Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-21 Thread tom ehlert via Freedos-devel
Hallo Herr Bernd Böckmann via Freedos-devel, am Donnerstag, 20. Februar 2025 um 00:11 schrieben Sie: > Hello Jeremy, > On 20.02.2025 00:03, perditi...@gmail.com wrote: >> Thanks for looking into it.  I will have a look at Watcom build and > see >> why it's  not patching header. > Have a look

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread Bernd Böckmann via Freedos-devel
Hello Jeremy, On 20.02.2025 00:03, perditi...@gmail.com wrote: Thanks for looking into it.  I will have a look at Watcom build and see why it's  not patching header. Have a look at https://github.com/FDOS/freecom/blob/fa674ea7b5aa4fcb0e4b4263d9ba31d79ccfc961/tools/ptchsize.c#L276 ;) While

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread perditionc--- via Freedos-devel
Thanks for looking into it. I will have a look at Watcom build and see why it's not patching header. Jeremy ___ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-devel

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread Bernd Böckmann via Freedos-devel
On 19.02.2025 23:10, Bernd Böckmann via Freedos-devel wrote: I saw that the word at 0x0A is 0x2B1 for v0.85a This was actually the GCC build of master, not the v0.85a Watcom build. I messed up my eight simultaneously open COMMAND.COM tabs in hexedit. But I think the conclusion I drew earlier

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread Bernd Böckmann via Freedos-devel
On 19.02.2025 22:35, Bernd Böckmann via Freedos-devel wrote: I saw that the word at 0x0A is 0x2B1 for v0.85a, where it is 0x118 for 0.86. But 0x118 is an impossible value for ptchsize +6KB This rather seems to be a result of ptchsize not patching the minimum needed additional memory when buil

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread Bernd Böckmann via Freedos-devel
Hello Tom, On 19.02.2025 22:17, tom ehlert via Freedos-devel wrote: Requiring minimum load size of 87K makes it load low The 87K I mentioned were the free UMA memory for which the unaltered 0.86 binary worked for me in connection with SHELLHIGH, after I instructed JEMMEX to include the memo

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread tom ehlert via Freedos-devel
Hallo Herr Bernd Böckmann via Freedos-devel, am Mittwoch, 19. Februar 2025 um 21:05 schrieben Sie: > I agree with Tom that changing the EXE header should be to way to go. > I did a quick test by patching the COMMAND.COM header by hand with a hex > editor. > When I patch offset 0x0A to be 0x400

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread Steve Nickolas via Freedos-devel
Replying here since I'm having trouble finding the place where exemod was actually mentioned. On Wed, 19 Feb 2025, Bernd Böckmann via Freedos-devel wrote: I agree with Tom that changing the EXE header should be to way to go. I wrote a replacement for exemod last year, but I never got around

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread Bernd Böckmann via Freedos-devel
I agree with Tom that changing the EXE header should be to way to go. I did a quick test by patching the COMMAND.COM header by hand with a hex editor. When I patch offset 0x0A to be 0x400 (16k additional memory), total initial allocation size becomes ~80K. That patched binary works, as it d

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread tom ehlert via Freedos-devel
Hallo Herr Bernd Böckmann via Freedos-devel, am Mittwoch, 19. Februar 2025 um 12:31 schrieben Sie: >> Am 19.02.2025 um 09:51 schrieb Bernd Böckmann via Freedos-devel >> : >> >> As previously stated, a significant portion of upper memory is not available >> under VMware. So we should further t

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread Bernd Böckmann via Freedos-devel
Forcing JEMMEX to use the memory region C900-DFFF results in 87K free in the UMA, and that is sufficient to make FreeCOM work when loaded via SHELLHIGH, meaning not terminating with an out of memory error. DEVICE=C:\FREEDOS\BIN\JEMMEX.EXE NOEMS I=C900-DFFF But this line is likely to have uninte

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread Bernd Böckmann via Freedos-devel
> Am 19.02.2025 um 09:51 schrieb Bernd Böckmann via Freedos-devel > : > > As previously stated, a significant portion of upper memory is not available > under VMware. So we should further try to figure out why this is the case. > > Screenshots are at: > https://nextcloud.iww.rwth-aachen.de/ind

Re: [Freedos-devel] What the AI says about the problem of Out of memory

2025-02-19 Thread Bernd Böckmann via Freedos-devel
I dumped memory maps for boot options 1 and 2 under VMware via INSTALL=C:\FREEDOS\BIN\MEM.EXE /D /P There is 69K of upper memory available. Seems that this is sufficient for the kernel to load the command interpreter high via SHELLHIGH. But most likely then FreeCOM fails to allocate additional

[Freedos-devel] What the AI says about the problem of Out of memory

2025-02-18 Thread Paul Dufresne via Freedos-devel
Since you did not reply much to my question... I asked my prefered A.I. Here what it says: This first question is less relevant... you could skip to the Me: In FreeDOS, I see that FreeCom seems to sometimes miss memory to allocate command line (40 bytes) in shell/init..