* [RFC] Documentation/filesystems/tmpfs
@ 2001-11-28 16:49 Christoph Rohland
2001-11-28 17:30 ` Padraig Brady
0 siblings, 1 reply; 5+ messages in thread
From: Christoph Rohland @ 2001-11-28 16:49 UTC (permalink / raw)
To: Linux Kernel Mailing List
[-- Attachment #1: Type: text/plain, Size: 319 bytes --]
Hi,
Apparently there is a lack of information about tmpfs out there.
So I would like to reduce the Configure.help entry and instead
introduce a file Documentation/filesystems/tmpfs with more details.
My proposal is appended. Feedback about content and english language
is highly appreciated.
Greetings
Christoph
[-- Attachment #2: Type: text/plain, Size: 3489 bytes --]
Tmpfs is a file system which keeps all files in virtual memory.
Everything is temporary in the sense that no files will be created on
your hard drive. If you reboot, everything in tmpfs will be lost.
In contrast to RAM disks, which get allocated a fixed amount of
physical RAM, tmpfs grows and shrinks to accommodate the files it
contains and is able to swap unneeded pages out to swap space.
Since tmpfs lives completely in the page cache and on swap, all in
memory tmpfs pages will show up as cached. It will not show up as
shared or something like that.
tmpfs has the following uses:
1) There is always a kernel internal mount which you will not see at
all. This is used for shared anonymous mappings and SYSV shared
memory.
This mount does not depend on CONFIG_TMPFS. If CONFIG_TMPFS is not
set the user visible part of tmpfs is not build, but the internal
mechanisms are always present.
2) glibc 2.2 and above expects tmpfs to be mounted at /dev/shm for
POSIX shared memory (shm_open, shm_unlink). Adding the following
line to /etc/fstab should take care of this:
tmpfs /dev/shm tmpfs defaults 0 0
Remember to create the directory that you intend to mount tmpfs on
if necessary (/dev/shm is automagically created if you use devfs).
This mount is _not_ needed for SYSV shared memory. The internal
mount is used for that. (In the 2.3 kernel versions it was
necessary to mount the predecessor of tmpfs (shm fs) to use SYSV
shared memory)
3) Some people (including me) find it very convenient to mount it
e.g. on /tmp and /var/tmp and have a big swap partition. But be
aware: loop mounts of tmpfs files do not work due to the internal
design. So mkinitrd shipped by most distributions will fail with a
tmpfs /tmp.
4) And probably a lot more I do not know about :-)
tmpfs has a couple of mount options:
size: The limit of allocated bytes for this tmpfs instance. The
default is half of your physical RAM without swap. If you
oversize your tmpfs instances the machine will deadlock
since the OOM handler will not be able to free that memory.
nr_blocks: The same as size, but in blocks of PAGECACHE_SIZE.
nr_inodes: The maximum number of inodes for this instance. The default
is half of the number of your physical RAM pages.
These parameters accept a suffix k, m or g for kilo, mega and giga and
can be changed on remount.
The initial permissions of the root directory can be set with the
mount option "mode". Later on you can change the permissions of the
root directory with chmod.
So 'mount -t tmpfs -o size=10G,nr_inodes=10k,mode=0700 tmpfs /mytmpfs'
will give you tmpfs instance on /mytmpfs which can allocate 10GB
RAM/SWAP in 10240 inodes and it is only accessible by root.
TODOs:
1) give the size option a percent semantic: If you give a mount option
size=50% the tmpfs instance should be able to grow to 50 percent of
RAM + swap. So the instance should adapt autatically if you add or
remove swap space.
2) loop mounts: This is difficult since loop.c relies on the readpage
operation. This operation gets a page from the caller to be filled
with the content of the file at that position. But tmpfs always has
the page and thus cannot copy the content to the given page. So it
cannot provide this operation. The VM had to be changed seriously
to achieve this.
3) Show the number of tmpfs pages. (As shared?)
Author:
Christoph Rohland, 27.11.01
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [RFC] Documentation/filesystems/tmpfs
2001-11-28 16:49 [RFC] Documentation/filesystems/tmpfs Christoph Rohland
@ 2001-11-28 17:30 ` Padraig Brady
2001-11-28 17:37 ` Christoph Rohland
0 siblings, 1 reply; 5+ messages in thread
From: Padraig Brady @ 2001-11-28 17:30 UTC (permalink / raw)
To: Christoph Rohland; +Cc: Linux Kernel Mailing List
Christoph Rohland wrote:
> Hi,
>
> Apparently there is a lack of information about tmpfs out there.
I aggree, and ramfs et al.
> So I would like to reduce the Configure.help entry and instead
> introduce a file Documentation/filesystems/tmpfs with more details.
>
> My proposal is appended. Feedback about content and english language
> is highly appreciated.
>
> Greetings
> Christoph
>
> ------------------------------------------------------------------------
>
> Tmpfs is a file system which keeps all files in virtual memory.
>
> Everything is temporary in the sense that no files will be created on
> your hard drive. If you reboot, everything in tmpfs will be lost.
>
> In contrast to RAM disks, which get allocated a fixed amount of
> physical RAM, tmpfs grows and shrinks to accommodate the files it
> contains and is able to swap unneeded pages out to swap space.
That isn't the case now since ramdisks were
integrated with the buffer cache:
$ dd if=/dev/zero of=/tmp/use_mem bs=1024 count=20000
$ cat /proc/meminfo
total: used: free: shared: buffers: cached:
Mem: 129966080 122720256 7245824 49152 65863680 28745728
Swap: 0 0 0
MemTotal: 126920 kB
MemFree: 7076 kB
MemShared: 48 kB
Buffers: 64320 kB
Cached: 28072 kB
SwapCached: 0 kB
Active: 10308 kB
Inact_dirty: 80196 kB
Inact_clean: 1936 kB
Inact_target: 26200 kB
HighTotal: 0 kB
HighFree: 0 kB
LowTotal: 126920 kB
LowFree: 7076 kB
SwapTotal: 0 kB
SwapFree: 0 kB
$ rm /tmp/use_mem
$ cat /proc/meminfo
total: used: free: shared: buffers: cached:
Mem: 129966080 101957632 28008448 49152 65863680 8290304
Swap: 0 0 0
MemTotal: 126920 kB
MemFree: 27352 kB
MemShared: 48 kB
Buffers: 64320 kB
Cached: 8096 kB
SwapCached: 0 kB
Active: 10352 kB
Inact_dirty: 60196 kB
Inact_clean: 1916 kB
Inact_target: 26200 kB
HighTotal: 0 kB
HighFree: 0 kB
LowTotal: 126920 kB
LowFree: 27352 kB
SwapTotal: 0 kB
SwapFree: 0 kB
Padraig.
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [RFC] Documentation/filesystems/tmpfs
2001-11-28 17:30 ` Padraig Brady
@ 2001-11-28 17:37 ` Christoph Rohland
2001-11-28 17:49 ` Padraig Brady
0 siblings, 1 reply; 5+ messages in thread
From: Christoph Rohland @ 2001-11-28 17:37 UTC (permalink / raw)
To: Padraig Brady; +Cc: Linux Kernel Mailing List
Hi Padraig,
On Wed, 28 Nov 2001, Padraig Brady wrote:
>> In contrast to RAM disks, which get allocated a fixed amount of
>> physical RAM, tmpfs grows and shrinks to accommodate the files it
>> contains and is able to swap unneeded pages out to swap space.
>
>
>
> That isn't the case now since ramdisks were integrated with the
> buffer cache:
What isn't the case any more?
> $ dd if=/dev/zero of=/tmp/use_mem bs=1024 count=20000
On what filesystem is /tmp/use_mem located? What do you want to show?
Greetings
Christoph
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [RFC] Documentation/filesystems/tmpfs
2001-11-28 17:37 ` Christoph Rohland
@ 2001-11-28 17:49 ` Padraig Brady
2001-11-28 20:38 ` [MOc]cda*mirabilos
0 siblings, 1 reply; 5+ messages in thread
From: Padraig Brady @ 2001-11-28 17:49 UTC (permalink / raw)
To: Christoph Rohland; +Cc: Linux Kernel Mailing List
Christoph Rohland wrote:
> Hi Padraig,
>
> On Wed, 28 Nov 2001, Padraig Brady wrote:
>
>>>In contrast to RAM disks, which get allocated a fixed amount of
>>>physical RAM, tmpfs grows and shrinks to accommodate the files it
>>>contains and is able to swap unneeded pages out to swap space.
>>>
>>
>>
>>That isn't the case now since ramdisks were integrated with the
>>buffer cache:
>>
>
> What isn't the case any more?
Because the RAM is now (de)allocated as required.
>
>>$ dd if=/dev/zero of=/tmp/use_mem bs=1024 count=20000
>>
>
> On what filesystem is /tmp/use_mem located? What do you want to show?
Filesystem? ext2 on /dev/ram1(rd.o)
meminfo shows the memory being reclaimed as the file is deleted.
Padraig.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [RFC] Documentation/filesystems/tmpfs
2001-11-28 17:49 ` Padraig Brady
@ 2001-11-28 20:38 ` [MOc]cda*mirabilos
0 siblings, 0 replies; 5+ messages in thread
From: [MOc]cda*mirabilos @ 2001-11-28 20:38 UTC (permalink / raw)
To: Padraig Brady, Christoph Rohland; +Cc: Linux Kernel Mailing List
> >>>In contrast to RAM disks, which get allocated a fixed amount of
> >>>physical RAM, tmpfs grows and shrinks to accommodate the files it
> >>>contains and is able to swap unneeded pages out to swap space.
> >>That isn't the case now since ramdisks were integrated with the
> >>buffer cache:
> > What isn't the case any more?
> Because the RAM is now (de)allocated as required.
> Filesystem? ext2 on /dev/ram1(rd.o)
> meminfo shows the memory being reclaimed as the file is deleted.
And this is transparent to {any|the ext2fs} underlying filesystem?
What if I use xfs, rasierfs, ntfs, $* ?
-mirabilos
--
Redistribution of this message body via AOL or the
Microsoft network strictly prohibited.
Quotation permitted if due credit is given.
(Excuse the X-Mailer, accusate my ISP for that)
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2001-11-28 20:39 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-11-28 16:49 [RFC] Documentation/filesystems/tmpfs Christoph Rohland
2001-11-28 17:30 ` Padraig Brady
2001-11-28 17:37 ` Christoph Rohland
2001-11-28 17:49 ` Padraig Brady
2001-11-28 20:38 ` [MOc]cda*mirabilos
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox