jumper

First two items were previously promised. Last fix is major--only hex return values were correctly parsed, unsigned decimals became zero. Known issue: Still not fully thread safe for stubs (okay for forwards); status: fixable I also use IPS Default. I'll try Opera 12.02 and FF 9.01; KM74_31 is way too slow! Thanks. :)

> the email-problem seem to be caused in Kexbases.17 ! When using v16 it works again. Good. Now try comparing api logs (DebugWindow.exe). In Filter Settings, include "(bfa" to limit output to kexbases apis. > I will look into the printing issue In Kstub823.ini: [Comdlg32.dll] PrintDlgExA=>Kexbasen:PrintDlgExA_new PrintDlgExW=>Kexbasen:PrintDlgExW_2 PrintDlgW=>Kexbasen:PrintDlgW_3 Check Kstub823.log or include filter "PrintDlg" to see which api to focus on.

@filth: Thanks. No problems have been reported for the current implementation, however. @MiKl: I use SM2.0.4 but not the email client. Please narrow the problem down to Kexbases.17 or Kexbasen.17. .18 is packaged and awaiting final review. Kexbasen.16 exported 11 new print dialog functions--access them via Kexstubs. @MERCURY127: due to site changes, I can no longer upload files. @deomsh: please debug things further with KexApiLog (DebugWindow.exe) and report back.

Kstub823.7z // 823 - search: Kstub823.ini, Kexstubs.ini, stubs.ini; e120; fix r<unsigned decimal value>

No, it means updates will be detected and automatically repatched without needing a manual reinstall of KernelEx.

- http://www.network-drivers.com/drivers/1241/1241700.htm Or use the built-in NIC or USB to connect an external wifi modem/router

"Most antivirus programs identify runouce.exe as malware—e.g. Microsoft identifies it as Virus:Win32/Chir.B@mm, and TrendMicro identifies it as PE_Chir. B-O or PE_CHIRUX.B." - https://www.file.net/process/runouce.exe.html Once removed, try creating a read-only folder in "C:\Windows\system32" named "runouce.exe" to prevent it from coming back.

Sounds like a DRM problem. The LGE may use a version that isn't supported by XP. What connection type did you use for each the Dell and LGE: VGA, DVI, HDMI? Try reconnecting the Dell without changing drivers. If this works, reconnect the LGE using the same connection type, else uninstall the new drivers and restore system settings from a backup.

KernelEx v4.5.2 patches Kernel32.dll in memory. KernelEx v4.5.1 patches Kernel32.dll on disk, however v4.5 Final Changelog.txt states: "KernelEx doesn't need to be reinstalled anymore if kernel32 is updated".

Any hex editor. (My ImportPatcher probably won't work; I haven't tested it on PE32+ files.)

GetTickCount can be called in place of GetTickCount64 for 49.7 days if the upper dword is zeroed. On XP X64 I would image this to be the case. Replace all "GetTickCount64" strings with "GetTickCount\0\0" to try.

Thanks for the direct links. Both file headers are "%PDF-1.5". SumatraPDF (all versions from 0.9.0 up) works with no complaints. I'll see about compiling the test code into an app for Win95+. Courtesy of Google: This is the html version of the file https://meltdownattack.com/meltdown.pdf. Google automatically generates html versions of documents as we crawl the web. This is the html version of the file https://spectreattack.com/spectre.pdf. Google automatically generates html versions of documents as we crawl the web. FLUSH+RELOAD: a High Resolution, Low Noise, L3 Cache Side-Channel Attack This is the html version of the file https://eprint.iacr.org/2013/448.pdf.

SpecuCheck doesn't check for the vulnerabilities, it checks for a patch. No patch for Win<7, so no reason for a Win<7 SpecuCheck. From https://github.com/ionescu007/SpecuCheck/blob/master/README.md:

> The only change to the system during this time span that I remember was updating Kernelex. However, uninstalling it did not solve the problem. Interesting. What was the previous installed version of KernelEx and to what version did you update?

Try disabling BIOS -> USB legacy once it is no longer needed. Try connecting the keyboard with and without a USB hub. Does a USB mouse work?

"AIO" = All In One Have you tested the scanner? "Please insert a disk into drive" may just mean that you haven't scanned anything yet so a file system hasn't been created in the EPSON WF-2750's internal memory yet. ...Or that you need to insert a thumb drive or SD card into an EPSON WF-2760 (as previously suggested by @bphlpt!).

Thanks for testing this new browser! Looks like we have a cascading failure. What to do: * The "box of stuff" is very important--please report it exactly. You probably don't have enough CPU (FineSSE might help). * Profile in Dependency Walker to see what module is failing to load. * Enable API logging on modules that fail to init and retest/profile with DebugWindow/KexApiLog open. * Test each (and every) module with Import Patcher. That version of Msvcr120 loads on my Via C3 system (no SSE or CMOV). I put it in C:\Windows\KernelEx\Msvc and enabled it with this registry setting: [HKEY_LOCAL_MACHINE\Software\KernelEx\KnownDLLs] "MSVCR120"="MSVC\\MSVCR120.DLL"

> How hard would it be to use Goanna engine in kmelon and back-port kmelon so it could work on 98se with kernelex. KM74_31 works enough to be usable, so maybe KM76+Goana 3.4.1 already does, too. I haven't had time to test it yet. > it complained that msvcr120 had missing exports in kernel32 Not for me. Should load using KernelEx 4.5.2016.17. Ask for help in Kext: DIY KernelEx extensions.

Every version of WMP seems to have a different way to enable captions: Accessibility for Windows Media Player Videos: Captions

Might be trying to write to a local configuration file. Try installing to a writable drive.

http://www.msfn.org/board/uploads/css_built_*_custom.css.* *custom.css* Working in Opera 10.63 with theme IPS Default. Font quality better at 90% scaling.

Agreed. Even my 656c never worked very well, but you might want to try its 9x drivers for USB. I haven't been able to find a timeline (or list) anywhere of HP printers (that supported 9x), so have no idea what was the last for 9x. If we knew, those would be the drivers to try.

> > > third_party/rust/kernel32-sys/src/lib.rs: pub fn K32GetModuleInformation( > > Most of these new dependencies are likely in third party code, not the core code. Or they might be generated by newer compilers. > How to avoid such dependency? Rather than create an OS stub, modify the source to assume failure (or success) and remove the call. If an api is called multiple times, it might be easier to redefine it: #define IsUserWin10Fanboy() FALSE if (IsUserWin10Fanboy()) {...} Or remove the third-party module entirely.

> NewMoonXP? It emphasizes New Moon's mission to serve XP, just as TenFourFox serves Mac OS X 10.4 users. A new moon can't be seen (it's below the horizon) so: FauxMoon -> FOMoon -> FiveOneMoon > third_party/rust/kernel32-sys/src/lib.rs: pub fn K32GetModuleInformation( Most of these new dependencies are likely in third party code, not the core code. Or they might be generated by newer compilers.

Must-read guide to getting USB printers recognized by Win98 and drivers installed (see third and fifth posts).