Re: max phy mem known working with FreeBSD 4.x
Nate Lawson wrote: > On Tue, 5 Nov 2002, Terry Lambert wrote: > > As far as PCI goes (or anything they publish, for that matter), the > > MindShare books are very, very good. But for the particular question > > of how much physical address space is eaten, you really have to go to > > the chipset spec. sheets to get the right answer these days. 8-(. > > The PCI book is good. The others less so. Bottom of the barrel is "plug > and play arch" -- terrible. Amazon's reviews are pretty close to reality > IMO. I actually really like the "Protected Mode Software Architecture", the "EISA", and the "ISA" books. The "Cardbus" and "Firewire" titles are also useful. Sorry to contradict you, but the "Plug and Play" is sufficient to let someone write "PnP OS" code, by manually scanning devices in the OS, instead of trusting the BIOS, which is good enough for me (I had an old box that had a built in bus mouse on IRQ 12 that wasn't in the BIOS; a PnP OS got this right; non-PnP OS's always failed to do the right thing, and reassigned IRQ 12 to a conflicting device (the second disk controller, on that particular box). -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
On Tue, 5 Nov 2002, Terry Lambert wrote: > As far as PCI goes (or anything they publish, for that matter), the > MindShare books are very, very good. But for the particular question > of how much physical address space is eaten, you really have to go to > the chipset spec. sheets to get the right answer these days. 8-(. The PCI book is good. The others less so. Bottom of the barrel is "plug and play arch" -- terrible. Amazon's reviews are pretty close to reality IMO. -Nate To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
"M. Warner Losh" wrote: > In message: <[EMAIL PROTECTED]> > Terry Lambert <[EMAIL PROTECTED]> writes: > : The CDROM and a full set of the specifications will run you US$1500, > : plus US$15 shipping in the US, or US$40 shipping, international. > : > : The price goes from $1500 to $75, if you can order through a company > : which is already a member of the PCI SIG. > > The cdrom was $50 when I ordered it with only the PCI 2.2 spec on > paper. Now that 3.0 is out, looks like they have jacked up the price > again. Looks like it costs a lot more now. The mindshare PCI book > has a good explaination. I paid $150 in the 2.x era, but I got the full boxed set, with the PCI-PCI bridge specifications, etc.. I think the thing that's bloating it now is that they have the PCI-X and some other stuff that no one will really use much. I've already spent a little over 2 feet of shelf space, just in Microsoft MSDN CDROM's and Windows SDK's and DDK's (~300 CDROMs). I expect the printed versions of the PCI specifications would probably come in around a foot now, if they are hard bound, when before they were around 4 inches wide. I've noticed that most people who sell standards sell them cheap to start, and then jump the price up after they get adopted as standards, thinking they have a captive market. One I can recommend is the ECMA standards CDROM; the ECMA people seem to actually be interested in people complying with their standards, so there was no cost for the CDROM, as long as you have a business address to get it sent to... As far as PCI goes (or anything they publish, for that matter), the MindShare books are very, very good. But for the particular question of how much physical address space is eaten, you really have to go to the chipset spec. sheets to get the right answer these days. 8-(. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
In message: <[EMAIL PROTECTED]> Terry Lambert <[EMAIL PROTECTED]> writes: : The CDROM and a full set of the specifications will run you US$1500, : plus US$15 shipping in the US, or US$40 shipping, international. : : The price goes from $1500 to $75, if you can order through a company : which is already a member of the PCI SIG. The cdrom was $50 when I ordered it with only the PCI 2.2 spec on paper. Now that 3.0 is out, looks like they have jacked up the price again. Looks like it costs a lot more now. The mindshare PCI book has a good explaination. Warner To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
Dinesh Nair wrote: > On Tue, 5 Nov 2002, Richard Sharpe wrote: > > Well, we have 4GB in a 4.6.2 system, and I think that we ran 4.3 on those > > systems for a while. > > > > However, you lose anywhere between 128M and 512M because of the PCI > > address space. > > what PCI address space ? could someone explain about this loss ? > curiousity beckons. :) Short answer Memory on devices on the PCI bus is mappes into the physical memory address space, and so "covers up" RAM in the window in which it is mapped. Good chipsets will only map in memory that actually exists for devices which are actually installed. Bad chipsets will reserve the entire window, and therefore take a much larger chunk out. Long answer --- See http://www.pcisig.com/specifications/ordering_information The CDROM and a full set of the specifications will run you US$1500, plus US$15 shipping in the US, or US$40 shipping, international. The price goes from $1500 to $75, if you can order through a company which is already a member of the PCI SIG. Your company can join the PCI SIG for a US$3000 per year fee, if it is not already a member. See http://www.pcisig.com/membership/join_pci_sig/new_member_forms/ for membership information. FWIW, I got my copy when I was an employeed of a member company; it's the way to go, if you can do it. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
On Tue, 5 Nov 2002, Richard Sharpe wrote: > Well, we have 4GB in a 4.6.2 system, and I think that we ran 4.3 on those > systems for a while. > > However, you lose anywhere between 128M and 512M because of the PCI > address space. what PCI address space ? could someone explain about this loss ? curiousity beckons. :) Regards, /\_/\ "All dogs go to heaven." [EMAIL PROTECTED](0 0)http://www.alphaque.com/ +==oOO--(_)--OOo==+ | for a in past present future; do| | for b in clients employers associates relatives neighbours pets; do | | echo "The opinions here in no way reflect the opinions of my $a $b." | | done; done | +=+ To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
PS: Richard Sharpe wrote: > Well, the P4 does have an address extension that allows addressing of up > to 64GB, EPA or something like that. It has segments, too. We don't use those, either. 8-). -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
Richard Sharpe wrote: > Well, the P4 does have an address extension that allows addressing of up > to 64GB, EPA or something like that. This is useless to discuss in the context of 4.3, or even 4.x, unless Paul Saab is doing his work in 4.7, and it makes it into 4.8. Given that we are pretty much forced to do all new work 0n -current in order to get committed (or everyone would be working on 4.x instead), it'll end up a forward port with no back-port (expect it after the 5.x release, if Yahoo gives it back to FreeBSD). > However, it requires some work. Would be nice in large Samba servers, > though, to be able to cache enormous amounts of file data :-) It would be useless for this purpose. Specifically, you would need to guarantee a locality of reference for clients that you could not guarantee, for this to be useful at all. Mostly, you would be swapping the PSE window around all the time, shooting down TLB's and cache lines, and you would burn up any benefits of having the data in main RAM instead of in RAM on the disk controller or the disk itself, very quickly. In addition, you would not be able to use "sendfile" or similar interfaces, and you would not be able to use scatter/gather DMA, unless you knew for certain that the card doing the DMA had 64 bit access to physical RAM. Otherwise, you would need to copy all data to be DMA'ed to/from any physical RAM above 4G to bounce buffers below 4G. The extra copying might as well be from the controller RAM or the RAM on the disk itself, than from/to RAM above 4G. The 2G limit on Alpha is basically the same issue, where the data needs to be bounced through lower memory in order to be accessible to the hardware DMA engines on the other side of the PCI bus (hence the need to convert the drivers to "bus space" to be able to use more than 2G in an Alpha box). Basically, the extra RAM makes assumptions about code that make us all realize that hardware designers don't really ask software designers why they ask for things like the ability to access more than 4G of RAM. 8-). If you want to discuss this further, please read the archives of the last 3 times the issue was discusees, on the -current and -arch lists. No matter how you look at it, though, unless you pay some schmuck to do the work, you're never going to see this in 4.3 anyway, so you might as well move the discussion over to the -current list. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
Julian Elischer wrote: > The hardware changes to do > 32 bit physical addresses > include a redefinition of how page tables and page directories are layed > out and to be able to use it we'd have to define > and turn on code for that differnt mode. iIt has not yet been written. > It wouldn't be hard to make a system that ONLY ran in that mode > but one that can run on new systems withthat mode and also run on old > 486 machines is much harder. It's easier to do it with PSE-36 than PAE, actually. The double indirect is also easy to setup, and then not use, if it comes to that, and 4K vs. 8K allocated in assembly is pretty much nothing. The big issue is the 2M vs. 4M "jumbo pages": you have to expect that, and FreeBSD, if PSE is present, will try to set up a 4M page for the kernel. What's really needed is a "remap4kto4m()" function that gets used for that; there's a trick you have to pull with CR3, CR0, and CR4 to make it not break, though, because there's a CPU bug, and it leaves you with a chicken-and-egg problem, otherwise. This would work for 2M pages, as well (just use two of them). If you understand the code, you should be able to write it in about two weeks (including the CPU bug workaround), once you have your head around things. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
On Tue, 5 Nov 2002, Terry Lambert wrote: > Matt wrote: > > > > Anyone knows the max physical mem that can be used with FreeBSD4.3? > [ ... ] > > any chance going more than 4G? > > Sure, if you want to install it to warm things up. > > No, if you want to access it; access is limited to 4G, because > that's 32 bits of address space, and your machine is a 32 bit > machine. Well, the P4 does have an address extension that allows addressing of up to 64GB, EPA or something like that. However, it requires some work. Would be nice in large Samba servers, though, to be able to cache enormous amounts of file data :-) Regards - Richard Sharpe, [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], http://www.richardsharpe.com To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
Matt wrote: > > > Anyone knows the max physical mem that can be used with FreeBSD4.3? [ ... ] > any chance going more than 4G? Sure, if you want to install it to warm things up. No, if you want to access it; access is limited to 4G, because that's 32 bits of address space, and your machine is a 32 bit machine. Unless you run IA-64 or SPARC-64, or run Alpha, and fix several of the drivers to use "bus space" properly. Of course, for these platforms, you wouldn't be running 4.3. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
On Mon, 4 Nov 2002, Matt wrote: > any chance going more than 4G? Not yet. (on x86) The hardware changes to do > 32 bit physical addresses include a redefinition of how page tables and page directories are layed out and to be able to use it we'd have to define and turn on code for that differnt mode. iIt has not yet been written. It wouldn't be hard to make a system that ONLY ran in that mode but one that can run on new systems withthat mode and also run on old 486 machines is much harder. Julian To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
Matt wrote: > Anyone knows the max physical mem that can be used with FreeBSD4.3? 2G on Alpha. 4G on Intel, if you tune your kernel and modify your KVA size to 3G. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
any chance going more than 4G? Best Regards Matt http://www.xgforce.com/product.html The Next Generation Server Clustering and Clustered Enterprise SQL database, and Firewall/VPN Solutions. - Original Message - From: "Richard Sharpe" <[EMAIL PROTECTED]> To: "Matt" <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Monday, November 04, 2002 10:50 PM Subject: Re: max phy mem known working with FreeBSD 4.x > On Mon, 4 Nov 2002, Matt wrote: > > > Anyone knows the max physical mem that can be used with FreeBSD4.3? > > Well, we have 4GB in a 4.6.2 system, and I think that we ran 4.3 on those > systems for a while. > > However, you lose anywhere between 128M and 512M because of the PCI > address space. > > Regards > - > Richard Sharpe, [EMAIL PROTECTED], [EMAIL PROTECTED], > [EMAIL PROTECTED], http://www.richardsharpe.com > To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: max phy mem known working with FreeBSD 4.x
On Mon, 4 Nov 2002, Matt wrote: > Anyone knows the max physical mem that can be used with FreeBSD4.3? Well, we have 4GB in a 4.6.2 system, and I think that we ran 4.3 on those systems for a while. However, you lose anywhere between 128M and 512M because of the PCI address space. Regards - Richard Sharpe, [EMAIL PROTECTED], [EMAIL PROTECTED], [EMAIL PROTECTED], http://www.richardsharpe.com To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
max phy mem known working with FreeBSD 4.x
Anyone knows the max physical mem that can be used with FreeBSD4.3? Thanks in advance To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message