There are different types of a fifo which can not handled in C without a
lot of overhead. So i decided to write the API as a set of macros, which
is the only way to do a kind of template meta programming without C++.
This macros handles the different types of fifos in a transparent way.
There are a lot of benefits:
- Compile time handling of the different fifo types
- Better performance (a save put or get of an integer does only generate
9 assembly instructions on a x86)
- Type save
- Cleaner interface, the additional kfifo_..._rec() functions are gone
- Easier to use
- Less error prone
- Different types of fifos: it is now possible to define a int fifo or
any other type. See below for an example.
- Smaller footprint for none byte type fifos
- No need of creating a second hidden variable, like in the old DEFINE_KFIFO
The API was not changed.
There are now real in place fifos where the data space is a part of the
structure. The fifo needs now 20 byte plus the fifo space. Dynamic
assigned or allocated create a little bit more code.
Most of the macros code will be optimized away and simple generate a
function call. Only the really small one generates inline code.
Additionally you can now create fifos for any data type, not only the
"unsigned char" byte streamed fifos.
There is also a new kfifo_put and kfifo_get function, to handle a single
element in a fifo. This macros generates inline code, which is lit bit
larger but faster.
I know that this kind of macros are very sophisticated and not easy to
maintain. But i have all tested and it works as expected. I analyzed the
output of the compiler and for the x86 the code is as good as hand written
assembler code. For the byte stream fifo the generate code is exact the
same as with the current kfifo implementation. For all other types of
fifos the code is smaller before, because the interface is easier to use.
The main goal was to provide an API which is very intuitive, save and easy
to use. So linux will get now a powerful fifo API which provides all what
a developer needs. This will save in the future a lot of kernel space,
since there is no need to write an own implementation. Most of the device
driver developers need a fifo, and also deep kernel development will gain
benefit from this API.
Here are the results of the text section usage:
Example 1:
kfifo_put/_get kfifo_in/out current kfifo
dynamic allocated 0x000002a8 0x00000291 0x00000299
in place 0x00000291 0x0000026e 0x00000273
kfifo.c new old
text section size 0x00000be5 0x000008b2
As you can see, kfifo_put/kfifo_get creates a little bit more code than
kfifo_in/kfifo_out, but it is much faster (the code is inline).
The code is complete hand crafted and optimized. The text section size is
as small as possible. You get all the fifo handling in only 3 kb. This
includes type safe fix size records, dynamic records and DMA handling.
This should be the final version. All requested features are implemented.
Note: Most features of this API doesn't have any users. All functions
which are not used in the next 9 months will be removed. So, please adapt
your drivers and other sources as soon as possible to the new API and post
it.
This are the features which are currently not used in the kernel:
kfifo_to_user()
kfifo_from_user()
kfifo_dma_....() macros
kfifo_esize()
kfifo_recsize()
kfifo_put()
kfifo_get()
The fixed size record elements, exclude "unsigned char" fifo's and the
variable size records fifo's
This patch:
User of the kernel fifo should never bypass the API and directly access
the fifo structure. Otherwise it will be very hard to maintain the API.
Signed-off-by: Stefani Seibold <stefani@seibold.net>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Sorry, one more fix, this one depends on the other, so this is rather 2/2.
--
tty->driver_data is used all over the code, but never set. This
results in oopses like:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000130
IP: [<ffffffff814a0040>] mutex_lock+0x10/0x40
...
Pid: 2157, comm: modem-manager Not tainted 2.6.34.1-0.1-desktop #1 2768DR7/2768DR7
RIP: 0010:[<ffffffff814a0040>] [<ffffffff814a0040>] mutex_lock+0x10/0x40
RSP: 0018:ffff88007b16fa50 EFLAGS: 00010286
RAX: 0000000000000000 RBX: 0000000000000130 RCX: 0000000000000003
RDX: 0000000000000003 RSI: 0000000000000286 RDI: 0000000000000130
RBP: 0000000000001000 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000000 R12: 0000000000000130
R13: 0000000000000001 R14: 0000000000000000 R15: ffff88007b16feb4
...
Call Trace:
[<ffffffffa077690d>] ntty_write_room+0x4d/0x90 [nozomi]
...
Set tty->driver_data to the computed port in .install to not recompute it in
every place where needed. Switch .open to use driver_data too.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Alan Cox <alan@linux.intel.com>
Cc: stable <stable@kernel.org> [.34, .35]
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Currently ntty_install omits to increment tty count and we get the
following warnings:
Warning: dev (noz2) tty->count(0) != #fd's(1) in tty_open
So to fix that, add one tty->count++ there.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Alan Cox <alan@linux.intel.com>
Cc: stable <stable@kernel.org> [.34, .35]
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
The original author didn't realise the kernel lock was a drop while sleep
lock so did clever (and wrong) things to work around the non need to avoid
deadlocks. Remove the cleverness and the comment (as we don't hold the BKL
now anyway in those paths)
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The Nozomi tty handling is very broken on the open/close side (See
http://bugzilla.kernel.org/show_bug.cgi?id=13024 for one example). In
particular it marks the tty as closed on the first close() not on the last.
Most of the logic is pretty solid except for the open/close path so switch
to the tty_port helpers and let them do all the heavy lifting. This is also
fixes all the POSIX behaviour violations in the open/close paths.
Begin by adding the tty port usage
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Nozomi goes wrong if you get the sequence
open
open
close
[stuff]
close
which turns out to occur on some ppp type setups.
This is a quick patch up for the problem. It's not really fixing Nozomi
which completely fails to implement tty open/close semantics and all the
other needed stuff. Doing it right is a rather more invasive patch set and
not one that will backport.
Signed-off-by: Alan Cox <alan@linux.intel.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Fix the "ignoring return value of '...', declared with attribute
warn_unused_result" compiler warning in several users of the new kfifo
API.
It removes the __must_check attribute from kfifo_in() and
kfifo_in_locked() which must not necessary performed.
Fix the allocation bug in the nozomi driver file, by moving out the
kfifo_alloc from the interrupt handler into the probe function.
Fix the kfifo_out() and kfifo_out_locked() users to handle a unexpected
end of fifo.
Signed-off-by: Stefani Seibold <stefani@seibold.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
rename kfifo_put... into kfifo_in... to prevent miss use of old non in
kernel-tree drivers
ditto for kfifo_get... -> kfifo_out...
Improve the prototypes of kfifo_in and kfifo_out to make the kerneldoc
annotations more readable.
Add mini "howto porting to the new API" in kfifo.h
Signed-off-by: Stefani Seibold <stefani@seibold.net>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Acked-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Move the pointer to the spinlock out of struct kfifo. Most users in
tree do not actually use a spinlock, so the few exceptions now have to
call kfifo_{get,put}_locked, which takes an extra argument to a
spinlock.
Signed-off-by: Stefani Seibold <stefani@seibold.net>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Acked-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is a new generic kernel FIFO implementation.
The current kernel fifo API is not very widely used, because it has to
many constrains. Only 17 files in the current 2.6.31-rc5 used it.
FIFO's are like list's a very basic thing and a kfifo API which handles
the most use case would save a lot of development time and memory
resources.
I think this are the reasons why kfifo is not in use:
- The API is to simple, important functions are missing
- A fifo can be only allocated dynamically
- There is a requirement of a spinlock whether you need it or not
- There is no support for data records inside a fifo
So I decided to extend the kfifo in a more generic way without blowing up
the API to much. The new API has the following benefits:
- Generic usage: For kernel internal use and/or device driver.
- Provide an API for the most use case.
- Slim API: The whole API provides 25 functions.
- Linux style habit.
- DECLARE_KFIFO, DEFINE_KFIFO and INIT_KFIFO Macros
- Direct copy_to_user from the fifo and copy_from_user into the fifo.
- The kfifo itself is an in place member of the using data structure, this save an
indirection access and does not waste the kernel allocator.
- Lockless access: if only one reader and one writer is active on the fifo,
which is the common use case, no additional locking is necessary.
- Remove spinlock - give the user the freedom of choice what kind of locking to use if
one is required.
- Ability to handle records. Three type of records are supported:
- Variable length records between 0-255 bytes, with a record size
field of 1 bytes.
- Variable length records between 0-65535 bytes, with a record size
field of 2 bytes.
- Fixed size records, which no record size field.
- Preserve memory resource.
- Performance!
- Easy to use!
This patch:
Since most users want to have the kfifo as part of another object,
reorganize the code to allow including struct kfifo in another data
structure. This requires changing the kfifo_alloc and kfifo_init
prototypes so that we pass an existing kfifo pointer into them. This
patch changes the implementation and all existing users.
[akpm@linux-foundation.org: fix warning]
Signed-off-by: Stefani Seibold <stefani@seibold.net>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Acked-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
After m68k's task_thread_info() doesn't refer to current,
it's possible to remove sched.h from interrupt.h and not break m68k!
Many thanks to Heiko Carstens for allowing this.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
This function does not have an error return and returning an error is
instead interpreted as having a lot of pending bytes.
Reported by Jeff Harris who provided a list of some of the remaining
offenders.
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Don't forget to drop a tty refererence on fail paths in
receive_data().
Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The kernel oopses if this flag is set.
[and neither driver should set it as they call tty_flip_buffer_push from IRQ
paths so have always been buggy]
Signed-off-by: Chuck Ebbert <cebbert@redhat.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Nozomi assumes the close method isn't called if open errors. The tty layer
is different to other drives in this respect however. Pointed out by Denis J
Barrow.
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
__FUNCTION__ is gcc-specific, use __func__
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Simple search/replace except for synclink.c where I noticed a real bug and
fixed it too. It was doing NULL + offset, then checking for NULL if the remap
failed.
Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Paul Fulghum <paulkf@microgate.com>
Acked-by: Jiri Slaby <jirislaby@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
- Stop drivers calling their own flush method indirectly, it obfuscates code
and it will change soon anyway
- A few more lock_kernel paths temporarily needed in some driver internal
waiting code
- Remove private put_char method that does a write call for one char - we
have that anyway
- Most but not yet all of the termios copy under lock fixing (some has other
dependencies to follow)
- Note a few locking bugs in drivers found in the process
- Kill remaining [ab]users of TIOCG/SSOFTCAR in the driver, these must go to
fix the termios locking
Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
aka if you see a force-cast, be very suspicious...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Acked-and-tested-by: Frank Seidel <fseidel@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Due to some flaws in the initialization and flow control
code kernel oopses could be triggered e.g. when accessing
the card too early after insertion.
See e.g. kernel.org bug #10077.
The main part of the fix is a trivial state management
making sure the card is realy ready to use before allowing
any access.
Signed-off-by: Frank Seidel <fseidel@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
nozomi: constify structures and annotate vars
Signed-off-by: Jan Engelhardt <jengelh@computergmbh.de>
Signed-off-by: Frank Seidel <fseidel@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Minor cleanups and removal of in-file changelog:
- Correction of misspellings and wrong encoded Name
- changed 'unsigned' to 'unsigned int' for better readability
- use of generic devicefile access macro
- fixed/added explanatory comment to ntty_put_char
Signed-off-by: Frank Seidel <fseidel@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This is a driver to control the cardbus wireless data card that works on
3g networks.
Greg Kroah-Hartman <gregkh@suse.de> did the initial driver cleanup.
Thanks to Arnaud Patard <apatard@mandriva.com> for help with bugfixing.
Thanks to Alan Cox for a lot of tty fixes.
Thanks to Satyam Sharma <satyam@infradead.org> for fixing buildbreakage.
Thanks to Frank Seidel <fseidel@suse.de> for a lot of bugfixes and
rewriting to make it a sane Linux driver
Thanks to Jiri Slaby <jirislaby@gmail.com> for a lot bugfixes, cleanups
and rewrites that make it much more readable.
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Frank Seidel <fseidel@suse.de>
Signed-off-by: Jiri Slaby <jirislaby@gmail.com>