x86: add handle_irq() to allow interrupt injection
Xen uses a different interrupt path, so introduce handle_irq() to
allow interrupts to be inserted into the normal interrupt path. This
is handled slightly differently on 32 and 64-bit.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Ingo Molnar [Mon, 9 Feb 2009 11:06:54 +0000 (12:06 +0100)]
tracing/blktrace: move the tracing file to kernel/trace, fix
Impact: build fix
The BLK_DEV_IO_TRACE entry used to be in block/Kconfig - which
file itself was dependent on CONFIG_BLOCK. But now the entry is
in kernel/trace/Kconfig - which is present even on !CONFIG_BLOCK.
Yu Zhao [Sun, 4 Jan 2009 08:28:52 +0000 (16:28 +0800)]
VT-d: handle Invalidation Queue Error to avoid system hang
When hardware detects any error with a descriptor from the invalidation
queue, it stops fetching new descriptors from the queue until software
clears the Invalidation Queue Error bit in the Fault Status register.
Following fix handles the IQE so the kernel won't be trapped in an
infinite loop.
Signed-off-by: Yu Zhao <yu.zhao@intel.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Ingo Molnar [Mon, 9 Feb 2009 10:57:45 +0000 (11:57 +0100)]
x86, pgtable.h: fix 2-level 32-bit build
- pmd_flags() needs to be available on 2-levels too
- provide pud_large() wrapper as well
- include page.h - it provides basic types relied on by pgtable.h
Ingo Molnar [Mon, 9 Feb 2009 10:42:57 +0000 (11:42 +0100)]
x86, pgtable.h: macro-ify *_page() methods
The p?d_page() methods still rely on highlevel types and methods:
In file included from arch/x86/kernel/early_printk.c:18:
/home/mingo/tip/arch/x86/include/asm/pgtable.h: In function ‘pmd_page’:
/home/mingo/tip/arch/x86/include/asm/pgtable.h:516: error: implicit declaration of function â\80\98__pfn_to_sectionâ\80\99
/home/mingo/tip/arch/x86/include/asm/pgtable.h:516: error: initialization makes pointer from integer without a cast
/home/mingo/tip/arch/x86/include/asm/pgtable.h:516: error: implicit declaration of function ‘__section_mem_map_addr’
/home/mingo/tip/arch/x86/include/asm/pgtable.h:516: error: return makes pointer from integer without a cast
So convert them to macros and document the type dependency.
arch/x86/kernel/early_printk.c: In function ‘early_dbgp_init’:
arch/x86/kernel/early_printk.c:827: error: ‘PAGE_KERNEL_NOCACHE’ undeclared (first use in this function)
arch/x86/kernel/early_printk.c:827: error: (Each undeclared identifier is reported only once
arch/x86/kernel/early_printk.c:827: error: for each function it appears in.)
x86: add clflush before monitor for Intel 7400 series
For Intel 7400 series CPUs, the recommendation is to use a clflush on the
monitored address just before monitor and mwait pair [1].
This clflush makes sure that there are no false wakeups from mwait when the
monitored address was recently written to.
[1] "MONITOR/MWAIT Recommendations for Intel Xeon Processor 7400 series"
section in specification update document of 7400 series
http://download.intel.com/design/xeon/specupdt/32033601.pdf
tracing/function-graph-tracer: drop the kernel_text_address check
When the function graph tracer picks a return address, it ensures this address
is really a kernel text one by calling __kernel_text_address()
Actually this path has never been taken.Its role was more likely to debug the tracer
on the beginning of its development but this function is wasteful since it is called
for every traced function.
Brian Gerst [Sun, 8 Feb 2009 14:58:40 +0000 (09:58 -0500)]
x86: fix abuse of per_cpu_offset
Impact: bug fix
Don't use per_cpu_offset() to determine if it valid to access a
per-cpu variable for a given cpu number. It is not a valid assumption
on x86-64 anymore. Use cpu_possible() instead.
Signed-off-by: Brian Gerst <brgerst@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Brian Gerst [Sun, 8 Feb 2009 14:58:39 +0000 (09:58 -0500)]
x86: use linker to offset symbols by __per_cpu_load
Impact: cleanup and bug fix
Use the linker to create symbols for certain per-cpu variables
that are offset by __per_cpu_load. This allows the removal of
the runtime fixup of the GDT pointer, which fixes a bug with
resume reported by Jiri Slaby.
x86: don't pretend that non-framepointer stack traces are reliable
Without frame pointers enabled, the x86 stack traces should not
pretend to be reliable; instead they should just be what they are:
unreliable.
The effect of this is that they have a '?' printed in the stacktrace,
to warn the reader that these entries are guesses rather than known
based on more reliable information.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
xfs: get rid of indirections in the quotaops implementation
Currently we call from the nicely abstracted linux quotaops into a ugly
multiplexer just to split the calls out at the same boundary again.
Rewrite the quota ops handling to remove that obfucation.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <david@fromorbit.com>
Get rid of various obsfucating wrappers for accessing the quota hash lock,
we only keep the accessors for accessing the mplist and freelist locks as
they encode a multi-level datastructure walk. But make sure all of them
are defined in the same way as simple macros.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <david@fromorbit.com>
xfs_create and xfs_mkdir only have minor differences, so merge both of them
into a sigle function. While we're at it also make the error handling code
more straight-forward.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Dave Chinner <david@fromorbit.com>
xfs_ialloc_btree.h has a a cuple of macros that only obsfucate the code
but don't provide any abstraction benefits. This patches removes those
and cleans up the reamaining defintions up a little.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <david@fromorbit.com>
Our default has been to always use 8 32KB log buffers for a while now, so
remove the special casing for larger block size filesystem to use the same
or even lower number of buffers.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <david@fromorbit.com>
ALSA: hda - Don't touch non-existent port f on 4-port 92hd71bxx codecs
When checking for input amps on pins 0x0a, 0x0d and 0x0f, and
initializing them for 92hd71xxx codec models, we must skip nid 0x0f
for 4-port models too like with 5-port models, as it is unused
(nid 0x0f is vendor reserved in 4-port models).
Signed-off-by: Herton Ronaldo Krzesinski <herton@mandriva.com.br> Signed-off-by: Takashi Iwai <tiwai@suse.de>
David S. Miller [Mon, 9 Feb 2009 06:32:31 +0000 (22:32 -0800)]
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>
David S. Miller [Mon, 9 Feb 2009 06:00:55 +0000 (22:00 -0800)]
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>
Herbert Xu [Sun, 8 Feb 2009 18:00:39 +0000 (18:00 +0000)]
gro: Optimise IPv4 packet reception
As this function can be called more than half a million times for
10GbE, it's important to optimise it as much as we can.
This patch does some obvious changes to use 2-byte and 4-byte
operations instead of byte-oriented ones where possible. Bit
ops are also used to replace logical ops to reduce branching.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
Herbert Xu [Sun, 8 Feb 2009 18:00:37 +0000 (18:00 +0000)]
gro: Optimise Ethernet header comparison
This patch optimises the Ethernet header comparison to use 2-byte
and 4-byte xors instead of memcmp. In order to facilitate this,
the actual comparison is now carried out by the callers of the
shared dev_gro_receive function.
This has a significant impact when receiving 1500B packets through
10GbE.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
Herbert Xu [Sun, 8 Feb 2009 18:00:36 +0000 (18:00 +0000)]
gro: Remember number of held packets instead of counting every time
This patch prepares for the move of the same_flow checks out of
dev_gro_receive. As such we need to remember the number of held
packets since doing a loop just to count them every time is silly.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 9 Feb 2009 03:24:13 +0000 (19:24 -0800)]
net: Increase default NET_SKB_PAD to 32.
Several devices need to insert some "pre headers" in front of the
main packet data when they transmit a packet.
Currently we allocate only 16 bytes of pad room and this ends up not
being enough for some types of hardware (NIU, usb-net, s390 qeth,
etc.)
So increase this to 32.
Note that drivers still need to check in their transmit routine
whether enough headroom exists, and if not use skb_realloc_headroom().
Tunneling, IPSEC, and other encapsulation methods can cause the
padding area to be used up.
Signed-off-by: David S. Miller <davem@davemloft.net>
o Cut down msi-x vectors from 8 to 1 since only one is used for now.
o Use separate handler for msi-x, that doesn't unnecessarily scrub
msi status register.
Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Alex Williamson [Mon, 9 Feb 2009 01:49:17 +0000 (17:49 -0800)]
tun: Fix unicast filter overflow
Tap devices can make use of a small MAC filter set via the
TUNSETTXFILTER ioctl. The filter has a set of exact matches
plus a hash for imperfect filtering of additional multicast
addresses. The current code is unbalanced, adding unicast
addresses to the multicast hash, but only checking the hash
against multicast addresses. This results in the filter
dropping unicast addresses that overflow the exact filter.
The fix is simply to disable the filter by leaving count set
to zero if we find non-multicast addresses after the exact
match table is filled.
Signed-off-by: Alex Williamson <alex.williamson@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Julia Lawall [Mon, 9 Feb 2009 01:00:49 +0000 (17:00 -0800)]
drivers/isdn: introduce missing kfree
Error handling code following a kmalloc should free the allocated data.
The semantic match that finds the problem is as follows:
(http://www.emn.fr/x-info/coccinelle/)
// <smpl>
@r exists@
local idexpression x;
statement S;
expression E;
identifier f,l;
position p1,p2;
expression *ptr != NULL;
@@
(
if ((x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...)) == NULL) S
|
x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...);
...
if (x == NULL) S
)
<... when != x
when != if (...) { <+...x...+> }
x->f = E
...>
(
return \(0\|<+...x...+>\|ptr\);
|
return@p2 ...;
)
Julia Lawall [Mon, 9 Feb 2009 01:00:02 +0000 (17:00 -0800)]
drivers/atm: introduce missing kfree
Error handling code following a kmalloc should free the allocated data.
The semantic match that finds the problem is as follows:
(http://www.emn.fr/x-info/coccinelle/)
// <smpl>
@r exists@
local idexpression x;
statement S;
expression E;
identifier f,l;
position p1,p2;
expression *ptr != NULL;
@@
(
if ((x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...)) == NULL) S
|
x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...);
...
if (x == NULL) S
)
<... when != x
when != if (...) { <+...x...+> }
x->f = E
...>
(
return \(0\|<+...x...+>\|ptr\);
|
return@p2 ...;
)
Hugh Dickins [Sun, 8 Feb 2009 20:56:58 +0000 (20:56 +0000)]
mm: fix error case in mlock downgrade reversion
Commit 27421e211a39784694b597dbf35848b88363c248, Manually revert
"mlock: downgrade mmap sem while populating mlocked regions", has
introduced its own regression: __mlock_vma_pages_range() may report
an error (for example, -EFAULT from trying to lock down pages from
beyond EOF), but mlock_vma_pages_range() must hide that from its
callers as before.
The XFS_QMOPT_DQLOCK flag introduces major complexity in the quota subsystem
but isn't actually used anywhere. So remove it and all the hazzles it
introduces.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Felix Blyakher <felixb@sgi.com>
Philipp Zabel [Thu, 5 Feb 2009 16:48:21 +0000 (17:48 +0100)]
ASoC: pxa2xx-i2s: remove I2S pin setup
This removes the calls to pxa_gpio_mode from the pxa2xx-i2s driver.
Pin setup should be done during board init via pxa2xx_mfp_config
instead.
Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com> Acked-by: Eric Miao <eric.miao@marvell.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Philipp Zabel [Thu, 5 Feb 2009 16:48:20 +0000 (17:48 +0100)]
pxa/spitz: Setup I2S pins for pxa2xx-i2s
The spitz has a WM8750 codec connected as I2S slave but doesn't use the
PXA I2S system clock.
Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com> Acked-by: Eric Miao <eric.miao@marvell.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Philipp Zabel [Thu, 5 Feb 2009 16:48:19 +0000 (17:48 +0100)]
pxa/h5000: Setup I2S pins for pxa2xx-i2s
The iPAQ h5000 has an AK4535 codec connected as I2S slave,
PXA I2S providing SYSCLK.
Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com> Acked-by: Eric Miao <eric.miao@marvell.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Robert Jarzmik [Sat, 7 Feb 2009 13:01:58 +0000 (14:01 +0100)]
ASoC: Add initial support of Mitac mioa701 device SoC.
This machine driver enables sound functions on Mitac mio
a701 smartphone. Build upon ASoC v1, it handles :
- rear speaker
- front speaker
- microphone
- GSM
A global "Mio Mode" switch is not yet provided to cope with
audio path setup. As balance on audio chip line is no more
assured, an incorrect setup can produce a lot of heat and
even fry the battery behind the wm9713 and the speaker
amplifier.
It doesn't cope with :
- headset jack
- mio master mode
- master volume control
This driver is backported from ASoc v2, and amputated from
scenario setups and master volume control.
[Minor mods for terminology in comments -- broonie]
Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
radeonfb: Fix resume from D3Cold on some platforms
For historical reason, this driver used its own saving/restoring
of the PCI config space, and used the state of it on resume as
an indication as to whether it needed to re-POST the chip or not.
This methods breaks with the later core changes since the core will
have restored things for us.
This patch fixes it by removing that custom code, using standard
core methods to save/restore state, and testing for the need to
re-POST by comparing the content of a few key PLL registers.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
aty128fb: Properly save PCI state before changing PCI PM level
This fixes aty128fb to properly save the PCI config space -before- it
potentially switches the PM state of the chip. This avoids a
warning with the new PM core and is the right thing to do anyway.
I also replaced the hand-coded switch to D2 with a call to the
genericc pci_set_power_state() and removed the code that switches it
back to D0 since the generic code is doing that for us nowadays.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
atyfb: Properly save PCI state before changing PCI PM level
This fixes atyfb to properly save the PCI config space -before- it
potentially switches the PM state of the chip. This avoids a
warning with the new PM core and is the right thing to do anyway.
I also slightly cleaned up the code that checks whether we are
running on a PowerMac to do a runtime check instead of a compile
check only, and replaced a deprecated number with the proper
symbolic constant.
Finally, I removed the useless switch to D0 from resume since
the core does it for us.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Cornelia Huck [Tue, 20 Jan 2009 14:31:31 +0000 (15:31 +0100)]
async: Rename _special -> _domain for clarity.
Rename the async_*_special() functions to async_*_domain(), which
describes the purpose of these functions much better.
[Broke up long lines to silence checkpatch]
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cornelia Huck [Mon, 19 Jan 2009 12:45:28 +0000 (13:45 +0100)]
async: Fix running list handling.
async_schedule() should pass in async_running as the running
list, and run_one_entry() should put the entry to be run on
the provided running list instead of always on the generic one.
Reported-by: Jonathan Corbet <corbet@lwn.net> Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Paul Walmsley [Wed, 28 Jan 2009 19:35:31 +0000 (12:35 -0700)]
[ARM] OMAP2/3 clock: don't tinker with hardirqs when they are supposed to be disabled
Clock rate change code executes inside a spinlock with hardirqs
disabled. The only code that should be messing around with the
hardirq state should be the plat-omap/clock.c code. In the
omap2_reprogram_dpllcore() case, this probably just wastes cycles, but
in the omap3_core_dpll_m2_set_rate() case, this is a nasty bug.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Based on a patch from Paul Walmsley <paul@pwsan.com>:
omap2_clk_enable() should enable a clock's clockdomain before
attempting to enable its parent clock's clockdomain. Similarly, in
the unlikely event that the parent clock enable fails, the clockdomain
should be disabled.
Paul Walmsley [Wed, 28 Jan 2009 19:35:09 +0000 (12:35 -0700)]
[ARM] OMAP clock: drop clk_get_usecount()
This function is race-prone and mistakenly conveys the impression to
drivers that it is part of the clock interface. Get rid of it: core
code that absolutely needs this can just check clk->usecount. Drivers
should not use it at all.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:35:06 +0000 (12:35 -0700)]
[ARM] OMAP2/3 clock: convert remaining MPU barriers into OCP barriers
Several parts of the OMAP2/3 clock code use wmb() to try to ensure
that the hardware write completes before continuing. This approach is
problematic: wmb() only ensures that the write leaves the ARM. It
does not ensure that the write actually reaches the endpoint device.
The endpoint device in this case - either the PRM, CM, or SCM - is
three interconnects away from the ARM - and the final interconnect is
low-speed. And the OCP interconnects will post the write, and who
knows how long that will take to complete. So the wmb() is not what
we want. Worse, the wmb() is indiscriminate; it causes the ARM to
flush any other unrelated buffered writes and wait for the local
interconnect to acknowledge them - potentially very expensive.
Fix this by converting the wmb()s into readbacks of the same PRM/CM/SCM
register. Since the PRM/CM/SCM devices use a single OCP thread, this
will cause the MPU to block while waiting for posted writes to that device
to complete.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Consolidate the commit code for DELAYED_APP clocks into a subroutine,
_omap2xxx_clk_commit(). Also convert the MPU barrier wmb() into an
OCP barrier, since with an MPU barrier, we have no guarantee that the
write actually reached the endpoint device.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:35:01 +0000 (12:35 -0700)]
[ARM] OMAP2/3 clock: don't use a barrier after clk_disable()
clk_disable() previously used an ARM barrier, wmb(), to try to ensure
that the hardware write completed before continuing. There are some
problems with this approach.
The first problem is that wmb() only ensures that the write leaves the
ARM -- not that it actually reaches the endpoint device. In this
case, the endpoint device - either the PRM, CM, or SCM - is three
interconnects away from the ARM, and the final interconnect is
low-speed. And the OCP interconnects will post the write, who knows
how long that will take to complete. So the wmb() is not really what
we want.
Worse, the wmb() is indiscriminate; it will cause the ARM to flush any
other unrelated buffered writes and wait for the local interconnect to
acknowledge them - potentially very expensive.
This first problem could be fixed by doing a readback of the same PRM/CM/SCM
register. Since these devices use a single OCP thread, this will cause the
MPU to wait for the write to complete.
But the primary problem is a conceptual one: clk_disable() should not
need any kind of barrier. clk_enable() needs one since device driver
code must not access a device until its clocks are known to be
enabled. But clk_disable() has no such restriction.
Since blocking the MPU on a PRM/CM/SCM write can be a very
high-latency operation - several hundred MPU cycles - it's worth
avoiding this barrier if possible.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Russell King [Sat, 31 Jan 2009 10:05:51 +0000 (10:05 +0000)]
[ARM] omap: create a proper tree of clocks
Traditionally, we've tracked the parent/child relationships between
clk structures by setting the child's parent member to point at the
upstream clock. As a result, when decending the tree, we have had
to scan all clocks to find the children.
Avoid this wasteful scanning by keeping a list of the clock's children.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Russell King [Thu, 29 Jan 2009 19:33:19 +0000 (19:33 +0000)]
[ARM] omap: clks: call recalc after any rate change
This implements the remainder of:
OMAP clock: move rate recalc, propagation code up to plat-omap/clock.c
from Paul Walmsley which is not covered by the previous:
[ARM] omap: move clock propagation into core omap clock code
[ARM] omap: remove unnecessary calls to propagate_rate()
[ARM] omap: move propagate_rate() calls into generic omap clock code
commits.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:27:48 +0000 (12:27 -0700)]
[ARM] OMAP2/3 clock: use standard set_rate fn in omap2_clk_arch_init()
Use the standard clk_set_rate() function in omap2_clk_arch_init()
rather than omap2_select_table_rate() -- this will ensure that clock
rates are recalculated and propagated correctly after those operations
are consolidated into clk_set_rate().
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tero Kristo [Wed, 28 Jan 2009 19:27:45 +0000 (12:27 -0700)]
[ARM] OMAP3: PM: Make sure clk_disable_unused() order is correct
Current implementation will disable clocks in the order defined in clock34xx.h,
at least DPLL4_M2X2 will hang in certain cases (and prevent retention / off)
if clocks are not disabled in correct order. This patch makes sure the parent
clocks will be active when disabling a clock.
Signed-off-by: Tero Kristo <tero.kristo@nokia.com> Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Add the omap3_core_dpll_m2_set_rate() function to the OMAP3 clock code,
which calls into the SRAM function omap3_sram_configure_core_dpll() to
change the CORE DPLL M2 divider. (SRAM code is necessary since rate changes
on clocks upstream from the SDRC can glitch SDRAM accesses.)
Use this function for the set_rate function pointer in the dpll3_m2_ck
struct clk. With this function in place, PM/OPP code should be able to
alter SDRAM speed via code similar to:
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
For a given SDRAM clock rate, SDRAM chips require memory controllers
to use a specific set of timing minimums and maximums to transfer data
reliably. These parameters can be different for different memory chips
and can also potentially vary by board.
This patch adds the infrastructure for board-*.c files to pass this
timing data to the SDRAM controller init function. The timing data is
specified in an 'omap_sdrc_params' structure, in terms of SDRC
controller register values. An array of these structs, one per SDRC
target clock rate, is passed by the board-*.c file to
omap2_init_common_hw().
This patch does not define the values for different memory chips, nor
does it use the values for anything; those will come in subsequent patches.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:27:37 +0000 (12:27 -0700)]
[ARM] OMAP2 SDRC: separate common OMAP2/3 code from OMAP2xxx code
Separate SDRC code common to OMAP2/3 from mach-omap2/sdrc2xxx.c to
mach-omap2/sdrc.c. Rename the OMAP2xxx-specific functions to use an
'omap2xxx' prefix rather than an 'omap2' prefix, and use "sdrc" in the
function names rather than "memory." Mark several functions
as static that should not be used outside the sdrc2xxx.c file.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:27:34 +0000 (12:27 -0700)]
[ARM] OMAP2 SDRC: rename memory.c to sdrc2xxx.c
Rename arch/arm/mach-omap2/memory.c to arch/arm/mach-omap2/sdrc2xxx.c, since
it contains exclusively SDRAM-related functions. Most of the functions
are also OMAP2xxx-specific - those which are common will be separated out
in a following patch.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:27:31 +0000 (12:27 -0700)]
[ARM] OMAP2 SDRC: move mach-omap2/memory.h into mach/sdrc.h
Move the contents of the arch/arm/mach-omap2/memory.h file to the
existing mach/sdrc.h file, and remove memory.h. Modify files which
include memory.h to include asm/arch/sdrc.h instead.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tony Lindgren [Wed, 28 Jan 2009 19:18:48 +0000 (12:18 -0700)]
[ARM] omap: Fix omap1 clock issues
This fixes booting, and is a step toward fixing things properly:
- Make enable_reg u32 instead of u16
[rmk: virtual addresses are void __iomem *, not u32]
- Get rid of VIRTUAL_IO_ADDRESS for clocks
- Use __raw_read/write instead of omap_read/write for clock registers
This patch adds a bunch of compile warnings until omap1 clock
also uses offsets.
Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:18:22 +0000 (12:18 -0700)]
[ARM] OMAP2 PRCM: clean up CM_IDLEST bits
This patch fixes a few OMAP2xxx CM_IDLEST bits that were incorrectly
marked as being OMAP2xxx-wide, when they were actually 2420-specific.
Also, originally when the PRCM register macros were defined, bit shift
macros used a "_SHIFT" suffix, and mask macros used none. This became
a source of bugs and confusion, as the mask macros were mistakenly
used for shift values. Gradually, the mask macros have been updated,
piece by piece, to add a "_MASK" suffix on the end to clarify. This
patch applies this change to the CM_IDLEST_* register bits.
The patch also adds a few bits that were missing, mostly from the 3430ES1
to ES2 revisions.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Jouni Hogander [Fri, 16 May 2008 10:58:18 +0000 (13:58 +0300)]
[ARM] OMAP34XX: Add miscellaneous definitions related to 34xx
Signed-off-by: Jouni Hogander <jouni.hogander@nokia.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:18:19 +0000 (12:18 -0700)]
[ARM] OMAP2/3 clock: clean up mach-omap2/clock.c
This patch rolls up several cleanup patches.
1. Some unnecessarily verbose variable names are used in several clock.c
functions; clean these up per CodingStyle.
2. Remove omap2_get_clksel() and just use clk->clksel_reg and
clk->clksel_mask directly.
3. Get rid of void __iomem * usage in omap2_clksel_get_src_field.
Prepend the function name with an underscore to highlight that it is a
static function.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:08:46 +0000 (12:08 -0700)]
[ARM] OMAP3 clock: disable DPLL autoidle while waiting for DPLL to lock
During _omap3_noncore_dpll_lock(), if a DPLL has no active downstream
clocks and DPLL autoidle is enabled, the DPLL may never lock, since it
will enter autoidle immediately. To resolve this, disable DPLL
autoidle while locking the DPLL, and unconditionally wait for the DPLL
to lock. This fixes some bugs where the kernel would hang when returning
from retention or return the wrong rate for the DPLL.
This patch is a collaboration with Peter de Schrijver
<peter.de-schrijver@nokia.com> and Kevin Hilman
<khilman@deeprootsystems.com>.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Peter de Schrijver <peter.de-schrijver@nokia.com> Cc: Kevin Hilman <khilman@deeprootsystems.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The DPLL FREQSEL jitter correction bits are set based on a table in
the 34xx TRM, Table 4-38, according to the DPLL's internal clock
frequency "Fint." Several Fint frequency ranges are missing from this
table. Previously, we allowed these Fint frequency ranges to be
selected in the rate rounding code, but did not change the FREQSEL bits.
Correspondence with the OMAP hardware team indicates that Fint values
not in the table should not be used. So, prevent them from being
selected during DPLL rate rounding. This removes warnings and also
can prevent the chip from locking up.
The first pass through the rate rounding code will update the DPLL max
and min dividers appropriately, so later rate rounding passes will run
faster than the first.
Peter de Schrijver <peter.de-schrijver@nokia.com> put up with several
test cycles of this patch - thanks Peter.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Peter de Schrijver <peter.de-schrijver@nokia.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The previous DPLL rate rounding algorithm counted the divider (N) down
from the maximum to 1. Since we currently use a broad DPLL rate
tolerance, and lower N values are more power-efficient, we can often
bypass several iterations through the loop by counting N upwards from
1.
Peter de Schrijver <peter.de-schrijver@nokia.com> put up with several
test cycles of this patch - thanks Peter.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Cc: Peter de Schrijver <peter.de-schrijver@nokia.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:08:14 +0000 (12:08 -0700)]
[ARM] OMAP3 clock: DPLL{1,2}_FCLK clksel can divide by 4
OMAP34xx ES2 TRM Delta G to H states that the divider for DPLL1_FCLK and
DPLL2_FCLK can divide by 4 in addition to dividing by 1 and 2. Encode this
into the OMAP3 clock framework.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Paul Walmsley [Wed, 28 Jan 2009 19:08:11 +0000 (12:08 -0700)]
[ARM] OMAP3 clock: fix DPLL jitter correction and rate programming
Fix DPLL jitter correction programming. Previously,
omap3_noncore_dpll_program() stored the FREQSEL jitter correction
parameter to the wrong register. This caused jitter correction to be set
incorrectly and also caused the DPLL divider to be programmed incorrectly.
Also, fix DPLL divider programming. An off-by-one error existed in
omap3_noncore_dpll_program(), causing DPLLs to be programmed with a higher
divider than intended.
Signed-off-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
projects / linux-2.6-omap-h63xx.git / log