Messages in this thread | | | From | (david parsons) | Subject | Re: Memory being incorrected sized at boot-up | Date | 15 Jul 1997 02:29:17 -0700 |
| |
In article <linux.kernel.19970715103725.60518@atrey.karlin.mff.cuni.cz>, Martin Mares <mj@atrey.karlin.mff.cuni.cz> wrote: >Hi, > >> > OK, I incorrectly assumed we call the old function first and continue >> >with the new one only if it reports 64M... >> >> That's what I did in 2.0.30, unfortuantely BIOS on Compaq's only >> reports a maximum of 16MB wiht the old BIOS call. > > Maybe call the new one for >=16MB? This will reduce the chance of crash >on a mis-behaved BIOS to minimum. If such broken BIOS'es exist, they are >probably on the 386's and most 386 machines have probably <16MB RAM.
Might it be a good idea to see if there are any old bioses out there that DO fail before embarking on some byzantine patch to deal with them? I know that the patch I put into the 2.1 kernels works on some positively ancient 386sx AMI and Phoenix bioses, as well as the usual bizarre outcroppings you find on old Compaq and IBM motherboards.
____ david parsons \bi/ Utterly confused by this discussion. \/
| |