Commit graph

64,840 commits

Author SHA1 Message Date
Alexey Kopytko
92a5de80e5 [ARM] Orion: Add Buffalo Linkstation Mini power-off method
This patch adds specific power-off method for Buffalo Linkstation Mini
board. The board has a hardware switch which should be monitored from
userspace. When the switch is in OFF position the board should be rebooted
and U-Boot will start in an idle mode and wait for the user to move the
power switch back to ON position.

Signed-off-by: Alexey Kopytko <alexey@kopytko.ru>
Signed-off-by: Nicolas Pitre <nico@marvell.com>
2008-09-25 23:54:19 -04:00
Russell King
01135d92c1 [ARM] dma: Reduce to one dma_map_sg()/dma_unmap_sg() implementation
No point having two of these; dma_map_page() can do all the work
for us.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 23:39:24 +01:00
Russell King
56f55f8b58 [ARM] dma: provide a better dma_map_page() implementation
We can translate a struct page directly to a DMA address using
page_to_dma().  No need to use page_address() followed by
virt_to_dma().

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 23:39:24 +01:00
Alexey Kopytko
7b0a26f2ca [ARM] Orion: Add Buffalo Linkstation Mini support
This patch adds support for Buffalo Linkstation Mini board.

Signed-off-by: Alexey Kopytko <alexey@kopytko.ru>
Signed-off-by: Nicolas Pitre <nico@marvell.com>
2008-09-25 17:00:01 -04:00
Ronen Shitrit
4360bb4192 [ARM] Kirkwood: add support for L2 cache WB/WT selection
Feroceon L2 cache can work in eighther write through or write back mode
on Kirkwood. Add the option to configure this mode according to Kconfig.

Signed-off-by: Ronen Shitrit <rshitrit@marvell.com>
Signed-off-by: Nicolas Pitre <nico@marvell.com>
2008-09-25 16:29:21 -04:00
Albert Aribaud
3d014b01e5 [ARM] EDMINIV2: add support for LED and power button
This patch provides standard GPIO LED control
for the ED Mini V2, with software blinking only
(CPLD hardware blinking capability is not used).

This patch also provides status of the power
button as a standard GPIO input event.

Signed-off-by: Albert Aribaud <albert.aribaud@free.fr>
Signed-off-by: Nicolas Pitre <nico@marvell.com>
2008-09-25 16:27:22 -04:00
Lennert Buytenhek
2d0c9e735a [ARM] Kirkwood: allow configuring mbus window for on-chip sram
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
2008-09-25 16:27:22 -04:00
Ronen Shitrit
3a64ebc9ce [ARM] Kirkwood: remove uart1 init calls for boards that don't expose uart1
Remove uart1 init calls for boards that use the physical pins onto
which the UART1 signals are multiplexed for different purposes.

Signed-off-by: Ronen Shitrit <rshitrit@marvell.com>
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
2008-09-25 16:27:22 -04:00
Ronen Shitrit
b2b3dc2fc4 [ARM] Kirkwood: add support for newer SoC models
Add support to the Kirkwood port for newer device models and silicon
revisions.  Instead of looking at the DEVICE_ID register, the device
version is now determined by looking at the PCI-Express device ID and
revision registers, as it is done for orion5x, and this information
is used to determine the TCLK frequency, again, as it is done for
orion5x.

Signed-off-by: Ronen Shitrit <rshitrit@marvell.com>
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
2008-09-25 16:27:21 -04:00
Ronen Shitrit
79d4dd77f0 [ARM] Kirkwood: prepare for runtime-determined timer tick rate
Currently, kirkwood uses a hardcoded timer tick rate of 166 MHz, but
the actual timer tick rate varies between different members of the SoC
family.

This patch prepares for runtime determination of the timer tick rate.

Signed-off-by: Ronen Shitrit <rshitrit@marvell.com>
2008-09-25 16:27:21 -04:00
Lennert Buytenhek
0b0740c3c1 [ARM] Orion: add RD88F6183AP-GE support
The RD88F6183AP-GE is an access point reference design for the
88F6183 SoC, with a 88E6161 six-port gigabit ethernet switch with
five PHYs (providing 1 WAN and 4 LAN ports and an interface to the
CPU), and a mini-PCIe slot for a wireless card.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
2008-09-25 16:27:21 -04:00
Lennert Buytenhek
d323ade13b [ARM] Orion: add 88F6183 (Orion-1-90) support
The Orion-1-90 (88F6183) is another member of the Orion SoC family,
which has a 16 bit DDR2 interface, one x1 PCIe port (configurable as
Root Complex or Endpoint), one 10/100/1000 ethernet interface, one
USB 2.0 port with PHY, one SPDIF/I2S interface, one SDIO interface,
one TWSI interface, two UARTs, one SPI interface, a NAND controller,
a crypto engine, and a 4-channel DMA engine.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
2008-09-25 16:26:44 -04:00
Sylver Bruneau
104ea0781a [ARM] Orion: add Buffalo Terastation Pro II/Live support
This patch adds support for the Buffalo Terastation Pro II/Live.

Signed-off-by: Sylver Bruneau <sylver.bruneau@googlemail.com>
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
2008-09-25 16:26:44 -04:00
Christopher Moore
0eb66974c0 [ARM] Orion: add LaCie Ethernet Disk mini V2 support
This patch adds support for the LaCie Ethernet Disk mini V2.

Signed-off-by: Albert Aribaud <albert.aribaud@free.fr>
Signed-off-by: Christopher Moore <moore@free.fr>
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
2008-09-25 16:26:44 -04:00
Lennert Buytenhek
ebe35aff88 [ARM] Orion: prepare for runtime-determined timer tick rate
Currently, orion5x uses a hardcoded timer tick rate of 166 MHz, but
the actual timer tick rate varies between different members of the SoC
family (and can vary based on strap pin settings).

This patch prepares for runtime determination of the timer tick rate.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
2008-09-25 16:26:44 -04:00
Lennert Buytenhek
1f8081f539 [ARM] mv78xx0: wire up ethernet error interrupt
Wire up the ethernet port's error interrupt so that the
mv643xx_eth driver can sleep for SMI event completion instead of
having to busy-wait for it.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Signed-off-by: Nicolas Pitre <nico@marvell.com>
2008-09-25 16:26:44 -04:00
Lennert Buytenhek
144f814a43 [ARM] Kirkwood: wire up ethernet error interrupt
Wire up the ethernet port's error interrupt so that the
mv643xx_eth driver can sleep for SMI event completion instead of
having to busy-wait for it.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
2008-09-25 16:26:44 -04:00
Lennert Buytenhek
eeff6d8600 [ARM] Orion: wire up ethernet error interrupt
Wire up the ethernet port's error interrupt so that the
mv643xx_eth driver can sleep for SMI event completion instead of
having to busy-wait for it.

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
2008-09-25 16:26:43 -04:00
Russell King
afd1a321c4 [ARM] Update dma_map_sg()/dma_unmap_sg() API
Update the ARM DMA scatter gather APIs for the scatterlist changes.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 20:48:45 +01:00
Russell King
0ddbccd118 [ARM] dma: rename consistent.c to dma-mapping.c
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 15:59:19 +01:00
Russell King
2b4ae1f1d2 [ARM] Print details relevant to how we handle the cache
This replaces the original cache type decoding printks.  We now
indicate how we're treating the cache which we found, rather
than what we found.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 15:39:20 +01:00
Russell King
c0e9587841 [ARM] Introduce new bitmask based cache type macros
Rather than trying to (inaccurately) decode the cache type from the
registers each time we need to decide what type of cache we have,
use a bitmask initialized early during boot.

Since the setup is a one-off initialization, we can be a little more
clever and take account of the CPU architecture as well.

Note that we continue to achieve the compactness on optimised kernels
by forcing tests to always-false or always-true as appropriate, thereby
allowing the compiler to do build-time code elimination.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 15:35:28 +01:00
Russell King
90f1e08478 [ARM] Remove cache type printks
The cache type register found in ARMv5 and later CPUs changes format
and meaning depending on the CPU architecture version.  Currently,
this code:
a) doesn't work for everything - Xscale's are identified as
   'unknown 5'.
b) is not able to tell whether the caches are VIVT or VIPT from the
   cache type.
c) prints rubbish on some ARMv6 and ARMv7+ CPUs.

The two solutions to this are:
1. Add yet more code to decode and print the various different register
   formats.
2. Remove the code altogther.

The code only exists to decode and print the cache parameters.
Increasing the complexity of it just for the sake of a few prinks
isn't worth it.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 14:45:02 +01:00
Russell King
73b610affe [ARM] pxa: remove references to pxa_gpio_mode() in comments
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 13:35:27 +01:00
Bill Nottingham
f6476774f1 x86_64: be less annoying on boot
Remove mostly useless message on every boot.

Signed-off-by: Bill Nottingham <notting@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-25 11:15:20 +02:00
Russell King
b4ffb0edf4 [ARM] Update mach-types
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 10:12:25 +01:00
FUJITA Tomonori
1615965e54 x86 gart: remove unnecessary initialization
There is no point to have such initialization in struct dma_mapping_ops.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-25 11:02:27 +02:00
FUJITA Tomonori
1d99088215 x86: restore old GART alloc_coherent behavior
Currently, GART alloc_coherent tries to allocate pages with GFP_DMA32
for a device having dma_masks > 24bit < 32bits. If GART gets an
address that a device can't access to, GART try to map the address to
a virtual I/O address that the device can access to.

But Andi pointed out, "The GART is somewhere in the 4GB range so you
cannot use it to map anything < 4GB. Also GART is pretty small."

http://lkml.org/lkml/2008/9/12/43

That is, it's possible that GART doesn't have virtual I/O address
space that a device can access to. The above behavior doesn't work for
a device having dma_masks > 24bit < 32bits.

This patch restores old GART alloc_coherent behavior (before the
alloc_coherent rewrite).

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-25 11:02:26 +02:00
FUJITA Tomonori
ecef533ea6 revert "x86: make GART to respect device's dma_mask about virtual mappings"
This reverts:

commit bee44f294e
Author: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date:   Fri Sep 12 19:42:35 2008 +0900

    x86: make GART to respect device's dma_mask about virtual mappings

I wrote the above commit to fix a GART alloc_coherent regression, that
can't handle a device having dma_masks > 24bit < 32bits, introduced by
the alloc_coherent rewrite:

  http://lkml.org/lkml/2008/8/12/200

After the alloc_coherent rewrite, GART alloc_coherent tried to
allocate pages with GFP_DMA32. If GART got an address that a device
can't access to, GART mapped the address to a virtual I/O address. But
GART mapping mechanism didn't take account of dma mask, so GART could
use a virtual I/O address that the device can't access to again.

Alan pointed out:

" This is indeed a specific problem found with things like older
  AACRAID where control blocks must be below 31bits and the GART
  is above 0x80000000. "

The above commit modified GART mapping mechanism to take care of dma
mask. But Andi pointed out, "The GART is somewhere in the 4GB range so
you cannot use it to map anything < 4GB. Also GART is pretty small."

http://lkml.org/lkml/2008/9/12/43

That means it's possible that GART doesn't have virtual I/O address
space that a device can access to. The above commit (to modify GART
mapping mechanism to take care of dma mask) can't fix the regression
reliably so let's avoid making GART more complicated.

We need a solution that always works for dma_masks > 24bit <
32bits. That's how GART worked before the alloc_coherent rewrite.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-25 11:02:25 +02:00
FUJITA Tomonori
9f6ac57729 x86: export pci-nommu's alloc_coherent
This patch exports nommu_alloc_coherent (renamed
dma_generic_alloc_coherent). GART needs this function.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-25 11:02:25 +02:00
Eric Miao
da1a3dc0eb [ARM] pxa: introduce cpu_is_pxa26x()
PXA26x (PXA261/262) is actually a PXA250 with stacked Intel(R)
StartaFlash. And this can be decided by bit 3 (PKG_TYPE) of
BOOT_DEF register.

Due to this extra I/O register access, make cpu_is_pxa26x() a
public function instead of a macro.

Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 09:54:56 +01:00
Eric Miao
0ffcbfd54e [ARM] pxa: make cpu_is_pxa2* macros more consistent
1. add a CPUID table in the comment

2. make cpu_is_pxa25x() true for PXA210/250/255/26x

3. PXA210 is treated as PXA25x, all related code modified to
   reflect this

Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 09:54:55 +01:00
Eric Miao
2b12797c59 [ARM] pxa: merge common cpu_is_pxa255() code together
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 09:54:54 +01:00
Eric Miao
ff7a4c7130 [ARM] corgi_lcd: use GPIO API for BACKLIGHT_ON and BACKLIGHT_CONT
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-09-25 09:38:14 +01:00
Hiroshi Shimamoto
4694d23912 x86: signal_32.c: introduce macro ia32_setup_frame and ia32_setup_rt_frame
Make 32-bit setup_rt_frame() look like 64-bit version for unification.

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-25 10:11:42 +02:00
Hiroshi Shimamoto
455edbc423 x86: signal: introduce helper macro is_ia32
Introduce new macro is_ia32 for unification of setup_rt_frame().
No effect in binary, compiler will optimize.

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-25 10:11:41 +02:00
Hiroshi Shimamoto
b94fd69827 x86: signal_64.c: introduce helper function signr_convert()
This helper function is for unification of setup_rt_frame().
No effect in binary.

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-25 10:11:40 +02:00
Hiroshi Shimamoto
8d8c13bdb5 x86: signal_32.c: introduce signr_convert()
Introduce signr_convert().
This function will help unification of setup_rt_frame().

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-09-25 10:11:38 +02:00
Jeff Garzik
ae19161e28 Merge branch 'for-2.6.28' of git://git.marvell.com/mv643xx_eth into upstream-next 2008-09-24 20:40:52 -04:00
Linus Torvalds
c0f4d6d4b1 Merge branch 'timers-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'timers-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86: prevent stale state of c1e_mask across CPU offline/online, fix
2008-09-24 16:39:50 -07:00
Linus Torvalds
1126388609 Merge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus
* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus:
  [MIPS] Fixe the definition of PTRS_PER_PGD
  [MIPS] au1000: Fix gpio direction
2008-09-24 16:38:52 -07:00
David Howells
08ec3c2d45 MN10300: Make sched_clock() report time since boot
Make sched_clock() report time since boot rather than time since last
timer interrupt.

Make sched_clock() expand and scale the 32-bit TSC value running at
IOCLK speed (~33MHz) to a 64-bit nanosecond counter, using cnt32_to_63()
acquired from the ARM arch and without using slow DIVU instructions
every call.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-09-24 16:38:17 -07:00
David Howells
b4f151ff89 MN10300: Move asm-arm/cnt32_to_63.h to include/linux/
Move asm-arm/cnt32_to_63.h to include/linux/ so that MN10300 can make
use of it too.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-09-24 16:38:17 -07:00
Becky Bruce
4ee7084eb1 POWERPC: Allow 32-bit hashed pgtable code to support 36-bit physical
This rearranges a bit of code, and adds support for
36-bit physical addressing for configs that use a
hashed page table.  The 36b physical support is not
enabled by default on any config - it must be
explicitly enabled via the config system.

This patch *only* expands the page table code to accomodate
large physical addresses on 32-bit systems and enables the
PHYS_64BIT config option for 86xx.  It does *not*
allow you to boot a board with more than about 3.5GB of
RAM - for that, SWIOTLB support is also required (and
coming soon).

Signed-off-by: Becky Bruce <becky.bruce@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
2008-09-24 16:29:44 -05:00
Kumar Gala
9a62c05180 powerpc/mm: Implement _PAGE_SPECIAL & pte_special() for 32-bit
Implement _PAGE_SPECIAL and pte_special() for 32-bit powerpc. This bit will
be used by the fast get_user_pages() to differenciate PTEs that correspond
to a valid struct page from special mappings that don't such as IO mappings
obtained via io_remap_pfn_ranges().

We currently only implement this on sub-arch that support SMP or will so
in the future (6xx, 44x, FSL-BookE) and not (8xx, 40x).

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2008-09-24 16:29:42 -05:00
Kumar Gala
9bf2b5cdc5 powerpc: Fixes for CONFIG_PTE_64BIT for SMP support
There are some minor issues with support 64-bit PTEs on a 32-bit processor
when dealing with SMP.

* We need to order the stores in set_pte_at to make sure the flag word
  is set second.
* Change pte_clear to use pte_update so only the flag word is cleared
* Added a WARN_ON to set_pte_at to ensure the pte isn't present for
  the 64-bit pte/SMP case (to ensure our assumption of this fact).

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Acked-by: Becky Bruce <becky.bruce@freescale.com>
2008-09-24 16:29:41 -05:00
Kumar Gala
0ba3418b8b powerpc: Introduce local (non-broadcast) forms of tlb invalidates
Introduced a new set of low level tlb invalidate functions that do not
broadcast invalidates on the bus:

_tlbil_all - invalidate all
_tlbil_pid - invalidate based on process id (or mm context)
_tlbil_va  - invalidate based on virtual address (ea + pid)

On non-SMP configs _tlbil_all should be functionally equivalent to _tlbia and
_tlbil_va should be functionally equivalent to _tlbie.

The intent of this change is to handle SMP based invalidates via IPIs instead
of broadcasts as the mechanism scales better for larger number of cores.

On e500 (fsl-booke mmu) based cores move to using MMUCSR for invalidate alls
and tlbsx/tlbwe for invalidate virtual address.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
2008-09-24 16:29:40 -05:00
Becky Bruce
b9579689ad powerpc: Make dma_addr_t a u64 if CONFIG_PHYS_64BIT is set
Signed-off-by: Becky Bruce <becky.bruce@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
2008-09-24 16:26:45 -05:00
Becky Bruce
4fc665b88a powerpc: Merge 32 and 64-bit dma code
We essentially adopt the 64-bit dma code, with some changes to support
32-bit systems, including HIGHMEM.  dma functions on 32-bit are now
invoked via accessor functions which call the correct op for a device based
on archdata dma_ops.  If there is no archdata dma_ops, this defaults
to dma_direct_ops.

In addition, the dma_map/unmap_page functions are added to dma_ops
because we can't just fall back on map/unmap_single when HIGHMEM is
enabled. In the case of dma_direct_*, we stop using map/unmap_single
and just use the page version - this saves a lot of ugly
ifdeffing.  We leave map/unmap_single in the dma_ops definition,
though, because they are needed by the iommu code, which does not
implement map/unmap_page.  Ideally, going forward, we will completely
eliminate map/unmap_single and just have map/unmap_page, if it's
workable for 64-bit.

Signed-off-by: Becky Bruce <becky.bruce@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
2008-09-24 16:26:45 -05:00
Becky Bruce
8fae035324 powerpc: Drop archdata numa_node
Use the struct device's numa_node instead; use accessor functions
to get/set numa_node.

Signed-off-by: Becky Bruce <becky.bruce@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
2008-09-24 16:26:43 -05:00