Commit Graph

59 Commits

Author SHA1 Message Date
David S. Miller
a52312b88c sparc32: Correct the return value of memcpy.
Properly return the original destination buffer pointer.

Signed-off-by: David S. Miller <davem@davemloft.net>
Tested-by: Kjetil Oftedal <oftedal@gmail.com>
2011-10-20 15:17:23 -07:00
David S. Miller
21f74d361d sparc32: Remove uses of %g7 in memcpy implementation.
This is setting things up so that we can correct the return
value, so that it properly returns the original destination
buffer pointer.

Signed-off-by: David S. Miller <davem@davemloft.net>
Tested-by: Kjetil Oftedal <oftedal@gmail.com>
2011-10-20 15:17:22 -07:00
David S. Miller
045b7de9ca sparc32: Remove non-kernel code from memcpy implementation.
Signed-off-by: David S. Miller <davem@davemloft.net>
Tested-by: Kjetil Oftedal <oftedal@gmail.com>
2011-10-20 15:17:22 -07:00
Josip Rodin
a61b582954 sparc: Fix __atomic_add_unless() return value.
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-08-04 02:47:40 -07:00
David S. Miller
56d205cc5c sparc: Use popc when possible for ffs/__ffs/ffz.
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-08-02 21:28:53 -07:00
David S. Miller
ef7c4d4675 sparc: Use popc if possible for hweight routines.
Just like powerpc, we code patch at boot time.

Signed-off-by: David S. Miller <davem@davemloft.net>
2011-08-02 21:28:50 -07:00
David S. Miller
e95ade0839 sparc: Minor tweaks to Niagara page copy/clear.
Don't use floating point on Niagara2, use the traditional
plain Niagara code instead.

Unroll Niagara loops to 128 bytes for copy, and 256 bytes
for clear.

Signed-off-by: David S. Miller <davem@davemloft.net>
2011-08-02 21:28:32 -07:00
Stephen Rothwell
678624e401 sparc: rename atomic_add_unless
Should have been done in commit 1af08a1407f4 ("This is in preparation
for more generic atomic").

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Arun Sharma <asharma@fb.com>
Cc: David Miller <davem@davemloft.net>
Cc: "Hans-Christian Egtvedt" <hans-christian.egtvedt@atmel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-07-27 12:53:36 -07:00
Arun Sharma
60063497a9 atomic: use <linux/atomic.h>
This allows us to move duplicated code in <asm/atomic.h>
(atomic_inc_not_zero() for now) to <linux/atomic.h>

Signed-off-by: Arun Sharma <asharma@fb.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-07-26 16:49:47 -07:00
David S. Miller
9fafbd8061 Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6 2011-05-20 12:59:54 -07:00
Tkhai Kirill
b1054282d7 sparc32: Fixed unaligned memory copying in function __csum_partial_copy_sparc_generic
When we are in the label cc_dword_align, registers %o0 and %o1 have the same last 2 bits,
but it's not guaranteed one of them is zero. So we can get unaligned memory access
in label ccte. Example of parameters which lead to this:
%o0=0x7ff183e9, %o1=0x8e709e7d, %g1=3

With the parameters I had a memory corruption, when the additional 5 bytes were rewritten.
This patch corrects the error.

One comment to the patch. We don't care about the third bit in %o1, because cc_end_cruft
stores word or less.

Signed-off-by: Tkhai Kirill <tkhai@yandex.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-05-11 21:35:04 -07:00
Daniel Hellstrom
1827237065 sparc32: removed unused code, implemented by generic code
Signed-off-by: Daniel Hellstrom <daniel@gaisler.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-04-21 16:44:44 -07:00
Linus Torvalds
0586bed3e8 Merge branch 'core-locking-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'core-locking-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  rtmutex: tester: Remove the remaining BKL leftovers
  lockdep/timers: Explain in detail the locking problems del_timer_sync() may cause
  rtmutex: Simplify PI algorithm and make highest prio task get lock
  rwsem: Remove redundant asmregparm annotation
  rwsem: Move duplicate function prototypes to linux/rwsem.h
  rwsem: Unify the duplicate rwsem_is_locked() inlines
  rwsem: Move duplicate init macros and functions to linux/rwsem.h
  rwsem: Move duplicate struct rwsem declaration to linux/rwsem.h
  x86: Cleanup rwsem_count_t typedef
  rwsem: Cleanup includes
  locking: Remove deprecated lock initializers
  cred: Replace deprecated spinlock initialization
  kthread: Replace deprecated spinlock initialization
  xtensa: Replace deprecated spinlock initialization
  um: Replace deprecated spinlock initialization
  sparc: Replace deprecated spinlock initialization
  mips: Replace deprecated spinlock initialization
  cris: Replace deprecated spinlock initialization
  alpha: Replace deprecated spinlock initialization
  rtmutex-tester: Remove BKL tests
2011-03-15 18:28:30 -07:00
Akinobu Mita
e637804c33 sparc: use bitmap_set()
Use bitmap_set() instead of calling __set_bit() each bit.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: sparclinux@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
2011-02-08 22:52:53 -08:00
Thomas Gleixner
24774fbdea sparc: Replace deprecated spinlock initialization
SPIN_LOCK_UNLOCK is deprecated. Use the lockdep capable variant
instead.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: David S. Miller <davem@davemloft.net>
2011-01-27 12:30:37 +01:00
David S. Miller
0f58189d4a sparc64: Make lock backoff really a NOP on UP builds.
As noticed by Mikulas Patocka, the backoff macros don't
completely nop out for UP builds, we still get a
branch always and a delay slot nop.

Fix this by making the branch to the backoff spin loop
selective, then we can nop out the spin loop completely.

Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-18 22:53:26 -07:00
Mikulas Patocka
6ec274750c sparc64: simple microoptimizations for atomic functions
Simple microoptimizations for sparc64 atomic functions:
Save one instruction by using a delay slot.
Use %g1 instead of %g7, because %g1 is written earlier.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-18 22:51:08 -07:00
David S. Miller
9b3bb86aca sparc64: Make rwsems 64-bit.
Basically tip-off the powerpc code, use a 64-bit type and atomic64_t
interfaces for the implementation.

This gets us off of the by-hand asm code I wrote, which frankly I
think probably ruins I-cache hit rates.

The idea was the keep the call chains less deep, but anything taking
the rw-semaphores probably is also calling other stuff and therefore
already has allocated a stack-frame.  So no real stack frame savings
ever.

Ben H. has posted patches to make powerpc use 64-bit too and with some
abstractions we can probably use a shared header file somewhere.

With suggestions from Sam Ravnborg.

Signed-off-by: David S. Miller <davem@davemloft.net>
2010-08-17 22:49:26 -07:00
David S. Miller
035df35d96 sparc64: Allocate sufficient stack space in ftrace stubs.
128 bytes is sufficient for the register window save area, but the
calling conventions allow the callee to save up to 6 incoming argument
registers into the stack frame after the register window save area.

This means a minimal stack frame is 176 bytes (128 + (6 * 8)).

This fixes random crashes when using the function tracer.

Reported-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-13 18:59:02 -07:00
David S. Miller
9960e9e894 sparc64: Add function graph tracer support.
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-12 22:37:26 -07:00
David S. Miller
a71d1d6bb1 sparc64: Give a stack frame to the ftrace call sites.
It's the only way we'll be able to implement the function
graph tracer properly.

A positive is that we no longer have to worry about the
linker over-optimizing the tail call, since we don't
use a tail call any more.

Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-12 22:37:15 -07:00
David S. Miller
ddacd0bc70 sparc64: Kill CONFIG_STACK_DEBUG code.
The generic stack tracer does this job just as well.

Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-12 22:36:03 -07:00
David S. Miller
63b7549573 sparc64: Add HAVE_FUNCTION_TRACE_MCOUNT_TEST and tidy up.
Check function_trace_stop at ftrace_caller

Toss mcount_call and dummy call of ftrace_stub, unnecessary.

Document problems we'll have if the final kernel image link
ever turns on relaxation.

Properly size 'ftrace_call' so it looks right when inspecting
instructions under gdb et al.

Signed-off-by: David S. Miller <davem@davemloft.net>
2010-04-12 22:35:24 -07:00
David S. Miller
4d14a45985 sparc: Stop trying to be so fancy and use __builtin_{memcpy,memset}()
This mirrors commit ff60fab71b
(x86: Use __builtin_memset and __builtin_memcpy for memset/memcpy)

Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-10 23:32:10 -08:00
David S. Miller
fb34035e7b sparc: Use __builtin_object_size() to validate the buffer size for copy_from_user()
This mirrors x86 commit 9f0cf4adb6
(x86: Use __builtin_object_size() to validate the buffer size for copy_from_user())

Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-10 23:05:23 -08:00
David S. Miller
166e553a57 sparc64: Fix stack debugging IRQ stack regression.
Commit 4f70f7a91b
(sparc64: Implement IRQ stacks.) has two bugs.

First, the softirq range check forgets to subtract STACK_BIAS
before comparing with %sp.  Next, on failure the wrong label
is jumped to, resulting in a bogus stack being loaded.

Reported-by: Igor Kovalenko <igor.v.kovalenko@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-12-09 01:43:45 -08:00
David S. Miller
6373fffc5d sparc64: Fix section attribute warnings.
CSUM copy to/from user assembler was missing allocatable and
executable attributes for .fixup

Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-29 16:12:02 -07:00
David S. Miller
aeb3987683 sparc64: Fix probe_kernel_{read,write}().
This is based upon a report from Chris Torek and his initial patch.
From Chris's report:

--------------------
This came up in testing kgdb, using the built-in tests -- turn
on CONFIG_KGDB_TESTS, then

    echo V1 > /sys/module/kgdbts/parameters/kgdbts

-- but it would affect using kgdb if you were debugging and looking
at bad pointers.
--------------------

When we get a copy_{from,to}_user() request and the %asi is set to
something other than ASI_AIUS (which is userspace) then we branch off
to a routine called memcpy_user_stub().  It just does a straight
memcpy since we are copying from kernel to kernel in this case.

The logic was that since source and destination are both kernel
pointers we don't need to have exception checks.

But for what probe_kernel_{read,write}() is trying to do, we have to
have the checks, otherwise things like kgdb bad kernel pointer
accesses don't do the right thing.

Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-08 22:32:31 -08:00
David S. Miller
40bdac7dbc sparc64: Kill .fixup section bloat.
This is an implementation of a suggestion made by Chris Torek:
--------------------
Something else I noticed in passing: the EX and EX_LD/EX_ST macros
scattered throughout the various .S files make a fair bit of .fixup
code, all of which does the same thing.  At the cost of one symbol
in copy_in_user.S, you could just have one common two-instruction
retl-and-mov-1 fixup that they all share.
--------------------

The following is with a defconfig build:

   text	   data	    bss	    dec	    hex	filename
3972767	 344024	 584449	4901240	 4ac978	vmlinux.orig
3968887	 344024	 584449	4897360	 4aba50	vmlinux

Signed-off-by: David S. Miller <davem@davemloft.net>
2009-02-08 22:00:55 -08:00
Sam Ravnborg
62dfcd336c sparc64: fix modpost failure
Previously PeeCeeI.o was a library but it
was always pulled in due to insw and friends being exported
(at least for a modular kernel).

But this resulted in modpost failures if there where no in-kernel
users because then insw & friends were not linked in.

Fix this by including PeeCeeI.o in the kernel unconditionally.

The only drawback for this solution is that a nonmodular kernel
will always include insw & friends no matter if they are in use or not.

Reported-by: Meelis Roos <mroos@linux.ee>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-17 23:12:27 -08:00
Sam Ravnborg
917c3660d6 sparc64: move EXPORT_SYMBOL to the symbols definition
Move all applicable EXPORT_SYMBOL()s to the file where the respective
symbol is defined.

Removed all the includes that are no longer needed in sparc_ksyms_64.c

Comment all remaining EXPORT_SYMBOL()s in sparc_ksyms_64.c

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>

Additions by Julian Calaby:
* Moved EXPORT_SYMBOL()s for prom functions to their rightful places.
* Made some minor cleanups to the includes and comments of sparc_ksyms_64.c
* Updated and tidied commit message.
* Rebased patch over sparc-2.6.git HEAD.
* Ensured that all modified files have the correct includes.

Signed-off-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-08 16:58:20 -08:00
Sam Ravnborg
45536ffc8d sparc: Create a new file lib/ksyms.c and add export of all symbols defined in assembler in lib/ to this file.
Remove the duplicate entries from kernel/sparc_ksyms_*.c

The rationale behind this is that the EXPORT_SYMBOL() should be close to
their definition and we cannot add designate a symbol to be exported in
assembler so at least put it in a file in the same directory.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>

Additions by Julian Calaby:
* Rebased over sparc-2.6.git HEAD

Signed-off-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2009-01-08 16:57:35 -08:00
David S. Miller
18cdae68e7 sparc: Commonize memcmp assembler.
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-09 04:09:07 -08:00
David S. Miller
ae984d72e0 sparc: Unify strlen assembler.
Use the new asm/asm.h header to help commonize the
strlen assembler between 32-bit and 64-bit

While we're here, use proper linux/linkage.h macros
instead of by-hand stuff.

Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-09 01:07:09 -08:00
David S. Miller
8bf68e4d90 sparc: Kill memcmp_32.S code which has been ifdef'd out for centuries.
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-08 16:05:49 -08:00
Sam Ravnborg
478b8fecda sparc,sparc64: unify lib/
o Renamed files in sparc64 to <name>_64.S when identical
  to sparc32 files.
o iomap.c were equal for sparc32 and sparc64
o adjusted sparc/Makefile now we have only one lib/

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-04 09:17:19 -08:00
Sam Ravnborg
18269c0fd4 sparc: prepare lib/ for unification
Identical named files renamed to <name>_32.S
Refactored Makefile to prepare for unification.

Linking order was altered slightly - but this is a lib.a file so
it should not matter.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-04 09:17:18 -08:00
Adrian Bunk
88278ca27a sparc: remove CVS keywords
This patch removes the CVS keywords that weren't updated for a long time
from comments.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-05-20 00:33:44 -07:00
Benjamin Herrenschmidt
b70d3a2c59 iomap: fix 64 bits resources on 32 bits
Almost all implementations of pci_iomap() in the kernel, including the generic
lib/iomap.c one, copies the content of a struct resource into unsigned long's
which will break on 32 bits platforms with 64 bits resources.

This fixes all definitions of pci_iomap() to use resource_size_t.  I also
"fixed" the 64bits arch for consistency.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-29 08:06:02 -07:00
Sam Ravnborg
c6d64c16bb [SPARC/SPARC64]: Fix usage of .section .sched.text in assembler code.
ld will generate an unique named section when assembler do not use
"ax" but gcc does. Add the missing annotation.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-01-31 19:32:43 -08:00
David S. Miller
6cc0735d0d [SPARC32]: Add __cmpdi2() libcall implementation ala. MIPS.
Device mapper generates calls to this with recent versions
of gcc.

Signed-off-by: David S. Miller <davem@davemloft.net>
2007-08-26 18:49:09 -07:00
Alexander Shmelev
f61698e648 [SPARC32]: Fix bug in sparc optimized memset.
Sparc optimized memset (arch/sparc/lib/memset.S) does not fill last
byte of the memory area, if area size is less than 8 bytes and start
address is not word (4-bytes) aligned.

Here is code chunk where bug located:
/* %o0 - memory address, %o1 - size, %g3 - value */
8:
     add    %o0, 1, %o0
    subcc    %o1, 1, %o1
    bne,a    8b
     stb %g3, [%o0 - 1]

This code should write byte every loop iteration, but last time delay
instruction stb is not executed because branch instruction sets
"annul" bit.

Patch replaces bne,a by bne instruction.

Error can be reproduced by simple kernel module:

--------------------
#include <linux/module.h>
#include <linux/config.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <string.h>

static void do_memset(void **p, int size)
{
        memset(p, 0x00, size);
}

static int __init memset_test_init(void)
{
    char fooc[8];
    int *fooi;
    memset(fooc, 0xba, sizeof(fooc));

    do_memset((void**)(fooc + 3), 1);

    fooi = (int*) fooc;
    printk("%08X %08X\n", fooi[0], fooi[1]);

    return -1;
}

static void __exit memset_test_cleanup(void)
{
    return;
}

module_init(memset_test_init);
module_exit(memset_test_cleanup);

MODULE_LICENSE("GPL");
EXPORT_NO_SYMBOLS;
--------------------

Signed-off-by: Alexander Shmelev <ashmelev@task.sun.mcst.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-07-24 13:41:44 -07:00
Andrew Morton
1fb8812ba5 [SPARC32]: Build fix.
Fix 6197fe4d72

arch/sparc/lib/atomic32.c: In function '__cmpxchg_u32':
arch/sparc/lib/atomic32.c:127: error: 'addr' undeclared (first use in this function)
arch/sparc/lib/atomic32.c:127: error: (Each undeclared identifier is reported only once
arch/sparc/lib/atomic32.c:127: error: for each function it appears in.)

I assume this is what was intended..

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: William Irwin <wli@holomorphy.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-05-31 01:52:51 -07:00
Kyle McMartin
6197fe4d72 [SPARC]: Emulate cmpxchg like parisc
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-05-29 02:51:13 -07:00
Randy Dunlap
e63340ae6b header cleaning: don't include smp_lock.h when not used
Remove includes of <linux/smp_lock.h> where it is not used/needed.
Suggested by Al Viro.

Builds cleanly on x86_64, i386, alpha, ia64, powerpc, sparc,
sparc64, and arm (all 59 defconfigs).

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-08 11:15:07 -07:00
Robert Reif
74e61dee2a [SPARC32]: Fix SMP build regression
commit b19cbe2a16 [BRIDGE]: Fix fdb RCU
race

breaks sparc SMP build because atomic_add_unless is not exported.

This patch exports atomic_add_unless and atomic_cmpxchg.

Signed-off-by: Robert Reif <reif@earthlink.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-03-26 21:49:13 -07:00
David S. Miller
8a8b836b91 [SPARC]: Make bitops use same spinlocks as atomics.
Recent workqueue changes basically make this a formal requirement.

Also, move atomic32.o from lib-y to obj-y since it exports symbols
to modules.

Signed-off-by: David S. Miller <davem@davemloft.net>
2006-12-17 16:18:47 -08:00
Al Viro
7a39f52202 [PATCH] sparc32 rwlock fix
read_trylock() is broken on sparc32 (doesn't build and didn't work
right, actually).  Proposed fix:

 - make "writer holds lock" distinguishable from "reader tries to grab
   lock"

 - have __raw_read_trylock() try to acquire the mutex (in LSB of lock),
   terminating spin if we see that there's writer holding it.  Then do
   the rest as we do in read_lock().

Thanks to Ingo for discussion...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-08 12:32:35 -07:00
David S. Miller
3a1d5c84ed [SPARC]: Don't zero out tail during copy_from_user_inatomic().
Actually, since we use the same code for all the copying
types in and out of userspace, we check at runtime whether
preemption is disabled.

Signed-off-by: David S. Miller <davem@davemloft.net>
2006-09-29 16:12:41 -07:00
Jörn Engel
6ab3d5624e Remove obsolete #include <linux/config.h>
Signed-off-by: Jörn Engel <joern@wohnheim.fh-wedel.de>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
2006-06-30 19:25:36 +02:00