYXRlRHJhd1RyYW5zZm9ybXNBbmRWaXNpYmlsaXR5KHJv b3QuZ2V0KCkpOworCisgICAgLy8gVGhlIGNoaWxkJ3MgZHJhd1RyYW5zZm9ybSBzaG91bGQgaGF2 ZSBiZWVuIHRha2VuIGJ5IGl0cyBzdXJmYWNlLgorICAgIEFTU0VSVF9UUlVFKGNoaWxkLT5yZW5k ZXJTdXJmYWNlKCkpOworICAgIEVYUEVDVF9UUkFOU0ZPUk1BVElPTl9NQVRSSVhfRVEoZXhwZWN0 ZWRDaGlsZERyYXdUcmFuc2Zvcm0sIGNoaWxkLT5yZW5kZXJTdXJmYWNlKCktPmRyYXdUcmFuc2Zv cm0oKSk7CisgICAgRVhQRUNUX1RSQU5TRk9STUFUSU9OX01BVFJJWF9FUShleHBlY3RlZENoaWxk U2NyZWVuU3BhY2VUcmFuc2Zvcm0sIGNoaWxkLT5yZW5kZXJTdXJmYWNlKCktPnNjcmVlblNwYWNl VHJhbnNmb3JtKCkpOworICAgIEVYUEVDVF9UUkFOU0ZPUk1BVElPTl9NQVRSSVhfRVEoaWRlbnRp dHlNYXRyaXgsIGNoaWxkLT5kcmF3VHJhbnNmb3JtKCkpOworICAgIEVYUEVDVF9UUkFOU0ZPUk1B VElPTl9NQVRSSVhfRVEoZXhwZWN0ZWRDaGlsZFNjcmVlblNwYWNlVHJhbnNmb3JtLCBjaGlsZC0+ c2NyZWVuU3BhY2VUcmFuc2Zvcm0oKSk7CisgICAgRVhQRUNUX1RSQU5TRk9STUFUSU9OX01BVFJJ WF9FUShleHBlY3RlZEdyYW5kQ2hpbGREcmF3VHJhbnNmb3JtLCBncmFuZENoaWxkLT5kcmF3VHJh bnNmb3JtKCkpOworICAgIEVYUEVDVF9UUkFOU0ZPUk1BVElPTl9NQVRSSVhfRVEoZXhwZWN0ZWRH cmFuZENoaWxkU2NyZWVuU3BhY2VUcmFuc2Zvcm0sIGdyYW5kQ2hpbGQtPnNjcmVlblNwYWNlVHJh bnNmb3JtKCkpOworfQorCiBURVNUKENDTGF5ZXJUcmVlSG9zdENvbW1vblRlc3QsIHZlcmlmeVJl bmRlclN1cmZhY2VMaXN0Rm9yUmVuZGVyU3VyZmFjZVdpdGhDbGlwcGVkTGF5ZXIpCiB7CiAgICAg UmVmUHRyPExheWVyQ2hyb21pdW0+IHBhcmVudCA9IExheWVyQ2hyb21pdW06OmNyZWF0ZSgpOwo= 156710 2012-08-06 09:40:03 -0700 2012-08-06 09:44:20 -0700 Patch for landing bug-85808-20120806093939.patch text/plain 1181 shawnsingh U3VidmVyc2lvbiBSZXZpc2lvbjogMTI0Nzc0CmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViQ29yZS9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViQ29yZS9DaGFuZ2VMb2cKaW5kZXggZjM3MzdkYjVlMzE3ZmMw NTY0NTZmNjc2M2VjMGVhODhiNDQ4NTczYi4uOWNiOTE0ZTU2OGMwNjY3OTllNjE3Y2JhMzJmZGVj OTA0ODM0NzJmOCAxMDA2NDQKLS0tIGEvU291cmNlL1dlYkNvcmUvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJDb3JlL0NoYW5nZUxvZwpAQCAtMyw3ICszLDcgQEAKICAgICAgICAgW2Nocm9taXVt XSBOb24tcHJlc2VydmVzLTNkIHJlcXVpcmVzIGV4cGxpY2l0IGZsYXR0ZW5pbmcgb2Ygc2NyZWVu LXNwYWNlIHRyYW5zZm9ybS4KICAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19i dWcuY2dpP2lkPTg1ODA4CiAKLSAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisg ICAgICAgIFJldmlld2VkIGJ5IEFkcmllbm5lIFdhbGtlci4KIAogICAgICAgICBXaGVuIGNvbXB1 dGluZyB0aGUgc2NyZWVuLXNwYWNlIHRyYW5zZm9ybXMsIHogdmFsdWVzIHdlcmUgbm90CiAgICAg ICAgIGJlaW5nIGZsYXR0ZW5lZCB3aGVuIHRoZXkgc2hvdWxkIGJlLiBUaGlzIGNhdXNlZCBpbmNv cnJlY3QKZGlmZiAtLWdpdCBhL1NvdXJjZS9XZWJLaXQvY2hyb21pdW0vQ2hhbmdlTG9nIGIvU291 cmNlL1dlYktpdC9jaHJvbWl1bS9DaGFuZ2VMb2cKaW5kZXggNzE0ZTFlNjgyZDU3OWQzZjdiNWQy M2QxZDM5M2U1Y2Q4YWU5NThlYy4uNGM2MGFmY2QxZTJiZjM3NjYxN2Q0MGYwYWE5OTM5YzZjZTA5 NjhmNyAxMDA2NDQKLS0tIGEvU291cmNlL1dlYktpdC9jaHJvbWl1bS9DaGFuZ2VMb2cKKysrIGIv U291cmNlL1dlYktpdC9jaHJvbWl1bS9DaGFuZ2VMb2cKQEAgLTMsNyArMyw3IEBACiAgICAgICAg IFtjaHJvbWl1bV0gTm9uLXByZXNlcnZlcy0zZCByZXF1aXJlcyBleHBsaWNpdCBmbGF0dGVuaW5n IG9mIHNjcmVlbi1zcGFjZSB0cmFuc2Zvcm0uCiAgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQu b3JnL3Nob3dfYnVnLmNnaT9pZD04NTgwOAogCi0gICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAo T09QUyEpLgorICAgICAgICBSZXZpZXdlZCBieSBBZHJpZW5uZSBXYWxrZXIuCiAKICAgICAgICAg KiB0ZXN0cy9DQ0xheWVyVHJlZUhvc3RDb21tb25UZXN0LmNwcDoKIAo= 156713 2012-08-06 09:44:23 -0700 2012-08-06 10:43:18 -0700 Patch for landing bug-85808-20120806094400.patch text/plain 9573 shawnsingh U3VidmVyc2lvbiBSZXZpc2lvbjogMTI0Nzc0CmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViQ29yZS9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViQ29yZS9DaGFuZ2VMb2cKaW5kZXggYmZhMTcxMmUzNzY4ZTc1 NjViODg1YWUxOGM1NWEzMTBkZGQ3MzkzZS4uOWNiOTE0ZTU2OGMwNjY3OTllNjE3Y2JhMzJmZGVj OTA0ODM0NzJmOCAxMDA2NDQKLS0tIGEvU291cmNlL1dlYkNvcmUvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJDb3JlL0NoYW5nZUxvZwpAQCAtMSwzICsxLDMxIEBACisyMDEyLTA4LTAyICBTaGF3 biBTaW5naCAgPHNoYXduc2luZ2hAY2hyb21pdW0ub3JnPgorCisgICAgICAgIFtjaHJvbWl1bV0g Tm9uLXByZXNlcnZlcy0zZCByZXF1aXJlcyBleHBsaWNpdCBmbGF0dGVuaW5nIG9mIHNjcmVlbi1z cGFjZSB0cmFuc2Zvcm0uCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVn LmNnaT9pZD04NTgwOAorCisgICAgICAgIFJldmlld2VkIGJ5IEFkcmllbm5lIFdhbGtlci4KKwor ICAgICAgICBXaGVuIGNvbXB1dGluZyB0aGUgc2NyZWVuLXNwYWNlIHRyYW5zZm9ybXMsIHogdmFs dWVzIHdlcmUgbm90CisgICAgICAgIGJlaW5nIGZsYXR0ZW5lZCB3aGVuIHRoZXkgc2hvdWxkIGJl LiBUaGlzIGNhdXNlZCBpbmNvcnJlY3QKKyAgICAgICAgY2xpcHBpbmcgZHVlIHRvIG9jY2x1c2lv biB0cmFja2luZyB0aGF0IHJlbGllZCBvbiB0aGUgc2NyZWVuLXNwYWNlCisgICAgICAgIHRyYW5z Zm9ybS4gVGhlIGZpeCBpcyB0byBmbGF0dGVuIHRoZSBzY3JlZW4tc3BhY2UgdHJhbnNmb3JtIGp1 c3QKKyAgICAgICAgbGlrZSB0aGUgInN1YmxheWVyTWF0cml4IiBpcyBmbGF0dGVuZWQsIHdoZW4g dGhlIGxheWVyIGRvZXMgbm90CisgICAgICAgIHByZXNlcnZlLTNkLgorCisgICAgICAgIEluIGFk ZGl0aW9uIHRvIG1ha2luZyB0aGUgc2ltcGxlIGZpeCwgaXQgd2FzIGNvbnZlbmllbnQgdG8gbWFr ZSBhCisgICAgICAgIGhlbHBlciBmdW5jdGlvbiBmb3IgdGhlIGZsYXR0ZW5pbmcgY29kZS4KKwor ICAgICAgICBBZGRpdGlvbmFsIHVuaXQgdGVzdCBhZGRlZCB0byBleGVyY2lzZSB0aGF0IGZsYXR0 ZW5pbmcgY29kZToKKyAgICAgICAgICBDQ0xheWVyVHJlZUhvc3RDb21tb25UZXN0LnZlcmlmeVRy YW5zZm9ybXNGb3JGbGF0dGVuaW5nTGF5ZXIKKworICAgICAgICAqIHBsYXRmb3JtL2dyYXBoaWNz L2Nocm9taXVtL2NjL0NDTGF5ZXJUcmVlSG9zdENvbW1vbi5jcHA6CisgICAgICAgIChXZWJDb3Jl OjpjYWxjdWxhdGVEcmF3VHJhbnNmb3Jtc0ludGVybmFsKToKKyAgICAgICAgKiBwbGF0Zm9ybS9n cmFwaGljcy9jaHJvbWl1bS9jYy9DQ01hdGhVdGlsLmNwcDoKKyAgICAgICAgKFdlYkNvcmU6OkND TWF0aFV0aWw6OmZsYXR0ZW5UcmFuc2Zvcm1UbzJkKToKKyAgICAgICAgKFdlYkNvcmUpOgorICAg ICAgICAqIHBsYXRmb3JtL2dyYXBoaWNzL2Nocm9taXVtL2NjL0NDTWF0aFV0aWwuaDoKKyAgICAg ICAgKENDTWF0aFV0aWwpOgorCiAyMDEyLTA4LTA2ICBIaXJvbm9yaSBCb25vICA8aGJvbm9AY2hy b21pdW0ub3JnPgogCiAgICAgICAgIFJvbGxpbmcgb3V0IG15IHIxMjMwNjcgYW5kIHIxMjM1NzIK ZGlmZiAtLWdpdCBhL1NvdXJjZS9XZWJLaXQvY2hyb21pdW0vQ2hhbmdlTG9nIGIvU291cmNlL1dl YktpdC9jaHJvbWl1bS9DaGFuZ2VMb2cKaW5kZXggMTc0MTNkZTg0ZTVlMzI2Y2VkNmI4ODUwY2Iw ODRhYTk3ZTUzZWZlYS4uNGM2MGFmY2QxZTJiZjM3NjYxN2Q0MGYwYWE5OTM5YzZjZTA5NjhmNyAx MDA2NDQKLS0tIGEvU291cmNlL1dlYktpdC9jaHJvbWl1bS9DaGFuZ2VMb2cKKysrIGIvU291cmNl L1dlYktpdC9jaHJvbWl1bS9DaGFuZ2VMb2cKQEAgLTEsMyArMSwxMiBAQAorMjAxMi0wOC0wMiAg U2hhd24gU2luZ2ggIDxzaGF3bnNpbmdoQGNocm9taXVtLm9yZz4KKworICAgICAgICBbY2hyb21p dW1dIE5vbi1wcmVzZXJ2ZXMtM2QgcmVxdWlyZXMgZXhwbGljaXQgZmxhdHRlbmluZyBvZiBzY3Jl ZW4tc3BhY2UgdHJhbnNmb3JtLgorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93 X2J1Zy5jZ2k/aWQ9ODU4MDgKKworICAgICAgICBSZXZpZXdlZCBieSBBZHJpZW5uZSBXYWxrZXIu CisKKyAgICAgICAgKiB0ZXN0cy9DQ0xheWVyVHJlZUhvc3RDb21tb25UZXN0LmNwcDoKKwogMjAx Mi0wOC0wNiAgWXVyeSBTZW1pa2hhdHNreSAgPHl1cnlzQGNocm9taXVtLm9yZz4KIAogICAgICAg ICBXZWIgSW5zcGVjdG9yOiByZW5hbWUgV29ya2VyQWdlbnQuc2V0V29ya2VySW5zcGVjdGlvbkVu YWJsZWQgdG8gV29ya2VyQWdlbnQuZW5hYmxlIGFuZCBtYWtlIGl0IHJldHVybiBlcnJvcgpkaWZm IC0tZ2l0IGEvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3JhcGhpY3MvY2hyb21pdW0vY2MvQ0NM YXllclRyZWVIb3N0Q29tbW9uLmNwcCBiL1NvdXJjZS9XZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNz L2Nocm9taXVtL2NjL0NDTGF5ZXJUcmVlSG9zdENvbW1vbi5jcHAKaW5kZXggMDA5NDZmNDBmNTYy MmVmNTMxYmExZTI0ODQ2ODJiNGE2ZTgyMGZiOS4uMmE3OWFlYWFjZWE1NTZiMGFhM2MwMWViY2Nm NjE3YjU1ZTkxZTBhNCAxMDA2NDQKLS0tIGEvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3JhcGhp Y3MvY2hyb21pdW0vY2MvQ0NMYXllclRyZWVIb3N0Q29tbW9uLmNwcAorKysgYi9Tb3VyY2UvV2Vi Q29yZS9wbGF0Zm9ybS9ncmFwaGljcy9jaHJvbWl1bS9jYy9DQ0xheWVyVHJlZUhvc3RDb21tb24u Y3BwCkBAIC01NTYsNiArNTU2LDggQEAgc3RhdGljIHZvaWQgY2FsY3VsYXRlRHJhd1RyYW5zZm9y bXNJbnRlcm5hbChMYXllclR5cGUqIGxheWVyLCBMYXllclR5cGUqIHJvb3RMYXkKIAogICAgIC8v IGxheWVyU2NyZWVuU3BhY2VUcmFuc2Zvcm0gcmVwcmVzZW50cyB0aGUgdHJhbnNmb3JtIGJldHdl ZW4gcm9vdCBsYXllcidzICJzY3JlZW4gc3BhY2UiIGFuZCBsb2NhbCBjb250ZW50IHNwYWNlLgog ICAgIFdlYlRyYW5zZm9ybWF0aW9uTWF0cml4IGxheWVyU2NyZWVuU3BhY2VUcmFuc2Zvcm0gPSBm dWxsSGllcmFyY2h5TWF0cml4OworICAgIGlmICghbGF5ZXItPnByZXNlcnZlczNEKCkpCisgICAg ICAgIENDTWF0aFV0aWw6OmZsYXR0ZW5UcmFuc2Zvcm1UbzJkKGxheWVyU2NyZWVuU3BhY2VUcmFu c2Zvcm0pOwogICAgIGxheWVyU2NyZWVuU3BhY2VUcmFuc2Zvcm0ubXVsdGlwbHkoZHJhd1RyYW5z Zm9ybSk7CiAgICAgbGF5ZXItPnNldFNjcmVlblNwYWNlVHJhbnNmb3JtKGxheWVyU2NyZWVuU3Bh Y2VUcmFuc2Zvcm0pOwogCkBAIC02NzksMTUgKzY4MSw4IEBAIHN0YXRpYyB2b2lkIGNhbGN1bGF0 ZURyYXdUcmFuc2Zvcm1zSW50ZXJuYWwoTGF5ZXJUeXBlKiBsYXllciwgTGF5ZXJUeXBlKiByb290 TGF5CiAgICAgfQogCiAgICAgLy8gRmxhdHRlbiB0byAyRCBpZiB0aGUgbGF5ZXIgZG9lc24ndCBw cmVzZXJ2ZSAzRC4KLSAgICBpZiAoIWxheWVyLT5wcmVzZXJ2ZXMzRCgpKSB7Ci0gICAgICAgIHN1 YmxheWVyTWF0cml4LnNldE0xMygwKTsKLSAgICAgICAgc3VibGF5ZXJNYXRyaXguc2V0TTIzKDAp OwotICAgICAgICBzdWJsYXllck1hdHJpeC5zZXRNMzEoMCk7Ci0gICAgICAgIHN1YmxheWVyTWF0 cml4LnNldE0zMigwKTsKLSAgICAgICAgc3VibGF5ZXJNYXRyaXguc2V0TTMzKDEpOwotICAgICAg ICBzdWJsYXllck1hdHJpeC5zZXRNMzQoMCk7Ci0gICAgICAgIHN1YmxheWVyTWF0cml4LnNldE00 MygwKTsKLSAgICB9CisgICAgaWYgKCFsYXllci0+cHJlc2VydmVzM0QoKSkKKyAgICAgICAgQ0NN YXRoVXRpbDo6ZmxhdHRlblRyYW5zZm9ybVRvMmQoc3VibGF5ZXJNYXRyaXgpOwogCiAgICAgLy8g QXBwbHkgdGhlIHN1YmxheWVyIHRyYW5zZm9ybSBhdCB0aGUgY2VudGVyIG9mIHRoZSBsYXllci4K ICAgICBzdWJsYXllck1hdHJpeC5tdWx0aXBseShsYXllci0+c3VibGF5ZXJUcmFuc2Zvcm0oKSk7 CmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFwaGljcy9jaHJvbWl1bS9j Yy9DQ01hdGhVdGlsLmNwcCBiL1NvdXJjZS9XZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNzL2Nocm9t aXVtL2NjL0NDTWF0aFV0aWwuY3BwCmluZGV4IGY5YmU3NjdhODE2NzFmYjQ5NWE3NWI3ZDdhZDUy MjY1NWY1OTRmMWMuLjRkNWVjOTMzZDdiMTAwY2Q0OWVjMzlhNjhkMjVlMzY1NDM5ZDJkNjAgMTAw NjQ0Ci0tLSBhL1NvdXJjZS9XZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNzL2Nocm9taXVtL2NjL0ND TWF0aFV0aWwuY3BwCisrKyBiL1NvdXJjZS9XZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNzL2Nocm9t aXVtL2NjL0NDTWF0aFV0aWwuY3BwCkBAIC0zNjAsNCArMzYwLDI1IEBAIEZsb2F0UG9pbnQgQ0NN YXRoVXRpbDo6cHJvamVjdFBvaW50KGNvbnN0IFdlYlRyYW5zZm9ybWF0aW9uTWF0cml4JiB0cmFu c2Zvcm0sIGNvCiAgICAgcmV0dXJuIGguY2FydGVzaWFuUG9pbnQyZCgpOwogfQogCit2b2lkIEND TWF0aFV0aWw6OmZsYXR0ZW5UcmFuc2Zvcm1UbzJkKFdlYlRyYW5zZm9ybWF0aW9uTWF0cml4JiB0 cmFuc2Zvcm0pCit7CisgICAgLy8gU2V0IGJvdGggdGhlIDNyZCByb3cgYW5kIDNyZCBjb2x1bW4g dG8gKDAsIDAsIDEsIDApLgorICAgIC8vCisgICAgLy8gT25lIHVzZWZ1bCBpbnRlcnByZXRhdGlv biBvZiBkb2luZyB0aGlzIG9wZXJhdGlvbjoKKyAgICAvLyAgLSBGb3IgeCBhbmQgeSB2YWx1ZXMs IHRoZSBuZXcgdHJhbnNmb3JtIGJlaGF2ZXMgZWZmZWN0aXZlbHkgbGlrZSBhbiBvcnRob2dyYXBo aWMKKyAgICAvLyAgICBwcm9qZWN0aW9uIHdhcyBhZGRlZCB0byB0aGUgbWF0cml4IHNlcXVlbmNl LgorICAgIC8vICAtIEZvciB6IHZhbHVlcywgdGhlIG5ldyB0cmFuc2Zvcm0gb3ZlcnJpZGVzIGFu eSBlZmZlY3QgdGhhdCB0aGUgdHJhbnNmb3JtIGhhZCBvbgorICAgIC8vICAgIHosIGFuZCBpbnN0 ZWFkIGl0IHByZXNlcnZlcyB0aGUgeiB2YWx1ZSBmb3IgYW55IHBvaW50cyB0aGF0IGFyZSB0cmFu c2Zvcm1lZC4KKyAgICAvLyAgLSBCZWNhdXNlIG9mIGxpbmVhcml0eSBvZiB0cmFuc2Zvcm1zLCB0 aGlzIGZsYXR0ZW5lZCB0cmFuc2Zvcm0gYWxzbyBwcmVzZXJ2ZXMgdGhlCisgICAgLy8gICAgZWZm ZWN0IHRoYXQgYW55IHN1YnNlcXVlbnQgKHBvc3QtbXVsdGlwbGllZCkgdHJhbnNmb3JtcyB3b3Vs ZCBoYXZlIG9uIHogdmFsdWVzLgorICAgIC8vCisgICAgdHJhbnNmb3JtLnNldE0xMygwKTsKKyAg ICB0cmFuc2Zvcm0uc2V0TTIzKDApOworICAgIHRyYW5zZm9ybS5zZXRNMzEoMCk7CisgICAgdHJh bnNmb3JtLnNldE0zMigwKTsKKyAgICB0cmFuc2Zvcm0uc2V0TTMzKDEpOworICAgIHRyYW5zZm9y bS5zZXRNMzQoMCk7CisgICAgdHJhbnNmb3JtLnNldE00MygwKTsKK30KKwogfSAvLyBuYW1lc3Bh Y2UgV2ViQ29yZQpkaWZmIC0tZ2l0IGEvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3JhcGhpY3Mv Y2hyb21pdW0vY2MvQ0NNYXRoVXRpbC5oIGIvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vZ3JhcGhp Y3MvY2hyb21pdW0vY2MvQ0NNYXRoVXRpbC5oCmluZGV4IDA0ZTg1MGViZDZiNDIzYzUzMzMwMGFk NWFkYmQ1MDExMWQ1ZTUzYzguLjNmYjU5MTQyOTBhMmE3NWJkNTI1ZjNlNDg0MzY1OTRjYThlODQ1 MGEgMTAwNjQ0Ci0tLSBhL1NvdXJjZS9XZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNzL2Nocm9taXVt L2NjL0NDTWF0aFV0aWwuaAorKysgYi9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9ncmFwaGljcy9j aHJvbWl1bS9jYy9DQ01hdGhVdGlsLmgKQEAgLTExMyw2ICsxMTMsOCBAQCBwdWJsaWM6CiAgICAg c3RhdGljIEZsb2F0UG9pbnQzRCBtYXBQb2ludChjb25zdCBXZWJLaXQ6OldlYlRyYW5zZm9ybWF0 aW9uTWF0cml4JiwgY29uc3QgRmxvYXRQb2ludDNEJiwgYm9vbCYgY2xpcHBlZCk7CiAgICAgc3Rh dGljIEZsb2F0UXVhZCBwcm9qZWN0UXVhZChjb25zdCBXZWJLaXQ6OldlYlRyYW5zZm9ybWF0aW9u TWF0cml4JiwgY29uc3QgRmxvYXRRdWFkJiwgYm9vbCYgY2xpcHBlZCk7CiAgICAgc3RhdGljIEZs b2F0UG9pbnQgcHJvamVjdFBvaW50KGNvbnN0IFdlYktpdDo6V2ViVHJhbnNmb3JtYXRpb25NYXRy aXgmLCBjb25zdCBGbG9hdFBvaW50JiwgYm9vbCYgY2xpcHBlZCk7CisKKyAgICBzdGF0aWMgdm9p ZCBmbGF0dGVuVHJhbnNmb3JtVG8yZChXZWJLaXQ6OldlYlRyYW5zZm9ybWF0aW9uTWF0cml4Jik7 CiB9OwogCiB9IC8vIG5hbWVzcGFjZSBXZWJDb3JlCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViS2l0 L2Nocm9taXVtL3Rlc3RzL0NDTGF5ZXJUcmVlSG9zdENvbW1vblRlc3QuY3BwIGIvU291cmNlL1dl YktpdC9jaHJvbWl1bS90ZXN0cy9DQ0xheWVyVHJlZUhvc3RDb21tb25UZXN0LmNwcAppbmRleCAz ZWQyZDYwZTJjNTk3NTUzNmJmYzRiNDZkNTQ5MTNiNmJhNjUxY2YxLi4wOGRjMThmYzRlM2E3NTBh YmQ4NGExNzgwZjY0YTQ4ZDhiOTUyMjI5IDEwMDY0NAotLS0gYS9Tb3VyY2UvV2ViS2l0L2Nocm9t aXVtL3Rlc3RzL0NDTGF5ZXJUcmVlSG9zdENvbW1vblRlc3QuY3BwCisrKyBiL1NvdXJjZS9XZWJL aXQvY2hyb21pdW0vdGVzdHMvQ0NMYXllclRyZWVIb3N0Q29tbW9uVGVzdC5jcHAKQEAgLTkwMyw2 ICs5MDMsNTEgQEAgVEVTVChDQ0xheWVyVHJlZUhvc3RDb21tb25UZXN0LCB2ZXJpZnlUcmFuc2Zv cm1zRm9yUmVuZGVyU3VyZmFjZUhpZXJhcmNoeSkKICAgICBFWFBFQ1RfRkxPQVRfRVEoNSwgZ3Jh bmRDaGlsZE9mUlMyLT5zY3JlZW5TcGFjZVRyYW5zZm9ybSgpLm00MigpKTsKIH0KIAorVEVTVChD Q0xheWVyVHJlZUhvc3RDb21tb25UZXN0LCB2ZXJpZnlUcmFuc2Zvcm1zRm9yRmxhdHRlbmluZ0xh eWVyKQoreworICAgIC8vIEZvciBsYXllcnMgdGhhdCBmbGF0dGVuIHRoZWlyIHN1YnRyZWUsIHRo ZXJlIHNob3VsZCBiZSBhbiBvcnRob2dyYXBoaWMgcHJvamVjdGlvbgorICAgIC8vIChmb3IgeCBh bmQgeSB2YWx1ZXMpIGluIHRoZSBtaWRkbGUgb2YgdGhlIHRyYW5zZm9ybSBzZXF1ZW5jZS4gTm90 ZSB0aGF0IHRoZSB3YXkgdGhlCisgICAgLy8gY29kZSBpcyBjdXJyZW50bHkgaW1wbGVtZW50ZWQs IGl0IGlzIG5vdCBleHBlY3RlZCB0byB1c2UgYSBjYW5vbmljYWwgb3J0aG9ncmFwaGljCisgICAg Ly8gcHJvamVjdGlvbi4KKworICAgIFJlZlB0cjxMYXllckNocm9taXVtPiByb290ID0gTGF5ZXJD aHJvbWl1bTo6Y3JlYXRlKCk7CisgICAgUmVmUHRyPExheWVyQ2hyb21pdW0+IGNoaWxkID0gTGF5 ZXJDaHJvbWl1bTo6Y3JlYXRlKCk7CisgICAgUmVmUHRyPExheWVyQ2hyb21pdW1XaXRoRm9yY2Vk RHJhd3NDb250ZW50PiBncmFuZENoaWxkID0gYWRvcHRSZWYobmV3IExheWVyQ2hyb21pdW1XaXRo Rm9yY2VkRHJhd3NDb250ZW50KCkpOworCisgICAgV2ViVHJhbnNmb3JtYXRpb25NYXRyaXggcm90 YXRpb25BYm91dFlBeGlzOworICAgIHJvdGF0aW9uQWJvdXRZQXhpcy5yb3RhdGUzZCgwLCAzMCwg MCk7CisKKyAgICBjb25zdCBXZWJUcmFuc2Zvcm1hdGlvbk1hdHJpeCBpZGVudGl0eU1hdHJpeDsK KyAgICBzZXRMYXllclByb3BlcnRpZXNGb3JUZXN0aW5nKHJvb3QuZ2V0KCksIGlkZW50aXR5TWF0 cml4LCBpZGVudGl0eU1hdHJpeCwgRmxvYXRQb2ludDo6emVybygpLCBGbG9hdFBvaW50Ojp6ZXJv KCksIEludFNpemUoMTAwLCAxMDApLCBmYWxzZSk7CisgICAgc2V0TGF5ZXJQcm9wZXJ0aWVzRm9y VGVzdGluZyhjaGlsZC5nZXQoKSwgcm90YXRpb25BYm91dFlBeGlzLCBpZGVudGl0eU1hdHJpeCwg RmxvYXRQb2ludDo6emVybygpLCBGbG9hdFBvaW50Ojp6ZXJvKCksIEludFNpemUoMTAsIDEwKSwg ZmFsc2UpOworICAgIHNldExheWVyUHJvcGVydGllc0ZvclRlc3RpbmcoZ3JhbmRDaGlsZC5nZXQo KSwgcm90YXRpb25BYm91dFlBeGlzLCBpZGVudGl0eU1hdHJpeCwgRmxvYXRQb2ludDo6emVybygp LCBGbG9hdFBvaW50Ojp6ZXJvKCksIEludFNpemUoMTAsIDEwKSwgZmFsc2UpOworCisgICAgcm9v dC0+YWRkQ2hpbGQoY2hpbGQpOworICAgIGNoaWxkLT5hZGRDaGlsZChncmFuZENoaWxkKTsKKyAg ICBjaGlsZC0+c2V0Rm9yY2VSZW5kZXJTdXJmYWNlKHRydWUpOworCisgICAgLy8gTm8gbGF5ZXJz IGluIHRoaXMgdGVzdCBzaG91bGQgcHJlc2VydmUgM2QuCisgICAgQVNTRVJUX0ZBTFNFKHJvb3Qt PnByZXNlcnZlczNEKCkpOworICAgIEFTU0VSVF9GQUxTRShjaGlsZC0+cHJlc2VydmVzM0QoKSk7 CisgICAgQVNTRVJUX0ZBTFNFKGdyYW5kQ2hpbGQtPnByZXNlcnZlczNEKCkpOworCisgICAgV2Vi VHJhbnNmb3JtYXRpb25NYXRyaXggZXhwZWN0ZWRDaGlsZERyYXdUcmFuc2Zvcm0gPSByb3RhdGlv bkFib3V0WUF4aXM7CisgICAgV2ViVHJhbnNmb3JtYXRpb25NYXRyaXggZXhwZWN0ZWRDaGlsZFNj cmVlblNwYWNlVHJhbnNmb3JtID0gcm90YXRpb25BYm91dFlBeGlzOworICAgIFdlYlRyYW5zZm9y bWF0aW9uTWF0cml4IGV4cGVjdGVkR3JhbmRDaGlsZERyYXdUcmFuc2Zvcm0gPSByb3RhdGlvbkFi b3V0WUF4aXM7IC8vIGRyYXdzIG9udG8gY2hpbGQncyByZW5kZXJTdXJmYWNlCisgICAgV2ViVHJh bnNmb3JtYXRpb25NYXRyaXggZXhwZWN0ZWRHcmFuZENoaWxkU2NyZWVuU3BhY2VUcmFuc2Zvcm0g PSByb3RhdGlvbkFib3V0WUF4aXMudG8yZFRyYW5zZm9ybSgpICogcm90YXRpb25BYm91dFlBeGlz OworCisgICAgZXhlY3V0ZUNhbGN1bGF0ZURyYXdUcmFuc2Zvcm1zQW5kVmlzaWJpbGl0eShyb290 LmdldCgpKTsKKworICAgIC8vIFRoZSBjaGlsZCdzIGRyYXdUcmFuc2Zvcm0gc2hvdWxkIGhhdmUg YmVlbiB0YWtlbiBieSBpdHMgc3VyZmFjZS4KKyAgICBBU1NFUlRfVFJVRShjaGlsZC0+cmVuZGVy U3VyZmFjZSgpKTsKKyAgICBFWFBFQ1RfVFJBTlNGT1JNQVRJT05fTUFUUklYX0VRKGV4cGVjdGVk Q2hpbGREcmF3VHJhbnNmb3JtLCBjaGlsZC0+cmVuZGVyU3VyZmFjZSgpLT5kcmF3VHJhbnNmb3Jt KCkpOworICAgIEVYUEVDVF9UUkFOU0ZPUk1BVElPTl9NQVRSSVhfRVEoZXhwZWN0ZWRDaGlsZFNj cmVlblNwYWNlVHJhbnNmb3JtLCBjaGlsZC0+cmVuZGVyU3VyZmFjZSgpLT5zY3JlZW5TcGFjZVRy YW5zZm9ybSgpKTsKKyAgICBFWFBFQ1RfVFJBTlNGT1JNQVRJT05fTUFUUklYX0VRKGlkZW50aXR5 TWF0cml4LCBjaGlsZC0+ZHJhd1RyYW5zZm9ybSgpKTsKKyAgICBFWFBFQ1RfVFJBTlNGT1JNQVRJ T05fTUFUUklYX0VRKGV4cGVjdGVkQ2hpbGRTY3JlZW5TcGFjZVRyYW5zZm9ybSwgY2hpbGQtPnNj cmVlblNwYWNlVHJhbnNmb3JtKCkpOworICAgIEVYUEVDVF9UUkFOU0ZPUk1BVElPTl9NQVRSSVhf RVEoZXhwZWN0ZWRHcmFuZENoaWxkRHJhd1RyYW5zZm9ybSwgZ3JhbmRDaGlsZC0+ZHJhd1RyYW5z Zm9ybSgpKTsKKyAgICBFWFBFQ1RfVFJBTlNGT1JNQVRJT05fTUFUUklYX0VRKGV4cGVjdGVkR3Jh bmRDaGlsZFNjcmVlblNwYWNlVHJhbnNmb3JtLCBncmFuZENoaWxkLT5zY3JlZW5TcGFjZVRyYW5z Zm9ybSgpKTsKK30KKwogVEVTVChDQ0xheWVyVHJlZUhvc3RDb21tb25UZXN0LCB2ZXJpZnlSZW5k ZXJTdXJmYWNlTGlzdEZvclJlbmRlclN1cmZhY2VXaXRoQ2xpcHBlZExheWVyKQogewogICAgIFJl ZlB0cjxMYXllckNocm9taXVtPiBwYXJlbnQgPSBMYXllckNocm9taXVtOjpjcmVhdGUoKTsK