Leaderboard

Unfortunately, Windows 2000 is unfortunate, did not find anyone to support it from these problems If you knew its value and the value of its logo and the feelings of its users towards it, it would not be left like this . WinWord2000 triste ma grazie a tutti !

This project make possible to compile ACPI 2.0 driver from leaked XP SP1 & W2003 RTM sources, has same functionality as existing acpi.sys v6666 (still missed integer/fields/memory 64 bit support/) Grab leaked XP SP1/W2003 RTM sources (google it) Use "XPSP1/NT" directory as basedir if you want to compile acpi.sys for Windows XP x32 Use "Win2K3/NT" directory as basedir for Windows 2003 x32 / Windows 2003 x64 / Windows XP x64 Download any GNU patch package for windows (gnuwin32.sourceforge.net, cygwin, mingw, msys2, ...) Open command console, change current dir to base\busdrv\acpi\ (Windows XP x32) Save text diff patch https://pastebin.com/C5NXwHbS (v7 update) to file base\busdrv\acpi\sp1_to_sp3(ACP2).patch (Windows 2003 x32 / Windows 2003 x64 / Windows XP x64) Save text diff patch https://pastebin.com/8QURrM49 (v7 update) to file base\busdrv\acpi\rtm_to_sp2(ACP2).patch (Windows 2003 x32 / Windows 2003 x64 / Windows XP x64) Rename Win2K3/NT/public to Win2K3/NT/public2 (Windows 2003 x32 / Windows 2003 x64 / Windows XP x64) Update compiler and headers to mix of W2003 DDK+WRK, unpack https://anonfiles.com/J1W9H1a8y1/W2003_tools_update_7z to basedir with overriding existing files Remove "read only" flag from base\busdrv\acpi directory including sub-dirs and files Apply patch to convert original SP1/RTM sources to SP3/SP2 with extended ACPi 2.0 syntax: patching file driver/amlinew/amlipriv.h patching file driver/amlinew/amlitest.c patching file driver/amlinew/data.c patching file driver/amlinew/misc.c patching file driver/amlinew/object.c patching file driver/amlinew/parser.c patching file driver/amlinew/proto.h patching file driver/amlinew/type1op.c patching file driver/amlinew/type2op.c patching file driver/inc/aml.h patching file driver/nt/debug.c patching file driver/nt/debug.h patching file driver/nt/devpower.c patching file driver/nt/internal.c patching file driver/nt/interupt.c patching file driver/nt/irqarb.c patching file driver/nt/osnotify.c patching file driver/nt/pciopregion.c patching file driver/nt/rangesup.c patching file driver/nt/root.c patching file driver/nt/wake.c Change current dir to basedir (Windows XP x32 / Windows 2003 x32) Run razzle environment setup: (Windows 2003 x64 / Windows XP x64) Run razzle environment setup Change current dir to base\busdrv\acpi\driver\ Complie ACPI driver: build /Dcegbw Compiled acpi.sys.sys will be in (x32) base\busdrv\acpi\driver\nt\obj\i386\ or (x64) base\busdrv\acpi\driver\nt\obj\amd64\ Project contains implementation of new ACPi 2.0 syntax: ToInteger ToString ToHexString Continue ConcatenateResTemplate ToDecimalString Mod ToBuffer CopyObject MidString QwordConst (inside ParseIntObj) Timer CreateQWordField(fake it as CreateDWordField) Know issues workarounds: BSOD 0xA5 (0x10006, ...) missing _DIS method for "PNP0C0F" (PCI Interrupt Link Devices) BSOD 0xA5 (0x02,xxx, 0x0, ...) ACPI vs E820 mem ranges conflict IOTRAPS I/O range 0xFF00-0xFFFF vs VGA (10-bit decode!) conflict BSOD 0xA5(0x03, ..., C0140008, ...) error in ValidateArgTypes() when reading 64-bit fields BSOD 0x7E(c0000005, ...) error in AcpiArbCrackPRT() when referencing null pointer BSOD 0xA5 (0x11, 0x08, ..., ...) error in _AMLILoadDDB() (zero lenght buffer) BSOD 0xA5 (0x11, 0x08, ..., ...) error in _AMLILoadDDB() (doubled device definition) BSOD 0xA5 (0x0000000D, ..., 0x4449555F, 0) absence _UID method BSOD 0xA5 (0x11, 0x08, ..., ...) error in _AMLILoadDDB() (Connection() opcode) CPU definition as Device with _HID=ACPI0007 BSOD 0xA5(0x03, ..., C0000034, ...) postponed SSDT loading on x64 platform (v8 update) Assertion Fail on loaddsdt.c, line 488 for x64 builds (v8 update) Unresolved issues: BSOD 0xA5 (0x0000000D, ..., ..., ...) duplicated/absence _HID/_UID method (AMD boards) BSOD 0xA5 (0x2001, 0x01, 0xC0000034, ...) failure to evaluate the _PIC method in NotifyHalWithMachineStates() Conflicted device names in Windows device manager (Code 42)

Yes, Windows 2000 is the best classic system and really needs this support

Sounds great, looking forward to trying it out!

AFAIK you still can't install later versions of KernelEx as they are. You have to first install the last version of the original KernelEx, 4.5.2, and then update it with the later files. Is that still correct @jumper?

In the slipstreamed MCE SP4 media, inside file hivesys.inf, search for string "POSReady WPA trick". Remove the registry entry below that. Reinstall Windows (clean) from the latest installation media. This may fix the product key issue. Otherwise, original pidgen.dll must be used from MCE2005 installation media. .NET Framework 4.0 can be unhidden in the sysoc.inf (replace ,hide,7 with ,,7 at the netfx40=... line) so that it can be installed by the Add/Remove Components Wizard if desired. @tpao12 Regarding .NET Framework 4.0 issue, please remove all .NET Frameworks from Add/Remove Components. Restart the system and reinstall .NET Framework 3.5 first, restart, then install .NET Framework 4.0 and reboot again.

Yes, exactly. When I installed the certificate trust list, it sometimes worked. But as I said, only by luck! At the beginning, I could not check for updates because the patch was not yet installed. Why I was able to open the page without a proxy at that time is still a big mystery to me today, because the page requires ECDHE ciphers and these are not actually supported by Xp.

oh, there is a missing change that you spotted. will be fixed in next build.

fixed

Just by the way I was just browsing my gallery and came across this video which I recorded on 26 December 2021. These were the first times I was able to see the page without the 0x80072EFF error and without any proxy or other programs. I had only installed any certificates and activated Tls 1.2. Strangely, I could only open the page by chance and if you opened the page in a new tab or in a new window it didn't work either. https://photos.google.com/share/AF1QipN6Tetz5NMAgHMWUgGFVMLlU-xPaNzpIB3XmtKVmvVGIHEt1WhV1mtClh2zpIi7YA/photo/AF1QipOMz2Mn0O7XKqc8-w3E14KLUVjhMglEhayOevmi?key=Y2wyZ2pQMGtqQ2R5cm0xeHNuT0phOV9oc0FCdkJB

It sounds like your company needs to re-evaluate their internals. At any manufacturing company, computers that run critical machines should either be segregated or firewalled. Or the company should be using a WSUS and only push updates to it that have been vetted beforehand. Alternatively, they may be using the wrong (but cost effective) SKUs. I see that a lot where a company will save costs by using a retail OS instead of an Embedded/IoT OS. Last year I got to visit a company and do a site inventory for them. Their critical machines controlled lasers for micro-manufacturing, and none of the computers attached to those systems were connected to the internet. For the same reason you outlined, because if one goes down, manufacturing stops and the company looses big dollars. So they had made it so those systems did not get updates. That company also were using the "wrong" OS for the job as a reason to cut costs (I don't want to think about how much those laser machines cost) and that was their mitigation strategy.