Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Clean up DMA timeout handling for HPT370:
- hpt370_lostirq_timeout() cleared the DMA status which made __ide_dma_end()
called afterwards return the incorrect result, and the DMA engine was reset
both before and after stopping DMA while the HighPoint drivers only do it
after (which seems logical) -- fix this and also rename the function;
- get rid of the needless mutual recursion in hpt370_ide_dma_end() and
hpt370_ide_dma_timeout();
- get rid of hpt370_lostirq_timeout() since hwif->ide_dma_end() called from
the driver's interrupt handler later does all its work.
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Continue with the driver rewrite:
- move the interrupt twiddling code from the speedproc handlers into the
init_hwif_hpt366 which allows to merge the two HPT37x speedproc handlers
into one;
- get rid of in init_hpt366 which solely consists of the duplicate code, then
fold init_hpt37x() into init_chipset_hpt366();
- fix hpt3xx_tune_drive() to always set the PIO mode requested, not the best
possible one, change hpt366_config_drive_xfer_rate() accordingly, simplify
it a bit;
- group all the DMA related code together init_hwif_hpt366(), and generally
clean up and beautify it.
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Begin the real driver redesign. For the starters:
- cache the offset of the IDE channel's MISC. control registers which are used
throughout the driver in hwif->select_data;
- only touch the relevant MCR when detecting the cable type on HPT374's
function 1;
- make HPT36x's speedproc handler look the same way as HPT37x ones; fix the
PIO timing register mask for HPT37x.
- rename all the HPT3xx register related variables consistently; clean up the
whitespace.
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Switch to using pci_get_slot() to get to the function 1 of HPT36x/374 chips --
there's no need for the driver itself to walk the list of the PCI devices, and
it also forgets to check the bus number of the device found.
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
- Rework the driver setup code so that it prefixes the driver startup
messages with the real chip name.
- Print the measured f_CNT value and the DPLL setting for non-HPT3xx
chips as well.
- Claim the extra 240 bytes of I/O space for all chips, not only for
those having PCI device ID of 0x0004.
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
- Rework hpt3xx_ratemask() and hpt3xx_ratefilter() so that the former
returns the max. mode computed at the load time and doesn't have to do
bad Ultra33 drive list lookups anymore; remove the duplicate code from
the latter function. Move the quirky drive list lookup into
hpt3xx_quirkproc() where it should have been from the start...
- Disable UltraATA/100 for HPT370 by default as the 33 MHz ATA clock
being used does not allow for it, and this *greatly* increases the
transfer speed.
- Save some space by using byte-wide fields in struct hpt_info; switch to
reading the 8-bit PCI revision ID reg. only, not the whole 32-bit reg.
- Start incrementing the driver version number with each patch (should
have been done from the first one posted).
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/shaggy/jfs-2.6:
JFS: Remove incorrect kgdb define
JFS: call io_schedule() instead of schedule() to avoid deadlock
JFS: Add lockdep annotations
JFS: Avoid BUG() on a damaged file system
* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw: (57 commits)
[GFS2] make gfs2_writepages() static
[GFS2] Unlock page on prepare_write try lock failure
[GFS2] nfsd readdirplus assertion failure
[DLM] fix softlockup in dlm_recv
[DLM] zero new user lvbs
[DLM/GFS2] indent help text
[GFS2] Fix unlink deadlocks
[GFS2] Put back semaphore to avoid umount problem
[GFS2] more CURRENT_TIME_SEC
[GFS2/DLM] fix GFS2 circular dependency
[GFS2/DLM] use sysfs
[GFS2] make lock_dlm drop_count tunable in sysfs
[GFS2] increase default lock limit
[GFS2] Fix list corruption in lops.c
[GFS2] Fix recursive locking attempt with NFS
[DLM] can miss clearing resend flag
[DLM] saved dlm message can be dropped
[DLM] Make sock_sem into a mutex
[GFS2] Fix typo in glock.c
[GFS2] use CURRENT_TIME_SEC instead of get_seconds in gfs2
...
On Mon, Jan 29, 2007 at 08:45:28PM -0800, Andrew Morton wrote:
>...
> Changes since 2.6.20-rc6-mm2:
>...
> git-gfs2-nmw.patch
>...
> git trees
>...
This patch makes the needlessly global gfs2_writepages() static.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
When the try lock of the glock failed in prepare_write we were
incorrectly exiting this function with the page still locked.
This was resulting in further I/O to this page hanging.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
When we do full FP emulation its possible that we need to post a SIGFPE based
on the results of the emulation. The previous code ignored this case completely.
Additionally, the Soft_emulate_8xx case had two issues. One, we should never
generate a SIGFPE since the code only does data movement. Second, we were
interpreting the return codes incorrectly, it returns 0 on success, 1 on
illop and -EFAULT on a data access error.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Anytime we are emulating an instruction we are going to be doing some form of
get_user() to get the instruction image to decode. Since get_user() might
sleep we need to ensure we have interrupts enabled or we might see something
like:
Debug: sleeping function called from invalid context at arch/powerpc/kernel/traps.c:697
in_atomic():0, irqs_disabled():1
Call Trace:
[D6023EB0] [C0007F84] show_stack+0x58/0x174 (unreliable)
[D6023EE0] [C0022C34] __might_sleep+0xbc/0xd0
[D6023EF0] [C000D158] program_check_exception+0x1d8/0x4fc
[D6023F40] [C000E744] ret_from_except_full+0x0/0x4c
--- Exception: 700 at 0x102a7100
LR = 0xdb9ef04
However, we want to ensure that interrupts are disabled when handling a trap
exception that might be used for a kernel breakpoint. This is why ProgramCheck
is marked as EXC_XFER_STD instead of EXC_XFER_EE.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Added kprobes to ppc32 platforms that have use single_step_exception. This
excludes 4xx and anything Book-E since their debug mechanisms for single stepping
are completely different.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
H_BULK_REMOVE lets us remove 4 entries from the MMU hash table with one
hypervisor call. This uses it in pSeries_lpar_hpte_invalidate so we
can tear down mappings with fewer hypervisor calls.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Some instruction tracing tools use the RI (recoverable interrupt) bit
in the MSR to indicate when it's safe to single-step. Currently we
clear RI after restoring r13 when returning to userspace. However,
if we single-step past the point where r13 is restored, we'll corrupt
r13 in the exception entry code and not restore it. This moves the
clearing of RI to just before r13 is restored so this doesn't happen.
Signed-off-by: Paul Mackerras <paulus@samba.org>
To the issue: some point during 2.6.20 development, Paul Mackerras
introduced the "lazy IRQ disabling" patch (very cool work, BTW).
In that patch, the performance monitor unit exception was marked as
"maskable", in the sense that if interrupts were soft-disabled, that
exception could be ignored. This broke my PowerPC profiling code.
The symptom that I see is that a varying number of interrupts
(from 0 to $n$, typically closer to 0) get delivered, when, in
reality, it should always be very close to $n$.
The issue stems from the way masking is being done. Masking in
this fashion seems to work well with the decrementer and external
interrupts, because they are raised again until "really" handled.
For the PMU, however, this does not apply (at least on my Xserver
machine with a 970FX processor). If the PMU exception is not handled,
it will _not_ be re-raised (at least on my machine). The documentation
states that the PMXE bit in MMCR0 is set to 0 when the PMU exception
is raised. However, software must re-set the bit to re-enable PMU
exceptions. If the exception is ignored (as currently) not only is
that interrupt lost, but because software does not re-set PMXE, the
PMU registers are "frozen" forever.
[This patch means that performance monitor exceptions are taken and
handled even if irqs are off, as long as some other interrupt hasn't
come along and caused interrupts to be hard-disabled. In this sense
the PMU exception becomes like an NMI. The oprofile code for most
powerpc processors does nothing that is unsafe in an NMI context, but
the Cell oprofile code does a spin_lock_irqsave. However, that turns
out to be OK because Cell doesn't actually use the performance
monitor exception; performance monitor interrupts come in as a
regular interrupt on Cell, so will be disabled when irqs are off.
-- paulus.]
Signed-off-by: Paul Mackerras <paulus@samba.org>
The new dcr code does not currently compile when configured for native
DCR access on ARCH=powerpc. This patch fixes the problems.
Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Some systems supported by the maple platform (e.g. JS2x blades running
SLOF) are able to use the mmio_nvram backend for reading and writing
nvram. This is an improvement over the current situation -- no nvram
access from userspace at all.
Select MMIO_NVRAM for the maple platform.
Initialize the mmio_nvram backend from maple setup code.
Signed-off-by: Nathan Lynch <ntl@pobox.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Base pasemi defconfig. Nothing special, just the native drivers plus
common PCI-express/PCI cards.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The DMA controller on PWRficient is somewhat special -- has a PCI header
so it looks like it's on the root PCI (-Express) root bus, but it uses
more than the default number of interrupts (and they are hardwired).
We need to wire up all interrupts for the DMA controller. The generic
IRQ code will only map the primary interrupt from the PCI header (128),
so add 129->211 by hand.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Timebase update is simple on PA6T, since global updates can be done from
one core by writing to an SPR.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Implement reset on platforms/pasemi. Default is just to reset the
cpu using the SDC registers.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Powersave support on PA6T. Right now it only uses 'doze' mode, and
will default to no savings (spin).
Signed-off-by: Olof Johansson <olof@lixom.net>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Early debug output for PA Semi UART. Uses the 2.05 CI real mode ops.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
G3 Blue & White is misconfigured by default so that CardBus controllers
in PCI slots don't work. The PCI bridge is programmed to only allow
access to bus 1 but not higher busses.
The patch forces the PCI busses to be reassigned if a Grackle controller
is found and the machine identifies itself as "PowerMac1,1"
Signed-off-by: Pavel Roskin <proski@gnu.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This patch adds base support for Celleb platform.
Signed-off-by: Kou Ishizaki <kou.ishizaki@toshiba.co.jp>
Acked-by: Arnd Bergmann <arnd.bergmann@de.ibm.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Spu management ops in arch/platforms/cell/spu_priv1_mmio.h can be used
commonly in of based platform. This patch separates spu management ops
from native cell code and uses on celleb platform.
Signed-off-by: Arnd Bergmann <arnd.bergmann@de.ibm.com>
Signed-off-by: Kou Ishizaki <kou.ishizaki@toshiba.co.jp>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This patch adds udbg support for Celleb platform.
Signed-off-by: Kou Ishizaki <kou.ishizaki@toshiba.co.jp>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This patch creates Celleb platform dependent file to support iommu.
Signed-off-by: Kou Ishizaki <kou.ishizaki@toshiba.co.jp>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
USB host controller in SCC requires enable sequence. It should be done
before USB host drivers start.
Signed-off-by: Kou Ishizaki <kou.ishizaki@toshiba.co.jp>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Default config file for mpc8272ads (powerpc port).Though relevant bits went
in, it is required to keep proper default configuration for the target,
which seems to be missed initially.
Signed-off-by: Vitaly Bordug <vbordug@ru.mvista.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Recent update of asm-powerpc/io.h caused cpm-related stuff to break in the
current kernel. Current patch fixes it, as well as other inconsistencies
expressed, that do not permit targets from working properly:
- Updated dts with a chosen node with interrupt controller,
- fixed messed device IDs among CPM2 SoC devices,
- corrected odd header name and fixed type in defines,
- Added 82xx subdir to the powerpc/platforms Makefile, missed during
initial commit,
- new solely-powerpc header file for 8260 family (was using one from
arch/ppc, this one cleaned up from the extra stuff), in fact for now
a placeholder to get the board-specific includes for stuff not yet
capable to live with devicetree peeks only
- Fixed couple of misprints in reference mpc8272 dts.
Signed-off-by: Vitaly Bordug <vbordug@ru.mvista.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This contains important fixes for the CPM2 PIC code. Eliminated
CPM_IRQ_OFFSET, pulling the respective interrupt numbers from the interrupt
mapping. Updated devicetree files to reflect that. Changed direct
IC-related IO accesses to the IO accessors. Fixed all the sense values to
keep coherency with ipic. In the current code, CPM2 stuff will have no IRQs
and hence could be hardly usable.
Signed-off-by: Vitaly Bordug <vbordug@ru.mvista.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Update ps3_defconfig to enable USB mass storage and VFAT.
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Acked-by: Arnd Bergmann <arnd.bergmann@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Add the missing pieces to support DMA scatter-gather on the PS3 system bus.
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Move the structures and routines needed for PS3 vuart port device registration
to asm-powerpc/ps3.h.
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Acked-by: Arnd Bergmann <arnd.bergmann@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Move the PS3 system bus routines from drivers/ps3 to
arch/powerpc/platforms/ps3.
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Arnd Bergmann <arnd.bergmann@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
