Leaderboard

@dmiranda What do you think about this custom button?

Thanks for replying! I will post the new version 2.0 of Restart & Purge soon. I have reworked the code considerably, and after a lot of testing, this custom button now works perfectly. The old version wasn't bad, but the new one is definitely better and more compatible. I think there should no longer be any problems that could have occurred in very rare cases. BTW, there are a lot of alternatives for the About-Menu custom button in Serpent 52.

The Custom Buttons version I use has been modified by me. The original version was actually multilingual, but the locales were not complete and therefore faulty. Many entities were missing, and I didn't feel like correcting and translating all that. Too much work. I prefer the English language in this area anyways. That's why I turned the multilingual version into a purely English one. I don't have the original link anymore, but maybe I'll write about this version here soon and upload my mod. I already posted a screenshot of this version:

I got a crash log with proper pdb file supplied in https://github.com/roytam1/basilisk55/issues/10#issuecomment-2408527740 so the problem is thread local storage introduced in https://github.com/libjpeg-turbo/libjpeg-turbo/commit/f579cc11b33e5bfeb9931e37cc74b4a33c95d2e6 , and I'm going to `#define THREAD_LOCAL ` to turn this off. EDIT: alright all gkmedias.dll are replaced in today's release

I have found the source of the problem with HDMI . In case you will ever use Nvidia's HDMI (Azalia) or any other HD drivers you have to install KB901105 x64 before the drivers. V3 worked for me . It's very important to install this KB before. Still , I do not know why the Nvidia CPL shows gen.1 PCI express . Could you look at yours ?

@UCyborg, I wanted to ask you , does Nvidia's HDMI work fine for you with XP64 and these drivers ? EDIT . My mobo has PCI express gen.2 . Windows Vista shows gen.2 , but with XP64 Nvidia CPL shows that it's gen.1 only. Any idea why is that ? Also , tried some DX9 games and the performance is much worse on the same hardware and with same driver versions. I'm not going to use it , of course , it's just out of curiosity . Thanks.

Thanks for the suggestion , actually GPUZ was the first thing I've tried . But any version of GPU-z starting from 0-8-5 hangs up the whole windows XP64 system to the point of hard reset . Please keep in mind that I'm talking about 64 bit , not x86. Older versions of GPU-z work fine , but they do not properly detect this GPU. By the way , Nvidia CPL must show the highest version of PCI express available, it shouldn't depend on the current load. And it shows gen. 2.0 on this exact PC with Vista x64. My card is not EVGA , I try to avoid EVGA as much as I can , I've had quite a few problems with their quality. It's just Nvidia (no other branding, from their website) , reference design. I do notice poor 3d performance with games on XP64, the same game is like 2x times faster with Vista. Hardware: GPU: Geforce GTX Titan 6GB mfg date February 2013, Motherboard: Fujitsu Siemens D3012-A11-GS3 DDR3 RAM 16GB with Intel chipset Q45 mfg date April 2011. CPU: Intel® Xeon® Processor X3370 12M Cache, 3.00 GHZ (Quad). PSU: FSP (Fujitsu Siemens Power) 600W ,rare design and size ,supplied with this PC, fits only Fujitsu/Siemens housings.

I'm not sure how GTX750ti is relevant to this discussion , being lower end and from another series. Yeah , I know it's the first gen. of Maxwell , but still another series . Btw, artifacts in Unique Heaven oftenly tell that somethig's wrong with the videocard itself . No such issue for me with high end 900 series and 344 drivers. What about the other issue with displays ? Besides , what version of XP you were using ? 32 or 64 bit ?

Well , that's at least 3 by you and a couple by me . I can confirm all mentioned by you and add these to the list. 1 - Significantly decreased performance starting with 35.x 2 - Problems with idle modes on GTX980/970 , GTX780 (new revision, reissue with another chip) , Titan Black. Though, the first GTX 780 (from May 2013) and the original Titan did not behave oddly during idle. Mind you , this happened on both , Vista and XP , so XP has nothing to do with it , it's the drivers themselves. My opinion : 344 series are the best for XP. If someone wants newer , 364 drivers seem to be OK , but not as fast 344.11 and they need to be modded to work with 9xx series. For some raging publishers : I didn't test with Titan X , so no comment. EDIT : I did not test with lower end cards .

I use: ytdl-format="bestvideo[height<=1080][ext=mp4][vcodec^=avc]+bestaudio[ext=m4a]/best[ext=mp4]/best" Only mp4 with avc codec plays hardware accelerated on my old PC... Pass hwdec=auto to MPV. Is your MPV player (or other player) up to date?

You'd need at least Outlook 2013. Also, Outlook 2013 and Outlook 2016 require registry edits to force-enable ModernAuth: https://learn.microsoft.com/en-us/exchange/troubleshoot/administration/modern-authentication-configuration From what I read, it also only works over IMAP and you need the auto-account-setup wizard as opposed to the old-style one in Control Panel. An app password is still required for POP3, in which case Outlook 2007 should work too. Edit: someone mentioned it still wouldn't work for them until they allowed Office to update itself.

BTW, it wasn't meant offensive. It was just an observation I made when reading your comments. I'm someone who likes to say what I think, sometimes very directly. And as for Rammstein, I'm sorry to say that I've never been a fan.

Does Rammstein's Adieu fit well when reading them? I keep running into issues when slightly deviating from few select sites and it doesn't inspire positivity, even though I'm sure I'm one of the most positive people here when it comes to UXP. How many of you have financially contributed to development? I have, and it wasn't just a couple of bucks. @Slavich You should take that reading after YT's interface is fully loaded.

Yesterday, I found this old button code in my archive. As the code of the button looked promising, I translated it and modified it completely. I changed the tooltip, totally reworked the menu of the important about: pages and added a help menu. I created a version for New Moon/Pale Moon and a version for Serpent/Basilisk/Firefox as some about: pages are different in these browsers. This button has a special feature. When selecting a string or copying a string to the clipboard, a middle-click onto the toolvar button opens it in the about:config page. All in all, it's a kind of recycling of old code so that it works again in our browsers. Thus, old code is not always useless and lost. And the code of this button is so universal that it seems to run in all Firefox-based browsers. Anyway! I will also post this forked custom button here very soon.

Yes, you can put that line in your youtube-dl config file. E.g.: C:\Documents and Settings\$USER\Application Data\youtube-dl\config.txt My config file for youtube-dl: --rm-cache-dir -i --no-mtime --add-metadata --embed-thumbnail -o "~/Desktop/%(title)s.%(ext)s" -f bestvideo[height<=1080][vcodec^=avc][ext=mp4]+bestaudio[ext=m4a]/best[ext=mp4]/best --user-agent "Mozilla/5.0 (Android 14; Mobile; rv:115.0) Gecko/115.0 Firefox/115.0"

The entire classic taskbar is, you might say euphemistically, heavily discouraged, having been removed in win11. Why should users who want classic functionality care about Microsoft's "discouragement"? Why would you help people restore classic functionality, and then "discourage" its use? Well, that's subjective. I've been using quick launch and its overflow menu from its inception and find it adequate for launching apps and opening certain frequent folders. I don't need a better way. I need the old way to keep working and to also work on the secondary when the primary is hidden by a full-screen app. A sane default for clone mode (where the secondary task bar already shows the same running tasks and same tray icons as the primary) is to also show the same toolbars. That's more logical, coherent and useful than the current partial clone. But for automatic moving of the primary taskbar when hidden by a full-screen app on primary, an option is needed, because that would be a new mode, which can't be forced on everyone.

That must be it . I first experienced a crash on this very forum. Roytam1 has uploaded new versions with gkmedias.dll replaced. Now everything is OK again.

I've been using multiple QuickLaunch toolbars practically since they were introduced with the IE4 desktop update – they are my main way of launching programs, specifically those that I use regularly, but not all the time (the programs that I use all the time have been pinned since that's been introduced in Windows 7). OTOH, I never missed the ability to have toolbars on secondary displays.

Personally, I use a much more recent Custom Buttons extension with a lot of additional features created by a well-known and very talented developer of custom buttons and extensions who was also very active in the old Custom Buttons forum that time. This Custom Buttons extension is of course particularly interesting for developers of custom buttons.

I have unfortunately noticed that strangely enough, the interest in custom buttons does not seem to be particularly high here. Therefore my question to all users of UXP browsers is how you feel about the custom buttons and whether you use them at all. TBH, if there is no real interest though, I do not need to post more custom buttons here, either. Especially the ones I created or modified myself. Personally, I am a fan of the legacy custom buttons and try to use them as much as possible. They use much less resources and cause much less conflict than extensions. Therefore, in my browsers, additional functions are first provided by custom buttons and only then the still missing ones by extensions.

Thanks guys.

Are you saying they shouldn't exist at all? They certainly do have a place, because that's been a feature of the classic taskbar, through win10. A quick launch toolbar, for example, is quite useful and it also has a useful overflow menu, which can have shortcuts to programs and folders and these can be organized into sub-folders which turn into sub-menus. Since the goal is to restore the classic taskbar, removing the toolbar feature does not make sense, that would not be a restoration. What does make sense is that a clone (if clone mode is configured) of the primary taskbar on the secondary screen should have the same functionality. Is my understanding of the specifics of what you proposed correct? That there would be an option for the primary taskbar to automatically move to the secondary screen, when an app (e.g. a vm) goes full-screen on the primary.

Confirmed. The gkmedias.dll file (8.496.128 Bytes) is faulty. I replaced it by the previous gkmedias.dll file (8.623.104 Bytes), and the error is history. I can confirm the same problem in the most recent release of New Moon 28 (2024-10-11). I replaced gkmedias.dll file (9.862.144 Bytes) by the previous one (9.989.632 Bytes), and the error is history.

The site is down until they can redo security and no data was lost. It was posted on x but I can't find the official posting. It doesn't help that one of archive's main admins blocks so many people on there.

This tutorial was created following the discussion in this other post. https://msfn.org/board/topic/181623-updates-office-2007/page/5/ We will be working with the WSUS Offline Update program. Until 2020 it had current versions and long-term maintenance ESR versions. https://download.wsusoffline.net/ The program is currently only relatively community-maintained with versions up to 2021 called CE with links to GitLab provided on the forum and no current version is compatible with Office 2003/2007/2010. https://gitlab.com/wsusoffline/wsusoffline/-/releases WSUS Offline Update 9.2.6 ESR (2019-11-08) is the latest version with the option to get updates for Office 2003/2007. The download link is not listed on the official page but the download is done from the official server. https://download.wsusoffline.net/wsusoffline926.zip WSUS Offline Update 11.9.6 ESR CE (2020-10-23) is the latest version with the option to get updates for Office 2010 and requires Windows Seven (maybe it works on Vista). https://gitlab.com/-/project/19504919/uploads/20adc1fc933dda0367de405e907da7cd/wsusofflineCE1196.zip STEPS PRIOR TO RUNNING WSUS OFFLINE UPDATE We download WSUS and unzip it to make some modifications prior to its execution. 1- Get WSUS to download old updates without "Secure Hash Algorithm". Using Notepad++ (v. 7.9.2 latest XP compatible) we modified the DownloadUpdates.cmd file in the "cmd" folder by adding "--no-check-certificate" in the following line: -WSUS 9.2.6 line 206 " set DLDR_PATH=..\bin\wget.exe --no-check-certificate" -WSUS 11.9.6 line 261 " set DLDR_PATH=%WGET_PATH% --no-check-certificate" Thanks @AstroSkipper 2- Add pause in the CMD window during the execution of WSUS to get an old wsusscn2.cab version to be pasted into the client/wsus folder. *In the case of WSUS 9.2.6 the last working wsusscn2.cab version is dated 2019-11-13 if you try to work with newer versions thousands of files (13 GiB) are downloaded that do not contain any updates for Microsoft Office, there are others that are not related to Office, such as random updates for Windows 2000/XP/Vista/Seven/10, .NET, etc. but others that are not related to Office such as random updates for Windows 2000/XP/Vista/Seven/10, .NET, etc. To work with an older version of wsusscn2.cab we will modify the following line in the "DownloadUpdates.cmd" file in the "cmd" folder by adding the "pause" command. -WSUS 9.2.6 line 978 -WSUS 11.9.6 line 1168 echo The script has stopped before extracting package.xml from the current wsusscn2.cab file. You can now manually replace it with an older one. pause Thanks @AstroSkipper 3- Add pause to CMD window before creating ISO file to add downloads that fail. We modified the following line in the "CreateISOImage.cmd" file in the "cmd" folder by adding the "pause" command. -WSUS 9.2.6 line 3 -WSUS 11.9.6 line 4 echo The script has been paused before creating the ISO file. Manually add the updates with failed download. pause Thanks @AstroSkipper 4- Get the files needed to run WSUS that are located in the main root bin folder and whose download fails. *There are 2 bin folders, I have not checked if it is also necessary to place them in the other bin folder located in client\bin. I have included them in both folders. -WSUS 9.2.6 \bin mkisofs exe sigcheck exe (v2.30 latest XP supported) http://wayback.archive.org/web/20160618051340/http://skwire.dcmembers.com/apps/sigcheckgui/SigcheckGUI.zip wget exe (the file that comes with WSUS is faulty) (v1.19.4 latest XP supported) https://eternallybored.org/misc/wget/1.19.4/32/wget.exe https://eternallybored.org/misc/wget/ -WSUS 11.9.6 \bin mkisofs exe Also in WSUS 9.2.6 we have to include the static download link for "sigcheck v2.30" inside the file "StaticDownloadLinks-sysinternals.txt" located in the "static" folder, replacing the current one which is not compatible with Windows XP and leaving the file: http://download.sysinternals.com/files/AutoLogon.zip http://wayback.archive.org/web/20160618051340/http://skwire.dcmembers.com/apps/sigcheckgui/SigcheckGUI.zip http://download.sysinternals.com/files/Streams.zip 5- Get static updates for the Office 2010 x64-bit version. Dynamic updates are downloaded in separate .CAB files for the 32-bit and 64-bit versions, but WSUS Offline Update does not have an option to download 64-bit static updates. There are two new .TXT files that go in the \static\custom folder. They contain the links to SP2 x64 and two more files. https://forums.wsusoffline.net/viewtopic.php?f=3&t=9818 *files versions for Office 2010 32-bit are automatically downloaded with WSUS 11.9.6 but are not downloaded with WSUS 9.2.6. NOTES: The steps above also help us to obtain updates for Windows XP or higher (previous versions do not work), .NET Framework and C++ if we check the corresponding options. Updates for .NET Framework: -If we want to avoid .NET updates that are not compatible with a version of Windows, we must exclude them in these 2 files in the "exclude" folder: ExcludeList-dotnet-x32.txt ExcludeList-dotnet-x64.txt For example, to avoid updates that are not compatible with XP, we must edit these 2 files, adding to each one: ndp45- ndp46- ndp47- ndp48- * the final dash is part of the text of the name of the excluded file. We must take into account that for any language, several files are also downloaded in German, such as "dotnetfx35langpack_x86de exe", which must be replaced manually with the corresponding one for our language or the following two files in the "static" folder must be modified, replacing the links in German: StaticDownloadLinks-dotnet-x32-glb.txt StaticDownloadLinks-dotnet-x64-glb.txt -All .NET Framework installers included in the two previous files and any additional language packs that are also included will be downloaded. In no WSUS Update Offline version is there a link to download the .NET Framework 1.1 installer, from which its updates are downloaded. Updates for Windows XP: -We must take into account that WSUS Offline Update 9.2.6 includes the updates for the XP Embedded version, if we do not want to download them we have to use WSUS 9.2.5. -The WindowsXP-KB968930-x86 file is downloaded in the UK English language version (ENG), if we want it in our language we have to replace it. *IMPORTANT: I have not checked, if After adding files with names not contained in the wsusscn2.cab downloads, if you run UpdateInstaller exe to install the updates, these new files are installed. FINAL STEPS AFTER PREVIOUS MODIFICATIONS After the previous steps... 1- Check that we have the files necessary to run WSUS in these folders \bin \client\bin \cmd \static Do not add the other downloads that fail yet, as they may not do so and the pasted files may be deleted during the download. 2- Paste the current wsusscn2.cab version in the client\wsus folder or let it download automatically when WSUS is run. 3- Run UpdateGenerator exe Uncheck all options except "ISO" and optionally Service Packs. Choose our desired latest update download and press "Start". The downloads performed are not linked to the operating system from which WSUS Offline Update is run or to whether we have one or another Microsoft Office version installed. These are independent, if the downloads number varies from one operating system to another it is due to an error in the program. 4- Paste an older wsusscn2.cab version when the message appears. (the message appears if you have "Pause" enabled in DownloadUpdates.cmd) For WSUS 9.2.6 for Office 2003 and 2007, version 2019-04-07 is recommended (the oldest version that contains all the updates) http://web.archive.org/web/20190420162709/http://download.windowsupdate.com/microsoftupdate/v6/wsusscan/wsusscn2.cab For WSUS 9.2.6 for Office 2010, version 2019-11-13 is recommended (the last version that works with WSUS 9.2.6). This does not contain the latest updates for Office 2010, if we want to get them it has to be from WSUS Offline Update 11.9.6 which is not compatible with Windows XP. http://web.archive.org/web/20191209214827/http://download.windowsupdate.com/microsoftupdate/v6/wsusscan/wsusscn2.cab For WSUS 11.9.6 for Office 2010, version 2023-11-14 (581 MiB) is recommended http://web.archive.org/web/20231114200328/http://download.windowsupdate.com/microsoftupdate/v6/wsusscan/wsusscn2.cab as the file is 215 MiB smaller than the older version that contains all updates for Office 2010 from 2021-04-13 (796 MiB) http://web.archive.org/web/20210504233947/http://download.windowsupdate.com/microsoftupdate/v6/wsusscan/wsusscn2.cab WSUS 9.2.6 running from Windows Seven asks for 800 dynamic downloads for Office 2003/2007. From Windows XP searching for updates for Office 2003/2007 may fail and WSUS 9.2.6 will ask for 2529 downloads. The error is due to the file "ExcludeList-superseded.txt" not being created correctly in the "exclude" folder. Once the problem is corrected WSUS 9.2.6 will also ask for 800 downloads from XP. You should check if the "ExcludeList-superseded.txt" file in the "exclude" folder is not empty after WSUS 9.2.6 checks the current "wsusscn2.cab" version. In this case, close the CMD download window but do not close the WSUS program, paste a correct copy of "ExcludeList-superseded.txt" and press Start again. If it fails, try repeating the process until it works. Sometimes everything fails, restart Windows and try again. The "ExcludeList-superseded.txt" file can be extracted by searching for other updates than Office 2003/2007 with WSUS 9.2.6. With Office 2010 from WSUS 9.2.6 there is no such problem, although more files are also downloaded than with WSUS 11.9.6, since dynamic updates contain updates for all Office versions that the WSUS version supports. 6- Look in the log for the failed downloads when the message to burn ISO appears. Open the download.log file located in the log folder and look for the links that say "failed". The correct file location folder is just above the "failed" message indicating the path to the folder. 7- Get the files whose download failed and paste them into the corresponding folders before creating the ISO file. (make a copy in a separate folder beforehand because these may disappear if we run WSUS again and it is not configured correctly) If many files appear with the error failed (more than 6 approx.) you should stop the operation and go back to step 4. So many failed downloads means that the connection has failed and some of these are actually available, you should try to download them again by running WSUS again. It is also possible that we have not added the "--no-check-certificate" command in the DownloadUpdates.cmd file. In my opinion, it is a good idea to delete the log before starting the process again to have a clean text or save the previous one as an old copy. 8- Before creating the ISO file, check if the text "unknow error" appears in the log. This means a connection error or similar during update downloads. In this case, we must stop the operation and return to step 4 (run UpdateGenerator.exe again) until the error disappears. 9- Continue with the ISO creation. *Remember that dynamic updates are downloaded for all Office versions that contain a WSUS Offline Update option, even if only Office 2003 (v.11), 2007 (v.12) or 2010 (v.14) is selected, updates for Office 2013 (v.15) and Office 2016/2019/2021 (all v.16) will be downloaded. If we only want updates for Office 2010 we have to open all the .CAB files one by one and see which version appears in the .XML file. 10- Before installing updates for Office 2007/2010 on XP with UpdateInstaller.exe it is necessary to first install the "Office 2xxx Upsourcer" patch with its prerequisites so that they are compatible with XP. https://msfn.org/board/topic/181623-updates-office-2007/?do=findComment&comment=1270419

Version history up to v5.30: [Ronen Tzur, Invincea and Sophos, last capture on 23Apr2020] https://web.archive.org/web/20200423140308/https://www.sandboxie.com/VersionChanges from v5.40.1 onwards [David Xanatos] https://github.com/xanasoft/Sandboxie-Plus/blob/master/CHANGELOG.md Registry fixes by modnar for Sandboxie v5.22 and v5.40 (special edition of Sandboxie by David Xanatos for WinXP, SSE2, 9Apr2020): 1) Rename in registry HKLM\SYSTEM\CurrentControlSet\Services\SbieDrv "DependsOnService" to "DependOnService" [no "s"] 2) Registry fixes to improve compatibility and efficiency, revised on 27Apr2025: GroupOrderList_FSFilter_Infrastructure_03_XP_USP4.regGroupOrderList_Filter_06_XP_USP4.reg Restart Windows after running the registry fixes. SbieDrv_Group_File_System.reg, also listed above earlier, was removed on 16Mar2025 because of the revisions and is not needed anymore. Below are corrections regarding earlier registry fixes by modnar, as of 22Feb2025: 1) The following are virgin backup settings to undo 2 registry patches which were posted here before 22Feb2025 and which have been rejected subsequently: GroupOrderList_Primary_Disk_05_XP_USP4.regGroupOrderList_SCSI_Class_03_XP_USP4.reg 2) If you have run NtfsDisableLastAccessUpdate_1.reg (to reduce the stress on the disk), previously posted here, please remove it from your system (delete it from HKeyLocalMachine/System/CurrentControlSet/Control/Filesystem so there are only 3 factory entries: NtfsDisable8dot3NameCreation (0), Win31FileSystem (0), Win95TruncatedExtensions (1) and reboot) Source - old versions https://web.archive.org/web/20170601000000*/http://www.sandboxie.com:80/SandboxieInstall.exe [best, only combined 32+64bit versions] https://web.archive.org/web/*/https://www.sandboxie.com/attic/* [32bit, 64bit and combined versions] https://sandboxie-website-archive.github.io/www.sandboxie.com/AllVersions.html [seems to be the same as https://web.archive.org/web/20200310035403/https://www.sandboxie.com/AllVersions ] Temporary postings:

Unfortunately not, and none of the usual hacks work, I mean DXVK and DX12on7. But the game has terrible performance even for those who have latest PC.

Lotus Lantern: Rescue Mother requires Windows 10 and up.

To sell more DX12 graphic cards? I dunno. What I'm sure of, it doesn't need DX12 with those graphics, agree. DX9, at best.

No. Star Wars: Battlefront is a series of classic games that was launched in 2004. (twenty years ago!) In 2024 they just added an artificial block to fend off windows 7 and below. https://en.wikipedia.org/wiki/Star_Wars:_Battlefront https://store.steampowered.com/app/2446550/STAR_WARS_Battlefront_Classic_Collection/ That's why it makes sense to post here.

STAR WARS: Battlefront Classic Collection Genres: Action, Shooter, First-person, Third-person, 3D Companies: Pandemic Studios, Aspyr Media Requires Windows 8 and above

New NewMoon 27 Build! 32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210508-32a8a4473-xpmod.7z 32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210508-32a8a4473-xpmod-sse.7z 32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210508-32a8a4473-xpmod-ia32.7z 64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20210508-32a8a4473-xpmod.7z source repo: https://github.com/roytam1/palemoon27 repo changes since my last build: - Revert recent dom/media related changes. This should hopefully restore stability as 27 March 2021 build. (c5b776d07) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1178508 ServiceWorker scripts should never be intercepted. r=baku (1dc29c560) - Bug 1169044 - Patch 4 - Set referrer on intercepted request. r=jdm (e400d51bb) - Bug 1169044 - Patch 5 - build fixes. a=bustage (ff54ccc97) - Bug 1158319 Cache should throw SecurityError if the principal is incorrect. r=ehsan (28713b9cf) - Bug 1160147 Improve Cache API WorkerFeature shutdown handling. r=baku (79ab46685) - Bug 1160227 Improve Cache API warnings. r=ehsan (0adda61ee) - Bug 1160138 P1 Update CacheStorage and Cache webidl to latest spec. r=ehsan (68025abcf) - Bug 1160138 P2 Add a [ChromeConstructor] to CacheStorage to support devtools. r=ehsan (8d3ea55b8) - Bug 1160138 P3 Test Cache chrome-only Constructor. r=ehsan (46d89d039) - Bug 1173467 P1 Modify CacheStorage to reject with SecurityErr instead of throwing on creation. r=ehsan (db32dffbf) - Bug 1140145 - Update web-platform-tests expected data to revision 9a2c04e06cb4e63b13b803722d345d085bf0debf, a=testonly (eb5c31e15) - Bug 1142501 - Disable webvtt/rendering/cues-with-video/processing-model/basic.html on Linux due to extremely frequent timeouts. No review. (8c197e60a) - Bug 1135107 - Enable pointer events for pointer events tests, a=testonly (03d19a853) - partial bug 1135107 - Update web-platform-tests expected data, rs=Ms2ger (794895f17) - Bug 1153521 - Update web-platform-tests expected data to revision 7311aa630534282885b9add15b1c30b2b59316dd, a=testonly (44a86bbb9) - Bug 1162594 - create build_resources.json even if psutil fails; r=ahal (44a1d1780) (cdb21a311) - import changes from `dev' branch of rmottola/Arctic-Fox and mozilla: - Bug 1156628 - Don't limit the number of unused dirty pages kept by jemalloc on B2G desktop. r=gsvelto, r=glandium (7b9878b36) - Bug 1172632 - Don't guess malloc_usable_size type of argument based on ANDROID_VERSION in mozjemalloc, but use the result of the configure test instead. r=njn (1f57d41d4) - bug 1172632 - Move some allocator related configure checks in a common location for both top-level and js/src to use. r=mshal (moz-hg: 5c0acaf8f) - Fixup for bug 1172632 on a CLOSED TREE. r=me (moz-hg: a1ea87558) - Bug 1186444 - part 1 - move uses of MODULE_OPTIMIZE_FLAGS to moz.build's CFLAGS; r=mshal (1b1e4e8cf) - Bug 1153566 - Incorrect re in _check_blacklisted_variables did not ignore comment. r=gps (ab563eb51) - goanna->gecko (034340741) - Bug 1162191 - Add |mach artifact| for installing downloaded Fennec binaries. r=gps (059a6ac32) - Bug 1185671 - Use BytesIO instead of StringIO as backing store for FileAvoidWrite, and handle writing unicode to it explicitly. r=gps (3381f4912) - Bug 1142127 - Fix typo in exception for "IDL already registered"; r=gps Bug 924187 - Deal with interfaces.manifest from the backend. r=mshal (32bf6ebbf) - goanna->gecko (7c91fd334) - Bug 1166538 - Make dozip.py a py_action. r=gps (828cdf7a1) - Bug 1166538 - Make it more straightforward, path-wise, to change $(ZIP) uses to the zip py_action. r=gps (b85adce8e) - Bug 1172800 - Move moz.build template handling in a separate class. r=gps (23a2d77bf) - Bug 1172800 - Avoid using inspect.getsourcelines() and inspect.getfile(). r=gps (13021cdf2) - Bug 1162826 - Display a proper error when a moz.build tries to set a "reserved keyword". r=mshal (6e0f1ca3c) - Bug 1172800 - Create actual functions to execute moz.build templates. r=gps (305c886e4) - Bug 1168607 - Add get method to finders; r=glandium (50f6c588b) - Bug 1168231 - Normalize file mode in jars. r=gps (456678caa) - Bug 1168607 - Add a read() method to File; r=glandium (8d3e3e066) - Bug 1168607 - Implement a finder that reads from a Mercurial repo; r=glandium (327d542a1) - Bug 1168607 - Use mozpack Finders for reading moz.build files; r=glandium (44e1b4cf2) - Bug 1168607 - Add mode to MercurialFileFinder to support non-relative paths; r=glandium (ba30a87e8) - Bug 1168607 - Add a native Mercurial finder; r=glandium (081934656) - Bug 1176642 - Use absolute_import in mozbuild; r=glandium (8aaf8d221) - Bug 1154687 - Restore EXTERNALLY_MANAGED_MAKE_FILE handling to the moz.build blacklist check. r=gps (099b2c496) - Bug 1191575 - Optimize the RecursiveMakeBackend._check_blacklisted_variables test. r=gps (3e1a9a03c) - Bug 1186444 - part 2 - remove MODULE_OPTIMIZE_FLAGS from config.mk; r=mshal (2576e322a) - Bug 1226907 - Part 2: Fix warnings in mozjemalloc and remove ALLOW_COMPILER_WARNINGS. r=glandium (154d35b0c) - Bug 1127618 (prereq) - Fix compile errors caused by files moving from one unified file to another. r=mcmanus (629aa7d47) - Bug 1185616 - remove optimization-related hacks in sqlite's Makefile.in; r=glandium (d445dd291) (3a8813078) - import changes from `dev' branch of rmottola/Arctic-Fox: - partial of Bug 1154325 - Disable all DOM Cache web-platform-tests until the intermittent failures have been fixed (e0672af20) - Bug 1154325 P3 Re-enable service-workers/cache-storage WPT tests. r=jgraham (e369b1bb7) - Another followup for bug 1142478, still with a CLOSED TREE (12d097e25) - And one more followup for bug 1142478. Still a CLOSED TREE (2dd25961c) - Bug 1146321 - Update web-platform-tests expected data to revision 1defdd7213b52589e4b7a259e53f4fff359c301a, a=testonly (2057d686f) - Bug 1121099. Make <basefont> be an HTMLElement to match my reading of the spec and Chrome. r=sicking (3f11b8dde) - partial of Bug 1127468 - Update web-platform-tests expected data to revision 83d9681ecf334f5fbe1c4b076601cf3c50f42176, a=testonly (f187379c5) - partial of Bug 1135107 - Update web-platform-tests expected data (1e307d251) - partial of Bug 1159756. Stop forcing the media.mediasource.enabled preference to true and the media.mediasource.whitelist preference to false in the test harness. r=kinetik (416e6fd57) - Bug 1161535 - Update web-platform-tests expected data to revision 40a9c4e9e4f99a738cd1a7602066c5e84d1b90b5, a=testonly (f7ef60633) - Bug 1161759 Remove stale wpt ini files for Cache API tests. r=jgraham (491a71c12) - Bug 1163829 - Set the necessary prefs for web-platform service worker tests; r=jgraham (9b6bd3b2c) - Bug 1173467 P1.5 Mark wpt sandboxed-iframes.https.html test as expected fail. r=jgraham (a1164a210) - Bug 1173467 P2 Add nsGlobalWindow utility method to get private browsing boolean. r=ehsan (d3fe4a931) - Bug 1154494 - patch 2 - Hold a ref to the sandbox. r=baku (783347586) - Bug 1152899 - Disallow the interception of third-party iframes using service workers when the third-party cookie preference is set. r=smaug,baku (cbd5bcc92) - Bug 1162344 - Use the exemptFromPerDomainMax pref in test_importscript.html; r=nsm (826325f4b) - Bug 982726 follow-up: Add exempt from worker limit pref to a test. (a8f64ded9) - Bug 1157692 - Add dom.caches.enabled pref to test. r=bkelly (1465e999a) - Remove the remnants of the dom.fetch.enabled pref, no bug (a73c86666) - Bug 1169210 - Can't call content service worker from chrome. r=nsm (9ca1c4d33) - Bug 1170822 - Fix controlled documents comparison. r=catalinb (e9ffeb6f5) - Bug 1173240 - Use the document URI for performing the service worker registration secure origin check; r=baku (01a330597) - Bug 1173361 Make ServiceWorkerRegisterJob::Start() call Done() async. r=nsm (ff01f64ba) - Bug 1171555 - Remove overly verbose ServiceWorker warnings. r=bkelly (5d5f6978a) - Bug 1172948 - Part 1: Refactor the logic for checking for authenticated origins into a helper function; r=nsm (883fef559) - Bug 1172948 - Part 2: Consider all non-chrome parent documents when checking the authenticity of an origin; r=nsm (ae6388831) - Bug 1143732 - Add exempt from limit for two service worker tests. r=bkelly (5462a0270) - Bug 1155987 - Unregister the service workers registered in tests; r=nsm (aefe9b374) - remaining of Bug 1137984. When parsing attribute selectors, treat EOF as ']' any place ']' is allowed. r=dbaron (d9a35d4c7) - Bug 1149535 - enable unicode-range for testing. r=heycam (e155d0f72) - Bug 1161036 part 1. Stop setting the layout.css.font-loading-api.enabled preference to true in the test harness. r=heycam (89e961c0e) - Bug 1161036 part 2. Only force-enable dom cache in test_serviceworker_interfaces if we're having to force-enable servicewceworkers too. r=bkelly (0fac219b3) - Bug 1172110 - Re-enable service workers on desktop and mobile for non-release builds; r=jst (43a9c4983) - Bug 1173389 - Disable ServiceWorker network interception by default. r=ehsan (730dd4da5) - Bug 1125412 - Add a pref-gated performance.mozMemory for accessing internal memory usage/threshold info, r=smaug, sr=smaug (c16590d7d) - Bug 1159794 - "fix the comment for dom.workers.websocket.enabled in all.js" . r=bz (fe38a2b94) - Bug 1159792 - get rid of dom.workers.websocket.enabled pref, r=smaug (a66597c65) - Bug 1160892 - Url.createObjectURL(blob) should support punycode origins, r=smaug (5f4e92b1a) (821f0fa10) - import changes from `dev' branch of rmottola/Arctic-Fox: - remaining part of Bug 1146339 - Do anchor scrolling right before dispatching popstate (1123b0133) - Bug 1170488 - Document URI should be updated before sending out onLocationChange. r=smaug (6f553c2ee) - Bug 1144820 - Use nsTArray<UniquePtr<>> to hold timeline markers. r=smaug, r=fitzgen (e4b878ec3) - Bug 1143004 - markers from console.timeStamp() should be rendered in the performance tool's timeline. r=vp (7d8d4cb0c) - Bug 1170671 - ProfileTimelineMarkers that use TRACING_TIMESTAMP should capture stack information. r=smaug (dc7982a30) - Bug 1141614 - Part 1: Maintain a list of docshells whose timeline markers are being observed; r=smaug (2dc04461c) - Bug 1159779 - mozilla::AutoTimelineMarker shouldn't be copy-able; r=smaug (24aadecd4) - Bug 1141614 - Part 2: Add mozilla::AutoGlobalTimelineMarker; r=smaug (c069191a5) - Bug 1141614 - Part 3: Trace cycle collection with AutoGlobalTimelineMarker; r=smaug (2cdccd0fc) - Bug 1160521 - The markers list sidebar should be wider by default, r=jsantell (c479c72c3) - Bug 1150696 - Move all timeline tests that are still relevent into the performance tools tests. r=vp (276572541) - Bug 1165504 - Refactor out marker details view into utilities, make marker definitions more declaritive. r=vp (42218191c) - Bug 1166139 - properly display the generic class name for markers in the marker filter popup. r=vp (711b0032a) - Bug 1157916 - Add tooltip for displaying start/end times on marker details' duration field. r=vp (c586757d5) (32a8a4473)

From February 8th, Roy said:

New regular/weekly KM-Goanna release: https://o.rthost.win/kmeleon/KM76.4.3-Goanna-20210501.7z Changelog: Out-of-tree changes: * update Goanna3 to git b4028f07d...202b273a2: - import changes from `dev' branch of rmottola/Arctic-Fox: - partial of Bug 1153658 - browser_compartments.js logspam. r=yoric (8e2c21aa5) - Bug 1178653 - Refactor construction code to use an interface consistent with the spec, with the one exception using an out-of-the-way, differently-named method. r=efaust (d316259d7) - Bug 1175098 - PerformanceStats for e10s. r=felipe, r=mconley (515acb8d7) - Bug 1147664 - Detailed mode for PerformanceStats (low-level). r=jandem (dda8d84de) - Bug 1147664 - Detailed mode for PerformanceStats (high-level). r=mossop (b86076568) - Bug 1164304 - Run all fetch tests in the service worker context as well; r=nsm (e20fa8bfd) - Bug 1143981 - Reroute all fetch tests through a transparent service worker; r=nsm (5196acc57) - Bug 1122161 - Redirected channels should respect skip service worker flag. r=nsm (f4288392e) - Bug 1170937 - Set the URL on the Response object created from a fetch() properly if the underlying channel gets redirected; r=baku (45febabb3) - Bug 1173029 - Remove mFinalURL from InternalResponse; r=baku a=KWierso (6bdc1083b) - Bug 1137683 - Use a loadgroup derived from the document's when updating a ServiceWorker; r=bkelly (fabaa2602) - Bug 1164397 - Part 1: Use the original channel URI for constructing the cache entry key when we're dealing with an intercepted channel; r=mcmanus (b20ab36c7) - Bug 1164397 - Part 2: Add an API for overriding the original URI on HttpChannelBase; r=mcmanus (20021722f) - Bug 1164397 - Part 3: Add an API for overriding the original URI on nsJARChannel; r=jdm (492b6fd6f) - Bug 1164397 - Part 4: Add infromation about whether a channel was redirected to ChannelInfo; r=jdm (e2ce84660) - Bug 1164397 - Part 5: Save the redirected flag and the redirected URI in the DOM cache; r=bkelly (7d2d1fc92) - Bug 1162018 - Add an automated test to ensure that a redirected Request won't be visible to a service worker if it had triggered the original fetch(); r=jdm (0397a073f) - Bug 1164397 - Part 6: Add a test case for the service worker responding with a redirected Response; r=jdm (e83e0bee4) - Bug 1164397 - Part 7: Add a test case for the redirected Response object being stored in the DOM Cache; r=jdm (7a82916d8) - Bug 1169296 - Intercepting top-level document loads is not working with JAR channels. Tests. r=jdm (fe8f128c5) - Bug 1171285 - Part 1: Add a script for regenerating the application.zip used by test_app_protocol.html; r=jdm (ec303b3b2) - Bug 1171285 - Part 2: Fix test_app_protocol.html to finish both index.html and controlled.html tests; r=jdm (2e68e6665) - Bug 1169613 - Use content type of synthesized response for JAR channel requests if available. Tests. r=jdm (b0095fc3b) - Bug 1164397 - Part 8: Add a test case for the service worker for an app:// URI responding with a redirected Response; r=jdm (460e834c9) - Bug 1169044 - Patch 3 - Store and set principal with script URI on ServiceWorkers. r=ehsan (6e0b0102a) - Bug 1164397 - Part 9: Add a test case for the service worker for an app:// URI responding with a redirected HTTPS response; r=jdm (1be195f5a) - Bug 1164397 - Part 10: Add a test case for the service worker for an app:// URI responding with cached HTTP and HTTPS responses; r=jdm# Please enter the commit message for your changes. Lines starting (56432b7b5) - Bug 1164397 - Part 11: Add a test case for the service worker responding to HTTPS normal and cached Responses; r=jdm (6ec238455) - Bug 1164397 - Part 12: Add a test case for the service worker responding to normal and cached HTTP->HTTPS responses; r=jdm (925a1970f) - Bug 1190074 - PerformanceGroup now uses mozilla::RefPtr;r=jandem (53dc0a640) - Bug 1169086 followup: Add missing 'override' annotation to VerifyTraceProtoAndIfaceCacheCalledTracer::trace() method decl. rs=ehsan (cdedce447) - Bug 1172824: Initialize a few members in CompartmentCheckTracer (CID 1304705); r=terrence (3db40160c) - missing bit of 1166678 (9fb0cceeb) - Bug 1139473: File some metadata for the js/src/ subdirectory; r=jorendorff (9dc4a29a7) - Bug 1173889 - Strongly type the CallbackTracer dispatch function; r=jonco, r=mccr8 (19b47dc70) (516d358f3) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1185749 - Implement a DynamicTraceable HashMap subclass that can be used with Rooted; r=jonco (3f84bdb30) - Bug 1186609 - Implement a DynamicTraceable Vector subclass for use with Rooted; r=sfink (fa5b47e5b) - Bug 1155985 - Set FieldInto::mType just before storing to reserved slot. r=jonco, a=abillings (34dbc3ca0) - Bug 1156295 - Refactor GC rooting in StructType::DefineInternal() r=terrence (ec40a1701) - Bug 1185755 - Use Rooted to simplify FieldInfoHash rooting; r=jonco (8dc66d7cd) - Bug 1189072 - Make DefaultTracer for struct types call T::trace; r=fitzgen (96f79bc6f) - Bug 1149294 - Part 1: Don't trace permanent atoms and well known symbols that are owned by a parent JSRuntime inside SimpeEdgeVectorTracer; r=terrence (7ce049a33) - Bug 1149294 - Part 2: expose ChromeUtils and HeapSnapshot to workers; r=bholley (5737cd923) - Bug 1147679 - Use PR_GetFileInfo64 when deserializing heap snapshots; r=jimb (1ce8cd241) - Bug 1147680 - Use PR_MemMap when deserializing heap snapshots; r=jimb (04e5a370d) - Bug 1149294 - Part 3: Split thread-safe methods on ChromeUtils out in to a new (4967637e2) - Bug 1173829. Fix some compile issues in code generated by bindings codegen for non-concrete interfaces. r=peterv (3e0f93640) - Bug 1170274 - A better string parser for nsGenericHTMLElement::GetURIListAttr, r=ehsan (8c1595d29) - Bug 1170274 - patch 2 - A better string parser for nsGenericHTMLElement::GetURIListAttr, r=ehsan CLOSED TREE (5626d631f) - Bug 1096550 - Update content scale when swapping remote tabs between windows. r=roc. (ed974edf4) - Bug 1180017: Give GCCellPtr is<T> and as<T> methods, replacing isFoo and toFoo methods. Fix callers. r=terrence (b95645bbb) - pointer style (a46b905fe) - pointer style (cb548fd44) - Bug 1151182 - Fix type for JSOP_THROWMSG documentation comment. r=efaust (f9bce9df7) - Bug 1077318 - Part 2: Fix the order of include in AsmJSModule.cpp. r=evilpie (a61716e33) - Bug 1155006: Fix unified build sensitivities in js/src. r=shu (b0f166ec9) - Bug 988463 - Add in support for tracelogging on ARM. r=h4writer (10345729b) - Bug 1190077 - Move RDTSC and prmjtime.* to vm/Time.*. r=nbp (b72c1f366) (6885f7ffc) - import changes from `dev' branch of rmottola/Arctic-Fox: - pointer style (d7087db63) - Bug 1186450 - Fix TraceableHashTable's GC template definitions; r=jonco (72b84c9b7) - Bug 1189722 - Fix const methods in MutableTraceableVectorOperations r=terrence (db8841981) - Bug 1182428 - Fix the ObjectGroup hazards, r=jonco (f3e9b1eff) - Bug 1028418 - Part 1: Make the InterpreterFrame::Flags enum typed the way it is used; r=shu (927cc66c5) - Bug 1028418 - Part 2: Make the BaselineFrame::Flags enum typed the way it is used; r=shu (488f2f6e9) - Bug 1028418 - Part 3: Take a bit on each of interpreter, baseline, and rematerialized frames for marking whether there is a js::SavedFrame for the given frame in the js::SavedStacks cache; r=shu (7aca9dd37) - Bug 1028418 - Part 4: Fix whitespace alignment in the InterpreterFrame::Flags enum; r=me (1bcce1eb9) - pointer and space style (9720a44a8) - Bug 1184235 - Add mozilla::Vector::emplaceBack; r=froydnj (98026070a) - Bug 1184238 - Use emplaceBack instead of growByUninitialized(1) and pByUninitialized(1) and placement-new in js::SavedStacks; r=shu (a6708e701) - Bug 1184318 - Use emplaceBack instead of growByUninitialized(1) and placement-new in js/src/jit/MIR.h; r=shu (2ad5e0e28) - pointer style (89342947f) - Bug 1149495 - SavedFrame objects should handle all uint32 values. r=shu (af0a5ced2) - Bug 1157456 - Re-wrap results in SavedFrame accessors; r=jorendorff (29a2d0ccc) - Bug 1155618 - Fix places where OOM errors are not reported to the context in js::DependentAddPtr r=terrence (9d5e98223) - Bug 1184839 - Provide an overloaded == operator for mozilla::Variant; r=Waldo (93d7d2c2d) - Bug 1195590 - Fix treating saved frame bit as part of the frame type. r=djvj (c2ff877dd) - Bug 1028418 - Part 5: Minimize stack walking when capturing SavedFrame stacks with a cache; r=shu (f5426807f) - Bug 1028418 - Part 6: Fix GC + SavedFrame test that made assumptions that are no longer true; r=shu (5cc355c30) - Bug 1028418 - Part 7: Tests for caching edge cases; r=shu (d608a7da4) - Bug 1028418 - Part 0: Fix typo in js/src/vm/Stack.h; r=me (b3084ac03) - Bug 1164294 - Move OrderedHashTable to ds (1767e38d8) - Bug 1173864 - Make MarkStack private in GCMarker; r=jonco (2f570c51e) - Bug 1166041 - Wait for nursery sweeping to finish before shutting down; r=jonco (761638a57) - Bug 1166041 - Add some assertions to help catch leaked slots memory r=terrence (8f522f335) - Bug 1166041 - Fix memory leaked by nursery when we hit OOM r=terrence (cd3095efa) - Bug 1150349 - Update the zone triggers when updating GC parameters; r=jonco (b17dee3f8) - Bug 1186156 - Move the PersistentRooted list heads into RootLists; r=sfink (9afc5d93f) (6964ded9a) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1175485 part 1 - Allow inner iterator of ReverseIterator deref to any type, and change IntegerIterator, EnumeratedRange, and nsFrameList::Iterator to return value type instead of a reference. r=roc,waldo (83c53e8e8) - Bug 1175485 part 2 - Add static_assert to MakeRange to ensure it is used with integers. r=waldo (2b6fc759a) - Bug 1175485 part 3 - Remove unused operators, typedefs and IteratorTraits. r=waldo (91bc5407d) - Bug 1175485 part 4 - Add unit test for integer range. r=waldo (ec2b7d4c1) - Bug 1188204 - Fix more constructors in MFBT; r=froydnj (099428523) - Bug 1187985 - Make PersistentRooted use rootKind to find its lists; r=sfink (0c223593a) - Bug 1184191 - Assert in release mode if we reenter the garbage collector r=terrence (3207d3631) - Bug 1188124 - Use rootKind to select the right PersistentRooted list head; r=sfink (47009c787) - Bug 1174849 - Remove "serviceworker" from RequestContext; r=smaug,bkelly (138e480ea) - Bug 1174868 - Avoid storing RequestContext inside InternalRequest; r=nsm,bkelly,smaug (779a05075) - Bug 1147668 - Correctly reflect video and track RequestContext values; r=smaug (90f265d5f) - Bug 1148818 - Re-enable the Cache part of test_request_context.html on Android now that it passes (782fe9777) - Bug 1148935 - Correctly reflect worker and sharedworker RequestContext values; r=smaug (fd40e01d8) (202b273a2) * Notice: the changelog above may not always applicable to XULRunner code which K-Meleon uses. A goanna3 source tree that has kmeleon adaption patch applied is available here: https://github.com/roytam1/palemoon27/tree/kmeleon76 Known Issue(s): - media playback function may be buggy. the issue may be relexed by setting media.mediasource.format-reader to true in about:config.

New NewMoon 27 Build! 32bit https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210501-202b273a2-xpmod.7z 32bit SSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210501-202b273a2-xpmod-sse.7z 32bit noSSE https://o.rthost.win/palemoon/palemoon-27.10.0.win32-git-20210501-202b273a2-xpmod-ia32.7z 64bit https://o.rthost.win/palemoon/palemoon-27.10.0.win64-git-20210501-202b273a2-xpmod.7z source repo: https://github.com/roytam1/palemoon27 repo changes since my last build: - import changes from `dev' branch of rmottola/Arctic-Fox: - partial of Bug 1153658 - browser_compartments.js logspam. r=yoric (8e2c21aa5) - Bug 1178653 - Refactor construction code to use an interface consistent with the spec, with the one exception using an out-of-the-way, differently-named method. r=efaust (d316259d7) - Bug 1175098 - PerformanceStats for e10s. r=felipe, r=mconley (515acb8d7) - Bug 1147664 - Detailed mode for PerformanceStats (low-level). r=jandem (dda8d84de) - Bug 1147664 - Detailed mode for PerformanceStats (high-level). r=mossop (b86076568) - Bug 1164304 - Run all fetch tests in the service worker context as well; r=nsm (e20fa8bfd) - Bug 1143981 - Reroute all fetch tests through a transparent service worker; r=nsm (5196acc57) - Bug 1122161 - Redirected channels should respect skip service worker flag. r=nsm (f4288392e) - Bug 1170937 - Set the URL on the Response object created from a fetch() properly if the underlying channel gets redirected; r=baku (45febabb3) - Bug 1173029 - Remove mFinalURL from InternalResponse; r=baku a=KWierso (6bdc1083b) - Bug 1137683 - Use a loadgroup derived from the document's when updating a ServiceWorker; r=bkelly (fabaa2602) - Bug 1164397 - Part 1: Use the original channel URI for constructing the cache entry key when we're dealing with an intercepted channel; r=mcmanus (b20ab36c7) - Bug 1164397 - Part 2: Add an API for overriding the original URI on HttpChannelBase; r=mcmanus (20021722f) - Bug 1164397 - Part 3: Add an API for overriding the original URI on nsJARChannel; r=jdm (492b6fd6f) - Bug 1164397 - Part 4: Add infromation about whether a channel was redirected to ChannelInfo; r=jdm (e2ce84660) - Bug 1164397 - Part 5: Save the redirected flag and the redirected URI in the DOM cache; r=bkelly (7d2d1fc92) - Bug 1162018 - Add an automated test to ensure that a redirected Request won't be visible to a service worker if it had triggered the original fetch(); r=jdm (0397a073f) - Bug 1164397 - Part 6: Add a test case for the service worker responding with a redirected Response; r=jdm (e83e0bee4) - Bug 1164397 - Part 7: Add a test case for the redirected Response object being stored in the DOM Cache; r=jdm (7a82916d8) - Bug 1169296 - Intercepting top-level document loads is not working with JAR channels. Tests. r=jdm (fe8f128c5) - Bug 1171285 - Part 1: Add a script for regenerating the application.zip used by test_app_protocol.html; r=jdm (ec303b3b2) - Bug 1171285 - Part 2: Fix test_app_protocol.html to finish both index.html and controlled.html tests; r=jdm (2e68e6665) - Bug 1169613 - Use content type of synthesized response for JAR channel requests if available. Tests. r=jdm (b0095fc3b) - Bug 1164397 - Part 8: Add a test case for the service worker for an app:// URI responding with a redirected Response; r=jdm (460e834c9) - Bug 1169044 - Patch 3 - Store and set principal with script URI on ServiceWorkers. r=ehsan (6e0b0102a) - Bug 1164397 - Part 9: Add a test case for the service worker for an app:// URI responding with a redirected HTTPS response; r=jdm (1be195f5a) - Bug 1164397 - Part 10: Add a test case for the service worker for an app:// URI responding with cached HTTP and HTTPS responses; r=jdm# Please enter the commit message for your changes. Lines starting (56432b7b5) - Bug 1164397 - Part 11: Add a test case for the service worker responding to HTTPS normal and cached Responses; r=jdm (6ec238455) - Bug 1164397 - Part 12: Add a test case for the service worker responding to normal and cached HTTP->HTTPS responses; r=jdm (925a1970f) - Bug 1190074 - PerformanceGroup now uses mozilla::RefPtr;r=jandem (53dc0a640) - Bug 1169086 followup: Add missing 'override' annotation to VerifyTraceProtoAndIfaceCacheCalledTracer::trace() method decl. rs=ehsan (cdedce447) - Bug 1172824: Initialize a few members in CompartmentCheckTracer (CID 1304705); r=terrence (3db40160c) - missing bit of 1166678 (9fb0cceeb) - Bug 1139473: File some metadata for the js/src/ subdirectory; r=jorendorff (9dc4a29a7) - Bug 1173889 - Strongly type the CallbackTracer dispatch function; r=jonco, r=mccr8 (19b47dc70) (516d358f3) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1185749 - Implement a DynamicTraceable HashMap subclass that can be used with Rooted; r=jonco (3f84bdb30) - Bug 1186609 - Implement a DynamicTraceable Vector subclass for use with Rooted; r=sfink (fa5b47e5b) - Bug 1155985 - Set FieldInto::mType just before storing to reserved slot. r=jonco, a=abillings (34dbc3ca0) - Bug 1156295 - Refactor GC rooting in StructType::DefineInternal() r=terrence (ec40a1701) - Bug 1185755 - Use Rooted to simplify FieldInfoHash rooting; r=jonco (8dc66d7cd) - Bug 1189072 - Make DefaultTracer for struct types call T::trace; r=fitzgen (96f79bc6f) - Bug 1149294 - Part 1: Don't trace permanent atoms and well known symbols that are owned by a parent JSRuntime inside SimpeEdgeVectorTracer; r=terrence (7ce049a33) - Bug 1149294 - Part 2: expose ChromeUtils and HeapSnapshot to workers; r=bholley (5737cd923) - Bug 1147679 - Use PR_GetFileInfo64 when deserializing heap snapshots; r=jimb (1ce8cd241) - Bug 1147680 - Use PR_MemMap when deserializing heap snapshots; r=jimb (04e5a370d) - Bug 1149294 - Part 3: Split thread-safe methods on ChromeUtils out in to a new (4967637e2) - Bug 1173829. Fix some compile issues in code generated by bindings codegen for non-concrete interfaces. r=peterv (3e0f93640) - Bug 1170274 - A better string parser for nsGenericHTMLElement::GetURIListAttr, r=ehsan (8c1595d29) - Bug 1170274 - patch 2 - A better string parser for nsGenericHTMLElement::GetURIListAttr, r=ehsan CLOSED TREE (5626d631f) - Bug 1096550 - Update content scale when swapping remote tabs between windows. r=roc. (ed974edf4) - Bug 1180017: Give GCCellPtr is<T> and as<T> methods, replacing isFoo and toFoo methods. Fix callers. r=terrence (b95645bbb) - pointer style (a46b905fe) - pointer style (cb548fd44) - Bug 1151182 - Fix type for JSOP_THROWMSG documentation comment. r=efaust (f9bce9df7) - Bug 1077318 - Part 2: Fix the order of include in AsmJSModule.cpp. r=evilpie (a61716e33) - Bug 1155006: Fix unified build sensitivities in js/src. r=shu (b0f166ec9) - Bug 988463 - Add in support for tracelogging on ARM. r=h4writer (10345729b) - Bug 1190077 - Move RDTSC and prmjtime.* to vm/Time.*. r=nbp (b72c1f366) (6885f7ffc) - import changes from `dev' branch of rmottola/Arctic-Fox: - pointer style (d7087db63) - Bug 1186450 - Fix TraceableHashTable's GC template definitions; r=jonco (72b84c9b7) - Bug 1189722 - Fix const methods in MutableTraceableVectorOperations r=terrence (db8841981) - Bug 1182428 - Fix the ObjectGroup hazards, r=jonco (f3e9b1eff) - Bug 1028418 - Part 1: Make the InterpreterFrame::Flags enum typed the way it is used; r=shu (927cc66c5) - Bug 1028418 - Part 2: Make the BaselineFrame::Flags enum typed the way it is used; r=shu (488f2f6e9) - Bug 1028418 - Part 3: Take a bit on each of interpreter, baseline, and rematerialized frames for marking whether there is a js::SavedFrame for the given frame in the js::SavedStacks cache; r=shu (7aca9dd37) - Bug 1028418 - Part 4: Fix whitespace alignment in the InterpreterFrame::Flags enum; r=me (1bcce1eb9) - pointer and space style (9720a44a8) - Bug 1184235 - Add mozilla::Vector::emplaceBack; r=froydnj (98026070a) - Bug 1184238 - Use emplaceBack instead of growByUninitialized(1) and pByUninitialized(1) and placement-new in js::SavedStacks; r=shu (a6708e701) - Bug 1184318 - Use emplaceBack instead of growByUninitialized(1) and placement-new in js/src/jit/MIR.h; r=shu (2ad5e0e28) - pointer style (89342947f) - Bug 1149495 - SavedFrame objects should handle all uint32 values. r=shu (af0a5ced2) - Bug 1157456 - Re-wrap results in SavedFrame accessors; r=jorendorff (29a2d0ccc) - Bug 1155618 - Fix places where OOM errors are not reported to the context in js::DependentAddPtr r=terrence (9d5e98223) - Bug 1184839 - Provide an overloaded == operator for mozilla::Variant; r=Waldo (93d7d2c2d) - Bug 1195590 - Fix treating saved frame bit as part of the frame type. r=djvj (c2ff877dd) - Bug 1028418 - Part 5: Minimize stack walking when capturing SavedFrame stacks with a cache; r=shu (f5426807f) - Bug 1028418 - Part 6: Fix GC + SavedFrame test that made assumptions that are no longer true; r=shu (5cc355c30) - Bug 1028418 - Part 7: Tests for caching edge cases; r=shu (d608a7da4) - Bug 1028418 - Part 0: Fix typo in js/src/vm/Stack.h; r=me (b3084ac03) - Bug 1164294 - Move OrderedHashTable to ds (1767e38d8) - Bug 1173864 - Make MarkStack private in GCMarker; r=jonco (2f570c51e) - Bug 1166041 - Wait for nursery sweeping to finish before shutting down; r=jonco (761638a57) - Bug 1166041 - Add some assertions to help catch leaked slots memory r=terrence (8f522f335) - Bug 1166041 - Fix memory leaked by nursery when we hit OOM r=terrence (cd3095efa) - Bug 1150349 - Update the zone triggers when updating GC parameters; r=jonco (b17dee3f8) - Bug 1186156 - Move the PersistentRooted list heads into RootLists; r=sfink (9afc5d93f) (6964ded9a) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1175485 part 1 - Allow inner iterator of ReverseIterator deref to any type, and change IntegerIterator, EnumeratedRange, and nsFrameList::Iterator to return value type instead of a reference. r=roc,waldo (83c53e8e8) - Bug 1175485 part 2 - Add static_assert to MakeRange to ensure it is used with integers. r=waldo (2b6fc759a) - Bug 1175485 part 3 - Remove unused operators, typedefs and IteratorTraits. r=waldo (91bc5407d) - Bug 1175485 part 4 - Add unit test for integer range. r=waldo (ec2b7d4c1) - Bug 1188204 - Fix more constructors in MFBT; r=froydnj (099428523) - Bug 1187985 - Make PersistentRooted use rootKind to find its lists; r=sfink (0c223593a) - Bug 1184191 - Assert in release mode if we reenter the garbage collector r=terrence (3207d3631) - Bug 1188124 - Use rootKind to select the right PersistentRooted list head; r=sfink (47009c787) - Bug 1174849 - Remove "serviceworker" from RequestContext; r=smaug,bkelly (138e480ea) - Bug 1174868 - Avoid storing RequestContext inside InternalRequest; r=nsm,bkelly,smaug (779a05075) - Bug 1147668 - Correctly reflect video and track RequestContext values; r=smaug (90f265d5f) - Bug 1148818 - Re-enable the Cache part of test_request_context.html on Android now that it passes (782fe9777) - Bug 1148935 - Correctly reflect worker and sharedworker RequestContext values; r=smaug (fd40e01d8) (202b273a2) Known Issue(s): - media playback function may be buggy. the issue may be relexed by setting media.mediasource.format-reader to true in about:config.

thats auto-detect in runtime

there is no such switch. https://docs.microsoft.com/en-us/cpp/build/reference/arch-x86?view=msvc-140

compatibility of installing 3rd party extensions should be on par with official PM28.

And what I'm trying to say , Aero buttons are still absent even in 2.2.8 (not to mention 3.x.x.) and even with the hex edit. (do you want me to repeat again?) Yes , I agree with Dixel , they somewhat look better after hexing , some curves are added , they don't look as flat as before , but nowhere near the previous ones , that you (and most of us) wanted to see . So hexing is NOT the ultimate solution , it's better to roll-back to earlier versions , 2.2.6 something. I don't quite understand how "testing again using a maximised window" will help with the aero look in the butoons . I'm just starting to think you don't understand English well enough , or have a poor LCD display and simply can't see the obviuos difference , excuse me if I'm wrong (in advance) , but it's pretty obvious and shown well enough at Dixel's picture with the patched 3.3.x and NOT patched 2.2.x . Also , his upper frame looks prettier than the one that you provided , I mean the shape , not the colour , since it is his own customized colour. In case you don't agree or understand again , please refrain from asking me 'cause I simply can't tell you anything new apart from what I've already said , please ask other members if they see the obvious differnce at the Dixel's picture .

1 - You mean the first picture in this topic where aero is totally absent ? I can't reproduce it no matter how hard I try , Ive tested on an a clean install , without any mods. I mean , yeah , I can find a newer version where the buttons will be flat , but the aero is always working with all versions ! Known or had it youself ? If it is easily reproduced , could you show us a pic. ? I'm curious. Thanks. 2 - The screen of the user named @Dixel clearly shows that aero is working and it is transparent , in both versions, patched and unpatched , don't you see ?

I have seen over the years different patches for syssetup.dll for Windows XP SP3 here and on ryanvm.net and most of them have some problems about driver warnings or breaking some functionality. Recently I replaced all of them with the patches published by Mr. Dusha on OSZone.net and have not had any problems since. http://forum.oszone.net/thread-119984.html. I had to register to download the files. The text file is very useful but despite its name does not contain all modifications made. Using the Fake Setup Advanced command line I was able to apply all the patches to syssetup.dll and setupapi.dll. Here's the complete list of patches and descriptions for Windows XP SP3. //=============// SETUPAPI.dll//=============//5.1.2600.5603 (KB948720)0001CD29: 74 EB // Disable DSP increase from 0 to 1,0001D509: 74 EB // if replaced SFC-protected signed driver00021B5F: 74 90 //00021B60: 0A 90 // Disable unsigned drivers decreasing rank00021B6E: 75 EB //000456A7: 74 EB // Disable error messages in Setupapi.log and pop-up driver install dialogs with patched syssetup.dll.....//5.1.2600.55120001CD29: 74 EB // Disable DSP increase from 0 to 1,0001D509: 74 EB // if replaced SFC-protected signed driver00021B7F: 74 90 //00021B80: 0A 90 // Disable unsigned drivers decreasing rank00021B8E: 75 EB //000456C7: 74 EB // Disable error messages in Setupapi.log and pop-up driver install dialogs with patched syssetup.dll.....//========================// SYSSETUP.dll PatchBlock//========================//5.1.2600.551200033B29: 73 EB // Disable Syssetup.inf protect 00033E00: 39 85 //00033E01: 5D DB // DefaultDrvSignPol = 0 (params in winnt.sif will be ignored)00033E02: 08 90 //000574FF: 74 90 // Disable OOBE00057500: 07 90 0005AB31: 6A EB // Disable SFC files scan in T-80005AB32: 46 0D ////===========// SFC_OS.dll//===========//5.1.2600.55120000EC87: 75: EB // SfcDisable=0xFFFFFF9D enable//=======================// UXTHEME.dll PatchBlock//=======================//6.0.2900.551200018727: 8B 33 //00018728: FF C0 //00018729: 55 C2 // Unsigned themes usage enable0001872A: 8B 04 //0001872B: EC 00 //