On Mon, 2004-06-21 at 16:37, Alain wrote:
> tom ehlert escreveu:
>
> >
> > if the a000 block is ever merged into the lower memory area,
> > interesting things might happen.
>
> I remember one for instance: it was possible with a MDA to have 704k
> lower memory. Not interesting anymore, after th
Arkady V.Belousov escreveu:
? _Now_ behavior of FD is same as with MS-DOS (ie. it not hangs when
base memory size is more than 640k after loading some driver).
I must agree that he has a point here ;-)
---
This SF.Net email sponsored by Bla
Hi!
22-Июн-2004 14:40 [EMAIL PROTECTED] ([EMAIL PROTECTED]) wrote to
[EMAIL PROTECTED]:
>>Ha! First, I don't know such (useful) programs, which use DMA and load
>>itself into "uppermost position" (LAST_FIT).
ase> Well, LBACACHE could eventually be a good candidate.
Do you mean, that LBA
>Hi!
>22-éÀÎ-2004 13:00 [EMAIL PROTECTED] ([EMAIL PROTECTED]) wrote to
>[EMAIL PROTECTED]:
>>>ASM> using EMM386 to provide UMBs, you may get troubles with DMA
>because of
>>>ASM> the mismatch of linear and physical addresses.
>>>FD currently compares buffer address (in dsk.c and blockio.c)
>w
Hi!
22-Июн-2004 13:00 [EMAIL PROTECTED] ([EMAIL PROTECTED]) wrote to
[EMAIL PROTECTED]:
>>ASM> using EMM386 to provide UMBs, you may get troubles with DMA because of
>>ASM> the mismatch of linear and physical addresses.
>>FD currently compares buffer address (in dsk.c and blockio.c) with
>>ex
2, 2004 11:41am
Asunto: Re: [Freedos-devel] bug in UMB support
>ASM> UMBs are different from other memory blocks. For instance, if
>you are
>ASM> using EMM386 to provide UMBs, you may get troubles with DMA
>because of
>ASM> the mismatch of linear and physical addr
Hi!
22-Июн-2004 09:46 [EMAIL PROTECTED] (Aitor Santamarэa Merino) wrote to
[EMAIL PROTECTED]:
ASM> One more opinion on this topic. I don't think it's a good idea that the
ASM> UMBs are chained into the main MCB chain.
This is property of the EMM386. But you may do (and undo) this also
manua
Hi,
One more opinion on this topic. I don't think it's a good idea that the
UMBs are chained into the main MCB chain.
UMBs are different from other memory blocks. For instance, if you are
using EMM386 to provide UMBs, you may get troubles with DMA because of
the mismatch of linear and physical a
Hi!
21-Июн-2004 23:20 [EMAIL PROTECTED] (Alain) wrote to
[EMAIL PROTECTED]:
>> A> I remember one for instance: it was possible with a MDA to have 704k
>> A> lower memory. Not interesting anymore, after the advent of VGA.
>> Why? Expanding base memory over VGA graphics segment is a nice featu
Arkady V.Belousov escreveu:
Hi!
21-Июн-2004 18:37 [EMAIL PROTECTED] (Alain) wrote to
[EMAIL PROTECTED]:
if the a000 block is ever merged into the lower memory area,
interesting things might happen.
A> I remember one for instance: it was possible with a MDA to have 704k
A> lower memory. Not intere
Hi!
21-Июн-2004 18:37 [EMAIL PROTECTED] (Alain) wrote to
[EMAIL PROTECTED]:
>> if the a000 block is ever merged into the lower memory area,
>> interesting things might happen.
A> I remember one for instance: it was possible with a MDA to have 704k
A> lower memory. Not interesting anymore, after t
tom ehlert escreveu:
if the a000 block is ever merged into the lower memory area,
interesting things might happen.
I remember one for instance: it was possible with a MDA to have 704k
lower memory. Not interesting anymore, after the advent of VGA.
So, my vote is that any workaround is as good as
Hi!
21-Июн-2004 12:31 [EMAIL PROTECTED] (tom ehlert) wrote to "Arkady V.Belousov"
<[EMAIL PROTECTED]>:
>> for example, as I understand, latest FD-EMM386 translates starting I= area
>> address A000 to A001 because there is bug in (current) FD. Isn't it?
te> right. ALL fd-emm386's translate a000 in
> for example, as I understand, latest FD-EMM386 translates
> starting I= area address A000 to A001 because there is bug in (current) FD.
> Isn't it?
right. ALL fd-emm386's translate a000 into a001; so it will never be
merged into the lower memory block.
if the a000 block is ever merged into the l
Hi!
20-Июн-2004 14:28 [EMAIL PROTECTED] (tom ehlert) wrote to Michael Devore
<[EMAIL PROTECTED]>:
>> I hope to have some free time later this weekend and will try to get caught
>> up on a few FreeDOS items, including your requested test, a stand-alone UMB
>> stress test,
te> whatever arkady sugge
Hello Michael,
> I hope to have some free time later this weekend and will try to get caught
> up on a few FreeDOS items, including your requested test, a stand-alone UMB
> stress test,
whatever arkady suggests - IMO its nonsense.
I coded the original implementation for 2 good reasons as it was
Hi!
20-Июн-2004 00:16 [EMAIL PROTECTED] (Michael Devore) wrote to
[EMAIL PROTECTED]:
MD> Unfortunately, there's been a death in the family,
My condolences to you.
MD> so all the planned tasks
MD> listed above will be pushed back another week or so.
Ok.
---
At 03:17 PM 6/18/2004 +0400, Arkady V.Belousov wrote:
17-éÀÎ-2004 22:40 [EMAIL PROTECTED] (Michael Devore) wrote to
[EMAIL PROTECTED]:
MD> I hope to have some free time later this weekend and will try to get
caught
MD> up on a few FreeDOS items, including your requested test, a
stand-alone UMB
MD
Hi!
17-Июн-2004 22:40 [EMAIL PROTECTED] (Michael Devore) wrote to
[EMAIL PROTECTED]:
MD> I hope to have some free time later this weekend and will try to get caught
MD> up on a few FreeDOS items, including your requested test, a stand-alone UMB
MD> stress test, and code up an XMS block auto-grabb
I hope to have some free time later this weekend and will try to get caught
up on a few FreeDOS items, including your requested test, a stand-alone UMB
stress test, and code up an XMS block auto-grabber/releaser for Erwin
Veermans' NWDSK if it proves feasible and fixes the immediate problem wit
Hi!
17-Июн-2004 09:47 [EMAIL PROTECTED] (Michael Devore) wrote to
[EMAIL PROTECTED]:
>> 2Michael Devore: Michael, try to initialize UMB blocks, returned to
>>EMM386 caller (DOS in case of with DOS=UMB), by 'M' letter - after this
>>FreeDOS should allocate UMB memory incorrectly (because prev
At 03:15 PM 6/18/2004 +0400, you wrote:
Hi!
2Michael Devore: Michael, try to initialize UMB blocks, returned to
EMM386 caller (DOS in case of with DOS=UMB), by 'M' letter - after this
FreeDOS should allocate UMB memory incorrectly (because prev_mcb() in
umb_init() will walk after latest previo
22 matches
Mail list logo
