Re: [Freedos-kernel] Re: test bootdisk
Hi! 15--2004 03:44 [EMAIL PROTECTED] (Bernd Blaauw) wrote to All: BB Bart, do you see the unnamed program, eating 48KB (probably just a BB viewing problem)? BB CTMOUSE 3,328(3K) 0(0K) 3,328(3K) BB 48,704 (48K) 0(0K) 48,704 (48K) BB Free 623,024 (608K)622,880 (608K)144(0K) Bernd, may you try: (1) MEM 1.6, (2) my MEM and (3) review complete lisintg (/F for Bart's MEM, /A in my)? --- This SF.Net email is sponsored by: thawte's Crypto Challenge Vl Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam Camcorder. More prizes in the weekly Lunch Hour Challenge. Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
Re: [Freedos-kernel] Re: test bootdisk
Arkady V.Belousov wrote: BB Bart, do you see the unnamed program, eating 48KB (probably just a BB viewing problem)? BB CTMOUSE 3,328(3K) 0(0K) 3,328(3K) BB 48,704 (48K) 0(0K) 48,704 (48K) BB Free 623,024 (608K)622,880 (608K)144(0K) Bernd, may you try: (1) MEM 1.6, (2) my MEM and (3) review complete lisintg (/F for Bart's MEM, /A in my)? I'll do that. Problem probably is in KEYB not loading completely successfull. Bernd --- This SF.Net email is sponsored by: thawte's Crypto Challenge Vl Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam Camcorder. More prizes in the weekly Lunch Hour Challenge. Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
Re: [Freedos-kernel] Re: test bootdisk
Hi! 15--2004 12:04 [EMAIL PROTECTED] (Bernd Blaauw) wrote to [EMAIL PROTECTED]: BB CTMOUSE 3,328(3K) 0(0K) 3,328(3K) BB 48,704 (48K) 0(0K) 48,704 (48K) BB Free 623,024 (608K)622,880 (608K)144(0K) Bernd, may you try: (1) MEM 1.6, (2) my MEM and (3) review complete lisintg (/F for Bart's MEM, /A in my)? BB I'll do that. Problem probably is in KEYB not loading completely successfull. ? --- This SF.Net email is sponsored by: thawte's Crypto Challenge Vl Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam Camcorder. More prizes in the weekly Lunch Hour Challenge. Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
Re: [Freedos-kernel] Re: test bootdisk
Hi, Bernd Blaauw escribió: Bernd Blaauw wrote: Hello all, I've put online a new bootdisk with which I, and you, can easily experiment. Download it from: http://fdos.org/ripcord/beta9-final/test/testing.zip [274KB, 1.44MB unzipped] OK, just uploaded a new version, now includes fixed autoexec.bat and mounting. program for small ISO file. http://fdos.org/ripcord/beta9-final/test/testing.zip Bart, do you see the unnamed program, eating 48KB (probably just a viewing problem)? Un-UPX-ed KEYB executable is around 20KBs, it is compiled for 2KB stack and no heap... Aitor --- This SF.Net email is sponsored by: thawte's Crypto Challenge Vl Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam Camcorder. More prizes in the weekly Lunch Hour Challenge. Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
Re: [Freedos-kernel] Re: test bootdisk
Bart Oldeman escribió: On Wed, 15 Sep 2004, Bernd Blaauw wrote: Bernd Blaauw wrote: Hello all, I've put online a new bootdisk with which I, and you, can easily experiment. Download it from: http://fdos.org/ripcord/beta9-final/test/testing.zip [274KB, 1.44MB unzipped] OK, just uploaded a new version, now includes fixed autoexec.bat and mounting. program for small ISO file. http://fdos.org/ripcord/beta9-final/test/testing.zip Bart, do you see the unnamed program, eating 48KB (probably just a viewing problem)? Yes, even in DOSEMU -- a few drivers can't be loaded but the effect is the same. It's a bug in mem (line 1124 of mem.c can be changed to if (mlist-owner (is_psp(mlist-seg) || mlist-owner == mlist-seg + 1)) to fix it -- here's a freed memory block but there's a PSP directly following the MCB and that confused mem). Well this weekend at the latest I'll try to release 1.7 with the German, Italian and new Dutch translations then. Bart, didn't you get my file with the Spanish translation to MEM? Aitor --- This SF.Net email is sponsored by: thawte's Crypto Challenge Vl Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam Camcorder. More prizes in the weekly Lunch Hour Challenge. Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
Re: [Freedos-kernel] Re: test bootdisk
alright, MEM display is a MEM bug as Bart indicated, Lucho fixed, as final public developers's work for his part, the 'remainig' - 'remaining' cosmetic bug. now only this strange bug of why DISPLAY loads high if MEM is first run, and loads low (and atapicdd/cdrcache load high instead) if MEM isn't run. Keyb seems alright afterall, but I thought 1) 'keyb partially loading -- unnamed block eating UMB' 2) 'keyb loaded with valid parameters' -- MEM /C shows alright. My conclusion then: must be KEYB. Conclusion now: must be MEM displaying something wrong (as Bart indicates). no idea if the DISPLAY binary has been UPX'd, and if that has any affect. Bart, I'm not distributing a 386/BorlandC/Apack kernel, this was just an experimental bootdisk image to show some troubles I experienced. bootdisk will be 8086+, and I hope the SHSUCDX/SHSUCDHD parts can also become 8086+ in no way is this a full-featured usable bootdisk, but it does show how to load the various drivers. I'm still confused by syntax for DISPLAY/MODE/KEYB.., so it's good to have working examples at hand. I'm planning to use a 8086/Openwatcom/UPX version of either Jeremy's or Lucho's sources, and to distribute a 2035A (the conservative tree) as kernel for harddisk. Haven't figured out why not everything (except UDMA ofcourse due to lack of VDS) loads high. Plenty of UMB space (48K). perhaps I should test against official 2035, to see if self-UMB-loading programs work there. All my testing is done in VMware 4.5.2, btw. No Bochs/Qemu/DosEMU/VirtualPC etc. Bernd --- This SF.Net email is sponsored by: thawte's Crypto Challenge Vl Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam Camcorder. More prizes in the weekly Lunch Hour Challenge. Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
Re: [Freedos-kernel] Re: test bootdisk
Bernd Blaauw escribió: no idea if the DISPLAY binary has been UPX'd, and if that has any affect. UPX first, COM2EXE next, produces smallest size. I'm still confused by syntax for DISPLAY/MODE/KEYB.., so it's good to have working examples at hand. I think what you are using is ok. It's equal than for MS, except that (both being fixed now) - DISPLAY is a COM/EXE and not a SYS - KEYB does not admit many layout bunch files (such as KEYBOARD.SYS, which is mostly a DATA file), but single files, such as SP.KL or IT.KL or GR.KL... If you have troubles, ask what troubles you or tell me what you want to get. Aitor --- This SF.Net email is sponsored by: thawte's Crypto Challenge Vl Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam Camcorder. More prizes in the weekly Lunch Hour Challenge. Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
Re: [Freedos-kernel] re: test bootdisk
Hi, Eric Auer escribió: About DISPLAY: UPX first, COM2EXE next, produces smallest size. bad idea. COM2EXE cannot detect de-UPX-ed size! So the exe header will tell how much space the compressed COM needs. But the whole idea of using COM2EXE was to let DOS know the de-UPX-ed size explicitly: As far as I recall, the idea of using COM2EXE is that there is a bug in kernel about loading high COM programs that resize their chunks (when there is no room, it aborts instead of loading low), and to avoid this problem. The experiments by Bernd showed that under MS-DOS it works fine (a growing COM file). UPXing EXE files has that ad*antage... ...that we don't need because the problem [might] doesn't happen for EXE files. If you load an UPXed COM into a too small memory block (or an com2exed UPXed com), you can end up in the situation 'upx stub sees that there is not enough memory, so it throws int 20 (exit with errorlevel 0, no message) instead of unpacking and running the program'... Well, maybe. But Bernd has showed in those strange-MEM experiments that DISPLAY loads high this way well. As far as for resident size, it doesn't change at all. So I could just try and revert the order of the steps for the next version. The only thing that will change is that DISPLAY.EXE will grow a bit (but it will be the LAST COM/EXE version anyway). All my testing is done in VMware 4.5.2, btw. No Bochs/Qemu/DosEMU/VirtualPC etc. By the way, does anyone know how to mount a drive or directory as a drive for VMWARE? (something like DOSEMU's lredir). Aitor --- This SF.Net email is sponsored by: thawte's Crypto Challenge Vl Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam Camcorder. More prizes in the weekly Lunch Hour Challenge. Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
Re: [Freedos-kernel] re: test bootdisk
Aitor wrote: By the way, does anyone know how to mount a drive or directory as a drive for VMWARE? (something like DOSEMU's lredir). I never had VMWARE myself (why spend US $189 when you don't need to?) but from what I've read about it has a virtual network card so you could use a SMB client in it (like MSNET) and use that to connect to the real machine and mount a drive. Bart --- This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170 Project Admins to receive an Apple iPod Mini FREE for your judgement on who ports your project to Linux PPC the best. Sponsored by IBM. Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
Re: [Freedos-kernel] re: test bootdisk
Hi! 15--2004 19:31 [EMAIL PROTECTED] (Eric Auer) wrote to [EMAIL PROTECTED]: EA About DISPLAY: UPX first, COM2EXE next, produces smallest size. EA bad idea. COM2EXE cannot detect de-UPX-ed size! So the exe header This is unimportnat, because (my) COM2EXE doesn't detects size at all. It simply points in header, that should be allocated 64k. _If_ you run COM2EXE with option, which reduces requested size, then _you_ should know, how much of memory should be allocated at program run (after unpacking). EA NCACHE2 only works if max 31.x MB XMS are free? No. Not works only autodection for ext option (without argument), but you may not allocate all memory (omit ext option) or point allocated memory explictily, as argument of ext. EA Then in must be quite old, 1994. :) --- This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170 Project Admins to receive an Apple iPod Mini FREE for your judgement on who ports your project to Linux PPC the best. Sponsored by IBM. Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
Re: [Freedos-kernel] re: test bootdisk
Hi! 16--2004 01:19 [EMAIL PROTECTED] (Aitor Santamara Merino) wrote to [EMAIL PROTECTED]: bad idea. COM2EXE cannot detect de-UPX-ed size! So the exe header will tell how much space the compressed COM needs. But the whole idea of using COM2EXE was to let DOS know the de-UPX-ed size explicitly: ASM As far as I recall, the idea of using COM2EXE is that there is a bug in ASM kernel about loading high COM programs that resize their chunks (when No. Idea of using COM2EXE is that _MS-DOS_, when loads .COM file into UMB (with LH), not necessarilly allocates (more than) 64k, if there is no enoughly big block (for example, there remained only 30k free). _And_, if program doesn't check how much of memory it have (for example, by examining value of SP), then it may/will overwrite own stack and following memory in this case (of reduced block). On the other side, DOS doesn't know how much of memory .COM program _really_ need (it not necessarily need all 64k), so it allocates biggest UMB block. So, COM2EXE allows to simplify program (not need to check available memory) and optimize program loading (it will/may be loaded into best suited block, not into biggest). ASM there is no room, it aborts instead of loading low), and to avoid this ASM problem. The experiments by Bernd showed that under MS-DOS it works fine ASM (a growing COM file). --- This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170 Project Admins to receive an Apple iPod Mini FREE for your judgement on who ports your project to Linux PPC the best. Sponsored by IBM. Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php ___ Freedos-kernel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-kernel
