Ключевые слова:linux, vm, memory, proccess, (найти похожие документы)
_ RU.OS.CMP (2:5077/15.22) _________________________________________ RU.OS.CMP _
From : Vadim Kolontsov 2:5020/400 17 Jan 99 22:28:02
Subj : Linux VM
________________________________________________________________________________
From: nospam.vadim@tversu.ru (Vadim Kolontsov)
Reply-To: nospam.vadim@tversu.ru
Привет,
для более аргументированного сравнения... приношу свои извинения, если
этот текст появлялся здесь неоднократно. Оригинал - на сайте автора,
http://apollo.backplane.com. Сразу за документом следует комментарий
к нему D.Dyson'а.
В предыдущем письме -- FreeBSD VM system overview.
V.
-------------------------------------------------------------------------------
Date: Wed, 13 Jan 1999 23:20:22 -0800 (PST)
From: Matthew Dillon <dillon@apollo.backplane.com>
Message-Id: <199901140720.XAA22609@apollo.backplane.com>
Cc: hackers@FreeBSD.ORG
Subject: Review and report of linux kernel VM
General Overview
I've been looking at the linux kernel VM - mainly just to see what they've
changed since I last looked at it. It's quite interesting... not bad at
all though it is definitely a bit more memory-resource-intensive then
FreeBSD's. However, it needs a *lot* of work when it comes to freeing
up pages.
I apologize in advance for any mistakes I've made!
Basically, the linux kernel uses persistent hardware-level page tables
in a mostly platform-independant fashion. The function of the persistent
page tables is roughly equivalent to the function of FreeBSD's vm_object's.
That is, the page tables are used to manage sharing and copy-on-write
functions for VM objects.
For example, when a process fork()'s, pages are duplicated literally by
copying pte's. Writeable MAP_PRIVATE pages are write-protected and marked
for copy-on-write. A global resident-page array is used to keep track
of shared reference counts.
Swapped-out pages are also represented by pte's and also marked for
copy-on-write as appropriate. The swap block is stored in the PFN
area of the pte (as far as I can tell). The swap system keeps a separate
shared reference count to manage swap usage. The overhead is around
3 bytes per swap page (whether it is in use or not), and another pte-sized
(int usually) field when storing the swap block in the pagetable.
Linux cannot swap out its page tables, mainly due to the direct use of
the page tables in handling VM object sharing.
In general terms, linux's VM system is much cleaner then FreeBSD's... and
I mean a *whole lot* cleaner, but at the cost of eating some extra memory.
It isn't a whole lot of extra memory - maybe a meg or two for a typical
system managing a lot of processes, and much less for typical 'small'
systems. They are able to completely avoid the vm_object stacking
(and related complexity) that we do, and they are able to completely
avoid most of the pmap complexity in FreeBSD as well.
Linux appears to implement a unified buffer cache. It's pretty
straight forward except the object relationship is stored in
the memory-map management structures in each process rather then
in a vm_object type of structure.
Linux appears to map all of physical memory into KVM. This avoids
FreeBSD's (struct buf) complexity at the cost of not being able to
deal with huge-memory configurations. I'm not 100% sure of this, but
its my read of the code until someone corrects me.
Problems
Swap allocation is terrible. Linux uses a linear array which it scans
looking for a free swap block. It does a relatively simple swap
cluster cache, but eats the full linear scan if that fails which can be
terribly nasty. The swap clustering algorithm is a piece of crap,
too -- once swap becomes fragmented, the linux swapper falls on its face.
It does read-ahead based on the swapblk which wouldn't be bad if it
clustered writes by object or didn't have a fragmentation problem.
As it stands, their read clustering is useless. Swap deallocation is
fast since they are using a simple reference count array.
File read-ahead is half-hazard at best.
The paging queues ( determing the age of the page and whether to
free or clean it) need to be written... the algorithms being used
are terrible.
* For the nominal page scan, it is using a one-hand clock algorithm.
All I can say is: Oh my god! Are they nuts? That was abandoned
a decade ago. The priority mechanism they've implemented is nearly
useless.
* To locate pages to swap out, it takes a pass through the task list.
Ostensibly it locates the task with the largest RSS to then try to
swap pages out from rather then select pages that are not in use.
From my read of the code, it also botches this badly.
Linux does not appear to do any page coloring whatsoever, but it would
not be hard to add it in.
Linux cannot swap-out its page tables or page directories. Thus, idle
tasks can eat a significant amount of memory. This isn't a big deal for
most systems ( small systems: no problem. Big systems: probably have lots
of memory anyway ). But, mmap()'d files can create a significant burden
if you have a lot of forked processes ( news, sendmail, web server,
etc...). Not only does Linux have to scan the page tables for all the
processes mapping the file, whether or not they are actively using the
page being checked for, but Linux's swapout algorithm scans page tables
and, effectively, makes redundant scans of shared objects.
What FreeBSD can learn
Well, the main thing is that the Linux VM system is very, very clean
compared to the FreeBSD implementation. Cleaning up FreeBSD's VM system
complexity is what I've been concentrating on and will continue to
concentrate on. However, part of the reason that FreeBSD's VM system
is more complex is because it does not use the page tables to store
reference information. Instead, it uses the vm_object and pmap modules.
I actually like this feature of FreeBSD. A lot.
The biggest thing we need to do to clean up our VM system is, basically,
to completely rewrite the struct buf filesystem buffering mechanism to
make it much, much less complex - basically it should only be used as
placeholders for read and write ops and not used to cache block number
mappings between the files and the VM system, nor should it be used to
map pages into KVM. Separating out these three mechanisms into three
different subsystems would simplify the code enormously, I think. For
example, we could implement a simple vm_object KVM mapping mechanism
using FreeBSD's existing vm_object stacking model to map portions of a
vm_object (aka filesystem partition) into KVM.
Linux demarks interrupts from supervisor code much better then we do.
If we move some of the more sophisticated operational capabilities
out of our interrupt subsystem, we could get rid of most of the spl*()
junk we currently have to do. This is a real sore spot in current
FreeBSD code. Interrupts are just too complex. I'd also get rid of
FreeBSD's intermediate 'software interrupt' layer, which is able to
do even more complex things then hard interrupt code. The latency
considerations just don't make any sense verses running pending software
interrupts synchronously in tsleep(), prior to actually sleeping. We
need to do this anyway ( or move softints to kernel threads ) to be able
to take advantage of SMP mechanisms. The *only* thing our interrupts
should be allowed to do is finish I/O on a page or use zalloc().
-Matt
Matthew Dillon
<dillon@backplane.com>
-------------------------------------------------------------------------------
From: "John S. Dyson" <dyson@dyson.iquest.net>
Subject: Re: Review and report of linux kernel VM
Cc: root@dyson.iquest.net, dg@root.com, jkh@FreeBSD.ORG,
hackers@FreeBSD.ORG
>
> In general terms, linux's VM system is much cleaner then FreeBSD's... and
> I mean a *whole lot* cleaner, but at the cost of eating some extra memory.
> It isn't a whole lot of extra memory - maybe a meg or two for a typical
> system managing a lot of processes, and much less for typical 'small'
> systems. They are able to completely avoid the vm_object stacking
> (and related complexity) that we do, and they are able to completely
> avoid most of the pmap complexity in FreeBSD as well.
>
IMO, the "cleaness" might be better described as "too simple."
>
> Linux appears to map all of physical memory into KVM. This avoids
> FreeBSD's (struct buf) complexity at the cost of not being able to
> deal with huge-memory configurations. I'm not 100% sure of this, but
> its my read of the code until someone corrects me.
>
I suggest that we should get rid of the (struct buf) complexity by
creating the concept of temporary kernel mappings. Such mappings
are a resource limited so that the system doesn't have to map all
of memory, yet have a cleaner, more consistant scheme than the current.
The vm_page_t's at the end of the struct bufs were only a first step
in that arena. There are about 5-10 more steps needed before it is
really fully realized.
>
> * For the nominal page scan, it is using a one-hand clock algorithm.
> All I can say is: Oh my god! Are they nuts? That was abandoned
> a decade ago. The priority mechanism they've implemented is nearly
> useless.
>
> * To locate pages to swap out, it takes a pass through the task list.
> Ostensibly it locates the task with the largest RSS to then try to
> swap pages out from rather then select pages that are not in use.
> From my read of the code, it also botches this badly.
>
Yep, and it has been very difficult for me not to "educate" them on
the right way to do it. Frankly, their code works really well until
it is overused. Given the Linux VM code, "overused" mostly means
used at all. :-).
>
> Linux does not appear to do any page coloring whatsoever, but it would
> not be hard to add it in.
>
It wasn't hard to add to FreeBSD, but the coloring should be moved to
a machine dependent section of the codebase.
>
> Linux cannot swap-out its page tables or page directories. Thus, idle
> tasks can eat a significant amount of memory. This isn't a big deal for
> most systems ( small systems: no problem. Big systems: probably have lots
> of memory anyway ). But, mmap()'d files can create a significant burden
> if you have a lot of forked processes ( news, sendmail, web server,
> etc...). Not only does Linux have to scan the page tables for all the
> processes mapping the file, whether or not they are actively using the
> page being checked for, but Linux's swapout algorithm scans page tables
> and, effectively, makes redundant scans of shared objects.
>
The key here is to NEVER swap out page tables or page directories. One
should FREE them when it is possible. The notion of a dirty page table,
or on that is on disk is meaningless. The FreeBSD code releases page
table pages when they are empty. Page directories should be freeable
when all descendants are no longer mapped (including page tables.) Of
course, in that evaulation, the kernel mappings should be ignored, and
when "swapping" page directories in, they are rebuilt from the kernel
requirments. (FreeBSD doesn't release page directories yet, unless
a process exits.)
> What FreeBSD can learn
>
> Well, the main thing is that the Linux VM system is very, very clean
> compared to the FreeBSD implementation. Cleaning up FreeBSD's VM system
> complexity is what I've been concentrating on and will continue to
> concentrate on. However, part of the reason that FreeBSD's VM system
> is more complex is because it does not use the page tables to store
> reference information. Instead, it uses the vm_object and pmap modules.
> I actually like this feature of FreeBSD. A lot.
>
IMO, the pmap level is super flexible, but also there is too much stratification
between the pmap code and the VM code. Layering is good for reference
implementations, but also adds overhead. It would be "nice" to be able
for the upper level VM code to simply modify page table entries sometimes,
wouldn't it? One thing that I did in the FreeBSD code was to minimize the
transitions between the pmap and VM layers. Perhaps that should be better
defined, but if you wanted to rework the interfaces, you might see a
major cleanup.
>
> The biggest thing we need to do to clean up our VM system is, basically,
> to completely rewrite the struct buf filesystem buffering mechanism to
> make it much, much less complex - basically it should only be used as
> placeholders for read and write ops and not used to cache block number
> mappings between the files and the VM system, nor should it be used to
> map pages into KVM. Separating out these three mechanisms into three
> different subsystems would simplify the code enormously, I think. For
> example, we could implement a simple vm_object KVM mapping mechanism
> using FreeBSD's existing vm_object stacking model to map portions of a
> vm_object (aka filesystem partition) into KVM.
>
I agree. Take a look at using a seperate kernel mapping concept (there
might even be some of that work still lying around somewhere.) You can
then change buffers into what they should really be: I/O requests. The
kernel mappings can be a dynamically allocated resource that are cached
LRU or somesuch. Part of the technology to support them ended up being
pmap_kenter/pmap_kremove and pmap_qenter/pmap_qremove. Without those, the
temporary kernel mappings would have been terribly expensive.
>
> Linux demarks interrupts from supervisor code much better then we do.
> If we move some of the more sophisticated operational capabilities
> out of our interrupt subsystem, we could get rid of most of the spl*()
> junk we currently have to do. This is a real sore spot in current
> FreeBSD code. Interrupts are just too complex. I'd also get rid of
> FreeBSD's intermediate 'software interrupt' layer, which is able to
> do even more complex things then hard interrupt code. The latency
> considerations just don't make any sense verses running pending software
> interrupts synchronously in tsleep(), prior to actually sleeping. We
> need to do this anyway ( or move softints to kernel threads ) to be able
> to take advantage of SMP mechanisms. The *only* thing our interrupts
> should be allowed to do is finish I/O on a page or use zalloc().
>
Note that Linux doesn't even handle IDE PIO correctly, and has historically
lost interrupts due to it (and they added a software interrupt scheme ot
allow it to work.) IDE DMA saved their a**.
John
-------------------------------------------------------------------------------
--- ifmail v.2.14dev2 * Origin: Tver State University NOC (2:5020/400@fidonet)
