87407 2012-05-24 11:11:23 -0700 window.devicePixelRatio should return the device's pixel ratio not a value dependent on the viewport 2012-06-15 20:54:22 -0700 1 1 1 Unclassified WebKit New Bugs 528+ (Nightly build) Unspecified Unspecified RESOLVED INVALID P2 Normal --- 66687 88114 1 abarth abarth aelias bdakin cmarcelo ddkilzer dev+webkit dino eric.carlson feature-media-reviews fsamuel joepeck johnme kenneth kpiascik macpherson menard peter pnormand simon.fraser tonikitoo webkit.review.bot oldest_to_newest 632964 0 abarth 2012-05-24 11:11:23 -0700 window.devicePixelRatio should return the device's pixel ratio, not the configured device scale factor 632965 1 143854 abarth 2012-05-24 11:11:51 -0700 Created attachment 143854 work in progress 632982 2 143859 abarth 2012-05-24 11:25:04 -0700 Created attachment 143859 Work in progres 632983 3 abarth 2012-05-24 11:25:48 -0700 Apparently this matches what iOS does more closely. 632991 4 143859 johnme 2012-05-24 11:30:26 -0700 Comment on attachment 143859 Work in progres This version looks great. Only question is whether all ports actually set defaultDeviceScaleFactor; I get the feeling it's currently somewhat optional. 632992 5 abarth 2012-05-24 11:33:39 -0700 > Only question is whether all ports actually set defaultDeviceScaleFactor; I get the feeling it's currently somewhat optional. My next step is to write a test, which will hopefully help answer this question. 633058 6 abarth 2012-05-24 12:34:13 -0700 Note to folks looking at this bug: I'm still very much learning about how we do scaling and/or pixel ratios. :) 633068 7 143859 simon.fraser 2012-05-24 12:41:14 -0700 Comment on attachment 143859 Work in progres I'm confused by this change. Why is the old way wrong? 633074 8 abarth 2012-05-24 12:50:11 -0700 > I'm confused by this change. Why is the old way wrong? Here is what johnme wrote when he made this in the chromium-android branch: ---8<--- Tweak device-pixel-ratio & device-width/height metrics. 1. Always return the default devicePixelRatio (e.g. 2 on Galaxy Nexus) instead of changing it based on the viewport tag. This matches both iOS and Android Browser behavior, as well as being more useful. 2. Use device-independent pixels for @media query & CSSOM View widths/heights. This matches iOS, and arguably better matches specs, as well as being more useful. --->8--- My understanding is that these locations shouldn't be affected by the viewport tag, which is why we need to use the defaultDeviceScaleFactor rather than the current deviceScaleFactor. As I wrote above, I'm just learning about these mechanisms, so please be gentle. :) 633081 9 abarth 2012-05-24 12:56:39 -0700 Note: http://trac.webkit.org/browser/trunk/LayoutTests/fast/viewport/viewport-133.html seems to imply that devicePixelRatio should be affected by the viewport tag. 633082 10 simon.fraser 2012-05-24 12:58:01 -0700 I see, but I think there's a misunderstanding about frame->page()->deviceScaleFactor(). This doesn't change with scaling; it's the screen pixel ratio (1 or 2), and nothing else. That test looks wrong. 633092 11 abarth 2012-05-24 13:05:21 -0700 http://trac.webkit.org/browser/trunk/Source/WebCore/page/Page.cpp#L674 seems to have a bunch of machinery for handling dynamic updates, which doesn't seem that useful if the value is a constant. I'm tracking down who calls that function now. 633104 12 abarth 2012-05-24 13:16:01 -0700 > I'm tracking down who calls that function now. It looks like at least Qt and Chromium call setDeviceScaleFactor when processing the viewport. How does PLATFORM(MAC) handle scaling things for the viewport? 633106 13 kpiascik 2012-05-24 13:18:31 -0700 (In reply to comment #12) > > I'm tracking down who calls that function now. > > It looks like at least Qt and Chromium call setDeviceScaleFactor when processing the viewport. > > How does PLATFORM(MAC) handle scaling things for the viewport? BlackBerry does so as well http://trac.webkit.org/browser/trunk/Source/WebKit/blackberry/Api/WebPage.cpp#L3338 633108 14 abarth 2012-05-24 13:21:05 -0700 > How does PLATFORM(MAC) handle scaling things for the viewport? Looks like PLATFORM(MAC) doesn't implement viewport: http://trac.webkit.org/browser/trunk/Source/WebKit2/UIProcess/API/mac/PageClientImpl.mm#L239 633109 15 abarth 2012-05-24 13:22:20 -0700 smfr: It looks like this value does change, which means that we'll need something like this patch to restore the unchanging nature of window.devicePixelRatio. It sounds like I should change viewport-133.html to expect the correct behavior. 633118 16 simon.fraser 2012-05-24 13:28:13 -0700 I guess we all have different notions of what deviceScaleFactor means :( The zooming that mobile devices and Safari do on desktop should not affect deviceScaleFactor. They show up in pageScaleFactor etc. 633125 17 abarth 2012-05-24 13:30:27 -0700 Fascinating. Should we change viewport to affect the pageScaleFactor rather than the deviceScaleFactor? We could potentially eliminate the defaultDeviceScaleFactor in that case. 633162 18 abarth 2012-05-24 13:41:59 -0700 Using the page scale factor for the viewport appear problematic because the page scale factor is settable via a public API. For example, WKPageSetScaleFactor appears to set the page scale factor directly. 633171 19 kpiascik 2012-05-24 13:48:08 -0700 (In reply to comment #8) > > I'm confused by this change. Why is the old way wrong? > > Here is what johnme wrote when he made this in the chromium-android branch: > > ---8<--- > Tweak device-pixel-ratio & device-width/height metrics. > > 1. Always return the default devicePixelRatio (e.g. 2 on Galaxy Nexus) > instead of changing it based on the viewport tag. This matches both > iOS and Android Browser behavior, as well as being more useful. Why do we want this change? A port can change its implementation to do what it wants with the results of computeViewportAttributes. > > 2. Use device-independent pixels for @media query & CSSOM View > widths/heights. This matches iOS, and arguably better matches specs, > as well as being more useful. Aren't device-independent pixels independent from the default devicePixelRatio. I thought they were supposed to be the viewport's value? 633181 20 abarth 2012-05-24 13:53:13 -0700 kpiascik, I'm sorry, I didn't understand your questions. > > 1. Always return the default devicePixelRatio (e.g. 2 on Galaxy Nexus) > > instead of changing it based on the viewport tag. This matches both > > iOS and Android Browser behavior, as well as being more useful. > > Why do we want this change? A port can change its implementation to do what it wants with the results of computeViewportAttributes. I'm not sure I understand. Presumably whether window.devicePixelRatio is affected by viewport meta tag is something that should work the same in every browser. My understanding is that window.devicePixelRatio should be a constant property of the device and not dependent on the configured viewport. smfr appears to confirm that understanding in Comment #10. > > 2. Use device-independent pixels for @media query & CSSOM View > > widths/heights. This matches iOS, and arguably better matches specs, > > as well as being more useful. > > Aren't device-independent pixels independent from the default devicePixelRatio. I thought they were supposed to be the viewport's value? I don't understand this question either, but I'm inclined to sort out window.devicePixelRatio first and then figure out what the right thing to do with media queries second. 633228 21 kenneth 2012-05-24 14:30:24 -0700 (In reply to comment #20) > kpiascik, I'm sorry, I didn't understand your questions. > > > > 1. Always return the default devicePixelRatio (e.g. 2 on Galaxy Nexus) > > > instead of changing it based on the viewport tag. This matches both > > > iOS and Android Browser behavior, as well as being more useful. > > > > Why do we want this change? A port can change its implementation to do what it wants with the results of computeViewportAttributes. > > I'm not sure I understand. Presumably whether window.devicePixelRatio is affected by viewport meta tag is something that should work the same in every browser. My understanding is that window.devicePixelRatio should be a constant property of the device and not dependent on the configured viewport. smfr appears to confirm that understanding in Comment #10. devicePixelRatio is basically the adjustment ratio. Like most mobile pages have been designed with a width of 320 in mind, and thus on iPhone4 it is 2x as 2x320 == 640. On many other devices such as android and firefox with device widths (in portrait mode) of 480, this value is often 1.5. When the target-densitydpi is set to something differently from 160 dpi, say device-dpi on a 480x... Android device, no scale adjustment is applied to the content, thus it might make sense to expose different device-pixel-ratio. I wonder what Android actually does here. John? > > > > 2. Use device-independent pixels for @media query & CSSOM View > > > widths/heights. This matches iOS, and arguably better matches specs, > > > as well as being more useful. > > > > Aren't device-independent pixels independent from the default devicePixelRatio. I thought they were supposed to be the viewport's value? > > I don't understand this question either, but I'm inclined to sort out window.devicePixelRatio first and then figure out what the right thing to do with media queries second. 633262 22 simon.fraser 2012-05-24 14:53:48 -0700 (In reply to comment #21) > (In reply to comment #20) > > kpiascik, I'm sorry, I didn't understand your questions. > > > > > > 1. Always return the default devicePixelRatio (e.g. 2 on Galaxy Nexus) > > > > instead of changing it based on the viewport tag. This matches both > > > > iOS and Android Browser behavior, as well as being more useful. > > > > > > Why do we want this change? A port can change its implementation to do what it wants with the results of computeViewportAttributes. > > > > I'm not sure I understand. Presumably whether window.devicePixelRatio is affected by viewport meta tag is something that should work the same in every browser. My understanding is that window.devicePixelRatio should be a constant property of the device and not dependent on the configured viewport. smfr appears to confirm that understanding in Comment #10. > > devicePixelRatio is basically the adjustment ratio. Like most mobile pages have been designed with a width of 320 in mind, and thus on iPhone4 it is 2x as 2x320 == 640. On many other devices such as android and firefox with device widths (in portrait mode) of 480, this value is often 1.5. This is not how iOS treats it. It's simply a multiplier between device pixels and "UI" pixels (== CSS pixels on an unscaled page). It does not change with zooming or viewport settings. 633273 23 kenneth 2012-05-24 15:00:04 -0700 > > devicePixelRatio is basically the adjustment ratio. Like most mobile pages have been designed with a width of 320 in mind, and thus on iPhone4 it is 2x as 2x320 == 640. On many other devices such as android and firefox with device widths (in portrait mode) of 480, this value is often 1.5. > > This is not how iOS treats it. It's simply a multiplier between device pixels and "UI" pixels (== CSS pixels on an unscaled page). It does not change with zooming or viewport settings. iOS also doesn't support the target-densitydpi argument, which is the only one that modified the device-pixel-ratio. Android supports that, but if it doesn't modify the device-pixel-ratio we should fix this, I just wonder whether that actually makes sense since target-densitydpi = device-dpi means that there is no upscaling of the content. 633276 24 kenneth 2012-05-24 15:01:25 -0700 > > This is not how iOS treats it. It's simply a multiplier between device pixels and "UI" pixels (== CSS pixels on an unscaled page). It does not change with zooming or viewport settings. I don't think anyone changes this with zooming, but only when the target-densitydpi is set as it affects the upscaling of the contents. 633283 25 abarth 2012-05-24 15:06:10 -0700 @kenneth: We're talking about this in #webkit now if you'd like to talk more interactively. 633309 26 abarth 2012-05-24 15:42:34 -0700 We had a long discussion about this topic in IRC, and the net result is that we believe this patch is incorrect and should be reverted from the chromium-android branch. 633318 27 kenneth 2012-05-24 15:50:09 -0700 http://menacingcloud.com/?c=highPixelDensityDisplays is a good resource for people not knowing what device-pixel-ratio is used for. 633320 28 kenneth 2012-05-24 15:52:07 -0700 Other good resources mentioning this: http://dev.opera.com/articles/view/an-introduction-to-meta-viewport-and-viewport/ http://developer.android.com/guide/webapps/targeting.html http://developer.android.com/reference/android/webkit/WebView.html http://dev.w3.org/csswg/css-device-adapt/ 633378 29 abarth 2012-05-24 16:52:36 -0700 More fun with pixel ratios in Bug 87440 for those who are interested. 636031 30 johnme 2012-05-29 08:04:41 -0700 Adam Barth wrote: > We had a long discussion about this topic in IRC, and the > net result is that we believe this patch is incorrect (...) Unfortunately I missed this due to timezones. Let me try to explain the reason for this change. On mobile pages with a "width=device-width" viewport, everyone agrees that window.devicePixelRatio and the corresponding media query should return the defaultDeviceScaleFactor, i.e. approximately (physical screen dpi / 160). That already works, with or without this patch. There are two controversial cases: 1. Desktop page with no viewport (or "width=980"). 2. Mobile page with "width=device-width, target-densitydpi=device=dpi". In both these cases WebCore currently always returns 1 for window.devicePixelRatio and the corresponding media query, but it would be better to return defaultDevicePixelRatio. Since there's no spec for devicePixelRatio, we need to choose our behavior based on: a) the de facto standard behavior b) usefulness A) First, let's try to work out the defacto standard behavior. I wrote a test page for both of the controversial cases above: 1. http://jsbin.com/urowoh/33#none 2. http://jsbin.com/urowoh/33#device-width,device-dpi and tested on high dpi devices with recent copies of Android Browser, Chrome for Android, Firefox, IE Mobile, Nokia N9 browser, Opera Mobile and Mobile Safari. On the 1st test page (no viewport): - IE9 Mobile doesn't support window.devicePixelRatio or the media query. - Android Browser, Chrome for Android, Opera Mobile and Mobile Safari all return the defaultDeviceScaleFactor (2.0 on a Samsung Galaxy Nexus, except Opera which returns 2.25 because they prefer to force their device-width metric to 320 instead of 360). - Firefox Mobile and Nokia N9 browser return 1 (Firefox only for the media query; it seems not to support window.devicePixelRatio). On the 2nd test page (width=device-width, target-densitydpi=device=dpi): - IE Mobile, Mobile Firefox and Mobile Safari don't support target-densitydpi. - Android Browser, Chrome for Android and Opera Mobile all return the defaultDeviceScaleFactor (2.0 on a Samsung Galaxy Nexus, except Opera which returns 2.25 because they prefer to force their device-width metric to 320 instead of 360). - Nokia N9 browser is the only browser that returns 1. In both cases it's clear that the de facto standard behavior is to return defaultDeviceScaleFactor, not 1. B) Now let's consider which behavior is actually useful to web developers. It's clear that always returning 1 when you are on a desktop page or a page with target-densitydpi=device-dpi provides no useful information. So what are we missing out on? Quite a lot, it turns out. On desktop pages that can be viewed on high dpi devices, authors use the defaultDeviceScaleFactor to determine whether to load high dpi images etc (window.devicePixelRatio can be used to provide polyfills for img srcset). On target-densitydpi=device-dpi pages, it's *crucial* for webpages to know the defaultDeviceScaleFactor, as they have taken on responsibility for scaling the page content appropriately for the device's dpi, e.g. they have to scale buttons in proportion to the defaultDeviceScaleFactor. Conclusion: both because it's the de facto standard implemented by 4 browsers (including 3 based on WebKit), and because it's much more useful to web developers, we should always return the fixed defaultDeviceScaleFactor value for window.devicePixelRatio and device-pixel-ratio media queries instead of returning 1. I'm ambivalent about the actual implementation details though. If we can make deviceScaleFactor constant so it doesn't depend on the viewport, then I'd be more than happy to get rid of defaultDeviceScaleFactor and use that instead. This might be cleaner, as I suspect there are several places where WebKit currently uses deviceScaleFactor where they intended to use defaultDeviceScaleFactor. 636037 31 kenneth 2012-05-29 08:10:51 -0700 > On desktop pages that can be viewed on high dpi devices, authors use the defaultDeviceScaleFactor to determine whether to load high dpi images etc (window.devicePixelRatio can be used to provide polyfills for img srcset). How does that help? You are painting one css units as one device pixel? How would loading higher resolution images and down-scaling them help? > On target-densitydpi=device-dpi pages, it's *crucial* for webpages to know the defaultDeviceScaleFactor, as they have taken on responsibility for scaling the page content appropriately for the device's dpi, e.g. they have to scale buttons in proportion to the defaultDeviceScaleFactor. That is a pretty good point. > Conclusion: both because it's the de facto standard implemented by 4 browsers (including 3 based on WebKit), and because it's much more useful to web developers, we should always return the fixed defaultDeviceScaleFactor value for window.devicePixelRatio and device-pixel-ratio media queries instead of returning 1. I am pretty OK with it, I just wanted to know the advantages of always returning it. Now, how can I make use of the second use case if I do target-densitydpi=low-dpi? How will I know the actual difference in this case? When we agree on this, can you send an email to the www-style? 636057 32 simon.fraser 2012-05-29 08:33:12 -0700 Did you not test iOS because it doesn't support target-densitydpi? 636208 33 johnme 2012-05-29 10:40:00 -0700 (In reply to comment #31) > > On desktop pages that can be viewed on high dpi devices, authors use the defaultDeviceScaleFactor to determine whether to load high dpi images etc (window.devicePixelRatio can be used to provide polyfills for img srcset). > > How does that help? You are painting one css units as one device pixel? How would loading higher resolution images and down-scaling them help? But we're not painting one CSS unit per device pixel: we're scaling the page by a pageScaleFactor which is less than one (e.g. 320/980 = 0.327) in overview mode, and often greater than one when zoomed in and reading content. In particular, when pageScaleFactor == defaultDeviceScaleFactor the content will appear the same size as it does on a desktop monitor, and so this is an important zoom level to allow the page to optimize for. This is complicated a little by Font Boosting / text size adjust, which means the user often doesn't need to zoom in so much on wide paragraphs, but defaultDeviceScaleFactor still provides a useful upper bound on how much a user is likely to need to zoom in. > Now, how can I make use of the second use case if I do target-densitydpi=low-dpi? How will I know the actual difference in this case? I never quite worked out why people would ever use values other than auto or device-dpi for this; I'd be keen to hear use cases where this is useful. In fact according to http://dev.w3.org/csswg/css-device-adapt/#the-target-densitydpi-property values other than device-dpi may be dropped. In any case, on browsers that default to target-densitydpi=160 (i.e. most phone browsers other than Opera on Galaxy Nexus) web developers can calculate desired ratios by noting that low-dpi, medium-dpi and high-dpi are fixed at 120, 160 and 240, so on low-dpi with width=device-width the ratio of CSS pixels to device pixels is approximately (window.devicePixelRatio * 160/120). I guess on devices with default target densities other than 160 (e.g. tablets and laptops, where distance from the screen is greater hence CSS pixels are larger due to http://www.w3.org/TR/css3-values/#reference-pixel) it might be useful to know what the default value of target-densitydpi=auto would have been if you hadn't overridden it with a fixed value, but generally if you're asking such questions you shouldn't have overridden it in the first place. So this may be a useful number to expose somehow, but it's pretty niche, and I don't think it's worth breaking the useful and commonplace uses of window.devicePixelRatio to achieve that. > When we agree on this, can you send an email to the www-style? Sure, I've been meaning to do that for a while. (In reply to comment #32) > Did you not test iOS because it doesn't support target-densitydpi? I did test iOS. Mobile Safari is included in the results for the no viewport test page (it returned the defaultDeviceScaleFactor, which was 2 on an iPod touch 4th gen). Obviously it wasn't possible to see what it would report when displaying a target-densitydpi page. 636232 34 dino 2012-05-29 10:55:14 -0700 (In reply to comment #30) > In both cases it's clear that the de facto standard behavior is to return defaultDeviceScaleFactor, not 1. [snip] > Conclusion: both because it's the de facto standard implemented by 4 browsers (including 3 based on WebKit), and because it's much more useful to web developers, we should always return the fixed defaultDeviceScaleFactor value for window.devicePixelRatio and device-pixel-ratio media queries instead of returning 1. I'm sorry, but we (Apple) can't change what we've been returning for devicePixelRatio. As Simon mentioned, for us that exposes the scaling from CSS pixels to device pixels under no zoom (1 on normal resolution, 2 when we do pixel doubling for "retina" screens). The value never changes on a device. If you want to expose something more useful to developers, then it would have to be a new property. 636354 35 johnme 2012-05-29 12:59:19 -0700 (In reply to comment #34) > I'm sorry, but we (Apple) can't change what we've been returning for devicePixelRatio. As Simon mentioned, for us that exposes the scaling from CSS pixels to device pixels under no zoom (1 on normal resolution, 2 when we do pixel doubling for "retina" screens). The value never changes on a device. We seem to be talking at cross purposes. I am calling for upstream WebKit to match downstream iOS whose behavior is to return the defaultDeviceScaleFactor, which is fixed per device (i.e. 2 on retina displays). Apple doesn't need to change iOS, we just need to change upstream WebKit to match what most of the ports are already doing. 636424 36 dino 2012-05-29 14:15:05 -0700 (In reply to comment #35) > (In reply to comment #34) > > I'm sorry, but we (Apple) can't change what we've been returning for devicePixelRatio. As Simon mentioned, for us that exposes the scaling from CSS pixels to device pixels under no zoom (1 on normal resolution, 2 when we do pixel doubling for "retina" screens). The value never changes on a device. > > We seem to be talking at cross purposes. I am calling for upstream WebKit to match downstream iOS whose behavior is to return the defaultDeviceScaleFactor, which is fixed per device (i.e. 2 on retina displays). Apple doesn't need to change iOS, we just need to change upstream WebKit to match what most of the ports are already doing. Excellent. Apologies for my confusion! 636558 37 143859 abarth 2012-05-29 16:45:47 -0700 Comment on attachment 143859 Work in progres Based on this discussion, it sounds like this patch is the one we should go with. There's a separate question of whether target-densitydpi is a good API, but that's something to worry about in a different bug. I'll spin up a new version of the patch that updates the test. 636578 38 144644 abarth 2012-05-29 17:05:55 -0700 Created attachment 144644 Patch 636581 39 144644 abarth 2012-05-29 17:06:43 -0700 Comment on attachment 144644 Patch Note: I'm not sure how to actually run that test, so I'll likely need to pull the right value off the bot after landing this patch. 636588 40 webkit.review.bot 2012-05-29 17:10:26 -0700 Attachment 144644 did not pass style-queue: Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'LayoutTests/ChangeLog', u'LayoutTests/fast..." exit_code: 1 Source/WebCore/css/MediaQueryEvaluator.cpp:297: Tests for true/false, null/non-null, and zero/non-zero should all be done without equality comparisons. [readability/comparison_to_zero] [5] Total errors found: 1 in 6 files If any of these errors are false positives, please file a bug against check-webkit-style. 636606 41 144644 buildbot 2012-05-29 17:20:36 -0700 Comment on attachment 144644 Patch Attachment 144644 did not pass win-ews (win): Output: http://queues.webkit.org/results/12846655 636610 42 144654 abarth 2012-05-29 17:25:12 -0700 Created attachment 144654 Patch 636618 43 webkit.review.bot 2012-05-29 17:29:30 -0700 Attachment 144654 did not pass style-queue: Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'LayoutTests/ChangeLog', u'LayoutTests/fast..." exit_code: 1 Source/WebCore/css/MediaQueryEvaluator.cpp:298: Tests for true/false, null/non-null, and zero/non-zero should all be done without equality comparisons. [readability/comparison_to_zero] [5] Total errors found: 1 in 6 files If any of these errors are false positives, please file a bug against check-webkit-style. 636619 44 144654 dino 2012-05-29 17:31:00 -0700 Comment on attachment 144654 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=144654&action=review > Source/WebCore/ChangeLog:16 > + There's been some confusing about this patch because the Android typo: confusion 636634 45 abarth 2012-05-29 17:37:25 -0700 Thanks Dean. 636635 46 144661 abarth 2012-05-29 17:38:50 -0700 Created attachment 144661 Patch for landing 636638 47 bdakin 2012-05-29 17:41:51 -0700 This patch looks like it will break the Mac because I don't see setDefaultDeviceScaleFactor() ever called in the Mac port. 636703 48 dino 2012-05-29 18:39:20 -0700 (In reply to comment #47) > This patch looks like it will break the Mac because I don't see setDefaultDeviceScaleFactor() ever called in the Mac port. Settings::Settings has m_defaultDeviceScaleFactor(1). We'd need to call setDefaultDeviceScaleFactor for retina. 636713 49 abarth 2012-05-29 18:50:24 -0700 I'm composing an email to webkit-dev about this topic. I think we need to delete Settings::defaultDeviceScaleFactor and store three floats on Page. 636971 50 144661 kenneth 2012-05-30 00:37:38 -0700 Comment on attachment 144661 Patch for landing Is it possible to make a test which sets it to 2.0 and then for a page without viewport meta tag, one with, etc? 637104 51 kenneth 2012-05-30 04:21:34 -0700 I guess we should remove the devicePixelRatio from the WebCore::ViewportAttributes as part of this patch, as well? 639568 52 johnme 2012-06-01 11:52:26 -0700 See also bug 88114 which covers how the (default) deviceScaleFactor is calculated in the first place. 650715 53 abarth 2012-06-15 20:54:22 -0700 defaultDeviceScaleFactor no longer exists 143854 2012-05-24 11:11:51 -0700 2012-05-24 11:25:00 -0700 work in progress bug-87407-20120524111150.patch text/plain 3164 abarth SW5kZXg6IFNvdXJjZS9XZWJDb3JlL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2Vi Q29yZS9DaGFuZ2VMb2cJKHJldmlzaW9uIDExODM5NCkKKysrIFNvdXJjZS9XZWJDb3JlL0NoYW5n ZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDIwIEBACisyMDEyLTA1LTI0ICBBZGFtIEJh cnRoICA8YWJhcnRoQHdlYmtpdC5vcmc+CisKKyAgICAgICAgd2luZG93LmRldmljZVBpeGVsUmF0 aW8gc2hvdWxkIHJldHVybiB0aGUgZGV2aWNlJ3MgcGl4ZWwgcmF0aW8sIG5vdCB0aGUgY29uZmln dXJlZCBkZXZpY2Ugc2NhbGUgZmFjdG9yCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3Jn L3Nob3dfYnVnLmNnaT9pZD04NzQwNworCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09Q UyEpLgorCisgICAgICAgIE5vIG5ldyB0ZXN0cy4gKE9PUFMhKQorCisgICAgICAgICogY3NzL01l ZGlhUXVlcnlFdmFsdWF0b3IuY3BwOgorICAgICAgICAoV2ViQ29yZTo6ZGV2aWNlX3BpeGVsX3Jh dGlvTWVkaWFGZWF0dXJlRXZhbCk6CisgICAgICAgICogcGFnZS9ET01XaW5kb3cuY3BwOgorICAg ICAgICAoV2ViQ29yZTo6RE9NV2luZG93OjpkZXZpY2VQaXhlbFJhdGlvKToKKyAgICAgICAgKiBw bGF0Zm9ybS9QbGF0Zm9ybVNjcmVlbi5oOgorICAgICAgICAoV2ViQ29yZTo6ZGVmYXVsdERldmlj ZVBpeGVsUmF0aW8pOgorICAgICAgICAoV2ViQ29yZSk6CisKIDIwMTItMDUtMjQgIEtpbnVrbyBZ YXN1ZGEgIDxraW51a29AY2hyb21pdW0ub3JnPgogCiAgICAgICAgIENsZWFudXA6IGludHJvZHVj ZSB0b0ZpbGUoKSB0byByZWR1Y2Ugc3RhdGljIGNhc3QgZnJvbSBCbG9iIHRvIEZpbGUKSW5kZXg6 IFNvdXJjZS9XZWJDb3JlL2Nzcy9NZWRpYVF1ZXJ5RXZhbHVhdG9yLmNwcAo9PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0t LSBTb3VyY2UvV2ViQ29yZS9jc3MvTWVkaWFRdWVyeUV2YWx1YXRvci5jcHAJKHJldmlzaW9uIDEx ODM5MCkKKysrIFNvdXJjZS9XZWJDb3JlL2Nzcy9NZWRpYVF1ZXJ5RXZhbHVhdG9yLmNwcAkod29y a2luZyBjb3B5KQpAQCAtMjkwLDExICsyOTAsMTIgQEAgc3RhdGljIGJvb2wgZGV2aWNlX2FzcGVj dF9yYXRpb01lZGlhRmVhdAogCiBzdGF0aWMgYm9vbCBkZXZpY2VfcGl4ZWxfcmF0aW9NZWRpYUZl YXR1cmVFdmFsKENTU1ZhbHVlICp2YWx1ZSwgUmVuZGVyU3R5bGUqLCBGcmFtZSogZnJhbWUsIE1l ZGlhRmVhdHVyZVByZWZpeCBvcCkKIHsKKyAgICBmbG9hdCBzY2FsZUZhY3RvciA9IGRlZmF1bHRE ZXZpY2VQaXhlbFJhdGlvKGZyYW1lLT52aWV3KCkpCiAgICAgaWYgKHZhbHVlKQotICAgICAgICBy ZXR1cm4gdmFsdWUtPmlzUHJpbWl0aXZlVmFsdWUoKSAmJiBjb21wYXJlVmFsdWUoZnJhbWUtPnBh Z2UoKS0+ZGV2aWNlU2NhbGVGYWN0b3IoKSwgc3RhdGljX2Nhc3Q8Q1NTUHJpbWl0aXZlVmFsdWUq Pih2YWx1ZSktPmdldEZsb2F0VmFsdWUoKSwgb3ApOworICAgICAgICByZXR1cm4gdmFsdWUtPmlz UHJpbWl0aXZlVmFsdWUoKSAmJiBjb21wYXJlVmFsdWUoc2NhbGVGYWN0b3IsIHN0YXRpY19jYXN0 PENTU1ByaW1pdGl2ZVZhbHVlKj4odmFsdWUpLT5nZXRGbG9hdFZhbHVlKCksIG9wKTsKKworICAg IHJldHVybiBzY2FsZUZhY3RvciAhPSAwOwogCi0gICAgcmV0dXJuIGZyYW1lLT5wYWdlKCktPmRl dmljZVNjYWxlRmFjdG9yKCkgIT0gMDsKLX0KIAogc3RhdGljIGJvb2wgZ3JpZE1lZGlhRmVhdHVy ZUV2YWwoQ1NTVmFsdWUqIHZhbHVlLCBSZW5kZXJTdHlsZSosIEZyYW1lKiwgTWVkaWFGZWF0dXJl UHJlZml4IG9wKQogewpJbmRleDogU291cmNlL1dlYkNvcmUvcGFnZS9ET01XaW5kb3cuY3BwCj09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT0KLS0tIFNvdXJjZS9XZWJDb3JlL3BhZ2UvRE9NV2luZG93LmNwcAkocmV2aXNpb24g MTE4MzkwKQorKysgU291cmNlL1dlYkNvcmUvcGFnZS9ET01XaW5kb3cuY3BwCSh3b3JraW5nIGNv cHkpCkBAIC0xMzc4LDExICsxMzc4LDExIEBAIGRvdWJsZSBET01XaW5kb3c6OmRldmljZVBpeGVs UmF0aW8oKSBjb24KICAgICBpZiAoIW1fZnJhbWUpCiAgICAgICAgIHJldHVybiAwLjA7CiAKLSAg ICBQYWdlKiBwYWdlID0gbV9mcmFtZS0+cGFnZSgpOwotICAgIGlmICghcGFnZSkKKyAgICBGcmFt ZVZpZXcqIHZpZXcgPSBtX2ZyYW1lLT52aWV3KCk7CisgICAgaWYgKCF2aWV3KQogICAgICAgICBy ZXR1cm4gMC4wOwogCi0gICAgcmV0dXJuIHBhZ2UtPmRldmljZVNjYWxlRmFjdG9yKCk7CisgICAg cmV0dXJuIGRlZmF1bHREZXZpY2VQaXhlbFJhdGlvKHZpZXcpOwogfQogCiB2b2lkIERPTVdpbmRv dzo6c2Nyb2xsQnkoaW50IHgsIGludCB5KSBjb25zdApJbmRleDogU291cmNlL1dlYkNvcmUvcGxh dGZvcm0vUGxhdGZvcm1TY3JlZW4uaAo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2ViQ29yZS9wbGF0 Zm9ybS9QbGF0Zm9ybVNjcmVlbi5oCShyZXZpc2lvbiAxMTgzOTApCisrKyBTb3VyY2UvV2ViQ29y ZS9wbGF0Zm9ybS9QbGF0Zm9ybVNjcmVlbi5oCSh3b3JraW5nIGNvcHkpCkBAIC01MSw2ICs1MSwx MyBAQCBuYW1lc3BhY2UgV2ViQ29yZSB7CiAgICAgRmxvYXRSZWN0IHNjcmVlblJlY3QoV2lkZ2V0 Kik7CiAgICAgRmxvYXRSZWN0IHNjcmVlbkF2YWlsYWJsZVJlY3QoV2lkZ2V0Kik7CiAKKyAgICBm bG9hdCBkZWZhdWx0RGV2aWNlUGl4ZWxSYXRpbyhXaWRnZXQqIHdpZGdldCkKKyAgICB7CisgICAg ICAgIC8vIGRldmljZVBpeGVsUmF0aW8gaXMgZGUgZmFjdG8gcmVsYXRpdmUgdG8gMTYwZHBpLiBT ZWUgYWxzbworICAgICAgICAvLyBjb21wdXRlVmlld3BvcnRBdHRyaWJ1dGVzIGluIFZpZXdwb3J0 QXJndW1lbnRzLmNwcC4KKyAgICAgICAgcmV0dXJuIHNjcmVlbkhvcml6b250YWxEUEkod2lkZ2V0 KSAvIDE2MC4wZjsKKyAgICB9CisKICNpZiBQTEFURk9STShNQUMpCiAgICAgTlNTY3JlZW4gKnNj cmVlbkZvcldpbmRvdyhOU1dpbmRvdyAqKTsKIAo= 143859 2012-05-24 11:25:04 -0700 2012-05-29 17:05:51 -0700 Work in progres bug-87407-20120524112503.patch text/plain 2271 abarth SW5kZXg6IFNvdXJjZS9XZWJDb3JlL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2Vi Q29yZS9DaGFuZ2VMb2cJKHJldmlzaW9uIDExODM5NCkKKysrIFNvdXJjZS9XZWJDb3JlL0NoYW5n ZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDE3IEBACisyMDEyLTA1LTI0ICBBZGFtIEJh cnRoICA8YWJhcnRoQHdlYmtpdC5vcmc+CisKKyAgICAgICAgd2luZG93LmRldmljZVBpeGVsUmF0 aW8gc2hvdWxkIHJldHVybiB0aGUgZGV2aWNlJ3MgcGl4ZWwgcmF0aW8sIG5vdCB0aGUgY29uZmln dXJlZCBkZXZpY2Ugc2NhbGUgZmFjdG9yCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3Jn L3Nob3dfYnVnLmNnaT9pZD04NzQwNworCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09Q UyEpLgorCisgICAgICAgIE5vIG5ldyB0ZXN0cy4gKE9PUFMhKQorCisgICAgICAgICogY3NzL01l ZGlhUXVlcnlFdmFsdWF0b3IuY3BwOgorICAgICAgICAoV2ViQ29yZTo6ZGV2aWNlX3BpeGVsX3Jh dGlvTWVkaWFGZWF0dXJlRXZhbCk6CisgICAgICAgICogcGFnZS9ET01XaW5kb3cuY3BwOgorICAg ICAgICAoV2ViQ29yZTo6RE9NV2luZG93OjpkZXZpY2VQaXhlbFJhdGlvKToKKwogMjAxMi0wNS0y NCAgS2ludWtvIFlhc3VkYSAgPGtpbnVrb0BjaHJvbWl1bS5vcmc+CiAKICAgICAgICAgQ2xlYW51 cDogaW50cm9kdWNlIHRvRmlsZSgpIHRvIHJlZHVjZSBzdGF0aWMgY2FzdCBmcm9tIEJsb2IgdG8g RmlsZQpJbmRleDogU291cmNlL1dlYkNvcmUvY3NzL01lZGlhUXVlcnlFdmFsdWF0b3IuY3BwCj09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT0KLS0tIFNvdXJjZS9XZWJDb3JlL2Nzcy9NZWRpYVF1ZXJ5RXZhbHVhdG9yLmNwcAko cmV2aXNpb24gMTE4MzkwKQorKysgU291cmNlL1dlYkNvcmUvY3NzL01lZGlhUXVlcnlFdmFsdWF0 b3IuY3BwCSh3b3JraW5nIGNvcHkpCkBAIC0yOTAsMTAgKzI5MCwxMSBAQCBzdGF0aWMgYm9vbCBk ZXZpY2VfYXNwZWN0X3JhdGlvTWVkaWFGZWF0CiAKIHN0YXRpYyBib29sIGRldmljZV9waXhlbF9y YXRpb01lZGlhRmVhdHVyZUV2YWwoQ1NTVmFsdWUgKnZhbHVlLCBSZW5kZXJTdHlsZSosIEZyYW1l KiBmcmFtZSwgTWVkaWFGZWF0dXJlUHJlZml4IG9wKQogeworICAgIGZsb2F0IHNjYWxlRmFjdG9y ID0gZnJhbWUtPnBhZ2UoKS0+c2V0dGluZ3MoKS0+ZGVmYXVsdERldmljZVNjYWxlRmFjdG9yKCk7 CiAgICAgaWYgKHZhbHVlKQotICAgICAgICByZXR1cm4gdmFsdWUtPmlzUHJpbWl0aXZlVmFsdWUo KSAmJiBjb21wYXJlVmFsdWUoZnJhbWUtPnBhZ2UoKS0+ZGV2aWNlU2NhbGVGYWN0b3IoKSwgc3Rh dGljX2Nhc3Q8Q1NTUHJpbWl0aXZlVmFsdWUqPih2YWx1ZSktPmdldEZsb2F0VmFsdWUoKSwgb3Ap OworICAgICAgICByZXR1cm4gdmFsdWUtPmlzUHJpbWl0aXZlVmFsdWUoKSAmJiBjb21wYXJlVmFs dWUoc2NhbGVGYWN0b3IsIHN0YXRpY19jYXN0PENTU1ByaW1pdGl2ZVZhbHVlKj4odmFsdWUpLT5n ZXRGbG9hdFZhbHVlKCksIG9wKTsKIAotICAgIHJldHVybiBmcmFtZS0+cGFnZSgpLT5kZXZpY2VT Y2FsZUZhY3RvcigpICE9IDA7CisgICAgcmV0dXJuIHNjYWxlRmFjdG9yICE9IDA7CiB9CiAKIHN0 YXRpYyBib29sIGdyaWRNZWRpYUZlYXR1cmVFdmFsKENTU1ZhbHVlKiB2YWx1ZSwgUmVuZGVyU3R5 bGUqLCBGcmFtZSosIE1lZGlhRmVhdHVyZVByZWZpeCBvcCkKSW5kZXg6IFNvdXJjZS9XZWJDb3Jl L3BhZ2UvRE9NV2luZG93LmNwcAo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2ViQ29yZS9wYWdlL0RP TVdpbmRvdy5jcHAJKHJldmlzaW9uIDExODM5MCkKKysrIFNvdXJjZS9XZWJDb3JlL3BhZ2UvRE9N V2luZG93LmNwcAkod29ya2luZyBjb3B5KQpAQCAtMTM4Miw3ICsxMzgyLDcgQEAgZG91YmxlIERP TVdpbmRvdzo6ZGV2aWNlUGl4ZWxSYXRpbygpIGNvbgogICAgIGlmICghcGFnZSkKICAgICAgICAg cmV0dXJuIDAuMDsKIAotICAgIHJldHVybiBwYWdlLT5kZXZpY2VTY2FsZUZhY3RvcigpOworICAg IHJldHVybiBwYWdlLT5zZXR0aW5ncygpLT5kZWZhdWx0RGV2aWNlU2NhbGVGYWN0b3IoKTsKIH0K IAogdm9pZCBET01XaW5kb3c6OnNjcm9sbEJ5KGludCB4LCBpbnQgeSkgY29uc3QK 144644 2012-05-29 17:05:55 -0700 2012-05-29 17:25:06 -0700 Patch bug-87407-20120529170554.patch text/plain 5105 abarth SW5kZXg6IFNvdXJjZS9XZWJDb3JlL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2Vi Q29yZS9DaGFuZ2VMb2cJKHJldmlzaW9uIDExODg1OSkKKysrIFNvdXJjZS9XZWJDb3JlL0NoYW5n ZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDMwIEBACisyMDEyLTA1LTI5ICBBZGFtIEJh cnRoICA8YWJhcnRoQHdlYmtpdC5vcmc+CisKKyAgICAgICAgd2luZG93LmRldmljZVBpeGVsUmF0 aW8gc2hvdWxkIHJldHVybiB0aGUgZGV2aWNlJ3MgcGl4ZWwgcmF0aW8gbm90IGEgdmFsdWUgZGVw ZW5kZW50IG9uIHRoZSB2aWV3cG9ydAorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9z aG93X2J1Zy5jZ2k/aWQ9ODc0MDcKKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMh KS4KKworICAgICAgICBUaGVyZSdzIGJlZW4gYSBsb3Qgb2YgZGlzY3Vzc2lvbiBhYm91dCB0aGlz IHBhdGNoLCBib3RoIG9uIHRoZSBidWcKKyAgICAgICAgcmVmZXJlbmNlZCBhYm92ZSBhbmQgb24g I3dlYmtpdC4KKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lk PTg3NDA3I2MzMCBoYXMgYSBnb29kIHN1bW1hcnkgb2YKKyAgICAgICAgdGhlIGRpc2N1c3Npb24g YW5kIG1vdGl2YXRpb24gYmVoaW5kIHRoaXMgY2hhbmdlLiAgRXNzZW50aWFsbHksIHRoaXMKKyAg ICAgICAgcGF0Y2ggbWFrZXMgV2ViS2l0J3MgYmVoYXZpb3Igb24gdHJ1bmsgbWF0Y2ggdGhhdCBv ZiB0aGUgY3VycmVudGx5CisgICAgICAgIHNoaXBwaW5nIGNvZGUgb24gaU9TIGFuZCBBbmRyb2lk LiAgT24gaU9TLCB0aGlzIGNoYW5nZSBpc24ndCBkZXRlY3RhYmxlCisgICAgICAgIGJlY2F1c2Ug aU9TIGRvZXNuJ3Qgc3VwcG9ydCB0YXJnZXQtZGVuc2l0eWRwaS4KKworICAgICAgICBUaGVyZSdz IGJlZW4gc29tZSBjb25mdXNpbmcgYWJvdXQgdGhpcyBwYXRjaCBiZWNhdXNlIHRoZSBBbmRyb2lk CisgICAgICAgIGRvY3VtZW50YXRpb24gc2VlbXMgdG8gaW1wbHkgdGhhdCB3ZSBzaG91bGRuJ3Qg bWFrZSB0aGF0IGNoYW5nZSwgYnV0CisgICAgICAgIHRoYXQgZG9jdW1lbnRhdGlvbiBpcyB3cm9u ZyBpbiB0aGF0IGl0IGRvZXNuJ3QgcmVmbGVjdCBBbmRyb2lkJ3MKKyAgICAgICAgaW1wbGVtZW50 YXRpb24gb2YgdGFyZ2V0LWRlbnNpdHlkcGkgYW5kIHRoZXJlIGFwcGVhcnMgdG8gYmUgY29udGVu dAorICAgICAgICB0aGF0IHJlbGllcyBvbiB0aGUgaW1wbGVtZW50YXRpb24gYmVoYXZpb3IgcmF0 aGVyIHRoYW4gdGhlIGRvY3VtZW50ZWQKKyAgICAgICAgYmVoYXZpb3IuCisKKyAgICAgICAgKiBj c3MvTWVkaWFRdWVyeUV2YWx1YXRvci5jcHA6CisgICAgICAgIChXZWJDb3JlOjpkZXZpY2VfcGl4 ZWxfcmF0aW9NZWRpYUZlYXR1cmVFdmFsKToKKyAgICAgICAgKiBwYWdlL0RPTVdpbmRvdy5jcHA6 CisgICAgICAgIChXZWJDb3JlOjpET01XaW5kb3c6OmRldmljZVBpeGVsUmF0aW8pOgorCiAyMDEy LTA1LTI5ICBTaGF3biBTaW5naCAgPHNoYXduc2luZ2hAY2hyb21pdW0ub3JnPgogCiAgICAgICAg IFtjaHJvbWl1bV0gbWFrZSBXZWJUcmFuc2Zvcm1hdGlvbk1hdHJpeCBvYmplY3QgdXNhYmxlIGJ5 IG5vbi13ZWJraXQgY29kZQpJbmRleDogU291cmNlL1dlYkNvcmUvY3NzL01lZGlhUXVlcnlFdmFs dWF0b3IuY3BwCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT0KLS0tIFNvdXJjZS9XZWJDb3JlL2Nzcy9NZWRpYVF1ZXJ5RXZh bHVhdG9yLmNwcAkocmV2aXNpb24gMTE4ODU5KQorKysgU291cmNlL1dlYkNvcmUvY3NzL01lZGlh UXVlcnlFdmFsdWF0b3IuY3BwCSh3b3JraW5nIGNvcHkpCkBAIC0yOTAsMTAgKzI5MCwxMSBAQCBz dGF0aWMgYm9vbCBkZXZpY2VfYXNwZWN0X3JhdGlvTWVkaWFGZWF0CiAKIHN0YXRpYyBib29sIGRl dmljZV9waXhlbF9yYXRpb01lZGlhRmVhdHVyZUV2YWwoQ1NTVmFsdWUgKnZhbHVlLCBSZW5kZXJT dHlsZSosIEZyYW1lKiBmcmFtZSwgTWVkaWFGZWF0dXJlUHJlZml4IG9wKQogeworICAgIGZsb2F0 IHNjYWxlRmFjdG9yID0gZnJhbWUtPnBhZ2UoKS0+c2V0dGluZ3MoKS0+ZGVmYXVsdERldmljZVNj YWxlRmFjdG9yKCk7CiAgICAgaWYgKHZhbHVlKQotICAgICAgICByZXR1cm4gdmFsdWUtPmlzUHJp bWl0aXZlVmFsdWUoKSAmJiBjb21wYXJlVmFsdWUoZnJhbWUtPnBhZ2UoKS0+ZGV2aWNlU2NhbGVG YWN0b3IoKSwgc3RhdGljX2Nhc3Q8Q1NTUHJpbWl0aXZlVmFsdWUqPih2YWx1ZSktPmdldEZsb2F0 VmFsdWUoKSwgb3ApOworICAgICAgICByZXR1cm4gdmFsdWUtPmlzUHJpbWl0aXZlVmFsdWUoKSAm JiBjb21wYXJlVmFsdWUoc2NhbGVGYWN0b3IsIHN0YXRpY19jYXN0PENTU1ByaW1pdGl2ZVZhbHVl Kj4odmFsdWUpLT5nZXRGbG9hdFZhbHVlKCksIG9wKTsKIAotICAgIHJldHVybiBmcmFtZS0+cGFn ZSgpLT5kZXZpY2VTY2FsZUZhY3RvcigpICE9IDA7CisgICAgcmV0dXJuIHNjYWxlRmFjdG9yICE9 IDA7CiB9CiAKIHN0YXRpYyBib29sIGdyaWRNZWRpYUZlYXR1cmVFdmFsKENTU1ZhbHVlKiB2YWx1 ZSwgUmVuZGVyU3R5bGUqLCBGcmFtZSosIE1lZGlhRmVhdHVyZVByZWZpeCBvcCkKSW5kZXg6IFNv dXJjZS9XZWJDb3JlL3BhZ2UvRE9NV2luZG93LmNwcAo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2Vi Q29yZS9wYWdlL0RPTVdpbmRvdy5jcHAJKHJldmlzaW9uIDExODg1OSkKKysrIFNvdXJjZS9XZWJD b3JlL3BhZ2UvRE9NV2luZG93LmNwcAkod29ya2luZyBjb3B5KQpAQCAtMTM4Miw3ICsxMzgyLDcg QEAgZG91YmxlIERPTVdpbmRvdzo6ZGV2aWNlUGl4ZWxSYXRpbygpIGNvbgogICAgIGlmICghcGFn ZSkKICAgICAgICAgcmV0dXJuIDAuMDsKIAotICAgIHJldHVybiBwYWdlLT5kZXZpY2VTY2FsZUZh Y3RvcigpOworICAgIHJldHVybiBwYWdlLT5zZXR0aW5ncygpLT5kZWZhdWx0RGV2aWNlU2NhbGVG YWN0b3IoKTsKIH0KIAogdm9pZCBET01XaW5kb3c6OnNjcm9sbEJ5KGludCB4LCBpbnQgeSkgY29u c3QKSW5kZXg6IExheW91dFRlc3RzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlvdXRUZXN0 cy9DaGFuZ2VMb2cJKHJldmlzaW9uIDExODg1OSkKKysrIExheW91dFRlc3RzL0NoYW5nZUxvZwko d29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDE2IEBACisyMDEyLTA1LTI5ICBBZGFtIEJhcnRoICA8 YWJhcnRoQHdlYmtpdC5vcmc+CisKKyAgICAgICAgd2luZG93LmRldmljZVBpeGVsUmF0aW8gc2hv dWxkIHJldHVybiB0aGUgZGV2aWNlJ3MgcGl4ZWwgcmF0aW8gbm90IGEgdmFsdWUgZGVwZW5kZW50 IG9uIHRoZSB2aWV3cG9ydAorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1 Zy5jZ2k/aWQ9ODc0MDcKKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKwor ICAgICAgICBVcGRhdGUgdGhpcyB0ZXN0IHRvIHNob3cgdGhhdCB3aW5kb3cuZGV2aWNlUGl4ZWxS YXRpbyBkb2Vzbid0IGRlcGVuZCBvbgorICAgICAgICB0aGUgdmlld3BvcnQgbWV0YSB0YWcuCisK KyAgICAgICAgKiBmYXN0L3ZpZXdwb3J0L3ZpZXdwb3J0LTEzMy1leHBlY3RlZC50eHQ6CisgICAg ICAgICogZmFzdC92aWV3cG9ydC92aWV3cG9ydC0xMzMuaHRtbDoKKwogMjAxMi0wNS0yOSAgRmls aXAgUGl6bG8gIDxmcGl6bG9AYXBwbGUuY29tPgogCiAgICAgICAgIERGRyBzaG91bGQga2VlcCBj YXB0dXJlZCB2YXJpYWJsZXMgYWxpdmUgdW50aWwgdGhlIChpbmxpbmUpIHJldHVybi4KSW5kZXg6 IExheW91dFRlc3RzL2Zhc3Qvdmlld3BvcnQvdmlld3BvcnQtMTMzLWV4cGVjdGVkLnR4dAo9PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09Ci0tLSBMYXlvdXRUZXN0cy9mYXN0L3ZpZXdwb3J0L3ZpZXdwb3J0LTEzMy1leHBlY3Rl ZC50eHQJKHJldmlzaW9uIDExODg1OSkKKysrIExheW91dFRlc3RzL2Zhc3Qvdmlld3BvcnQvdmll d3BvcnQtMTMzLWV4cGVjdGVkLnR4dAkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDMgQEAKIHZp ZXdwb3J0IHNpemUgNDgweDUzNCBzY2FsZSAxLjAwMDAwMCB3aXRoIGxpbWl0cyBbMS4wMDAwMDAs IDUuMDAwMDAwXSBhbmQgdXNlclNjYWxhYmxlIC0xLjAwMDAwMAotUEFTUyB3aW5kb3cuZGV2aWNl UGl4ZWxSYXRpbyBpcyB3aXRoaW4gMC4wMDEgb2YgMC43MDgzMzMzMTM0NjUxMTg0CitQQVNTIHdp bmRvdy5kZXZpY2VQaXhlbFJhdGlvIGlzIHdpdGhpbiAwLjAwMSBvZiAxCiAKSW5kZXg6IExheW91 dFRlc3RzL2Zhc3Qvdmlld3BvcnQvdmlld3BvcnQtMTMzLmh0bWwKPT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PQotLS0gTGF5 b3V0VGVzdHMvZmFzdC92aWV3cG9ydC92aWV3cG9ydC0xMzMuaHRtbAkocmV2aXNpb24gMTE4ODU5 KQorKysgTGF5b3V0VGVzdHMvZmFzdC92aWV3cG9ydC92aWV3cG9ydC0xMzMuaHRtbAkod29ya2lu ZyBjb3B5KQpAQCAtNyw3ICs3LDcgQEAKICAgICAgICAgICAgIGlmICh3aW5kb3cubGF5b3V0VGVz dENvbnRyb2xsZXIpIHsKICAgICAgICAgICAgICAgICBsYXlvdXRUZXN0Q29udHJvbGxlci5kdW1w QXNUZXh0KCk7CiAgICAgICAgICAgICAgICAgbGF5b3V0VGVzdENvbnRyb2xsZXIuZHVtcENvbmZp Z3VyYXRpb25Gb3JWaWV3cG9ydCgyNDAsIDQ4MCwgNTY0LCA0ODAsIDUzNCk7Ci0gICAgICAgICAg ICAgICAgc2hvdWxkQmVDbG9zZVRvKCJ3aW5kb3cuZGV2aWNlUGl4ZWxSYXRpbyIsIDAuNzA4MzMz MzEzNDY1MTE4NCwgMC4wMDEsIGZhbHNlKTsKKyAgICAgICAgICAgICAgICBzaG91bGRCZUNsb3Nl VG8oIndpbmRvdy5kZXZpY2VQaXhlbFJhdGlvIiwgMSwgMC4wMDEsIGZhbHNlKTsKICAgICAgICAg ICAgIH0KICAgICAgICAgfQogICAgIDwvc2NyaXB0Pgo= 144654 2012-05-29 17:25:12 -0700 2012-05-29 17:38:43 -0700 Patch bug-87407-20120529172511.patch text/plain 5282 abarth SW5kZXg6IFNvdXJjZS9XZWJDb3JlL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2Vi Q29yZS9DaGFuZ2VMb2cJKHJldmlzaW9uIDExODg1OSkKKysrIFNvdXJjZS9XZWJDb3JlL0NoYW5n ZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDMwIEBACisyMDEyLTA1LTI5ICBBZGFtIEJh cnRoICA8YWJhcnRoQHdlYmtpdC5vcmc+CisKKyAgICAgICAgd2luZG93LmRldmljZVBpeGVsUmF0 aW8gc2hvdWxkIHJldHVybiB0aGUgZGV2aWNlJ3MgcGl4ZWwgcmF0aW8gbm90IGEgdmFsdWUgZGVw ZW5kZW50IG9uIHRoZSB2aWV3cG9ydAorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9z aG93X2J1Zy5jZ2k/aWQ9ODc0MDcKKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMh KS4KKworICAgICAgICBUaGVyZSdzIGJlZW4gYSBsb3Qgb2YgZGlzY3Vzc2lvbiBhYm91dCB0aGlz IHBhdGNoLCBib3RoIG9uIHRoZSBidWcKKyAgICAgICAgcmVmZXJlbmNlZCBhYm92ZSBhbmQgb24g I3dlYmtpdC4KKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lk PTg3NDA3I2MzMCBoYXMgYSBnb29kIHN1bW1hcnkgb2YKKyAgICAgICAgdGhlIGRpc2N1c3Npb24g YW5kIG1vdGl2YXRpb24gYmVoaW5kIHRoaXMgY2hhbmdlLiAgRXNzZW50aWFsbHksIHRoaXMKKyAg ICAgICAgcGF0Y2ggbWFrZXMgV2ViS2l0J3MgYmVoYXZpb3Igb24gdHJ1bmsgbWF0Y2ggdGhhdCBv ZiB0aGUgY3VycmVudGx5CisgICAgICAgIHNoaXBwaW5nIGNvZGUgb24gaU9TIGFuZCBBbmRyb2lk LiAgT24gaU9TLCB0aGlzIGNoYW5nZSBpc24ndCBkZXRlY3RhYmxlCisgICAgICAgIGJlY2F1c2Ug aU9TIGRvZXNuJ3Qgc3VwcG9ydCB0YXJnZXQtZGVuc2l0eWRwaS4KKworICAgICAgICBUaGVyZSdz IGJlZW4gc29tZSBjb25mdXNpbmcgYWJvdXQgdGhpcyBwYXRjaCBiZWNhdXNlIHRoZSBBbmRyb2lk CisgICAgICAgIGRvY3VtZW50YXRpb24gc2VlbXMgdG8gaW1wbHkgdGhhdCB3ZSBzaG91bGRuJ3Qg bWFrZSB0aGF0IGNoYW5nZSwgYnV0CisgICAgICAgIHRoYXQgZG9jdW1lbnRhdGlvbiBpcyB3cm9u ZyBpbiB0aGF0IGl0IGRvZXNuJ3QgcmVmbGVjdCBBbmRyb2lkJ3MKKyAgICAgICAgaW1wbGVtZW50 YXRpb24gb2YgdGFyZ2V0LWRlbnNpdHlkcGkgYW5kIHRoZXJlIGFwcGVhcnMgdG8gYmUgY29udGVu dAorICAgICAgICB0aGF0IHJlbGllcyBvbiB0aGUgaW1wbGVtZW50YXRpb24gYmVoYXZpb3IgcmF0 aGVyIHRoYW4gdGhlIGRvY3VtZW50ZWQKKyAgICAgICAgYmVoYXZpb3IuCisKKyAgICAgICAgKiBj c3MvTWVkaWFRdWVyeUV2YWx1YXRvci5jcHA6CisgICAgICAgIChXZWJDb3JlOjpkZXZpY2VfcGl4 ZWxfcmF0aW9NZWRpYUZlYXR1cmVFdmFsKToKKyAgICAgICAgKiBwYWdlL0RPTVdpbmRvdy5jcHA6 CisgICAgICAgIChXZWJDb3JlOjpET01XaW5kb3c6OmRldmljZVBpeGVsUmF0aW8pOgorCiAyMDEy LTA1LTI5ICBTaGF3biBTaW5naCAgPHNoYXduc2luZ2hAY2hyb21pdW0ub3JnPgogCiAgICAgICAg IFtjaHJvbWl1bV0gbWFrZSBXZWJUcmFuc2Zvcm1hdGlvbk1hdHJpeCBvYmplY3QgdXNhYmxlIGJ5 IG5vbi13ZWJraXQgY29kZQpJbmRleDogU291cmNlL1dlYkNvcmUvY3NzL01lZGlhUXVlcnlFdmFs dWF0b3IuY3BwCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT0KLS0tIFNvdXJjZS9XZWJDb3JlL2Nzcy9NZWRpYVF1ZXJ5RXZh bHVhdG9yLmNwcAkocmV2aXNpb24gMTE4ODU5KQorKysgU291cmNlL1dlYkNvcmUvY3NzL01lZGlh UXVlcnlFdmFsdWF0b3IuY3BwCSh3b3JraW5nIGNvcHkpCkBAIC00NSw2ICs0NSw3IEBACiAjaW5j bHVkZSAiUGxhdGZvcm1TY3JlZW4uaCIKICNpbmNsdWRlICJSZW5kZXJWaWV3LmgiCiAjaW5jbHVk ZSAiUmVuZGVyU3R5bGUuaCIKKyNpbmNsdWRlICJTZXR0aW5ncy5oIgogI2luY2x1ZGUgIlN0eWxl UmVzb2x2ZXIuaCIKICNpbmNsdWRlIDx3dGYvSGFzaE1hcC5oPgogCkBAIC0yOTAsMTAgKzI5MSwx MSBAQCBzdGF0aWMgYm9vbCBkZXZpY2VfYXNwZWN0X3JhdGlvTWVkaWFGZWF0CiAKIHN0YXRpYyBi b29sIGRldmljZV9waXhlbF9yYXRpb01lZGlhRmVhdHVyZUV2YWwoQ1NTVmFsdWUgKnZhbHVlLCBS ZW5kZXJTdHlsZSosIEZyYW1lKiBmcmFtZSwgTWVkaWFGZWF0dXJlUHJlZml4IG9wKQogeworICAg IGZsb2F0IHNjYWxlRmFjdG9yID0gZnJhbWUtPnBhZ2UoKS0+c2V0dGluZ3MoKS0+ZGVmYXVsdERl dmljZVNjYWxlRmFjdG9yKCk7CiAgICAgaWYgKHZhbHVlKQotICAgICAgICByZXR1cm4gdmFsdWUt PmlzUHJpbWl0aXZlVmFsdWUoKSAmJiBjb21wYXJlVmFsdWUoZnJhbWUtPnBhZ2UoKS0+ZGV2aWNl U2NhbGVGYWN0b3IoKSwgc3RhdGljX2Nhc3Q8Q1NTUHJpbWl0aXZlVmFsdWUqPih2YWx1ZSktPmdl dEZsb2F0VmFsdWUoKSwgb3ApOworICAgICAgICByZXR1cm4gdmFsdWUtPmlzUHJpbWl0aXZlVmFs dWUoKSAmJiBjb21wYXJlVmFsdWUoc2NhbGVGYWN0b3IsIHN0YXRpY19jYXN0PENTU1ByaW1pdGl2 ZVZhbHVlKj4odmFsdWUpLT5nZXRGbG9hdFZhbHVlKCksIG9wKTsKIAotICAgIHJldHVybiBmcmFt ZS0+cGFnZSgpLT5kZXZpY2VTY2FsZUZhY3RvcigpICE9IDA7CisgICAgcmV0dXJuIHNjYWxlRmFj dG9yICE9IDA7CiB9CiAKIHN0YXRpYyBib29sIGdyaWRNZWRpYUZlYXR1cmVFdmFsKENTU1ZhbHVl KiB2YWx1ZSwgUmVuZGVyU3R5bGUqLCBGcmFtZSosIE1lZGlhRmVhdHVyZVByZWZpeCBvcCkKSW5k ZXg6IFNvdXJjZS9XZWJDb3JlL3BhZ2UvRE9NV2luZG93LmNwcAo9PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3Vy Y2UvV2ViQ29yZS9wYWdlL0RPTVdpbmRvdy5jcHAJKHJldmlzaW9uIDExODg1OSkKKysrIFNvdXJj ZS9XZWJDb3JlL3BhZ2UvRE9NV2luZG93LmNwcAkod29ya2luZyBjb3B5KQpAQCAtMTM4Miw3ICsx MzgyLDcgQEAgZG91YmxlIERPTVdpbmRvdzo6ZGV2aWNlUGl4ZWxSYXRpbygpIGNvbgogICAgIGlm ICghcGFnZSkKICAgICAgICAgcmV0dXJuIDAuMDsKIAotICAgIHJldHVybiBwYWdlLT5kZXZpY2VT Y2FsZUZhY3RvcigpOworICAgIHJldHVybiBwYWdlLT5zZXR0aW5ncygpLT5kZWZhdWx0RGV2aWNl U2NhbGVGYWN0b3IoKTsKIH0KIAogdm9pZCBET01XaW5kb3c6OnNjcm9sbEJ5KGludCB4LCBpbnQg eSkgY29uc3QKSW5kZXg6IExheW91dFRlc3RzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlv dXRUZXN0cy9DaGFuZ2VMb2cJKHJldmlzaW9uIDExODg1OSkKKysrIExheW91dFRlc3RzL0NoYW5n ZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDE2IEBACisyMDEyLTA1LTI5ICBBZGFtIEJh cnRoICA8YWJhcnRoQHdlYmtpdC5vcmc+CisKKyAgICAgICAgd2luZG93LmRldmljZVBpeGVsUmF0 aW8gc2hvdWxkIHJldHVybiB0aGUgZGV2aWNlJ3MgcGl4ZWwgcmF0aW8gbm90IGEgdmFsdWUgZGVw ZW5kZW50IG9uIHRoZSB2aWV3cG9ydAorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9z aG93X2J1Zy5jZ2k/aWQ9ODc0MDcKKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMh KS4KKworICAgICAgICBVcGRhdGUgdGhpcyB0ZXN0IHRvIHNob3cgdGhhdCB3aW5kb3cuZGV2aWNl UGl4ZWxSYXRpbyBkb2Vzbid0IGRlcGVuZCBvbgorICAgICAgICB0aGUgdmlld3BvcnQgbWV0YSB0 YWcuCisKKyAgICAgICAgKiBmYXN0L3ZpZXdwb3J0L3ZpZXdwb3J0LTEzMy1leHBlY3RlZC50eHQ6 CisgICAgICAgICogZmFzdC92aWV3cG9ydC92aWV3cG9ydC0xMzMuaHRtbDoKKwogMjAxMi0wNS0y OSAgRmlsaXAgUGl6bG8gIDxmcGl6bG9AYXBwbGUuY29tPgogCiAgICAgICAgIERGRyBzaG91bGQg a2VlcCBjYXB0dXJlZCB2YXJpYWJsZXMgYWxpdmUgdW50aWwgdGhlIChpbmxpbmUpIHJldHVybi4K SW5kZXg6IExheW91dFRlc3RzL2Zhc3Qvdmlld3BvcnQvdmlld3BvcnQtMTMzLWV4cGVjdGVkLnR4 dAo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09Ci0tLSBMYXlvdXRUZXN0cy9mYXN0L3ZpZXdwb3J0L3ZpZXdwb3J0LTEzMy1l eHBlY3RlZC50eHQJKHJldmlzaW9uIDExODg1OSkKKysrIExheW91dFRlc3RzL2Zhc3Qvdmlld3Bv cnQvdmlld3BvcnQtMTMzLWV4cGVjdGVkLnR4dAkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDMg QEAKIHZpZXdwb3J0IHNpemUgNDgweDUzNCBzY2FsZSAxLjAwMDAwMCB3aXRoIGxpbWl0cyBbMS4w MDAwMDAsIDUuMDAwMDAwXSBhbmQgdXNlclNjYWxhYmxlIC0xLjAwMDAwMAotUEFTUyB3aW5kb3cu ZGV2aWNlUGl4ZWxSYXRpbyBpcyB3aXRoaW4gMC4wMDEgb2YgMC43MDgzMzMzMTM0NjUxMTg0CitQ QVNTIHdpbmRvdy5kZXZpY2VQaXhlbFJhdGlvIGlzIHdpdGhpbiAwLjAwMSBvZiAxCiAKSW5kZXg6 IExheW91dFRlc3RzL2Zhc3Qvdmlld3BvcnQvdmlld3BvcnQtMTMzLmh0bWwKPT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PQot LS0gTGF5b3V0VGVzdHMvZmFzdC92aWV3cG9ydC92aWV3cG9ydC0xMzMuaHRtbAkocmV2aXNpb24g MTE4ODU5KQorKysgTGF5b3V0VGVzdHMvZmFzdC92aWV3cG9ydC92aWV3cG9ydC0xMzMuaHRtbAko d29ya2luZyBjb3B5KQpAQCAtNyw3ICs3LDcgQEAKICAgICAgICAgICAgIGlmICh3aW5kb3cubGF5 b3V0VGVzdENvbnRyb2xsZXIpIHsKICAgICAgICAgICAgICAgICBsYXlvdXRUZXN0Q29udHJvbGxl ci5kdW1wQXNUZXh0KCk7CiAgICAgICAgICAgICAgICAgbGF5b3V0VGVzdENvbnRyb2xsZXIuZHVt cENvbmZpZ3VyYXRpb25Gb3JWaWV3cG9ydCgyNDAsIDQ4MCwgNTY0LCA0ODAsIDUzNCk7Ci0gICAg ICAgICAgICAgICAgc2hvdWxkQmVDbG9zZVRvKCJ3aW5kb3cuZGV2aWNlUGl4ZWxSYXRpbyIsIDAu NzA4MzMzMzEzNDY1MTE4NCwgMC4wMDEsIGZhbHNlKTsKKyAgICAgICAgICAgICAgICBzaG91bGRC ZUNsb3NlVG8oIndpbmRvdy5kZXZpY2VQaXhlbFJhdGlvIiwgMSwgMC4wMDEsIGZhbHNlKTsKICAg ICAgICAgICAgIH0KICAgICAgICAgfQogICAgIDwvc2NyaXB0Pgo= 144661 2012-05-29 17:38:50 -0700 2012-05-30 00:37:37 -0700 Patch for landing bug-87407-20120529173849.patch text/plain 5278 abarth SW5kZXg6IFNvdXJjZS9XZWJDb3JlL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2Vi Q29yZS9DaGFuZ2VMb2cJKHJldmlzaW9uIDExODg1OSkKKysrIFNvdXJjZS9XZWJDb3JlL0NoYW5n ZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwzICsxLDMwIEBACisyMDEyLTA1LTI5ICBBZGFtIEJh cnRoICA8YWJhcnRoQHdlYmtpdC5vcmc+CisKKyAgICAgICAgd2luZG93LmRldmljZVBpeGVsUmF0 aW8gc2hvdWxkIHJldHVybiB0aGUgZGV2aWNlJ3MgcGl4ZWwgcmF0aW8gbm90IGEgdmFsdWUgZGVw ZW5kZW50IG9uIHRoZSB2aWV3cG9ydAorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9z aG93X2J1Zy5jZ2k/aWQ9ODc0MDcKKworICAgICAgICBSZXZpZXdlZCBieSBEZWFuIEphY2tzb24u CisKKyAgICAgICAgVGhlcmUncyBiZWVuIGEgbG90IG9mIGRpc2N1c3Npb24gYWJvdXQgdGhpcyBw YXRjaCwgYm90aCBvbiB0aGUgYnVnCisgICAgICAgIHJlZmVyZW5jZWQgYWJvdmUgYW5kIG9uICN3 ZWJraXQuCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVnLmNnaT9pZD04 NzQwNyNjMzAgaGFzIGEgZ29vZCBzdW1tYXJ5IG9mCisgICAgICAgIHRoZSBkaXNjdXNzaW9uIGFu ZCBtb3RpdmF0aW9uIGJlaGluZCB0aGlzIGNoYW5nZS4gIEVzc2VudGlhbGx5LCB0aGlzCisgICAg ICAgIHBhdGNoIG1ha2VzIFdlYktpdCdzIGJlaGF2aW9yIG9uIHRydW5rIG1hdGNoIHRoYXQgb2Yg dGhlIGN1cnJlbnRseQorICAgICAgICBzaGlwcGluZyBjb2RlIG9uIGlPUyBhbmQgQW5kcm9pZC4g IE9uIGlPUywgdGhpcyBjaGFuZ2UgaXNuJ3QgZGV0ZWN0YWJsZQorICAgICAgICBiZWNhdXNlIGlP UyBkb2Vzbid0IHN1cHBvcnQgdGFyZ2V0LWRlbnNpdHlkcGkuCisKKyAgICAgICAgVGhlcmUncyBi ZWVuIHNvbWUgY29uZnVzaW9uIGFib3V0IHRoaXMgcGF0Y2ggYmVjYXVzZSB0aGUgQW5kcm9pZAor ICAgICAgICBkb2N1bWVudGF0aW9uIHNlZW1zIHRvIGltcGx5IHRoYXQgd2Ugc2hvdWxkbid0IG1h a2UgdGhhdCBjaGFuZ2UsIGJ1dAorICAgICAgICB0aGF0IGRvY3VtZW50YXRpb24gaXMgd3Jvbmcg aW4gdGhhdCBpdCBkb2Vzbid0IHJlZmxlY3QgQW5kcm9pZCdzCisgICAgICAgIGltcGxlbWVudGF0 aW9uIG9mIHRhcmdldC1kZW5zaXR5ZHBpIGFuZCB0aGVyZSBhcHBlYXJzIHRvIGJlIGNvbnRlbnQK KyAgICAgICAgdGhhdCByZWxpZXMgb24gdGhlIGltcGxlbWVudGF0aW9uIGJlaGF2aW9yIHJhdGhl ciB0aGFuIHRoZSBkb2N1bWVudGVkCisgICAgICAgIGJlaGF2aW9yLgorCisgICAgICAgICogY3Nz L01lZGlhUXVlcnlFdmFsdWF0b3IuY3BwOgorICAgICAgICAoV2ViQ29yZTo6ZGV2aWNlX3BpeGVs X3JhdGlvTWVkaWFGZWF0dXJlRXZhbCk6CisgICAgICAgICogcGFnZS9ET01XaW5kb3cuY3BwOgor ICAgICAgICAoV2ViQ29yZTo6RE9NV2luZG93OjpkZXZpY2VQaXhlbFJhdGlvKToKKwogMjAxMi0w NS0yOSAgU2hhd24gU2luZ2ggIDxzaGF3bnNpbmdoQGNocm9taXVtLm9yZz4KIAogICAgICAgICBb Y2hyb21pdW1dIG1ha2UgV2ViVHJhbnNmb3JtYXRpb25NYXRyaXggb2JqZWN0IHVzYWJsZSBieSBu b24td2Via2l0IGNvZGUKSW5kZXg6IFNvdXJjZS9XZWJDb3JlL2Nzcy9NZWRpYVF1ZXJ5RXZhbHVh dG9yLmNwcAo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09Ci0tLSBTb3VyY2UvV2ViQ29yZS9jc3MvTWVkaWFRdWVyeUV2YWx1 YXRvci5jcHAJKHJldmlzaW9uIDExODg1OSkKKysrIFNvdXJjZS9XZWJDb3JlL2Nzcy9NZWRpYVF1 ZXJ5RXZhbHVhdG9yLmNwcAkod29ya2luZyBjb3B5KQpAQCAtNDUsNiArNDUsNyBAQAogI2luY2x1 ZGUgIlBsYXRmb3JtU2NyZWVuLmgiCiAjaW5jbHVkZSAiUmVuZGVyVmlldy5oIgogI2luY2x1ZGUg IlJlbmRlclN0eWxlLmgiCisjaW5jbHVkZSAiU2V0dGluZ3MuaCIKICNpbmNsdWRlICJTdHlsZVJl c29sdmVyLmgiCiAjaW5jbHVkZSA8d3RmL0hhc2hNYXAuaD4KIApAQCAtMjkwLDEwICsyOTEsMTEg QEAgc3RhdGljIGJvb2wgZGV2aWNlX2FzcGVjdF9yYXRpb01lZGlhRmVhdAogCiBzdGF0aWMgYm9v bCBkZXZpY2VfcGl4ZWxfcmF0aW9NZWRpYUZlYXR1cmVFdmFsKENTU1ZhbHVlICp2YWx1ZSwgUmVu ZGVyU3R5bGUqLCBGcmFtZSogZnJhbWUsIE1lZGlhRmVhdHVyZVByZWZpeCBvcCkKIHsKKyAgICBm bG9hdCBzY2FsZUZhY3RvciA9IGZyYW1lLT5wYWdlKCktPnNldHRpbmdzKCktPmRlZmF1bHREZXZp Y2VTY2FsZUZhY3RvcigpOwogICAgIGlmICh2YWx1ZSkKLSAgICAgICAgcmV0dXJuIHZhbHVlLT5p c1ByaW1pdGl2ZVZhbHVlKCkgJiYgY29tcGFyZVZhbHVlKGZyYW1lLT5wYWdlKCktPmRldmljZVNj YWxlRmFjdG9yKCksIHN0YXRpY19jYXN0PENTU1ByaW1pdGl2ZVZhbHVlKj4odmFsdWUpLT5nZXRG bG9hdFZhbHVlKCksIG9wKTsKKyAgICAgICAgcmV0dXJuIHZhbHVlLT5pc1ByaW1pdGl2ZVZhbHVl KCkgJiYgY29tcGFyZVZhbHVlKHNjYWxlRmFjdG9yLCBzdGF0aWNfY2FzdDxDU1NQcmltaXRpdmVW YWx1ZSo+KHZhbHVlKS0+Z2V0RmxvYXRWYWx1ZSgpLCBvcCk7CiAKLSAgICByZXR1cm4gZnJhbWUt PnBhZ2UoKS0+ZGV2aWNlU2NhbGVGYWN0b3IoKSAhPSAwOworICAgIHJldHVybiBzY2FsZUZhY3Rv ciAhPSAwOwogfQogCiBzdGF0aWMgYm9vbCBncmlkTWVkaWFGZWF0dXJlRXZhbChDU1NWYWx1ZSog dmFsdWUsIFJlbmRlclN0eWxlKiwgRnJhbWUqLCBNZWRpYUZlYXR1cmVQcmVmaXggb3ApCkluZGV4 OiBTb3VyY2UvV2ViQ29yZS9wYWdlL0RPTVdpbmRvdy5jcHAKPT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PQotLS0gU291cmNl L1dlYkNvcmUvcGFnZS9ET01XaW5kb3cuY3BwCShyZXZpc2lvbiAxMTg4NTkpCisrKyBTb3VyY2Uv V2ViQ29yZS9wYWdlL0RPTVdpbmRvdy5jcHAJKHdvcmtpbmcgY29weSkKQEAgLTEzODIsNyArMTM4 Miw3IEBAIGRvdWJsZSBET01XaW5kb3c6OmRldmljZVBpeGVsUmF0aW8oKSBjb24KICAgICBpZiAo IXBhZ2UpCiAgICAgICAgIHJldHVybiAwLjA7CiAKLSAgICByZXR1cm4gcGFnZS0+ZGV2aWNlU2Nh bGVGYWN0b3IoKTsKKyAgICByZXR1cm4gcGFnZS0+c2V0dGluZ3MoKS0+ZGVmYXVsdERldmljZVNj YWxlRmFjdG9yKCk7CiB9CiAKIHZvaWQgRE9NV2luZG93OjpzY3JvbGxCeShpbnQgeCwgaW50IHkp IGNvbnN0CkluZGV4OiBMYXlvdXRUZXN0cy9DaGFuZ2VMb2cKPT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PQotLS0gTGF5b3V0 VGVzdHMvQ2hhbmdlTG9nCShyZXZpc2lvbiAxMTg4NTkpCisrKyBMYXlvdXRUZXN0cy9DaGFuZ2VM b2cJKHdvcmtpbmcgY29weSkKQEAgLTEsMyArMSwxNiBAQAorMjAxMi0wNS0yOSAgQWRhbSBCYXJ0 aCAgPGFiYXJ0aEB3ZWJraXQub3JnPgorCisgICAgICAgIHdpbmRvdy5kZXZpY2VQaXhlbFJhdGlv IHNob3VsZCByZXR1cm4gdGhlIGRldmljZSdzIHBpeGVsIHJhdGlvIG5vdCBhIHZhbHVlIGRlcGVu ZGVudCBvbiB0aGUgdmlld3BvcnQKKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hv d19idWcuY2dpP2lkPTg3NDA3CisKKyAgICAgICAgUmV2aWV3ZWQgYnkgRGVhbiBKYWNrc29uLgor CisgICAgICAgIFVwZGF0ZSB0aGlzIHRlc3QgdG8gc2hvdyB0aGF0IHdpbmRvdy5kZXZpY2VQaXhl bFJhdGlvIGRvZXNuJ3QgZGVwZW5kIG9uCisgICAgICAgIHRoZSB2aWV3cG9ydCBtZXRhIHRhZy4K KworICAgICAgICAqIGZhc3Qvdmlld3BvcnQvdmlld3BvcnQtMTMzLWV4cGVjdGVkLnR4dDoKKyAg ICAgICAgKiBmYXN0L3ZpZXdwb3J0L3ZpZXdwb3J0LTEzMy5odG1sOgorCiAyMDEyLTA1LTI5ICBG aWxpcCBQaXpsbyAgPGZwaXpsb0BhcHBsZS5jb20+CiAKICAgICAgICAgREZHIHNob3VsZCBrZWVw IGNhcHR1cmVkIHZhcmlhYmxlcyBhbGl2ZSB1bnRpbCB0aGUgKGlubGluZSkgcmV0dXJuLgpJbmRl eDogTGF5b3V0VGVzdHMvZmFzdC92aWV3cG9ydC92aWV3cG9ydC0xMzMtZXhwZWN0ZWQudHh0Cj09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT0KLS0tIExheW91dFRlc3RzL2Zhc3Qvdmlld3BvcnQvdmlld3BvcnQtMTMzLWV4cGVj dGVkLnR4dAkocmV2aXNpb24gMTE4ODU5KQorKysgTGF5b3V0VGVzdHMvZmFzdC92aWV3cG9ydC92 aWV3cG9ydC0xMzMtZXhwZWN0ZWQudHh0CSh3b3JraW5nIGNvcHkpCkBAIC0xLDMgKzEsMyBAQAog dmlld3BvcnQgc2l6ZSA0ODB4NTM0IHNjYWxlIDEuMDAwMDAwIHdpdGggbGltaXRzIFsxLjAwMDAw MCwgNS4wMDAwMDBdIGFuZCB1c2VyU2NhbGFibGUgLTEuMDAwMDAwCi1QQVNTIHdpbmRvdy5kZXZp Y2VQaXhlbFJhdGlvIGlzIHdpdGhpbiAwLjAwMSBvZiAwLjcwODMzMzMxMzQ2NTExODQKK1BBU1Mg d2luZG93LmRldmljZVBpeGVsUmF0aW8gaXMgd2l0aGluIDAuMDAxIG9mIDEKIApJbmRleDogTGF5 b3V0VGVzdHMvZmFzdC92aWV3cG9ydC92aWV3cG9ydC0xMzMuaHRtbAo9PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBM YXlvdXRUZXN0cy9mYXN0L3ZpZXdwb3J0L3ZpZXdwb3J0LTEzMy5odG1sCShyZXZpc2lvbiAxMTg4 NTkpCisrKyBMYXlvdXRUZXN0cy9mYXN0L3ZpZXdwb3J0L3ZpZXdwb3J0LTEzMy5odG1sCSh3b3Jr aW5nIGNvcHkpCkBAIC03LDcgKzcsNyBAQAogICAgICAgICAgICAgaWYgKHdpbmRvdy5sYXlvdXRU ZXN0Q29udHJvbGxlcikgewogICAgICAgICAgICAgICAgIGxheW91dFRlc3RDb250cm9sbGVyLmR1 bXBBc1RleHQoKTsKICAgICAgICAgICAgICAgICBsYXlvdXRUZXN0Q29udHJvbGxlci5kdW1wQ29u ZmlndXJhdGlvbkZvclZpZXdwb3J0KDI0MCwgNDgwLCA1NjQsIDQ4MCwgNTM0KTsKLSAgICAgICAg ICAgICAgICBzaG91bGRCZUNsb3NlVG8oIndpbmRvdy5kZXZpY2VQaXhlbFJhdGlvIiwgMC43MDgz MzMzMTM0NjUxMTg0LCAwLjAwMSwgZmFsc2UpOworICAgICAgICAgICAgICAgIHNob3VsZEJlQ2xv c2VUbygid2luZG93LmRldmljZVBpeGVsUmF0aW8iLCAxLCAwLjAwMSwgZmFsc2UpOwogICAgICAg ICAgICAgfQogICAgICAgICB9CiAgICAgPC9zY3JpcHQ+Cg==