Re: New new enhanced memory detection patch; testers wanted

Guest section DW (dwguest@win.tue.nl)
Sat, 20 Feb 1999 16:31:52 +0100 (MET)


From: o.r.c@p.e.l.l.p.o.r.t.l.a.n.d.o.r.u.s (david parsons)

____
david parsons \bi/ I sure hope nobody is using 1e4->1eb in the transfer
\/ block.

I am today, but don't let that stop you.

#define PARAM empty_zero_page
#define EXT_MEM_K (*(unsigned short *) (PARAM+2))
+ #ifndef STANDARD_MEMORY_BIOS_CALL
+ #define E801_MEM_K (*(unsigned long *) (PARAM+0x1e0))
+ #define E820_MEM (*(unsigned long *) (PARAM+0x1e4))
+ #endif
#ifdef CONFIG_APM
#define APM_BIOS_INFO (*(struct apm_bios_info *) (PARAM+64))
#endif

Maybe it is bad to conditionally take some space.
That is, I would omit these #ifdef's here, even if nobody uses
the defined constants.

I have been playing with IBM-Phoenix BIOS extensions, that give
82 bytes of information for each disk. For a modest limit like
16 disks, one already needs more space than is available.
And no doubt there will come many other devices that we'll want
to interrogate at boot time.

This points at the necessity to precisely define the memory use
at boot time, and to set aside a fragment of memory substantially
larger than what we have now.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/