unsatisfied symbols

Hi,

it is not possible to compile the latest CVS-Kernels. I tried various configurations, but I still get unsatisfied symbol __delay. Is anybody out there, who compiles the actual sources succesfully ? make menuconfig seems to be broken too (since 4 weeks)

Ingo

Ingo Matthaes
HP-UX Net und Linux Support
HP Support Center Ratingen

Ingo:
When you say make menuconfig seems to be broken, what exactly doesn't work? As I understand the current method for compiling kernels takes place using HP-UX with gcc. Does it also include ncurses or is the group attempting to use the native HP-UX curses? If the later then understand that HP-UX has at last count at least three or four different curses libraries. Some of them are badly broken while others are only semi broken. My advice is to get ncurses and use it instead of the various native curses implementations.

If the make menuconfig problem isn't a problem with the curses implementation on HP-UX I could be barking up the wrong tree here.

Steven A. DuChene   Linux Fan!  http://www.mindspring.com/~sduchene

Outside of a dog, a book is a man's best friend. Inside a dog it's
too dark to read.
                -- Groucho Marx

our config.in was broken and Ingo sent a patch that corrected the problem.

Am Tue, 03 Aug 1999 schrieb Ingo Matthaes:

Hi Ingo.

The unsatisfied symbol __delay was caused by the inclusion of pdc_cons.c by the bootloader. I fixed it some minutes ago in the cvs.
The actual kernel from cvs compiles, links and boots (!!) now cleanly (I just tried it), if you add the line
extern pgd_t swapper_pg_dir[1024];
to the file pgtable.h. I didn�t wanted to fix that, since I think, that Philipp is actually working on that code....

NB: If you want faster screen-output, then remove the delay() in pdc_cons.c.

Ciao,
Helge.

O.k. it works fine on my 715/33. I will try it on some 712 and, if available, on a a-class

Why should I remove that classic feeling :-)

Ingo

Ingo Matthaes
HP-UX Net und Linux Support
HP Support Center Ratingen

Hi!

I just noticed that Linus is doing feature freeze for 2.4 in few weeks. That raised couple of questions.

Do we have any estimate on parisc/linux timetables? When is the pa-risc Linux going to be "released"? Do we target it to 2.5 or 2.6?

We are now based on 2.2.1. How big effort would be required to merge pa-risc linux (2.2.1) to 2.3/2.4?

- Goodi
"The linuX Files -- The Source is Out There." ø,¸¸,ø¤º°°º¤ø,¸¸,ø¤º°°º¤ø,¸¸,ø¤º°°º¤ø,¸¸,ø¤º°°º¤ø¤º°`°º¤ø,¸¸,ø¤º°

I believe we're aiming at having a 2.2 version ready for the end of the year -- approximately the same time as Linus expects to have a 2.4 release out. Updating to 2.4 shouldn't be a big problem, and we don't have to be ready for the release of 2.4.0 -- Linus merged alpha, sparc, ppc and mips trees at several stages after 2.2.0 was released. (Indeed, there is a big mips merge in the 2.2.11 prepatches.)

Don't worry about 2.4 yet :-)

Matthew Wilcox <[email protected]>
"Windows and MacOS are products, contrived by engineers in the service of
specific companies. Unix, by contrast, is not so much a product as it is a
painstakingly compiled oral history of the hacker subculture." - N Stephenson

I've talked to Linus about merging the changes into his mainstream kernel tree, and he's agreed, but only once our kernel actually does something. I agree with this.

It's very difficult to suggest a timeline to a project that's done only on volunteer time, so I don't think anyone's ready to commit to a schedule.

My feeling is we'll address the kernel merge once we have a kernel to merge. A LOT of the architecture ports had their own seperate tree for many years. MIPS, for instance, has only just merged.

- alex

Alex deVries
Vice President of Engineering
The Puffin Group

we do "something" right now. I think we should wait with the merge until we are really willing to start mainstream testing of the kernel, with a more-or-less running userspace and all.

That is not exactly true. My time is paid for by SuSE.

Philipp Rumpf