Messages in this thread | | | From | "Khimenko Victor" <> | Date | Sun, 24 Jan 1999 19:50:20 +0300 (MSK) | Subject | Re: VM performance |
| |
In <Pine.LNX.4.05.9901241610350.1171-100000@ns.snowman.net> Stephen Frost (sfrost@ns.snowman.net) wrote: > On Mon, 25 Jan 1999, Ed Lang wrote:
>> On Sun, Jan 24, 1999 at 02:38:39PM +0000, Alan Cox wrote: >> > > > "Is this something that normal people would ever really care deeply >> > > > about?" >> > > >> > > Probably, if there are people other than me that are still using systems >> > > with 16mb of physical RAM. There probably are some. Maybe. >> >> I, for one, have a box with 16M of RAM.
> I have a number with 8M, a couple with 16M, and one with 4M.
>> > On 32Mb 2.2.0pre9 works ok but not well [1] - I can build a kernel without >> > much swapping while doing other stuff, on 2.1.132 I can build two >> > kernels in parallel with the same kind of feel while 2.2.0x chokes up >> > horribly at this point. >> > >> > [1] It still beats 2.0.x I should point out 8) >> > >> > >> > On 8Mb 2.2.0pre6, pre8, pre9 are basically unusable while 2.1.132 was fine >> > >> >> Ye gods. If that's the case, 2.2.0 sure ain't going on my 386 with 6M of RAM >> (which is all I have at the moment, but that's a long story).
> I also find this very distressing as two of my routers have 8M of ram > while another has 4M, and both perform fairly well while doing any number of > things. My secondary name server and secondary mail server also has 8M of ram.
> If this can't be changed for 2.2.0 (Anyone know what's causing it?), > perhaps in 2.2.1?
Neither in 2.2.0 nor in 2.2.1. Use 2.0.37 :-))
> Or maybe some option ("low mem?") that will turn off memory hogs and go back > to prior slower but less memory eating routines? If this is possible?
No. Problems with 8-16Mb could be resolved (and I hope will be resolved -- may be some 2.2.x will work with 8Mb even better then 2.0.x -- wild dream, I know) but 4Mb is "no way". The whole design of a lot of subsystems was changed and thus there are no chances to get "normal" work on 4Mb systems.
- 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/
| |