Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Rich Walsh

Pages: 1 ... 10 11 [12] 13 14 ... 23
166
Programming / Re: Qt6
« on: January 13, 2023, 07:51:27 am »
@Paul,

PMFJI... attached is the util I mentioned ('mk83.exe') that will rename DLLs with long names.

Code: [Select]

mk83   - changes a DLL's file and module names to be 8.3-compliant
       - also sets/clears the HIMEM flag for executable segments

usage: mk83 [/H | /L] [/D] filename.dll [modname | lookup.txt]
  or   mk83 [/H | /L] [/D] /B lookup.txt

       [/H] set flag  [/L] clear flag  [/D] delete original DLL
       [/B] batch update all DLLs in lookup.txt

       [lookup.txt] a text file with one "filename.dll modname"
         pair per line; if the DLL name is qualified, its path is
         ignored for lookups but used for batch updates.


FYI... this is actually a REXX script attached to an executable header ('runrx.exe'). To see the script, open 'mk83.exe' in Text Editor, turn on word-wrap, then scroll down to the beginning of the script. You can recover the script by deleting everything from the beginning of the file down to the first /* then saving as 'mk83.cmd'.

167
Programming / Re: Qt6
« on: January 10, 2023, 08:05:00 pm »
could be one of those cases where due to our 8.3 DLL convention, we have multiple DLL's with the same name
Wouldn't it be easier to let these packages use whatever name they want, then in a post-processing step, update the file and module names as needed? This could be done at build-time or later in a batch coordinated by a control file.

Definitely interested - but this approach would also require changes to cmake to make cmake NOT shorten the names. I also suspect this might not work for some of the Qml issues, where Qt seems to be looking for the 'original' name.

The goal is to be helpful across a range of projects without requiring you to go out of your way to use it. Perhaps one of these use cases will fit your needs now or later:

* Basic usage:    mk83 qt6webengine.dll qt6weben

* Avoid name collisions:    mk83 qt6webee.dll qt6weben

* Coordinate renames using a control file:    mk83 qt6webengine.dll /F qt6dlls.txt
(where 'qt6dlls.txt' is a simple plain-text file with entries like "qt6webengine.dll qt6weben").

* Batch rename after a build:    mk83 /X qt6dlls.txt
(this assumes the control file contains qualified filenames so they can be found)

These are the scenarios that come to mind. If it seems worth the effort, I'll proceed.

168
Programming / Re: Qt6
« on: January 09, 2023, 01:44:17 pm »
could be one of those cases where due to our 8.3 DLL convention, we have multiple DLL's with the same name

Just a thought... It often seems like getting a usable 8.3 DLL name out of some of these build systems is quite a chore. Wouldn't it be easier to let these packages use whatever name they want, then in a post-processing step, update the file and module names as needed? This could be done at build-time or later in a batch coordinated by a control file.

If you're interested, let me know. I'd probably implement it in REXX (which is pretty good for patching binaries) but deliver it as an .exe so you wouldn't have to run 'cmd.exe'. And... since I'll probably recycle some of the REXX I wrote for MozTurbo, the util could also mark the DLL for high-mem, if desired.

169
Setup & Installation / Re: Virtual memory, a curious phenomenon
« on: January 04, 2023, 09:40:07 pm »
I just did a lot of testing on this. I found that the value shown for "Available virtual memory" depends on the size of your swap partition and nothing else. If your swap partition is less than at least 4gb, then the value shown is likely to be useless.

Here are the numbers for my normal setup using  "SWAPPATH=N:\ 2048 2048", where N: is the 4.9gb boot drive.

Code: [Select]
Total physical memory:     16,171 MB
Accessible to system:       2,355 MB
Additional (PAE) memory:   13,816 MB

Resident memory:              130 MB
Available virtual memory:   1,215 MB

Available process memory:
  Private low memory:         379 MB
  Private high memory:      1,792 MB
  Shared low memory:          347 MB
  Shared high memory:       1,763 MB

Moving the swappath to a 1gb ramdisk _or_ harddrive gives me:
Code: [Select]
Available virtual memory:   3,157 MB

Increase the partition to 2gb and I get:
Code: [Select]
Available virtual memory:      67 MB

Move the swappath to a 50gb harddrive:
Code: [Select]
Available virtual memory:      2,179 MB

Turn off swapping (MEMMAN=NOSWAP,PROTECT):
Code: [Select]
Available virtual memory:      2,101 MB

The values that you see here are returned by a call to DosQuerySysInfo(QSV_TOTAVAILMEM). I don't think anyone has ever questioned its output before, but it looks like there is good reason to believe it's buggy (or at least _very_ counter-intuitive).

170
I ONLY want to change the font size for the Drives/Drive/Folder WPS objects...I do not want to modify anything else.

To change the default font used by the WPS for icon titles and details-view text, edit os2.ini->PM_SystemFonts->IconText.
The default is normally 9.WarpSans but at 1920*1080, 10.WarpSans is more appropriate, particularly if you use Xview a lot (it always uses the default).

171
I noticed that the " Windows List Widget" on the xCenter is not showing the windows that are open or minimized on the desktop. The bar just remains always empty.

The problem is likely not in the widget itself but in the XWP Daemon (xwpdaemn.exe) which notifies it of changes. If the daemon isn't running or has some problem, the WinList widget won't get updated.

If you're running XWP Full, you have to enable the daemon yourself in XWP Setup, though enabling XPager should also do it, IIRC. Under XWP Lite, it's always enabled. If that doesn't fix it, perhaps the "Windows Miracle Cure" will - just reboot and hope for the best...

172
Applications / Re: PDF Printer
« on: December 20, 2022, 05:01:58 am »
I was able to print to a PS file from Firefox and then convert to PDF.

...but Firefox has the built-in feature to print directly to a PDF...so why aren't you using that instead of the PS=>PDF pathway?

What version of FF are we talking about here?

Does it work for you? Doesn't here.

I've never had it not work! Save the print job to file, then give the filename a ".PDF" extension.

173
Events / Re: Warpstock 2022 - Personal Remarks
« on: December 10, 2022, 10:02:15 pm »
2) The GUI and High Resolutions.
- A need that jumped out from this presentation was that we need to make bigger icons for xCenter. The xCenter size can be increased by the icons there remains small. It may not very difficult to make it but we need a developer to tweak the xCenter code and turn it into a reality.

Have you considered talking to XWP's current maintainer about it?

As it happens, it's already on my agenda because...

Quote
- The need to improve the size of fonts in one shot (a centralized GUI) for high resolutions was covered by Alex on his presentation, and I hope to see a release of his software for it next year.

I'm working with Alex on this because there's more to  getting fonts changed than a centralized GUI (though it's obviously needed). There's the matter of getting stuff to pay attention to those settings - most notably, XWP. That's why I've spent the last week  ensuring that it's next release will be 100% "font-compliant" and will honor those settings rather than always using 9.WarpSans. I've also updated DTProgram and some other WPS add-ons.

Anyway, implementing full-sized icons in XCenter as part of expanded support for higher resolutions is an obvious enhancement that will happen if at all possible. OTOH, PNG support is highly unlikely (would be nice, but...)

174
Programming / Re: Qt6
« on: November 15, 2022, 05:01:46 pm »
G:/qt6-os2/src/plugins/platforms/os2/main.cpp:49:1: error: Parse error at "IID"

The code in question is:
Code: [Select]
Q_OBJECT Q_PLUGIN_METADATA(IID QPlatformIntegrationFactoryInterface_iid FILE "os2.json")

'IID' identifies the datatype of "QPlatformIntegrationFactoryInterface_iid" but the compiler has no idea of what that may be because 'IID' hasn't been defined (or typedef'd).

From the context, I'd guess that 'IID' is a very fundamental type that should have been defined in a header #included by some ancestor of this class - you shouldn't have to #include it yourself. Since other platforms use this same function, maybe the definition of 'IID' was conditionally-compiled out for the OS/2 platform. Just a WAG...

175
Programming / Re: UEFI and GPT not developed by ArcaNoae - by whom else?
« on: November 08, 2022, 02:39:05 pm »
If I understand you correct: if it is possible to boot eCs on UEFI only it should be possible to boot an with AN Drivers updated MCP 4.52 (what is my main system) as well, shouldn't it?

If - and only if - you own a license for AOS 5.1 because you need AN's proprietary UEFI bootloader files installed in your EFI System Partition (ESP). Don't expect to use QSINIT - as I said, I believe that UEFI is AN-only.

Quote
Would you share these two lines wich have to be changed in config.sys? My guess is that one may be able to use an updated QSINIT and need the filter driver.

I described 2 very different scenarios. The one that you want to emulate requires NO changes to eCS _at all_. The other one that required config.sys changes involved a licensed 5.1 beta BIOS/MBR install that I converted to UEFI/GPT using files already part of that system (i.e. no license violations).

Quote
I wonder why these informations have not been included in the Presentation.

Either they don't want to have to deal with it (possible) or they simply don't know about it (probable).

FYI... I worked for AN for 5 years before quitting last February.

176
Programming / Re: UEFI and GPT not developed by ArcaNoae - by whom else?
« on: November 08, 2022, 01:05:23 pm »
I have no clue who is doing the work.

The primary developer for UEFI and GPT is Dmitry Zavalskov (Dixie), the author of QSINIT and one of the best low-level coders OS/2 has ever seen. Roderick Klein tested countless DOS games to help Dixie get the video right, while I worked with him on designing GPT and some of the boot-up "plumbing"; others at AN rearranged the deck chairs and the like. AFAIK, all of the work is proprietary to AN.

While the implementation of UEFI and GPT is really phenomonal on a functional level (I'm posting this from a UEFI-only machine with a GPT-formatted NVME disk running an AOS 5.1 beta and it all "just works"), a little perspective is needed to avoid getting too, too impressed with this accomplishment.

I'd estimate it took the equivalent of one full-time employee 2 years to implement almost everything needed. In the grand scheme of IT projects, that's small potatoes. And, out of all the subsystems that comprise OS/2, only two were changed: loader and video. Stuff like networking, DASD/LVM, PM, etc. are all unchanged.

Strangely enough, it's the fact that these changes are so limited that allows AOS to do things that I don't think are possible with any other OS. Before my old BIOS/UEFI machine died last week, I was able to boot eCS v2.2rc2 under UEFI with absolutely NO changes to the eCS installation. After it died, I xcopied my primary BIOS-based installation from the old machine to this UEFI-only machine, changed 2 lines in my config.sys, and then booted up the fully UEFI-compliant installation that I'm using now.

I missed Lewis' AOS 5.1 upgrade presentation at last weekend's Warpstock. Did he mention these as upgrade options?

177
Hardware / Re: Turn off Laptop Display
« on: October 21, 2022, 05:38:14 pm »
So how about any INT10h calls that do not change the resolution but actually want to write something to the screen like int 10h / AH = 0Ah or int 10h / AH = 13h ? Will these work ? Or is AN int10h support limited to only the calls that control/query the resolution ?

I did not participate in the development of the driver, so I can only comment on implementation details at the level of "informed observer". My understanding is that any call that is required to boot was implemented. So, if the kernel needs to output a string while still in real mode, the specific subfunction it uses is operational. Anything else is a no-op.

Among those no-ops is the one to change video mode. PM still issues it as always, thinks it succeeded, then adjusts other settings accordingly. The fact that it doesn't actually do anything is of no consequence because the loader has already set the mode using the exact same parameters that PM will use. True emulation: the code thinks it's doing one thing, the underlying system does something completely different, but the end result is the same so everyone is happy :)

FYI... OEMHLP has been greatly extended for UEFI to provide boot and video info that may not be available elsewhere. Refer to 'oemhelpex.h' for details.

178
Hardware / Re: Turn off Laptop Display
« on: October 20, 2022, 09:38:05 am »
It does raise the question of how resolutions are changed in UEFI mode. Does it depend on the hardware or does the mini driver take care of it?

When using Panorama in UEFI mode, AOS 5.1 is incapable of changing any video setting itself because it doesn't contain a bona-fide video driver - i.e. code that knows how to manipulate a video card's registers.. This "mini-driver" is a really thin BIOS emulation that returns data cached by os2ldr at boot time and (I believe) allows you to issue video commands that will be ignored. Instead, AOS uses a workaround that capitalizes on the fact that OS/2 has never been able to change resolutions on-the-fly.

When 'aosldr' starts up, it uses the UEFI firmware's boot-time services to read 'os2.ini' and determine the desired resolution / refresh rate / color depth. It then uses the firmware to set the video accordingly. Later, when it loads the OS/2 kernel, it loses access to those boot-time video services, but the video settings remain in place, unchanged and unchangable for the rest of the session.

Consequently, if turning off the display requires any sort of direct manipulation of the video card, you are S-O-L.

179
Applications / Re: Installing programs
« on: September 11, 2022, 06:25:36 pm »
Would you like my eCS 1.1 registration key? I no longer use any version of eCS, so the license should be transferrable.

180
Applications / Re: Boicot Discussion
« on: May 27, 2022, 01:38:04 am »
I took these words as your opinion and FUD against the OS4 development.

This is OUTRAGEOUS!  For anyone to make such an accusation with no basis in fact would be out of line. For you to do so - especially against someone who has done more to promote OS/2 development (e.g. Dooble) than nearly anyone - calls into question whether you are really suitable to be an os2world moderator. You should apologize immediately.

Pages: 1 ... 10 11 [12] 13 14 ... 23