It is currently Wed Nov 20, 2019 5:34 am




 Page 3 of 4 [ 54 posts ]  Go to page Previous  1, 2, 3, 4  Next
Author Message
 Post subject: Re: Fast Reads - how?
PostPosted: Sat Sep 18, 2010 3:18 am 

Joined: Wed Mar 04, 2009 7:27 am
Posts: 98
Location: United Kingdom
I understand: what I mean is, if the CF card was set up in 16-bit mode (for example, to work with older SDX versions which don't support 512 byte sectors), the driver needs to know NOT to switch to 8-bit mode at startup. There's NO WAY for the driver to know this until it gets the information from the disk (to discern how it was partitioned). So all that's required is for the sector size info to be readable in both modes.



_________________
The Last Word, MA65 Assembler, RIF SpartaDOS Utilities, XEDIT
Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Sat Sep 18, 2010 4:52 am 

Joined: Mon Jan 23, 2006 10:49 am
Posts: 187
Location: Salzburg, Austria
flashjazzcat wrote:
I understand: what I mean is, if the CF card was set up in 16-bit mode (for example, to work with older SDX versions which don't support 512 byte sectors), the driver needs to know NOT to switch to 8-bit mode at startup. There's NO WAY for the driver to know this until it gets the information from the disk (to discern how it was partitioned). So all that's required is for the sector size info to be readable in both modes.

There's another problem:

At the point of driver initialisation a CF card might either be in 8bit or 16bit mode, so you can't reliably read the partition sector (you don't know if you have to read 256 or 512 bytes). You have to put the card into a defined state first either by resetting it (then it will be in 16bit mode, but I'm not sure if this is possible with the MyIDE interface, IIRC this requires access to the alternate status/device control register) or by issuing the $EF/$01 or $EF/$81 command.

I'd strongly recommend doing all transfers in 16bit mode (except when accessing 512 byte partitions) so that the physical data layout is identical for both harddrives and cf cards. Mixing 8bit and 16bit modes (i.e. always accessing CF cards in 8bit mode) has some really nasty consequences:

First, you have to support 2 different physical data layouts in your driver. When accessing the partition table or SD/DD partitions you have to guess if it was written in 8bit or 16bit mode. If you are accessing a SD/DD partition (or the partition table) in 8bit mode you have to add read/write dummies for every other byte to maintain compatibility with 16bit drives.

Keep in mind that people might dump the whole content of their harddrive / cf cards to a file on the PC (using "dd", "Winhex" or any other program that allows raw disk access) and then later dump them back onto another device (which might be either a harddrive or a CF card). I use this a lot when testing, files on my PC are a lot faster to access than the CF card in my cardreader.

If you use the same physical layout for both CF cards and harddrives, cross-dumping is no problem at all, you don't have to worry about anything. If you use a different physical layout there's a chance of getting garbage.

And, last but not least, supporting 2 different layouts (a dummy every second byte vs. dummies at the end of the sector) will be a real mess in myidetool. I really wouldn't want to have to implement this...

so long,

Hias


Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Sat Sep 18, 2010 5:34 am 

Joined: Wed Mar 04, 2009 7:27 am
Posts: 98
Location: United Kingdom
OK - so we read the partition table in 16 bit mode, and if a partition is flagged up as having 512 byte sectors, we use CFA mode to access it. The partition data is always read/written in 16 bit mode. That sounds fine to me!



_________________
The Last Word, MA65 Assembler, RIF SpartaDOS Utilities, XEDIT
Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Sat Sep 18, 2010 6:27 am 

Joined: Mon Jan 23, 2006 10:49 am
Posts: 187
Location: Salzburg, Austria
flashjazzcat wrote:
OK - so we read the partition table in 16 bit mode, and if a partition is flagged up as having 512 byte sectors, we use CFA mode to access it. The partition data is always read/written in 16 bit mode. That sounds fine to me!

Yes, I guess this should be the easiest way to implement.

We can also see it this way: let the drive do all necessary conversion, by using the 8bit/16bit modeswitch commands.

so long,

Hias


Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Sat Sep 18, 2010 12:24 pm 

Joined: Wed Mar 04, 2009 7:27 am
Posts: 98
Location: United Kingdom
Something I forgot: I'd like to have a write-lock flag on individual partitions.

FDISK is racing along: 1,200 lines already (most of this is the UI):

Attachment:
fdisk.png


What's the best way to read the disk geometry (i.e. the number of LBA sectors)? Using the $EC Identify Drive command, we find the LBA capacity in words 60-61 (32 bits). Of course, in 16-bit mode we can't get at all the bits.

Must we resort to the lower-order bits of the CHS words at words 1, 3 and 6? I believe Sijmen uses a brute-force method of repeated CHS reads in FDISK.BAS (not sure if this is true of the OS-based FDISK). What would be the formula to convert the CHS geometry to a maximum LBA block value?


There are images and/or files attached to this posting. You must be logged-in to view or download this content.



_________________
The Last Word, MA65 Assembler, RIF SpartaDOS Utilities, XEDIT
Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Sun Sep 19, 2010 5:06 am 

Joined: Mon Jan 23, 2006 10:49 am
Posts: 187
Location: Salzburg, Austria
Hi Jon!

flashjazzcat wrote:
Something I forgot: I'd like to have a write-lock flag on individual partitions.

Excellent idea!

We could merge this into the (previous) "active flag", turning it into a "partition flags" entry. Maybe something like this:
bit 0: partition active (0=inactive, 1=active)
bit 1: partition writable (0=read only, 1=writable)
bit 2..7: reserved, must be 0

BTW: I'd like to finish a first testing version of MyIDE tool today, it would be great if we used identical definitions and values so our programs are compatible :-)
Here are my current definitions:
        // offsets in partition table
        enum {
                ePartitionOffset        = 0x10,         // start of partition entry #1
                eImageSpaceOffset       = 0xf0          // partition entry #15
        };

        // offsets within global header block
        enum {
                eOfsTotalLBASize        = 0,            // 4 bytes LO..HI
                eOfsLBAIdentificationr  = 4,            // must be $10
                eOfsPartitions          = 5,            // ?
                eOfsBootNo              = 6,            // ?
                eOfsOptionByte          = 7             // ?
        };

        // partition block size
        enum {
                ePartitionBlockSize     = 0x10
        };
        // offsets within partition information block
        enum {
                eOfsPartitionStart      = 0,            // 4 bytes LO..HI
                eOfsPartitionSize       = 4,            // 4 bytes LO..HI
                eOfsPartitionType       = 8,            // see below
                eOfsPartitionFlags      = 9,            // see below
                eOfsPartitionDrive      = 10            // drive number (1..15)
        };

        // partition types
        enum {
                ePartTypeNone           = 0,
                ePartTypeSD             = 0x01,         // 128 bytes/sector
                ePartTypeDD             = 0x02,         // 256 bytes/sector
                ePartTypeQD             = 0x03,         // 512 bytes/sector
                ePartTypeImageSpace     = 0x80
        };

        // partition flags
        enum {
                ePartFlagActive         = 0x01,
                ePartFlagWritable       = 0x02
        };

Just drop me a line if you'd like to have some other layout/values/flags/... and I'll change them (it's really no big deal, as you see it only requires a few changes to my header file).

Quote:
FDISK is racing along: 1,200 lines already (most of this is the UI):

Wow, you are really fast! The UI looks nice, can't wait to see the final version!

Quote:
What's the best way to read the disk geometry (i.e. the number of LBA sectors)? Using the $EC Identify Drive command, we find the LBA capacity in words 60-61 (32 bits). Of course, in 16-bit mode we can't get at all the bits.

Must we resort to the lower-order bits of the CHS words at words 1, 3 and 6? I believe Sijmen uses a brute-force method of repeated CHS reads in FDISK.BAS (not sure if this is true of the OS-based FDISK).

In 16bit mode this is a problem, we have to use some trial and error or let the user enter the number (most harddrives have the number of LBA sectors usually printed on the label).

Quote:
What would be the formula to convert the CHS geometry to a maximum LBA block value?

Just multiply them, the result is usually a little bit lower than the maximum LBA value (LBA usually doesn't divide without remainder to C/H/S, so the last < H*S sector aren't available in CHS mode as they wouldn't form a full cylinder).

so long,

Hias


Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Sun Sep 19, 2010 5:34 am 

Joined: Wed Mar 04, 2009 7:27 am
Posts: 98
Location: United Kingdom
Thanks Hias! The draft definition looks good: I have to go out this afternoon, but I'll get back to it this evening. If I need any changes, I'll try and let you know ASAP. I must admit I need a few hours away from this today so I can come back refreshed later. I've basically written a UI framework in two evenings... :) Form over function or what!

But today we should get some meat on the bones. Priorities:

1. Get FDISK to recognize the interface and produce a partition table (biggest job is writing the code to display the table and add new entries: should be about 2-3 hours' work).
2. Make the driver pick up (basic) geometry from the partition table (just enough for now to get a working system)

Beyond this:

3. Code to detect uninitialized disks
4. Code to test for CFA capability
5. Generally beefed-up retry and error checking
6. Surface-scan (on Tools menu) to determine whether device can run with the fast (60KB/s) sector I/O loop

...plus other niceties. I'm 90% confident I'll have something running tonight.



_________________
The Last Word, MA65 Assembler, RIF SpartaDOS Utilities, XEDIT
Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Sun Sep 19, 2010 6:07 am 

Joined: Mon Jan 23, 2006 10:49 am
Posts: 187
Location: Salzburg, Austria
flashjazzcat wrote:
I must admit I need a few hours away from this today so I can come back refreshed later.

I can understand this - I needed (and did) this yesterday :-)

Met with my friends and took a guided tour through the Almkanal, I thing I always wanted to do for some 20 years by now.

Quote:
But today we should get some meat on the bones. Priorities:

1. Get FDISK to recognize the interface and produce a partition table (biggest job is writing the code to display the table and add new entries: should be about 2-3 hours' work).
2. Make the driver pick up (basic) geometry from the partition table (just enough for now to get a working system)

Sounds fine!

I have some (currently dead) code in myidetool to setup a basic partition table from scratch. No fancy stuff, everything hardcoded but it might be of help to you. ATM it creates a 720 sectors SD partition #1 (as D1:) and a 65535 sectors DD partition #2. I guess for your tests a 65535 sectors QD partition #1 would be better, any suggestions?

BTW: So far I the "partitions#" and "boot#" fields in the header aren't handled in myidetool. Should I set them to some specific values?

so long,

Hias


Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Tue Sep 21, 2010 1:29 am 
User avatar

Joined: Sun Nov 02, 2003 8:15 am
Posts: 1376
Location: netherlands
Hi Guys,

Just a short post from me. You are doing a great job.
Although I haven't read all posts ( not enough time :-( )
The direction is OK.
If possible, the LBA layout already set for MyIDE would be nice to support.
That way the myidetool from Matthtias works, since it include LBA support when ID=$51 ;-)

For reading sector 0 I would suggest to use 8 bit mode (like the atari does on DD 1050 drives).
Extract the partitiontable and switch to 512 bytes/sector when set as QD in the drive-partition entry.

Any idea of adding a checksum for error-checking?
For 128 bytes/sector this would be easy.....
Hmmmm. This is something for the MyIDE-routines too.

Later,
Sijmen.

Here is my last layout code:
Version 24 nov 2008.

Supported densities are S/M/D/Q = 128/128/256/512 (bytes per sector).
IDE disabled when identifier $51 is not found.

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Partition table (10x8 bytes):

1x8 bytes for drive settings.
a1 a2 a3 a4 a5 a6 a7 a8
a1 a2 a3 a4 : size in LBA sectors
a5          : $51   '5.1' identification (IF>8 is new format, changed matthias)
a6          : Drive bits
a7          : Optionbyte
a8          : free
(reference to previous CHS: CL CH HD SC Dn DB IM IS)

8x8 bytes for drives.
b1 b2 b3 b4 b5 b6 b7 b8
b1          : density+drive# ($0x,SD+MD $2x,DD $8x,QD / DRIVES $x1-$x8)
b2 b3 b4 b5 : Partition location. LBA Sector L, M, H, U byte
b6 b7 b8    : Partition size L, M, H.

1x8 bytes for image-setting.
c1 c2 c3 c4 c5 c6 c7 c8
c1 c2 c3 c4 : Start of Image location. LBA Sector L, M, H, U byte
              =$00000000 then images are disabled/not set.
              =$value, this limits the extension of logical drives.
c5 c6       : Numbers of Images L, H.
c7 c8       : free

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

MyIDE-keys:

RESET-related;

+SELECT    powerup  (resets MyIDE and cartridge)
+OPTION    enable internal BASIC-rom
+START     boot in image mode (cold/powerup only)
+SHIFT     coldstart
+HELP      disable IDE.
+CNTR+HELP disable HSIO and IDE
all combinations of above.

LOGICAL/IMAGE MODE [shift+control+key];
+D             Disable IDE
+D then number Disable IDE-partition D#
+E             Enable  IDE, all partitions
+E then number Enable  IDE-partition D#
+P             Protect IDE from writing
+U             Unprotect IDE
+A             Toggle onscreen activity
+R             Reset MyIDE, load partition-table.
+H             Enable  HSIO, get speedbytes.
+N             Disable HSIO

IMAGE ONLY MODE (extra shift+control+keys);
(default: HDOPT = $81. A, no activity and P,write protected set)
+number        Load relative image-slot [number-1] as D1:

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

All sectors #0 are reserved, images sector(s) $0000 is reserved for image_header.
Will be protected from read/write. Set $FF in $307 for access!
Imageheader: 31 characters: Name, SPACE, density.
Name, 29 characters
SPACE (#$00), hide name/density when non zero.
Density: SD, $B3, must be valid value, hide name/density when false.
         MD, $AD
         DD, $A4
         QD, $B0

All images are $410 / #1040 sectors (sector 1 through 1040)
Were the 'hidden' sector #0 has the label (not the last sector of the image).

LBA-sector addressing needs 3.5 bytes.
L, M, H, U (U, top half is used for ,master/slave and CHS/LBA addressing)
256*256*256*16=268435456 sectors (128 Gbyte)
Byte U is always set to $Ex, were $E stands for LBA@master
and $x for the highest 0.5 byte LBA sector-address.

EXAMPLE OF SECTOR 0:
-0 -1 -2 -3 -4 -5 -6 -7
00 00 20 00 51 03 01 00
01 01 00 00 00 D0 02 00
22 D2 02 00 00 FF FF 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
D2 02 01 00 90 07 00 00

MEANING:
DSKSIZE = $00200000 (~1 Gb)
MYIDE   = 5.1 VERSION
DRIVES  = $03 = %00000011 ( D1 D2 PRESENT)
OPTION  = $01
IMAGES  @ $00102D2 FIXED SIZE $410
#Images   $0790
D1 (SD) @ $0000001 SIZE $02D0 ( 720 SECTOR  810 DISK)
D2 (DD) @ $00002D2 SIZE $FFFF

OPTIONBYTE, (normally all zero’s):
Bit 7:   HD is write protect when set
Bit 4-6: Bootdrive 1 through 8
Bit 3:   Start IMAGE MODE when set.
Bit 2:   Disable HSIO when set
Bit 1:   Disable HD-patch when set
Bit 0:   Disable on-screen activity when set


Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Tue Sep 21, 2010 5:04 am 

Joined: Wed Mar 04, 2009 7:27 am
Posts: 98
Location: United Kingdom
Hi Sijmen

If you mean read sector 0 using 16-bit mode and switch to 8-bit PIO mode when dealing with 512bps partitions: that's what we're doing. This way the partition table is the same on any media.

Looking at your last layout code, I see a few limitations (which is why we're using a 16 byte header and 16 byte entries at the moment).

1. 8 byte partition entries leave no scope for 32 bit (or at least 28 bit) LBA partition sizes (which will be supported eventually by SDX, yielding 4GB partitions). We could probably pack the bits together, but I don't want to introduce a lot of bit shifting, and anyway we have plenty of room. We also need space for a partition lock flag, plus drive # and density, and scope for additional flags in the future. With 16 byte entries we have lots of space and the LBA start and size are 4 bytes each.
2. Matthias had the excellent idea of keeping image entries in specially flagged partition entries. The SDX driver will simply ignore these, and it keeps the system simple and flexible.

If you look a couple of posts up, you can see Matthias has posted the exact specification I'm working with at the moment.

As for option bytes, partitions can be individually locked using their own option bytes. They can also be active or inactive. As for boot drives, we have scope for drive 1 through 15 (we have room for 15 partitions + the header record).

Instead of hotkeys, the SDX driver will support a number of additional methods (probably via new SIO commands) such as reload partition table, raw I/O, etc. This is how the FDISK program will force the driver to activate the new partition table without a coldstart. There's nothing to prevent the creation of additional drivers and apps which interact with the base code via the SIO. Since MYIDE.SYS will be on the cart, it could have an "/R" switch which would simply reload the partition table when the driver is already installed. Hotkeys, etc, won't be part of the core system: they interfere with apps such as The Last Word, and in any case the purpose of the SDX driver is to simulate as closely as possible a PBI compliant HDD, although any app which tries to access the HDD through the OS vector will still fail, of course (and that includes one or two exisiting SDX utilities). However, in day-to-day use with "serious" apps, this system has already proved fast, flexible and reliable, and has caused fewer incompatibilities and crashes with SDX apps than the alternative (removing and replacing OS routines in ROM).

The system will require new utility apps specific to SDX, since the MyIDE ROM vector table doesn't exist. Read and writes through the SDX SIOV are plenty fast, and RAW I/O will be handled by special SIO commands (or a similar relocatable system).

As for error-checking and checksums, we're not there yet. I've found the error correction in the IDE device to be 100% reliable (the beta version of the driver has zero errors at 60KB/s after two weeks of use), but I can easily build in more (optional) bullet-proofing later on.



_________________
The Last Word, MA65 Assembler, RIF SpartaDOS Utilities, XEDIT
Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Wed Sep 22, 2010 12:42 am 
User avatar

Joined: Sun Nov 02, 2003 8:15 am
Posts: 1376
Location: netherlands
Quote:
If you mean read sector 0 using 16-bit mode and switch to 8-bit PIO mode when dealing with 512bps partitions: that's what we're doing. This way the partition table is the same on any media.

Yes, that is what I ment.

Quote:
Looking at your last layout code, I see a few limitations (which is why we're using a 16 byte header and 16 byte entries at the moment).

No problem, we will see what the future brings.
I hoop Matthias MyIDE tool will support CHS, LBA (MyIDE $51) as well.

Quote:
Instead of hotkeys, the SDX driver will support a number of additional methods (probably via new SIO commands) such as reload partition table, raw I/O, etc. This is how the FDISK program will force the driver to activate the new partition table without a coldstart. There's nothing to prevent the creation of additional drivers and apps which interact with the base code via the SIO. Since MYIDE.SYS will be on the cart, it could have an "/R" switch which would simply reload the partition table when the driver is already installed. Hotkeys, etc, won't be part of the core system: they interfere with apps such as The Last Word, and in any case the purpose of the SDX driver is to simulate as closely as possible a PBI compliant HDD, although any app which tries to access the HDD through the OS vector will still fail, of course (and that includes one or two exisiting SDX utilities). However, in day-to-day use with "serious" apps, this system has already proved fast, flexible and reliable, and has caused fewer incompatibilities and crashes with SDX apps than the alternative (removing and replacing OS routines in ROM).

In MyIDE the hotkeys are not in the MyIDE-code.
They are in or patched to the KEYIRQ of the atari-OS.
But I'm sure any other way of supporing the user needs will do fine.

Quote:
As for error-checking and checksums, we're not there yet. I've found the error correction in the IDE device to be 100% reliable (the beta version of the driver has zero errors at 60KB/s after two weeks of use), but I can easily build in more (optional) bullet-proofing later on.

Hmmm, interesting, will you share the I/O-code with me? Perhaps I there is something I can use in my I/O-core.
(I only need the small read/write routine that talks directly with the drive)

Quote:
Some days ago you asked me how I use the IDENTIFY command in FDISK:

First I try to set PIO-8 and use the result as a true/false flag.
Then I use IDENTIFY and depending on the flag read 256/512 bytes and extract the low bytes of CHS.
When PIO-8 is true, I read the C hi-byte and store them.
When PIO-8 is false, I start with the low CHS bytes and add $FF as hi-byte for C.
Read the drive and check the error-register, continue reading [hi-byte]-1 until succes.

I'm not into Sparta, so my knowledge there is very limited.
The SDX is nice, I tested a version on my MyIDE/1mb flash.
I forced to use the SIO.sys, so the MyIDE-OS still works ;-)
But I will surely checkout the new version. I asume there will be a MyIDE/1mb flash download?

Question,
Will there be a conversion tool voor CHS <-> LBA-SDX?
So users won't loose or have to redo all there data?

Later,
Sijmen.


Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Wed Sep 22, 2010 5:43 am 

Joined: Mon Jan 23, 2006 10:49 am
Posts: 187
Location: Salzburg, Austria
Hi Sijmen!

mr-atari wrote:
I hoop Matthias MyIDE tool will support CHS, LBA (MyIDE $51) as well.

CHS is still supported in MyIDE tool, but I dropped the old LBA code. The are still a few things missing in the new LBA code (for example the moviewriter code) and I use $10 instead of $51 to identify the partition table. So the old $51 partition tables won't be recognized at all.

If you continue your work on the LBA OS it would be great if you could adapt it to use the same partition table layout that we designed (it's basically just an extension of the old $51 LBA format). There's still room for additional parameters (both in the header and in the partition table / image space entries) so it should be no problem to adapt it to your needs (for example I we could store the size of a partition slot and the total number of slots in the image space entry, in addition to just start and size).

Quote:
Will there be a conversion tool voor CHS <-> LBA-SDX?

Converting a CHS partition table into LBA would be no problem. Read the CHS table, multiply with C with H*S and you have the LBA values. Converting image space would also be possible, we'd just have to set the size of each slot to be H*S*number_of_cylinders_per_slot.

The other way round doesn't work - LBA partitions may begin at any arbitrary sector, not only cylinder boundaries. But users still could use myidetool on their PC, backup all data, then repartition it to CHS and write back all data.

so long,

Hias


Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Wed Sep 22, 2010 9:04 am 

Joined: Wed Mar 04, 2009 7:27 am
Posts: 98
Location: United Kingdom
Glad everything seems to be agreeable, Sijmen. I realize the hotkeys are part of the keyboard interrupt code (where else could they be? :) ), but I only meant to explain that the driver won't have such features in the core code. I may well implement <Shift+Reset> to re-read the partition table(s) however.

Is MyIDE supposed (or proven) to work with master/slave setups?

I had an idea for a foolproof way to read the LBA drive geometry in 16 bit mode: just use a binary chop, starting with LBA 0 through <arbitrarily high value>. I would imagine the upper LBA range would be found after a dozen or so disk reads.

I tend to agree with Matthias that the easiest way for users to transfer data between the old and new formats is via the PC (using MyIDE Tool).

Anyway - FDISK is nearly finished and I should have the revised driver working with the partition tables by Thursday evening.



_________________
The Last Word, MA65 Assembler, RIF SpartaDOS Utilities, XEDIT
Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Wed Sep 22, 2010 10:37 pm 

Joined: Sat Sep 13, 2003 12:21 am
Posts: 251
Hi,

Will you have seperate drivers for internal/external hardware? I tend to use my internal MyIDE mostly... since I don't have an intSDX.



_________________
MyIDE Tools
Offline
 Profile  
 
 Post subject: Re: Fast Reads - how?
PostPosted: Thu Sep 23, 2010 12:41 am 
User avatar

Joined: Sun Nov 02, 2003 8:15 am
Posts: 1376
Location: netherlands
Hi All,

Ok to all :-) No problem in using the new LBA layout in future.
I wasn't busy continuing the LBA-mode, still working on the rewritten OS from scratch......
That uses CHS-mode, still the smallest code for the OS. The 16k isn't much and almost full.
I managed to put the 2 vectors +code back.

Later,
Sijmen.


Offline
 Profile  
 
Display posts from previous:  Sort by  
 Page 3 of 4 [ 54 posts ]  Go to page Previous  1, 2, 3, 4  Next


Who is online

Users browsing this forum: No registered users and 15 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  

cron