Jump to content

Leaderboard

Popular Content

Showing content with the highest reputation on 08/02/2024 in all areas

  1. 5 points
  2. @Cixert I only use WSUS Offline Update 9.2.6 for downloading Windows XP updates. The download of Office 2007 updates was just for testing purpose to help you. As I already mentioned, I downloaded all of them in the past. There is nothing new to download anymore. Regarding Windows XP, I deleted all references to files which are not XP-compatible. "I deleted the default download links pointing to .NET Frameworks installer files higher than 4.0 inside WSUS Offline Update 9.2.6" means to remove all these references in the StaticDownloadLinks-dotnet.txt and "and then enabled their download." actually means that I checked the box in the main programme window with the option to download C++ Libraries and .NET Frameworks for the next download session. I wrote that before I found out to better set exclusions. Setting exclusions is very effective to avoid unwanted downloads. And the last pause gives the user the possibility to add missing or remove unwanted files before creating the iso image. Please let me know if now all is working for you regarding the Office 2007 updates as it should! Whether it also works for Office 2003, I can't say anything about as I didn't test it. You have to check this on your own. Greetings from Germany, AstroSkipper
    4 points
  3. Windows 7 never supported D3D11.1. It's a rumour/fake news. D3D11.1 is windows 8+ only. Some DXGI 1.2 API and the ID3DUserDefinedAnnotation were ported on windows 7, but most of the really useful functionality, no. You can get only partial/miserable/fake Direct3D 11.1 by installing the Platform Update for Windows 7 And, for example, this component is not available: DirectComposition (DComp) All information is taken from Microsoft website. https://learn.microsoft.com/en-gb/windows/win32/direct3darticles/platform-update-for-windows-7
    3 points
  4. Maybe, this one is already too new. Try one of the versions I linked to! Start with the version from 29.01.2019! You can't manually pause the script as you will never match the right time. WSUS Offline Update actually starts extracting the wsusscn2.cab file immediately after downloading it. Pausing must be carried out in the script using the Pause command. Apart from that, the older version of the wsusscn2.cab file must be inserted several times. With each extraction process, WSUS Offline Update checks whether it is still up-to-date. If not, it downloads the latest version again, which we can't avoid at this point. Otherwise you would have to completely adapt the script for downloading older Office products, which I will not do, of course. Therefore, the Pause command must be placed in the correct position in the DownloadUpdates.cmd script. I have inserted the following in line 978, right before the extracting of Microsoft's update catalog file package.xml: echo The script has paused. Now, you can manually replace the file wsusscn2.cab by an older one. When done, press a key to continue! pause Each time this message is displayed, the wsusscn2.cab file must be replaced. And don't forget to also modify the CreateISOImage.cmd script, as described in one of my previous posts! A positive side effect of the Pause command is of course that you can simply leave the script running unattended and just insert the file and press a key from time to time, no matter when. If none of this works for you again, you're probably doing something wrong. Because it was tested by me and definitely works here.
    3 points
  5. Just for the sake of comparison, I tested the last version 10.9.2 of WSUS Offline Update, which still supports downloading Office 2007 updates, on a Windows 10 computer. As with Windows XP and WSUS Offline Update 9.2.6, the result is disastrous. Tons of updates are downloaded that have nothing to do with Office 2007. So my method of replacing the wsusscn2.cab file with a suitable one is probably the only way to download Office 2007 updates with WSUS Offline Update today. I suspect that Microsoft has changed the download links for old Office products from a certain version of wsusscn2.cab onwards, although most of the files are still on their servers. All that means we have to be tricky in these days to get such old updates from Microsoft's servers.
    3 points
  6. https://msfn.org/board/topic/185918-arcticfoxienotheretoplaygames-360chrome-v1351030-redux/page/7/#comment-1260747
    2 points
  7. No, it's not. Here's a MonoBleedingEdge (Engine version from 2019) log file for you from: OS-Version: 6.1.0 (win 7). Renderer: NVIDIA GeForce GTX 1070 (ID=0x1b81) GTX 1070 supports all the way up tp 12_1. Yet the log for Windows 7 says: "Forcing GfxDevice: Direct3D 11 GfxDevice: creating device client; threaded=1 d3d11: QueryInterface(IDXGIFactory5) failed (80004002). Direct3D: Version: Direct3D 11.0 [level 11.0]" Source: https://steamcommunity.com/app/251570/discussions/1/2646361461399067327/ It's from 2020! Eight years after the alleged 11.1 update KB came out.
    2 points
  8. Nothing weird, I always suspected win32 stood behind all Kernel releases to begin with, it's just my personal opinion after looking at the code. Not seeking an argument over it.
    2 points
  9. From this first sentence it's clear as day that you implied that Microsoft actually did bring "D3D11.1 API to Win 7" And no matter how many games I tried to run on Win 7, even after the aforementioned update applied, they all fall back to just 11.0. So, D3D11.1 on Win 7 is just a stub, it's not "API" like you wrote. What it does, it simply tells a game, go on, run, it runs, but with 11.0 since the 11.1 features are absent. One would need to rewrite Win 7 to actually add the real D3D 11,1. That's what D.Draker tried to explain to you. As of now, Vista and Win 7 have the same 11.0. Is it more clear now?
    2 points
  10. It doesn't let me to download. https://web.archive.org/web/20240727151625/https://github.com/i486/VxKex/releases/tag/Version1.1.1.1375
    2 points
  11. Good! Thanks for coming out with the confession, then the supervisor will keep an eye on you, too.
    2 points
  12. Yes, thank you very much, if it weren't for your help I wouldn't have achieved anything. After everything works correctly I will make a summary of all the necessary steps. Let's go by systems, for the moment the question is as follows: WSUS 9.2.6 option XP x86 with .NET Framework and C++: Adding these lines to ExcludeList-dotnet-x86.txt & ExcludeList-dotnet-x64.txt does work. ndp35- ndp45- ndp46- ndp47- ndp48- The .NET 3.5 installer is normally downloaded, what is not downloaded are the updates for .NET 3.5 but are for .NET 2.0, 3.0... The problem was that on my computer WSUS does not work well with the temporary files that contain the download lists, the strange thing is that downloading the XP updates without .NET did work but with .NET it failed. I have run it on another computer and it worked with the latest wsusscn2.cab version. The following files are not downloaded but I add them manually thanks to your pause functions on the correct lines: \client\dotnet -NDP40-KB2600211-x86.exe \client\win\glb -Silverlight.exe \client\wxp\esn -wmp11-windowsxp-x86-ES-ES.exe \client\wxp\glb -WindowsXP-KB942288-v3-x86.exe Also I delete for x64: \client\cpp -vc_redist_x64.exe (all) \client\win\glb -Silverlight64.exe and... dotnetfx35langpack_x86de.exe (only for German language) Now I have this problem in the log and I don't know how to add these missing downloads or what the message exactly means, I see in some old forums that there are more than 2 downloads: 2024-08-02 22:09:12,83 - Info: Downloaded/validated 2 statically defined updates for dotnet x86-glb failed: Unknown error. failed: Unknown error. 2024-08-02 22:15:09,45 - Info: Downloaded/validated 14 statically defined updates for wxp esn failed: Unknown error. failed: Unknown error. All other files download correctly, how do I guess which ones are missing? The error does not occur with the XP option without .NET I don't know what "statically defined updates" means. I think it's the links contained in the "static" folder. EDIT: I have tried with old wsusscn2.cab (2019-04-06) and these files have been downloaded correctly. The pity is that this file has 54 dynamic updates less. I will try to make a mix between both. And also the pity is that this wsusscn2.cab is dated 3 days before the end of support for Windows XP Embedded. It seems to make no sense that if they are URLs contained in a .txt file, they are downloaded or not depending on the version wsusscn2.cab and whether or not the .NET option is checked in WSUS.
    1 point
  13. "Russian Businesses Can Now Legally Steal Intellectual Property from ‘Unfriendly Countries’" "As of Monday, March 7, 2022, the Russian government has legalised intellectual property (IP) theft. With this move, businesses in Russia can now violate IP rights, as they no longer need to compensate patent holders from “unfriendly countries.” The list of “unfriendly countries” includes the United States, Canada, Australia, Switzerland, Norway, Iceland, the United Kingdom, Japan, South Korea, New Zealand, Singapore, Taiwan and all 27 European Union (EU) member countries." That includes software. https://ipwatchdog.com/2022/03/16/russian-businesses-can-now-legally-steal-intellectual-property-unfriendly-countries/id=147528/
    1 point
  14. Hello! No, no solution. Switch to Nvidia only in the BIOS. Intel card is weak and crappy anyways.
    1 point
  15. In this weird modern World, it doesn't work the way you assume. The US hosts a super large portion of openly anti-American population, most of them aren't even citizens, they broke all the laws, yet they aren't deported, they are even subsidised. It's the same with the github since they are an American entity, they resonate on the same (current) agenda frequencies. Also, regarding github, github works for the international market, they can't just throw away whomever they don't like. Or they would've needed to implement the only US members demand.
    1 point
  16. What is POSReady2009? And how it's related to Win7 Kernel? Win7 was released only at the end of 2009.
    1 point
  17. Some folks are like children, do the opposite. Or may be he really is, we can't see his age.
    1 point
  18. My developer friend on github said they usually send an official letter demanding to remove the content in question, so not likely the case.
    1 point
  19. It can't be classified as an "attack" in either of the cases, Microsoft has the right to defend what's theirs. Closed source Kernel modifications aren't allowed in the US,
    1 point
  20. You confuse hardware level with software level. "11_1 feature level" is a hardware level, it can't be "implemented" in just "soft" mode. Here you can educate yourself regarding the topic. Microsoft article with precise and simple explanations. https://learn.microsoft.com/en-us/windows/win32/direct3d12/hardware-feature-levels
    1 point
  21. I follow your instructions @AstroSkipper I put download .NET and C++ for Windows XP in Spanish and it tries to download .NET 4.6.2 in German. Maybe because the program WSUS is made by Germans... EDIT: FIXED: I have deleted most the links contents in StaticDownloadLinks-dotnet.txt Which most contains links in German. Delete for all languages except German.
    1 point
  22. I have exactly the same problem. All shortcuts from the taskbar on the first monitor also appeared on the taskbar on the second monitor and cannot be deleted on the second monitor. Because of this, I had to revert to an older version of StartAllBack.
    1 point
  23. Philippe, hello and welcome to the neighbour from the South of France!
    1 point
  24. First thanks! And now just for clarification, I am not the creator of ProxHTTPSProxy. But I am the creator of the package ProxHTTPSProxy's PopMenu TLS 1.3 3V3. And as you can see inside the name of this package, it is based on a TLS 1.3 proxy since 2022. Unfortunately, the rest of your comment does not match our recent discussion about the tool WSUS Offline Update and how one can get Office 2007 updates today. To be honest, I don't understand at all what your comment here is actually referring to.
    1 point
  25. Did you replace the wsusscn2.cab file by an older one which is dated short after the EoS of Office 2007 in 2017 in the way I described in my previous posts? I checked this method, and it works. Even under Windows XP. No 2529 updates but 580 related to Office 2007 ofc glb. For me, the wsusscn2.cab file from 12.02.2019 also worked fine. In any case, the most current wsusscn2.cab files do not properly work at all. Too much files are downloaded which are not related to Office 2007. I have the impression that you haven't read my posts properly. The proof that my method works is the download log I provided in this post: https://msfn.org/board/topic/181623-updates-office-2007/?do=findComment&comment=1270043 Try, for example, this one from 29.01.2019: https://web.archive.org/web/20190129233558/http://download.windowsupdate.com/microsoftupdate/v6/wsusscan/wsusscn2.cab or this one from 05.10.2017: https://web.archive.org/web/20171005192358/http://download.windowsupdate.com/microsoftupdate/v6/wsusscan/wsusscn2.cab
    1 point
  26. These files are only temporarily generated for further processing to get a final csv list of updates which is then saved inside the corresponding folder of WSUS Offline Update and whose links have to be downloaded in the end. For example, the UpdateTable-ofc-glb.csv file I already mentioned. After processing, these files are immediately removed from the temp folder. The temp folder is, so to speak, the working directory of WSUS Offline Update when the links to be loaded are extracted from the wsusscn2.cab file.
    1 point
  27. Here's the manual https://www.tyan.com/manuals/m_s5360_100.pdf It says 184-pin DDR 333/266 Registered ECC with single chip sizes of 256, 512 or 1024Mb. (their nomenclature) Whether an unregistered will work, it is not going to be a 100% answer because all boards and RAM are different. You can go further by looking up the spec sheets of the RAM itself and compare them. But my experience tells me that there are going to be some boards where using UDIMM on RDIMM is going to either a) work fine, b) not work, c) not work fine or d) cause damage. So unless you can find any post online of someone who has tried the same exact RAM that you have in the same board rev and they say it works, then it will probably work. But real world isn't so handy to be able to find that info, so you are likely to have to try it yourself to see if it works or not. I've also not experienced any situation where having an unsupported type of memory in a board has caused damage. Either with old Pentium systems or modern server boards. I suspect those instances of damage were likely caused by something other than mismatched memory, such as dirt or some other situation.
    1 point
  28. The wmp11-windowsxp-x86-DE-DE.exe file is of course the German version and must be replaced by the desired language version. And yes, they can be downloaded from other sources and easily inserted before the iso image is created. If the user enables the programme option "Include C++ Runtime Libraries and .Net Frameworks", the download script should also be paused before the iso image is created just to remove tons of .Net Frameworks 4.5, 4.6, 4.7. and 4.8 files which are no longer work under Windows XP. I have found a way to make it even easier, faster and, above all, more convenient. You can set exclusions in WSUS Offline Update 9.2.6. To do this, you must insert the following strings in the ExcludeList-dotnet-x86.txt and ExcludeList-dotnet-x64.txt files in the exclude folder: ndp45- ndp46- ndp47- ndp48- Doing so, the download of files containing these strings will be skipped. And in order not to miss the right moment to press the pause button when adding the missing updates, you can add the following command in line 3 of the CreateISOImage.cmd file in the cmd folder: echo The script has paused. Now, you can manually add all updates whose download has failed. When done, press a key to continue! pause Cheers, AstroSkipper
    1 point
  29. I own a GTX Titan X (Top Maxwell card), the iconic Spellforce series doesn't support MSAA, and with CSAA it would've looked fantastic, I know this because I also own a 780Ti. Old games don't need VRAM economy. The OP wanted to run vry old games (securom protected), so we can safely assume his games are approx. up to 2006-2007.
    1 point
  30. Yep. And thanks! Reviving, modifying and fixing programmes that no longer work properly is one of my favorite things to do. Just for clarification. The Wget option --no-check-certificate has nothing to do with the download option "Verify downloads updates" in the main programme window. In my first download logs generated by WSUS Offline Update 9.2.6, I noticed some errors related to certificate checks when connecting to servers. The Wget option --no-check-certificate is meant to not check these server certificates. The programme option "Verify downloads updates" on the other hand, switches on or off the signature verification of downloaded files by the Sysinternals programme Sigcheck. This is the full list of files that can no longer be downloaded by WSUS Offline Update 9.2.6 for Windows XP: The wmp11-windowsxp-x86-DE-DE.exe file is of course the German version and must be replaced by the desired language version. And yes, they can be downloaded from other sources and easily inserted before the iso image is created. If the user enables the programme option "Include C++ Runtime Libraries and .Net Frameworks", the download script should also be paused before the iso image is created just to remove tons of .Net Frameworks 4.5, 4.6, 4.7. and 4.8 files which are no longer work under Windows XP.
    1 point
  31. No, on the contrary, your English is good. I wish many others had at least your level. Today, I wanted to stick forks in my eyes when I saw "maded" written by a dude who claims to be a "developer", lol. So, I just made you a compliment, as you see, I'm not here "against you", either. What I wanted to tell, we know that Kernel developer from here, he is a member, I don't trust github either, I trust MSFN. Edit. Nothing personal, it's just we had some MSFN haters joined as of lately. So some folks may be a bit agitated and over-protective.
    1 point
  32. 1 point
  33. Compare with what? What if the user is new, and doesn't have any of these from before? Since the original repo is gone, as you wrote yourself.
    1 point
  34. No one is "paranoid", we are just being on the safe side. I mean, do you realise how serious is to install kernel from unknown, non-developer sources!? MSFN only hosts and provides 100% clean originals. The same with external links. We can't just take someone's word for it, sorry.
    1 point
  35. New regular/weekly KM-Goanna release: https://o.rthost.win/kmeleon/KM76.5.5-Goanna-20240727.7z Changelog: Out-of-tree changes: * update Goanna3 to git be4fef94e8...cbc917768b: - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1262361 - Use gfxCriticalError to collect useful info when device reset. r=milan (a565047753) - Bug 1265282 - Annotate reports instead of crashing when setting an incompatible compositor (D3D11). r=dvander (ed04e9e8eb) - Bug 1235407 - Part 3: Force a reset when OpenSharedHandle fails. r=milan (a2f5c656eb) - Bug 1264142 - Add ImageLayerComposite::GetFullyRenderedRegion() r=mattwoodrow (3a5f479820) - Bug 1224199 - Remove some unused code in TiledLayerBufferComposite - r=nical (b3c60126ce) - Bug 1262328 - Add Add ClientLayer::HandleMemoryPressure() r=nical (31bb89613d) - Bug 1224833 - Check explicitly if image has TextureClient r=nical (34044dc195) - Bug 1240867 - Add missing include. r=nical (abd9a8876e) - Bug 1240800: When we've reallocated our buffer client side and fail to track the proper invalid region always upload the bounds of the visible region. r=mattwoodrow (7a7cc1928c) - Bug 1253386 - Double buffer in CanvasClient2D - r=nical (a0fe10f1a6) - Bug 1266380 - fix offset of untransformed surface when 3d transforming in BasicLayerManager. r=jrmuizel (3d3931346f) - Bug 1257288 - Fix a bug in APZCTM logging code. r=kats (3c73c99077) - Bug 1265424 - Record if the target was confirmed via a timeout or not. r=botond (e25c0c016d) - Bug 1265424 - Ensure that HasReceivedAllContentNotifications doesn't start returning true if the target APZC was confirmed by timeout. r=botond (6c4e57cb8f) - Bug 1229039 - If a PanGesture input block gets interrupted, just start a new block instead of not sending the rest of the events through the APZ. r=mstange (534b6691f7) - Bug 1056356 - Remove the hand-rolled mechanism used to get nsRefPtr<const OverscrollHandoffChaiin> to work. r=kats (9d692fa32f) - Bug 1253617 - Fix non-unified build bustage in OverscrollHandoffStat.cpp r=BenWa (38a53f7521) - Bug 1262432 - Remove assertion that may be legitimately false sometimes. r=tnikkel (36bbb36476) - Bug 1169802 - Temporary workaround for a deeper bug, to prevent an assertion from firing. r=botond (b1edd69e49) - bits of Bug 1223296 (182a1ee34e) - move around (bug 1014691) and align some more tests (19db30010c) - Bug 944164 - Update jorendb to current function names, and much else: (b96a636c60) - Bug 1244222 - Tests. r=bz (9cfbe54859) - Bug 888969 - Make XPCJSID instanceof comparisons work correctly when [[GetPrototypeOf]] on the [[Prototype]] chain of the instance being tested throws an exception. r=bz (ff450de2ef) - Bug 1256306 - Bump the Windows stack limit. r=bholley (5f4f0c3b1c) - Bug 1257234 - Detect main thread's stack size at runtime, on Windows. r=ted (108608ce73) - bug 1264651 - remove dom.max_child_script_run_time pref r=billm (443bc36d84) - Bug 1266630 - Make fallible the orphan node table used during memory reporting. r=mccr8. (b7c6da934c) - Bug 1259699 - Adjust Windows stack limits to account for large PGO stack frames. r=bholley (87ff9dbef9) - bug 1250486 - get rid of the static ctor for XPCShellImpl.cpp r=bz (748def73ea) - Bug 1231309 - guard sz with assert. r=bobbyholley (34080a1022) - Bug 1255817 part 5. Remove the now-unused xpc::SystemErrorReporter. r=bholley (c014f7fa7f) - Bug 1257888 - Link chromium mutex-based atomics implementation to webrtc signaling tests. r=froydnj (fadb9764b2) - Bug 1253123 - Remove ipc_sync_message (r=jld) a=kwierso (7cc8a8fe0e) - Bug 1253123 - Remove ipc_channel_proxy (r=jld) (4000ea80f3) - Bug 1253123 - Remove message_router (r=jld) a=kwierso (8b41859a17) - Bug 1235633 - IPC OOM mitigation by eliminating buffer copying (r=jld) (ecd6c12bec) - Bug 1263314: Remove NonThreadSafe. r=jld (06a6331dee) - Bug 1261567 - Include compat dir in libevent include path. r=billm (b496d2ad1f) - Bug 1258312 - Add crash annotation to EnumSerializer r=jld (7face45d63) - Bug 1262463 - part 1 - turn NS_RUNTIMEABORTs in protocols into FatalErrors; r=jld (731b35dc68) - Bug 1262463 - part 2 - don't pass the other process pid into FatalError; r=jld (6f54c58494) - Bug 1262463 - part 3 - out-of-line NS_RUNTIMEABORT calls in IPDL-generated code; r=jld (0728f61a5f) - Bug 1259428 - part 1 - don't call Log methods of generated method classes; r=jld (938c6ff705) - Bug 1259428 - part 2 - remove dodgy static_cast downcasts from logging statements; r=jld (4fa0a5228d) - Bug 1259428 - part 3 - remove Log() methods from generated message subclasses; r=jld (964a0c2487) - Bug 1259428 - part 4 - remove prtime.h from generated protocol headers; r=jld (8548b37b91) - Bug 1259428 - part 5 - convert Message subclasses to constructor functions; r=jld (63f195ffce) - Bug 1259428 - part 6 - remove unneeded MessageDecl methods; r=jld (fc8cd72bd7) - Bug 1262458 - rename {msg,reply}Class-related things to reflect new functional reality; r=jld (8122887df4) - Bug 1258663 - Crash annotate system call failures in the IPC transport. r=gabor (5f483a17a5) - Bug 1258317 - Part 1: Annotate the crash report with process information on failure to transfer an IPC transport to another process. r=krizsa (107947c7c3) - Bug 1258317 - cross compilation fixup. (534c214182) - Bug 1258317 - Part 2: Remove unused privilege in mozilla::ipc::TransferHandleTorProcess(). r=krizsa (82fc8ca4cc) - Bug 1252246 - Try to use PTHREAD_PROCESS_SHARED for CrossProcessMutex on more Unices. r=glandium f=kats (195742a508) - now we can add crashreporter (2cbb6f2d1a) - Bug 1251226 - Avoid passing a std::wstring through the variadic method; r=bobowen (80b2dd8034) - Bug 1260736 - Let the client to filter out its interested messages to lower the number of times entering the monitor in PeekMessages(). r=dvander (3aea1ed77a) - Bug 1264662 - Record IPC message capacity instead of size. r=billm (fbfff4e5b8) - Bug 1261099 - Avoid two Message copies in MaybeUndeferIncall. r=billm (ebb7fe2b47) (05670874b5) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1266578 - OOM crash if malloc fails in ProcessIncomingMessages(). r=billm (5eca51be20) - Bug 1261340: Don't refer to |ScopedFreePtrTraits| in |RawDBusConnection|, r=shuang (61996db57c) - Bug 1246931: Move |DBusWatcher| into its own files, r=shuang (a7fc4a68c3) - Bug 1246931: Add support for |RefPtr<DBusConnection>| and convert callers, r=shuang (6c07d3e8d3) - Bug 1246931: Add support for |RefPtr<DBusMessage>|, r=shuang (c0ed14cc2b) - Bug 1246931: Add support for |RefPtr<DBusPendingCall>|, r=shuang (a007a15699) - Bug 1246931: Use |DBusConnection| in |DBusWatcher|, r=shuang (11b5eaf535) - Bug 1246931: Add DBus I/O helpers, r=shuang (be249fa65b) - Bug 1264398 - Avoid extra assign() on windows in IPC code (r=jld) (0f31b798a3) - Bug 1261231: Fix shutdown leak in imgLoader::GetInstance. r=gabor (96e5143c41) - Bug 1256999 - Use nsIDocument for ImageCacheKey. r=bz r=seth (cd4765ce34) - Bug 1257101. imgFrame::IsImageComplete says whether we've had pixels decoded to the whole image rect, but it's used to check if the frame is finished decoding. These are different things when the image has more than one progress pass. r=seth (06d619410f) - Bug 1222596. If RasterImage::LookupFrame does (some) sync decoding and encouters an error we don't want to return the surface with an error. r=seth (c4dbc8e0d5) - Bug 763784 - Make VectorImage::GetAnimated check for CSS animations. r=dholbert (4d0e5b88eb) - Bug 1210745 - Update CheckProgressConsistency() to match current ImageLib behavior. r=tn (2317b24fcc) - Bug 1249576. Add crashtest. (485f03120b) - Bug 1251091. Add crashtest. (7e1682a1e6) - Bug 1253362. SVGDocumentWrapper::IsAnimated can be called after SVGDocumentWrapper::DestroyViewer so null check mViewer. r=dholbert (2b4a1c4619) - Bug 1210745. Change image progress asserts to allow transparency to be posted after the size is posted. r=seth (699f3c5496) - Bug 1209780. Use the DrawResult return value of imgIContainer::Draw in the cocoa code. r=seth (abaea789e3) - No Bug - Remove some unnecessary SVGImageContext.h includes and add comments. r=sparky (8232661a23) - cleanup style (de33e4bfa8) - Bug 860857, support custom datatransfer types using a special type, r=smaug,jmathies,mstange (cf7e19deb6) - Bug 1248459 - Don't query out-of-bounds selection; r=masayuki (86c127f143) - Bug 1261671 - ContentEventHandler::ConvertToRootRelativeOffset() should return the root-relative result in the frame's own appUnits, not the root's appUnits in the case when they're different. r=masayuki (2cd72b5ebc) - Bug 1266702 - Clean up formatting in dom/events/DataTransfer.* and mark some methods const, r=jwatt (bd439cdad5) (cbc917768b) * 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
    1 point
  36. Probably, the same case as with the forceful takedown of Supermium. Guys, someone knows, does it support Win7 without the bloated SP1?
    1 point
  37. New regular/weekly KM-Goanna release: https://o.rthost.win/kmeleon/KM76.5.5-Goanna-20240720.7z Changelog: In-tree changes: - KmAppInfo: restore GetLastRunCrashID() API (00118dd9) - bump KM version (52f44b04) Out-of-tree changes: * update Goanna3 to git 696c195395...be4fef94e8: - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1257418 - Remove USE_DEPENDENT_LIBS. r=gps (fbe746751a) - Bug 1255817 part 3. Remove the now-unreachable JS_ReportPendingException call in nsXPCWrappedJSClass::CallQueryInterfaceOnJSObject. r=bholley (dbffe901a9) - Bug 1255817 part 4. Make XPCJSRuntime use MozCrashErrorReporter by default, now that there should be no stray JS_ReportPendingException calls. r=bholley (68d82e4b68) - Bug 1245000 part 1 - Add a preference for enabling Element.animate; r=bz (d94ab6f1d9) - Bug 1245000 part 2 - Don't ship Animation.ready; r=bz (129a052a16) - Bug 1245000 part 3 - Turn on Element.animate in release channels too; r=bz (cc6743c31d) - Bug 1245000 part 4 - Enable the Animation constructor when Element.animate is enabled; r=bz (8a65bdedcb) - Bug 1247685 - WebIDL and DOM implementation changes for app server keys. r=mt,baku (69870bd979) - Bug 1248565 - Introduce MOZ_LOG_* variables for mozilla logging. r=erahm (ea236d3b26) - Bug 1258231 - Lock while iterating console messages. r=mccr8 (2b54be114d) - bug 1252104 - make NS_ERROR_GET_CODE() and NS_ERROR_GET_MODULE() constexpr r=froydnj (f29ae0f6de) - Bug 1221160 - fix AutoTraceLogLock deadlock on Windows; r=froydnj (11f1c2a071) - Bug 1251895 - don't race on nsTraceRefcnt's object serial number tables; r=mccr8 (2d52aceffe) (8375688356) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 1244228 - fix DrawTargetSkia::OptimizeSourceSurface to still create Skia surfaces for GPU canvases even if creating a GPU surface failed. r=jmuizelaar (fbc02e4dc2) - Bug 1259621 - use AlphaBoxBlur in DrawTargetSkia::DrawSurfaceWithShadow. r=mchang (2c9df0d317) - Bug 1257717 - throw InvalidStateError when CreatePattern fails to snapshot source. r=jrmuizel (95fc5c4101) - Bug 1264246 - verify that SkCanvas::getClipBounds succeeds before using result. r=mchang (20de1e897d) - Bug 1249033 - Enable Subpixel AA Text on Skia for Dwrite and GDI Fonts. r=jrmuizel (f7402a4c9d) - Bug 1250787 - ensure that subpixel text is disabled for Skia's Cairo font host so that font rendering matches our font metrics. r=mchang (4d77670f3f) - Bug 1260454. Disable LCD font smoothing if we have grayscale AA on OS X. r=lsalzman (44bf859df5) - bug 1263685 - Check if paint has a shader before modifying local matrix in DrawTargetSkia::MaskSurface. r=mchang (7937265238) - missing bit for bump version (e3b14b6854) - Bug 1252650 - Add data parameter to NS_CreateServicesFromCategory; r=froydnj (3a3d392669) - Bug 1252650 - char16_t*/char16ptr_t mismatch fixup. (66d4bb1d1b) - Bug 1261382 - Add ReadUntil method to Tokenizer r=mayhemer (3c0b05a359) - Bug 1242963 - nsID stringification helper for logging; r=froydnj (966c73478a) - Bug 1262400 - remove outer pointers from nsCreateInstance* classes; r=erahm (16680909c0) - Bug 1257331 - Use LPWSTR instead of LPVOID to prevent C4477 warning; r=froydnj (456b677e25) - bits of Bug 868814 - Fold mozallo (3580f23f5a) - Bug 1256027 - Cast a few things in varargs to the expected types for MSVC 2015 warning-compatibility. r=froydnj (6e3e5f1763) - Bug 1257262 - Removed invalid nsAString:Assign variant. r=bsmedberg (1f4b303f7a) - bug 761909 - crash reporter plumbing for gonk. r=bsmedberg (e27b276147) - Bug 1223800 - Accept BHR reports from 50% of beta clients. Up from 1%. r=vladan (fb67094a56) - Bug 1241507 - Enable BHR reporting on 100% of the Beta population. r=rvitillo (6b0abc7767) - Bug 1234618. Fix misdetected BHR hangs. r=jchen (4adc7d57a1) - bits of Bug 1128768: Part 2 - Refactor (f75559fe5b) - Bug 1168167 - Mark LazyIdleThread::mIdleObserver with MOZ_UNSAFE_REF. r=froydnj (9750c9686d) - Bug 1250829 - add customized assertions for completion promises to facilitate promise chaining. r=bobbyholley. (66fe29ec68) - Bug 1191063 - Followup comments. r=me DONTBUILD (a627c0bb62) - bits of Bug 1172769 (6ce96f8829) - Bug 1241656 - Lazify AutoTaskDispatcher::mDirectTasks. r=bholley. (48fab6f3fb) - Bug 1202667 - make TaskQueue dispatching slightly more efficient; r=mccr8 (0c9f19fae5) - Bug 1202667 - make TaskQueue task running slightly more efficient; r=mccr8 (cee8821a9f) - Bug 1207031 - Suppress -Wshadow warnings from google-breakpad headers in xpcom/threads. r=froydnj (e2426fcf8a) - Bug 1222101 - Reorder some thread code (r=jld) a=kwierso (4b816f501e) - Bug 1257862 - Allow running things from the delayed work queue during a nested event loop. r=khuey,dvander (066ce6ab76) - Bug 1236789. Part 2 - fix assertions when |aFlags == NS_DISPATCH_TAIL|. r=bholley (d6500cceb0) - Bug 1254218 - Rely on old-configure.in to detect dladdr(). r=njn (8beefa2792) - Bug 1257786 - Proper init of LogModule for Nuwa r=khuey (5170b5250b) - Bug 1236108: Add support for exception-context annotations for content processes to the crash reporter; r=bsmedberg (3441aa3773) - Bug 1256541: Fix incorrect generation of path for child process .extra files when content sandboxing is disabled; r=bsmedberg (ae8ba562ca) - bug 1264242 - Write memory info streams for Windows Minidumps of child processes. r=bsmedberg (245b797566) - Bug 1257486 - Annotate global memory status in the crash reporter for child process; r=bsmedberg (9af844a7dd) - Bug 1257209 - record uptime in crash metadata/pings, r=gfritzsche (bfe0ea8759) - Bug 1241613 - Only pass MOZ_BUILDID to the one source that requires it in toolkit/xre. r=mshal (944abe034f) - Bug 1255655 - Heap-allocate _progname. r=froydnj. (3cce05ccaa) - Bug 828704: Avoid duplicate -foreground argument when launching Firefox on OSX. r=mstange (2e5c12bdda) - bug 1166759 - force annotateCrashReport arguments to be UTF-8. r=bsmedberg (5f2c30676a) - Bug 1261751 - Problems with OS X Sandboxed TempDir and Rules. r=bobowen r=gcp (9ad9272b92) (ca069b707a) - import changes from `dev' branch of rmottola/Arctic-Fox: - Bug 711670: Setup the Mac Application Delegate earlier to respond to URLs clicked in external applications when updates are pending. r=mstange (2471469680) - Bug 1252889: On Windows use a 10 second pause instead of trying to invoke the JIT debugger with MOZ_DEBUG_CHILD_PROCESS. r=jimm (228cd75e4d) - Bug 1253850: On Windows change MOZ_DEBUG_CHILD_PROCESS back to JIT and add MOZ_DEBUG_CHILD_PAUSE for a 10 second pause. r=mchang (83cdaec0a7) - Bug 1259809 - Make CrashReporter::InitChildProcessTmpDir conditional on MOZ_CRASHREPORTER. r=aklotz (19672176dd) - Bug 1176703 - attempt to fix define issues with some compilers/envs, rs=me, CLOSED TREE, bustage (8e6f3a6b94) - Bug 1266292 - Remove unnecessary compiler checks for MSVS 2010 in gfx/2d/. r=bas (9c3acd9f5d) - Bug 1195788 - Set arithmetic composite clamp to true for D2D filter node. r=bas (5ba4191512) - Bug 1248223 - verify that stroke widths are finite for Skia. r=jmuizelaar (1f676f5c24) - Bug 1241918 - remove workarounds for zero-length dash intervals in HelpersSkia.h, r=jrmuizel (ce0665ab77) - crashreporter includes (c64d834706) - Add a driver crash guard to D3D11 DXVA. (bug 1253666, r=mattwoodrow) (8589eef61b) - Bug 1189918 - Theme overrides should work in safe mode for comm-central applications too. r=bsmedberg (535486b4c0) - Disable the WebGL crash guard on Android. (bug 1256371, r=snorp) a=merge (855cc66aef) - Bug 1265273 - Add two missing get_PBrowserParent() null checks. r=jimm. (b5f9ac0967) - Bug 1261599 - null check SendPBrowserConstructor return in ContentParent to avoid crashes on content process shutdown. r=smaug (28e2ad71b5) - Fix D3D11VideoCrashGuard missing a switch case in ContentParent. (bug 1263725, r=mattwoodrow) (07a5b63633) - Bug 1216371 - Load default theme from XPI in safe mode. r=bsmedberg (596bc54a13) - Bug 1216371 - Load default theme from XPI in safe mode. r=bsmedberg (596bc54a13) - Bug 1162965 - Use /dev/shm instead of /tmp for sandbox chroot if possible. r=kang (4fa3c77756) - Bug 1264226: Don't use '_COARSE' Posix clocks if not defined, r=jld (a173b195eb) - Bug 1259273 - Add sys_unlink to seccomp-bpf whitelist. r=jld (72ca2626f6) - Bug 1259283 - Add sys_fchown to seccomp whitelist. r=jld (628f3cd627) - Bug 1253194: Suppress -Wimplicit-fallthrough clang warning for intentional fallthrough in icu_utf.cc (which is imported code). r=bobowen (a503963ff2) - Bug 1257031 - Return more informative error code when encountering invalid integers rather than SEC_ERROR_BAD_DER. r=keeler (7f4955614a) - Bug 1153446 - Replace instances of double spacing with single spacing in nsserrors.properties. r=dkeeler (166f0bd1c4) - Bug 1256530 - Fix compiler warnings about casting to larger size. r=hsivonen (53020d1543) - Bug 1251235 - changed from naked pointers to UniquePtr to prevent resource leak. r=froydnj (953040a377) - Bug 1264887: |nsWifiScannerDBus|: Hold DBusConnection reference in |RefPtr|, r=josh (3d8b9b3f52) - Bug 1212056 - Add virtual dtor to WindowsWifiScannerInterface. r=jdm (ce00cf01e9) - code style (d7fea73856) - Bug 1222170 - fix initialization order in ScopedWLANObject; r=jimm (b515ea0273) - Bug 1204919 - Build osx_corewlan.mm with -Wno-error=objc-method-access to avoid bustage on some machines in warnings-on-errors builds. r=glandium. (f88b02868d) - Bug 1204277 - Disallow warnings in six more directories. r=glandium. (7e249cf4ab) - Bug 1248761 - Destroy VRHMDInfo objects on the compositor thread r=daoshengmu (ccf6af1fec) - Bug 1260530 - Add support for Oculus 1.3 runtime r=daoshengmu (f7ed5abb4f) - Bug 1242005 - Disable C4359 to allow compilation on VS2015; r=kip (71db7148f4) - Bug 1176158 - "Remove unused lang atoms (ko-xxx and x-symbol)" r=vladimir (25e46d1150) - Bug 1176158 - Part 2. Remove unused ko_xxx and x_symbols from gfxGDIFontList. r=jdaggett (9b213333e0) - Bug 1262304 - Add a pref to dump display lists for the content process. r=BenWa (c0a4e56542) - Bug 1180942 - Use cairo's image backend for drawing on linux. r=jrmuizel (83deed801b) - Bug 1254400 - Handle device reset for d3d9. r=dvander (170aaeb272) - Bug 1254400 - Make access to d3d9 device manager thread-safe. r=dvander (a6112d8b84) - Bug 1256547: When the DWM is disabled don't force presentation. r=milan (0fe31bfc50) - Bug 1229988 - verify that reference draw target is created after device reset and log failure in gfxWindowsPlatform::UpdateRenderMode. r=milan (32084a6688) - Bug 1250669. Don't init dwrite fonts if content device creation fails. r=dvander (4a55599c8b) - Include vendor and device ids in content process DXGI adapter matching. (bug 1228905, r=jrmuizel) (c9403d3798) - Bug 1233576. Ensure vsync timestamps are monotonic on Windows 10. r=kats (d075fb7230) - Bug 1249036 Check that DWMFlush executes correctly. r=kats (f460b09853) - Bug 1259248. Follow up to fix static analysis bustage (ab462300ff) - Bug 1259248. Add a comment about mfbt/Range.h (a78f7c2110) - Bug 1262681 - Draw target also has to be valid for us to demand a snapshot. r=bas (c4c0bfcaca) - Bug 1258313 - Allow snapping of 'pseudo' 3d transforms. r=jrmuizel (fc85ac9147) - Bug 1241060 - Skip non-2d sibling layers when calculating plugin window clipping under e10s. r=mstange (fc6836ab11) - Bug 1257288 - Have APZ hit testing respect scroll clips. r=kats (5f8db91909) - Bug 1255961 - Avoid zeroing out shmems (r=dvander) (cc0e856f83) - Bug 1251523 - Remove TextureFlags::DEALLOCATE_CLIENT from AndroidSurfaceTextureData r=nical (406a27e229) - Regenerate SyncObjects on device resets. (bug 1255711, r=bas) (8a83c23cdf) - Bug 1240659 - Add GonkNativeHandle::CreateDupNhObj() r=nical (3da777655e) - Bug 1238602 - Update ParamTraits<GonkNativeHandle>::Read() r=nical (969f8177fd) - Bug 1248923 - Store result in Update ParamTraits<GonkNativeHandle>::Read() r=nical (0052a996a0) (be4fef94e8) * 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
    1 point
  38. BTW it seems to be possible to create a small kernel32 wrapper (stubbing EncodePointer/DecodePointer/SetDllDirectoryW/GetProcessHandleCount, wrapping GetProcessId, hex-editing VC2010 compiled EXE/DLL replacing KERNEL32 to KERNELXP i.e. our little wrapper) to make it work on XP RTM: kernelxp src+bin+util: http://o.rthost.win/gpc/files1.rt/kernelxp.7z
    1 point
  39. ~23mins full build on Ryzen 5800X and WD SN750 SSD on Win7 (with `-j14` parallel make jobs).
    1 point
  40. trying to import NSS 3.42 Beta in this, it builds but CORS seems to be broken and it refuses to load many https sites like ddg. EDIT: the problem is not in NSS, but 38esr's HTTP/2 encoding is too old. EDIT2: changes pushed. https://github.com/roytam1/firefox38-vc2010 and works even on a 486: binaries for testing: original NSS from 38esr: http://o.rthost.win/gpc/files1.rt/firefox-38.8.0-vc10-nss319.7z updated NSS from retrozilla (TLS 1.2): http://o.rthost.win/gpc/files1.rt/firefox-38.8.0-vc10-nss321.7z updated NSS for TLS 1.3: http://o.rthost.win/gpc/files1.rt/firefox-38.8.0-vc10-tls13.7z
    1 point
  41. with lots of fixes, it works. changes: https://github.com/roytam1/firefox38-vc2010/compare/dd0315735742429dbaae309babccfc08583f9310...HEAD with a js rdtsc hackfix, it can work on a 486. https://x.com/roytam1/status/1797637584493101276
    1 point
  42. https://codeberg.org/Nicholas_McAnespy/FxVC10Mods/releases
    1 point
  43. and also created a VC6 build for replacing NSS in Firefox 1.0 - 1.5 and K-Meleon 0.9-1.1: https://msfn.org/board/topic/185966-my-browser-builds-part-5/?do=findComment&comment=1266992
    1 point
  44. and have fun making NS9 working with TLS 1.3! updated files: http://o.rthost.win/gpc/files1.rt/ns9-nss-update-tls13.7z
    1 point
  45. since @ClassicNick provided a newer NSS port, I'm trying to fix/port it to VC7.1 and VC6 for retrozilla use. EDIT: progress - VC7.1 works, VC6 needs more fixing: in NT 3.51: and in Win95: EDIT2: VC6 port is finished. PR created: https://github.com/rn10950/RetroZilla/pull/68 Bin (suite): http://o.rthost.win/gpc/files1.rt/retrozilla-suite-tls13-20240526.7z browser-only (rzbrowser): http://o.rthost.win/gpc/files1.rt/rzbrowser-tls13-20240526.7z
    1 point
  46. Yes, looks very weird it's still not there, I mean for a fifteen years old totally out-of-support buggy OS.
    1 point
×
×
  • Create New...