Jack Ren and Eric Miao tracked down the following long standing
problem in the NOHZ code:
scheduler switch to idle task
enable interrupts
Window starts here
----> interrupt happens (does not set NEED_RESCHED)
irq_exit() stops the tick
----> interrupt happens (does set NEED_RESCHED)
return from schedule()
cpu_idle(): preempt_disable();
Window ends here
The interrupts can happen at any point inside the race window. The
first interrupt stops the tick, the second one causes the scheduler to
rerun and switch away from idle again and we end up with the tick
disabled.
The fact that it needs two interrupts where the first one does not set
NEED_RESCHED and the second one does made the bug obscure and extremly
hard to reproduce and analyse. Kudos to Jack and Eric.
Solution: Limit the NOHZ functionality to the idle loop to make sure
that we can not run into such a situation ever again.
cpu_idle()
{
preempt_disable();
while(1) {
tick_nohz_stop_sched_tick(1); <- tell NOHZ code that we
are in the idle loop
while (!need_resched())
halt();
tick_nohz_restart_sched_tick(); <- disables NOHZ mode
preempt_enable_no_resched();
schedule();
preempt_disable();
}
}
In hindsight we should have done this forever, but ...
/me grabs a large brown paperbag.
Debugged-by: Jack Ren <jack.ren@marvell.com>,
Debugged-by: eric miao <eric.y.miao@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Compiling ppc64_defconfig with gcc 4.3 gives thes warnings:
arch/powerpc/sysdev/mpic.c: In function 'mpic_irq_get_priority':
arch/powerpc/sysdev/mpic.c:1351: warning: 'is_ipi' may be used uninitialized in this function
arch/powerpc/sysdev/mpic.c: In function 'mpic_irq_set_priority':
arch/powerpc/sysdev/mpic.c:1328: warning: 'is_ipi' may be used uninitialized in this function
It turns out that in the cases where is_ipi is uninitialized, another
variable (mpic) will be NULL and it is dereferenced. Protect against
this by returning if mpic is NULL in mpic_irq_set_priority, and removing
mpic_irq_get_priority completely as it has no in tree callers.
This has the nice side effect of making the warning go away.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Since commit "85xx: Add support for relocatable kernel (and
booting at non-zero)" (37dd2badcf),
PHYSICAL_START is #defined as kernstart_addr if RELOCATABLE
and FLATMEM is enabled.
PHYSICAL_START is used in prom_init.c and so kernstart_addr
needs to be added to the list of allowed symbols that
prom_init.c can access.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
__set_fixmap() in pgtable_32.c currently fails to compile if
STRICT_MM_TYPECHECKS is defined. This fixes it.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
A change was made to walk_memory_resource() in commit
4b119e21d0 that added a
check of find_lmb(). Add the coresponding lmb_add()
call to ps3_mm_add_memory() so that that check will
succeed.
This fixes the condition where the PS3 boots up with
only the 128 MiB of boot memory, and doesn't see the
other 128MiB that is available.
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
dtbImage.* and several zImage. targets get created but never cleaned up.
Also, moved zImage to the clean-files line associated with all other image
results (was previously duplicated).
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
[PATCH] return to old errno choice in mkdir() et.al.
[Patch] fs/binfmt_elf.c: fix wrong return values
[PATCH] get rid of leak in compat_execve()
[Patch] fs/binfmt_elf.c: fix a wrong free
[PATCH] avoid multiplication overflows and signedness issues for max_fds
[PATCH] dup_fd() part 4 - race fix
[PATCH] dup_fd() - part 3
[PATCH] dup_fd() part 2
[PATCH] dup_fd() fixes, part 1
[PATCH] take init_files to fs/file.c
This removes a CVS keyword that wasn't updated for a long time from a
comment.
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The PowerPC 440EP, 440GR, 440EPx, and 440GRx chips have an issue that
causes the PLB3-to-PLB4 bridge to wait indefinitely for transaction
requests that cross the end-of-memory-range boundary. Since the DDR
controller only returns the valid portion of a read request, the bridge
will prevent other PLB masters from completing their transactions.
This implements the recommended workaround for this errata for chips that
use older versions of firmware that do not already handle it. The last
4KiB of memory are hidden from the kernel to prevent the problem
transactions from occurring.
Signed-off-by: Josh Boyer <jwboyer@linux.vnet.ibm.com>
Acked-by: Stefan Roese <sr@denx.de>
Signed-off-by: Josh Boyer <jwboyer@linux.vnet.ibm.com>
The old device/driver matching scheme is going away so stop using it.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Olof Johansson <olof@lixom.net>
Fix following warnings:
WARNING: arch/powerpc/platforms/cell/built-in.o(.devinit.text+0x9c): Section mismatch in reference from the function .cell_setup_phb() to the function .init.text:.iowa_register_bus()
WARNING: arch/powerpc/platforms/cell/built-in.o(.devinit.text+0xa4): Section mismatch in reference from the function .cell_setup_phb() to the function .init.text:.io_workaround_init()
Signed-off-by: Kou Ishizaki <kou.ishizaki@toshiba.co.jp>
Signed-off-by: Paul Mackerras <paulus@samba.org>
With CONFIG_VIRT_CPU_ACCOUNTING disabled, I got the following error:
linux-2.6/arch/powerpc/platforms/cell/spufs/file.c: In function 'spu_switch_log_notify':
linux-2.6/arch/powerpc/platforms/cell/spufs/file.c:2542: error: implicit declaration of function 'get_tb'
make[4]: *** [arch/powerpc/platforms/cell/spufs/file.o] Error 1
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This changes vmemmap to use a different region (region 0xf) of the
address space, and to configure the page size of that region
dynamically at boot.
The problem with the current approach of always using 16M pages is that
it's not well suited to machines that have small amounts of memory such
as small partitions on pseries, or PS3's.
In fact, on the PS3, failure to allocate the 16M page backing vmmemmap
tends to prevent hotplugging the HV's "additional" memory, thus limiting
the available memory even more, from my experience down to something
like 80M total, which makes it really not very useable.
The logic used by my match to choose the vmemmap page size is:
- If 16M pages are available and there's 1G or more RAM at boot,
use that size.
- Else if 64K pages are available, use that
- Else use 4K pages
I've tested on a POWER6 (16M pages) and on an iSeries POWER3 (4K pages)
and it seems to work fine.
Note that I intend to change the way we organize the kernel regions &
SLBs so the actual region will change from 0xf back to something else at
one point, as I simplify the SLB miss handler, but that will be for a
later patch.
Signed-off-by: Paul Mackerras <paulus@samba.org>
If victim (not ctx) is in spu_run, add victim to rq.
Signed-off-by: Luke Browning <lukebrowning@us.ibm.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The following adds local bus, flash and MTD partition nodes for
sbc8548. As well, a compatible field for the soc node, so that
of_platform_bus_probe() will pick it up.
Something that is provided through this newly added epld node
is the Hardware Revision which is now being utilized.
Signed-off-by: Jeremy McNicoll <jeremy.mcnicoll@windriver.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Zhang Wei <wei.zhang@freescale.com>
Cc: "Nelson, Shannon" <shannon.nelson@intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
DIU platform code should not just write to the PIXIS' BRDCFG0 register,
it should set and clear its own bits only, otherwise it will break
firmware setup (in fact it breaks second uart).
Also get rid of magic numbers in the related code.
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
This patch adds device tree nodes for NOR and NAND flashes and places
board-control node inside the localbus.
defconfig and board file updated appropriately.
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
The 8568 MDS needs some configuration changes to the PHY in order to
work properly. These are done in the firmware, normally, but Linux
shouldn't need to rely on the firmware running such things (someone
could disable the PHY support in the firmware to save space, for instance).
Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
As of current mainline tree, TULIP driver is unusable on MPC8610HPCD
boards. There is a patch[1] floating around (and also included in the
BSP), which tries to heal the situation, though the ethernet is still
unusable. Practically it takes ages to mount NFS filesystem:
VFS: Mounted root (nfs filesystem).
Freeing unused kernel memory: 180k init
nfs: server 10.0.0.2 not responding, still trying
nfs: server 10.0.0.2 OK
nfs: server 10.0.0.2 not responding, still trying
nfs: server 10.0.0.2 not responding, still trying
nfs: server 10.0.0.2 not responding, still trying
nfs: server 10.0.0.2 not responding, still trying
nfs: server 10.0.0.2 OK
nfs: server 10.0.0.2 not responding, still trying
So, instead of trying to add uli526x functionality into TULIP driver
(which is already bloated enough), I fixed existing ULI526X driver
and now it works perfectly well here.
[1] http://www.bitshrine.org/gpp/0024-MPC8610-ETH-Lyra-native-ethernet.txt
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
This fixes a few more miscellaneous compile problems with ARCH=ppc.
1. Don't compile devres.c on ARCH=ppc, it doesn't have ioremap_flags.
2. Include <asm/irq.h> in setup.c for the __DO_IRQ_CANON definition.
3. Include <linux/proc_fs.h> in residual.c for the
definition of create_proc_read_entry.
4. Fix xchg_ptr to be a static inline to eliminate a compiler warning.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Commit 76bc080ef5 ("POWERPC] Make default
cputable entries reflect selected CPU family") added default entries
for the e200 and e500 families, but missed a closing brace on those
entries, as pointed out by David Gibson. This adds the closing braces.
Signed-off-by: Paul Mackerras <paulus@samba.org>
* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc: (23 commits)
[POWERPC] Remove leftover printk in isa-bridge.c
[POWERPC] Remove duplicate #include
[POWERPC] Initialize lockdep earlier
[POWERPC] Document when printk is useable
[POWERPC] Fix bogus paca->_current initialization
[POWERPC] Fix of_i2c include for module compilation
[POWERPC] Make default cputable entries reflect selected CPU family
[POWERPC] spufs: lockdep annotations for spufs_dir_close
[POWERPC] spufs: don't requeue victim contex in find_victim if it's not in spu_run
[POWERPC] 4xx: Fix PCI mem in sequoia DTS
[POWERPC] 4xx: Add endpoint support to 4xx PCIe driver
[POWERPC] 4xx: Fix problem with new TLB storage attibute fields on 440x6 core
[POWERPC] spufs: spu_create should send inotify IM_CREATE event
[POWERPC] spufs: handle faults while the context switch pending flag is set
[POWERPC] spufs: fix concurrent delivery of class 0 & 1 exceptions
[POWERPC] spufs: try to route SPU interrupts to local node
[POWERPC] spufs: set SPU_CONTEXT_SWITCH_PENDING before synchronising SPU irqs
[POWERPC] spufs: don't acquire state_mutex interruptible while performing callback
[POWERPC] spufs: update master runcntl with context lock held
[POWERPC] spufs: fix post-stopped update of MFC_CNTL register
...
This printk() appears twice in the same function. Only the latter one
in the inval_range: section appears to be legitimate.
Signed-off-by: Nate Case <ncase@xes-inc.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Remove duplicate #include of <asm/prom.h> in
arch/powerpc/kernel/btext.c.
Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This moves lockdep_init() to before udbg_early_init() as the later
can call things that acquire spinlocks etc... This also makes printk
safer to use earlier.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
When debugging early boot problems, it's common to sprinkle printk's
all over the place. However, on 64-bit powerpc, this can lead to
memory corruption if done too early due to the PACA pointer and
lockdep core not being initialized.
This adds some comments to early_setup() that document when it is
safe to do so in order to save time for whoever has to debug that
stuff next.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
When doing lockdep, I had two patches to initialize paca->_current
early, one bogus, and one correct. Unfortunately both got merged
as the bad one ended up being part of the main lockdep patch by
mistake. This causes memory corruption at boot. This removes
the offending code.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Changes the cputable so that various CPU families that have an exclusive
CONFIG_ option have a more sensible default entry to use if the specific
processor hasn't been identified.
This makes the kernel more generally useful when booted on an unknown
PVR for things like new 4xx variants.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (32 commits)
net: Added ASSERT_RTNL() to dev_open() and dev_close().
can: Fix can_send() handling on dev_queue_xmit() failures
netns: Fix arbitrary net_device-s corruptions on net_ns stop.
netfilter: Kconfig: default DCCP/SCTP conntrack support to the protocol config values
netfilter: nf_conntrack_sip: restrict RTP expect flushing on error to last request
macvlan: Fix memleak on device removal/crash on module removal
net/ipv4: correct RFC 1122 section reference in comment
tcp FRTO: SACK variant is errorneously used with NewReno
e1000e: don't return half-read eeprom on error
ucc_geth: Don't use RX clock as TX clock.
cxgb3: Use CAP_SYS_RAWIO for firmware
pcnet32: delete non NAPI code from driver.
fs_enet: Fix a memory leak in fs_enet_mdio_probe
[netdrvr] eexpress: IPv6 fails - multicast problems
3c59x: use netstats in net_device structure
3c980-TX needs EXTRA_PREAMBLE
fix warning in drivers/net/appletalk/cops.c
e1000e: Add support for BM PHYs on ICH9
uli526x: fix endianness issues in the setup frame
uli526x: initialize the hardware prior to requesting interrupts
...
We need to acquire the parent i_mutex with I_MUTEX_PARENT to keep
lockdep happy.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
We should not requeue the victim context in find_victim if the owner is
not in spu_run. It's first not needed because leaving the context on
the spu is an optimization and second is harmful because it means the
owner could re-enter spu_run when the context is on the runqueue and
trip the BUG_ON in __spu_update_sched_info.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
This patch is fixes the sequoia.dts device tree file to use the values defined
in the 440Epx data sheet from AMCC.
That fixes an issue where some devices, including graphics cards, would not
initialize properly because the PCI resource space was not big enough.
Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
Signed-off-by: Josh Boyer <jwboyer@linux.vnet.ibm.com>
This patch adds basic endpoint support to the 4xx PCIe driver.
This is done by checking the device_type property of the PCIe
device node ("pci" for root-complex and "pci-endpoint" for endpoint
configuration).
Note: Currently we map a fixed 64MByte window to PLB address 0 (SDRAM).
This should probably be configurable via a dts property.
Signed-off-by: Stefan Roese <sr@denx.de>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Josh Boyer <jwboyer@linux.vnet.ibm.com>
The new 440x6 core used on AMCC 460EX/GT introduces new storage attibure
fields to the TLB2 word. Those are:
Bit 11 12 13 14 15
WL1 IL1I IL1D IL2I IL2D
With these bits the cache (L1 and L2) can be configured in a more flexible
way, instruction- and data-cache independently now. The "old" I and W bits
are still available and setting these old bits will automically set these
new bits too (for backward compatibilty).
The current code does not clear these fields resulting in disabling the cache
by chance. This patch now makes sure that these new bits are cleared when
the TLB2 word is written.
Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Josh Boyer <jwboyer@linux.vnet.ibm.com>
Creating a spufs context or gand using spu_create should send an inotify
event so that things like performance monitors have an easy way to find
out about newly created contexts.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
Simply replace proc_create and further data assigned with proc_create_data.
No need to check for data!=NULL after that.
Signed-off-by: Denis V. Lunev <den@openvz.org>
Cc: Alexey Dobriyan <adobriyan@openvz.org>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
We provide an ioremap_flags, so this provides a corresponding
devm_ioremap_prot. The slight name difference is at Ben
Herrenschmidt's request as he plans on changing ioremap_flags to
ioremap_prot in the future.
Signed-off-by: Emil Medve <Emilian.Medve@Freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Acked-by: Tejun Heo <htejun@gmail.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Currently, page fault handlers don't issue a mfc restart if the context
switch pending flag is set, which can leave us with a hanging DMA after
a context restore.
This patch introduces fault pending flag that is set by the fault
handler and read by the context switch code, so that the latter can add
the restart bit at the right spot, after it has successfuly saved the
state of the mfc control register.
Signed-off-by: Luke Browning <lukebr@linux.vnet.ibm.com>
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
SPU class 0 & 1 exceptions may occur in parallel, so we may end up
overwriting csa.dsisr.
This change adds dedicated fields for each class to the spu and the spu
context so that fault data is not overwritten.
Signed-off-by: Luke Browning <lukebr@linux.vnet.ibm.com>
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
