aeBIOS Test Request
in [ASM]
|
From: s_dubrovich on 9 Oct 2007 08:47 On Oct 9, 7:29 am, "Rod Pemberton" <do_not_h...(a)nohavenot.cmm> wrote: > <s_dubrov...(a)yahoo.com> wrote in message > > news:1191889022.329681.96310(a)57g2000hsv.googlegroups.com... > > > > > > > On Oct 7, 3:09 pm, Mike Gonta <mikego...(a)gmail.com> wrote: > > > On Oct 5, 8:22 am, s_dubrov...(a)yahoo.com wrote: > > > > > Thanks for the source.. > > > > Thanks Steve, > > > I've made some major changes. > > > Please give the latest version another try. > > > > The latest aeBIOS version is here > > > >http://mikegonta.com/aeBIOS > > > > Mike Gonta > > > > look and see - many look but few see > > > Hi Mike, > > > V20071007 > > Boots to 'Hello World!" on.. > > > Pentium MMX Laptop (dell) Phoenix Bios. (c) 1985-1988. > > Pentium MMX Desktop (compaq presario). > > Pentium 4 Desktop (dell dimension 8250). > > :-) > > Some good news: > > PhoenixBIOS 4.0 Revision 6.0 (previously untested) > 1004 works - no problems > 1007 works - no problems > > :-) > > And, the bad: > > Award Modular BIOS v4.51PG > 1004 & 1007 works - but fails to wait for keypresses (existing issue) > 1007 works - but first boot says can't enable A20 then reboots and works > (new issue) > > Award Modular BIOS v4.51PG > 1004 fails - 01 > 1007 fails - 06 (existing issue, except 1004) > > Rod Pemberton- Hide quoted text - > > - Show quoted text - Hi Rod, That's an interesting result. I'm wondering if the Award Modular Bios is flashable with an upgrade? I browsed Mike's source to see if there was an unmatched CLI|STI, but there doesn't seem to be. I did have that as an issue on some example code I tried once, which was hanging the floppy (drv lite on), because STI was missed. Just perhaps, some such issue is present in the Award Modular Bios itself? Steve
From: Maxim S. Shatskih on 9 Oct 2007 12:07 > That's an interesting result. I'm wondering if the Award Modular Bios > is flashable with an upgrade? Yes it is. v4.51PG is something like the i815 chipset for Pentium-III/800 CPU of the year 2000 (mobos like Asus CUSL2). -- Maxim Shatskih, Windows DDK MVP StorageCraft Corporation maxim(a)storagecraft.com http://www.storagecraft.com
From: Rod Pemberton on 9 Oct 2007 13:48 <s_dubrovich(a)yahoo.com> wrote in message news:1191934063.190619.291800(a)o3g2000hsb.googlegroups.com... > On Oct 9, 7:29 am, "Rod Pemberton" <do_not_h...(a)nohavenot.cmm> wrote: > > <s_dubrov...(a)yahoo.com> wrote in message > > news:1191889022.329681.96310(a)57g2000hsv.googlegroups.com... > > > > > V20071007 > > > Boots to 'Hello World!" on.. > > > > > Pentium MMX Laptop (dell) Phoenix Bios. (c) 1985-1988. > > > Pentium MMX Desktop (compaq presario). > > > Pentium 4 Desktop (dell dimension 8250). > > > :-) > > > > Some good news: > > > > PhoenixBIOS 4.0 Revision 6.0 (previously untested) > > 1004 works - no problems > > 1007 works - no problems > > > > :-) > > > > And, the bad: > > > > Award Modular BIOS v4.51PG > > 1004 & 1007 works - but fails to wait for keypresses (existing issue) > > 1007 works - but first boot says can't enable A20 then reboots and works > > (new issue) > > > > Award Modular BIOS v4.51PG > > 1004 fails - 01 > > 1007 fails - 06 (existing issue, except 1004) > > > > That's an interesting result. I'm wondering if the Award Modular Bios > is flashable with an upgrade? > Yes. The problem system had it's BIOS flashed to the final version. I might've flashed the other, but I believe it's original since I don't have any ROM images for it my motherboards directory. Both the Award PC's are older. The one which mostly works is an Abit MB with an Intel 233 and Intel chipset, IIRC. The other is compact sized PC with a PC Chips MB, VIA chipset, and AMD K6-2 500. Despite being an "unknown" MB, it's been very reliable. The Abit MB was a good one too. It was one of the first jumperless boards and the first brand name board I owned. I got it real cheap from a small PC store that went out of business. They bought it for a custom PC build for another customer, but found out he wanted a different version of the MB that had USB. At the time, all they could do was swap MB's to get USB. > I browsed Mike's source to see if there was an unmatched CLI|STI, but > there doesn't seem to be. > There doesn't seem to be any problem in the aeBIOS code. I've done quite a bit of modification, testing, and a wide variety of temporary code changes to aeBIOS to expose the problem, but nothing yet. It'd be awful if I have to resort to a debugger... ;) [snip, BIOS error?] Yes, I'm beginning to think the problem is "outside" aeBIOS. I have a utility which I wrote that restores the BIOS vectors from DOS and boots the A drive. It's somewhat flawed - using undocumented features of DOS to restore some vectors instead of restoring an exact copy of the default BIOS vectors. It'll boot FYSOS but not DOS. DOS locks up at random places. With aeBIOS, it gets one good boot. Then it happens to fail on int 0x1c. So, DOS is doing something helpful which gets "changed" by aeBIOS such as a vector, memory, BDA, irq, ports. My thoughts have run to PM code in the BIOS, SMM, interrupts, stack data mismatch, NMI, destroyed flags/reg in the PM/RM switches, BDA, corrupted registers for int 0x13 since it gets called a few times, etc. But, I haven't found anything to clue me in yet. Rod Pemberton
From: Mike Gonta on 9 Oct 2007 18:28 On Oct 9, 8:47 am, s_dubrov...(a)yahoo.com wrote: > I browsed Mike's source to see if there was an unmatched CLI|STI, but > there doesn't seem to be. Hi Steve, Not on purpose, however the interrupt handler that displays the interrupt number does hang in a spin loop with interrupts disabled. I've added code to shut the floppy drives off. There is also now an indicator "PM" or "RM" to indicate the mode of the interrupt. The latest version is here http://mikegonta.com/aeBIOS Mike Gonta look and see - many look but few see
From: s_dubrovich on 10 Oct 2007 16:44
On Oct 9, 5:28 pm, Mike Gonta <mikego...(a)gmail.com> wrote: > On Oct 9, 8:47 am, s_dubrov...(a)yahoo.com wrote: > > > I browsed Mike's source to see if there was an unmatched CLI|STI, but > > there doesn't seem to be. > > Hi Steve, > > Not on purpose, however the interrupt handler that displays the > interrupt number > does hang in a spin loop with interrupts disabled. > I've added code to shut the floppy drives off. There is also now an > indicator > "PM" or "RM" to indicate the mode of the interrupt. > > The latest version is here > > http://mikegonta.com/aeBIOS > > Mike Gonta > > look and see - many look but few see Hi Mike, V20071010 Boots to 'A:\>' on.. Pentium MMX Laptop (dell) Phoenix Bios. (c) 1985-1988. Pentium MMX Desktop (compaq presario 4770). Pentium 4 Desktop (dell dimension 8250). :-) app hello works on all three. app binclock doesn't seem to work on any.. on the compaq it displays some red squares, gives PM 0D, and locks up. On the other two, the machines reboot, tho on the laptop some of the squares are briefly displayed. Steve |