Here you will find information on the current state of the development
effort. It will be updated periodically and will give you a
snapshot of the current state of Linux on the PA-RISC. For more information
on the current state of the PA-RISC project see Status Reports and
check out our list of Contributors.
Current Report
2000 Report Archive
1999 Report Archive
2000 REPORT ARCHIVE
February 11, 2000
State of the Port
It's been a while since we put one of these out, so here goes:
What's happened recently?
- Grant has contributed support for the C3000/J5000 and similar machines
which have an I/O SAPIC.
- A cast of thousands hacked on the Tulip (onboard DINO/IOSAPIC based
machines) driver. Eventually, Thomas persuaded it to work.
- Sammy & Helge made LASI ethernet work.
- Thomas massaged some C IP checksum code until it worked. NFS now works!
- Our kernel tree is currently based on Linux 2.3.42. Michael
Ang shepherded us through this merge with contributions from
(alphabetically), Paul, Thomas, Grant, Philipp and Matthew.
- Naturally, Linus put out 2.3.43 within a few hours of the merge
being completed.
- Grant has also got some signs of life from the Symbios 896
- Martin fixed a console problem that made sash crash when using the
serial console.
- John Marvin has been working on Linux syscalls.
- Sammy has a 32-bit ELF kernel booted [but hasn't committed the code
yet, to give us enough time to get an ELF toolchain to everyone].
So what does it currently work on?
- C3000/J5000 (no serial input yet)
- A180
- C360
- 712
These are the machines developers are mainly using right now.
Other machines simply aren't being tested and may work anyway.
The future
- We'll try to stay more current with Linus' codebase. On the other hand,
we don't want to be merging all the time. The compromise we've decided to
try is to merge approximately once a month, depending on exactly what's
been changed. 2.3.43 breaks half the network drivers, so we don't want
to merge yet. On the other hand, it does have the new `coherent DMA'
interface, which we will want to take advantage of.
- We need to start thinking about moving the kernel to 64-bit ELF.
There are some places where we've assumed sizeof(long) == sizeof(int),
and some of the assembler needs changing.
- We need to start working on making the system self-hosting. That means
getting a C compiler running, porting glibc, etc.
Page last modified on 2000-02-29
|