XiP Filesystem Primps For Linux 2.6.28 33
nerdyH writes "The Linux-Embedded discussion list has been abuzz the last two days over a flash filesystem designed to support binary code execution (sometimes called XIP, or "execute-in-place"). When combined with forthcoming "Phase Change Memory" products from Intel/ST JV Numonyx, the "Advanced XiP Filesystem" (AXFS) could radically change Linux systems of all kinds, replacing Flash, hard drives, and even DRAM with a single chunk of low-cost, non-volatile memory that can both store files and serve as a platform for program execution."
Huh? (Score:1)
Didn't read the article...
So this acts as both memory and "hard drive?"
If true, would data read times be affected at all? I mean, if the operating system is simultaneously accessing a file, and using something in memory, would there be a performance hit?
Re: (Score:1)
Re:Huh? (Score:5, Informative)
XiP has been supported in other filesystems, this is more like a small improvement - nothing to get excited about. Moreover, you won't see it in use in any desktops - this filesystem is read-only; more like an archive format understood by the kernel.
More useful discussion can be found at kerneltrap [kerneltrap.org].
Re: (Score:3, Informative)
I think that the concept is that you have a processor with a boat-load of some type of RAM-look-alike (presumably non-volatile). This RAM is used as a RAM-drive (well, if the RAM is non-volatile, then you loose no data). Imagine a linux-based PDA that does not have the usual RAM/FLASH combination, but instead has a few gig of RAM that keeps its memory when powered off. So, you use the same memory for both "main memory" applications (loading programs, stack, heap, etc.) and also holds a file system.
So, yo
Re: (Score:3, Funny)
I like nice tight data myself.
Re: (Score:2)
Imagine a linux-based PDA that does not have the usual RAM/FLASH combination, but instead has a few gig of RAM that keeps its memory when powered off.
You mean, like Palm was doing for most of it early years until recently (5.4), because some patent troll had control on PDA-using-flash and a non volatile file system ?
Yup this has already been done. The only differences are :
- The old Palm OS used read/write storage, whereas according to TFA this file system is read-only.
- The old Palm devices used plain simple DRAM, that was kept powered when the rest of the device was shut down, simply because then, there wasn't such a thing as non-volatile RAM in produc
Re: (Score:2)
Re: (Score:3, Informative)
The idea is for embedded systems that don't have a hard drive. Right now, at least with a normal file system, you have to take the data from flash and copy it into RAM. Even if the flash is mapped into memory, you still have to do this.
Execute in Place means letting the processor read the code directly from the flash (when available) so you don't have to make the (rather pointless) copy in RAM. It obviously wouldn't work in some circumstances (self-modifying code being the obvious one) but it would save em
Re: (Score:1)
Re: (Score:3, Informative)
AXFS is read-only. The usage case tends to go as follows:
- Build a compressed filesystem
- Run a test case, interacting with the filesystem to generate page hits on your binaries.
- Take the profile the filesystem generates and rebuild the filesystem.
Precisely. CramFS works by setting the
Re: (Score:2)
aha (Score:1)
my new choice for bootloader
So we're going back to the roots (Score:2)
This upgrade brings the computer closer back to the Turing Machine :)
For people unfamiliar to XIP: (Score:3, Informative)
Re: (Score:2)
The thing I'm confused about is - how is this different from MMIO? Or is the new thing that we can do MMIO for programs when they're on flash?
Must be memory mapped (Score:2, Informative)
Re: (Score:2)
Classic Palm OS (Score:4, Informative)
It's block rather than record oriented, but this seems similar to how classic Palm OS worked. In the Handspring Visor, which had directly addressable flash memory, this was used to execute code directly out of flash.
Primps? (Score:1)
Read headline slowly (Score:2)
"XiP" looks entirely too much like "XP". They should have gone with something like "XNP".
Re: (Score:2)
Well, XIP as a TLA predates XP by a *long* time.
XIP, however, is not the same as XiP, although from across the room both probably look like XP.
Interesting question.... (Score:2, Interesting)
6mb? what the heck is a milli-bit? (Score:1, Flamebait)
6mb? What the heck is a milli-bit?
Perhaps you meant 6Mb, so 0.000715255737 megabytes?
Or just maybe 6MB or 50 million bits.
Ooo, did I just invent a new type of /. nazism - the SI-Nazi??
Re: (Score:1)
To quote an excellent post I saw a few days ago;
people who insist on seeing only the "literal" meaning (a nebulous concept to anyone who has studied language from any perspective) are basically illiterate.
Re: (Score:2)
Yes, I'm illiterate! Netcraft confirms it.
If the parent post had just used email-case (aka lowercase) throughout I wouldn't have bothered. But it seems perverse to go the trouble of using capitals at the start of your sentences, where they are largely redundant, and then use lowercase where it changes the meaning entirely.
Oh, and could you read the above back to me because I can't ;0)>
---
Yes, I'm a fan of run-on sentences.
Re: (Score:3, Insightful)
That idea doesn't really have anything to do with XiP support in the file system. You could run the kernel directly from flash as well. And since that is probably simpler, I'd be surprised if that isn't already done. As for running everything from the CPU cache, that is not controlled by the kernel but rather by the CPU itself. And it will try to use the cache for the mostly used parts of memory regardless of whethe
nothing to see here (Score:2, Interesting)
unless you're an embedded dev, nothing to see, move along.
the slowest part of the intel cpu cycle is decode - but before your cpu processes streams of op codes for decode you have to load the executable image into ram and then run the dynamic linker. the dynamic linker needs to parse every the segment to be worthwhile, otherwise you'll hit the linker each time you load a segment that's slightly interesting (weaker the symbols/IAT, the more likely) only to find out you didn't need to causing jitter.
its one o
Re: (Score:2)
This comment is not very well wrote. (Score:3, Interesting)