fu@eecs.nwu.edu (Chih-Hua Fu) (11/12/88)
I am planning to put Metrabyte's PDMA-16, a high speed 16 bit parallel
digital interface card with DMA, into the APOLLO Domain 3000 AT-compatible bus.
Then, I will write a device driver to access a memory bank device through the
APOLLO's DMA channel. However, I have some questions:
1. Is APOLLO's AT-compatible bus *REALLY* compatible? Can any interface working
in the IBM PC-AT be directly inserted into the APOLLO's AT-compatible bus
without any problems ?
2. In the AEGIS operating system, the GPI/O library supports PBU-$DMA_START
and PBU_$DMA_STOP for DMA. However, the buffer must be page aligned.
One option is to have the user pass page aligned buffers. This can be done
relatively easily by the user and avoids the overhead of one copy (since
the DMA can be done directly to the user supplied buffer).
However, for ease of use, I would like the ability to have the user supply
a non page aligned buffer and my driver to handle it by grabbing a page
aligned buffer from somewhere in the kernel and do the DMA on this buffer.
Then the kernel buffer can be copied into the user supplied buffer.
In most BSD UNIX systems you can usually get a page aligned buffer in
the kernel by getting an mbuf or raiding some other page aligned memory
buffer pool. How do you do it in AEGIS ? How can my driver get a
page aligned buffer that can be returned to the kernel pool once the copy
into user space is done.
3. I do not have the manual about the APOLLO's AT-compatible bus protocol.
Some documents say "Refer to the Domain 3000 Technical reference (008778)".
Does that manual describe the bus in detail ?
4. This is my first experience with writing a device driver. If you have any
hints about writing device drivers for AEGIS or things that I should be
watching out for, please drop me a note.
Your help is very much appreciated. Thank you.
Chih-Hua Fu,
Electrical Engineering and Computer Science,
Northwestern University.
Internet: fu@eecs.nwu.edu
Uucp: {att, chinet, oddjob}!nucsrl!fukrowitz@RICHTER.MIT.EDU (David Krowitz) (11/15/88)
Some notes from my own struggles with DMA on the DN3000 ... 1) the buffer does not necessarily have to be page alligned. The actual working requirement is that the interrupt routine (and the actuall DMA operation), must not cause a page fault. As long as the DMA does not cross a page boundry and the page is wired into physical memory, you are OK. You can perform DMA to an arbitrary buffer by simply breaking up the operation into chunks which each fit within a single page. 2) Some of the routines which wire pages in memory may be the cause of the page allignment requirement. You can allign a buffer by putting it in a seperate data section (in Pascal, you would use the extension to the VAR statement to specify a new section) and then using the -ALIGN PAGE option with the binder. You can also simply allocate a buffer which is 1024 bytes larger than what you really need, and then calculate where the first page break would be (ie. if PTR points to you buffer, then (again in Pascal) the page aligned buffer can be found with NEW_PTR := NEW_POINTER_TYPE(((INTEGER32(PTR) DIV 1024)+1)*1024), which gives you the first page break after the start of the buffer -- note that all of the arithmetic is 32-bit integer operations) 3) At least some of the earlier DN3000's have a problem with their DMA hardware. Running two DMA devices at the same time will cause them to interfere with each other's I/O. This is why the hard disk and the cartridge tape drives don't use DMA on the DN3000. The floppy disk does use DMA, and I can not write floppies while my printer interface is sending data to the printer. Later model DN3000's are supposed to have this fixed. 4) You might be able to use the pub_$wire_buffer_special call to get your buffer wired into physical memory with physcially consecutive pages. This would allow you to do DMA to the entire buffer without having to worry about crossing a page boundry. I haven't tested this yet, though. -- David Krowitz krowitz@richter.mit.edu (18.83.0.109) krowitz%richter@eddie.mit.edu krowitz%richter@athena.mit.edu krowitz%richter.mit.edu@mitvma.bitnet (in order of decreasing preference)