Ingo Molnar [Wed, 21 Jan 2009 09:24:27 +0000 (10:24 +0100)]
x86: uv cleanup, build fix
Fix:
arch/x86/mm/srat_64.c: In function ‘acpi_numa_processor_affinity_init’:
arch/x86/mm/srat_64.c:141: error: implicit declaration of function ‘get_uv_system_type’
arch/x86/mm/srat_64.c:141: error: ‘UV_X2APIC’ undeclared (first use in this function)
arch/x86/mm/srat_64.c:141: error: (Each undeclared identifier is reported only once
arch/x86/mm/srat_64.c:141: error: for each function it appears in.)
A couple of UV definitions were moved to asm/uv/uv.h, but srat_64.c did
not include that header. Add it.
Merge it here because both the cpumask changes and the ongoing percpu
work is touching the TLB code. The percpu changes take precedence, as
they eliminate tlb_32.c altogether.
Tejun Heo [Wed, 21 Jan 2009 08:26:06 +0000 (17:26 +0900)]
x86: make x86_32 use tlb_64.c
Impact: less contention when issuing invalidate IPI, cleanup
Make x86_32 use the same tlb code as 64bit. The 64bit code uses
multiple IPI vectors for tlb shootdown to reduce contention. This
patch makes x86_32 allocate the same 8 IPIs as x86_64 and share the
code paths.
Note that the usage of asmlinkage is inconsistent for x86_32 and 64
and calls for further cleanup. This has been noted with a FIXME
comment in tlb_64.c.
Tejun Heo [Wed, 21 Jan 2009 08:26:06 +0000 (17:26 +0900)]
x86: prepare for tlb merge
Impact: clean up, ipi vector number reordering for x86_32
Make the following changes to prepare for tlb merge.
* reorder x86_32 ip vectors
* adjust tlb_32.c and tlb_64.c such that their logics coincide exactly
- on spurious invalidate ipi, tlb_32 acks the irq
- tlb_64 now has proper memory barriers around clearing
flush_cpumask (no change in generated code)
* unexport flush_tlb_page from tlb_32.c, there's no user
Tejun Heo [Wed, 21 Jan 2009 08:26:06 +0000 (17:26 +0900)]
x86: uv cleanup
Impact: cleanup
Make the following uv related cleanups.
* collect visible uv related definitions and interfaces into uv/uv.h
and use it. this cleans up the messy situation where on 64bit, uv
is defined properly, on 32bit generic it's dummy and on the rest
undefined. after this clean up, uv is defined on 64 and dummy on
32.
* update uv_flush_tlb_others() such that it takes cpumask of
to-be-flushed cpus as argument, instead of that minus self, and
returns yet-to-be-flushed cpumask, instead of modifying the passed
in parameter. this interface change will ease dummy implementation
of uv_flush_tlb_others() and makes uv tlb flush related stuff
defined in tlb_uv proper.
Brian Gerst [Wed, 21 Jan 2009 08:26:06 +0000 (17:26 +0900)]
x86: merge mmu_context.h
Impact: cleanup
tj: * changed cpu to unsigned as was done on mmu_context_64.h as cpu
id is officially unsigned int
* added missing ';' to 32bit version of deactivate_mm()
Signed-off-by: Brian Gerst <brgerst@gmail.com> Signed-off-by: Tejun Heo <tj@kernel.org>
Brian Gerst [Wed, 21 Jan 2009 08:26:05 +0000 (17:26 +0900)]
x86: set %fs to __KERNEL_PERCPU unconditionally for x86_32
Impact: cleanup
%fs is currently set to __KERNEL_DS at boot, and conditionally
switched to __KERNEL_PERCPU for secondary cpus. Instead, initialize
GDT_ENTRY_PERCPU to the same attributes as GDT_ENTRY_KERNEL_DS and
set %fs to __KERNEL_PERCPU unconditionally.
Signed-off-by: Brian Gerst <brgerst@gmail.com> Signed-off-by: Tejun Heo <tj@kernel.org>
Brian Gerst [Wed, 21 Jan 2009 08:26:05 +0000 (17:26 +0900)]
x86: fix percpu_write with 64-bit constants
Impact: slightly better code generation for percpu_to_op()
The processor will sign-extend 32-bit immediate values in 64-bit
operations. Use the 'e' constraint ("32-bit signed integer constant,
or a symbolic reference known to fit that range") for 64-bit constants.
Signed-off-by: Brian Gerst <brgerst@gmail.com> Signed-off-by: Tejun Heo <tj@kernel.org>
Brian Gerst [Mon, 19 Jan 2009 03:21:28 +0000 (12:21 +0900)]
x86: move stack_canary into irq_stack
Impact: x86_64 percpu area layout change, irq_stack now at the beginning
Now that the PDA is empty except for the stack canary, it can be removed.
The irqstack is moved to the start of the per-cpu section. If the stack
protector is enabled, the canary overlaps the bottom 48 bytes of the irqstack.
tj: * updated subject
* dropped asm relocation of irq_stack_ptr
* updated comments a bit
* rebased on top of stack canary changes
Signed-off-by: Brian Gerst <brgerst@gmail.com> Signed-off-by: Tejun Heo <tj@kernel.org>
Brian Gerst [Mon, 19 Jan 2009 03:21:27 +0000 (12:21 +0900)]
x86: remove pda_init()
Impact: cleanup
Copy the code to cpu_init() to satisfy the requirement that the cpu
be reinitialized. Remove all other calls, since the segments are
already initialized in head_64.S.
Signed-off-by: Brian Gerst <brgerst@gmail.com> Signed-off-by: Tejun Heo <tj@kernel.org>
Brian Gerst [Sun, 18 Jan 2009 15:38:59 +0000 (00:38 +0900)]
x86-64: Use absolute displacements for per-cpu accesses.
Accessing memory through %gs should not use rip-relative addressing.
Adding a P prefix for the argument tells gcc to not add (%rip) to
the memory references.
Signed-off-by: Brian Gerst <brgerst@gmail.com> Signed-off-by: Tejun Heo <tj@kernel.org>
Brian Gerst [Sun, 18 Jan 2009 15:38:58 +0000 (00:38 +0900)]
x86-64: Convert irqstacks to per-cpu
Move the irqstackptr variable from the PDA to per-cpu. Make the
stacks themselves per-cpu, removing some specific allocation code.
Add a seperate flag (is_boot_cpu) to simplify the per-cpu boot
adjustments.
tj: * sprinkle some underbars around.
* irq_stack_ptr is not used till traps_init(), no reason to
initialize it early. On SMP, just leaving it NULL till proper
initialization in setup_per_cpu_areas() works. Dropped
is_boot_cpu and early irq_stack_ptr initialization.
* do DECLARE/DEFINE_PER_CPU(char[IRQ_STACK_SIZE], irq_stack)
instead of (char, irq_stack[IRQ_STACK_SIZE]).
Signed-off-by: Brian Gerst <brgerst@gmail.com> Signed-off-by: Tejun Heo <tj@kernel.org>
Tejun Heo [Sat, 17 Jan 2009 05:42:50 +0000 (14:42 +0900)]
linker script: add missing VMLINUX_SYMBOL
The newly added PERCPU_*() macros define and use __per_cpu_load but
VMLINUX_SYMBOL() was missing from usages causing build failures on
archs where linker visible symbol is different from C symbols
(e.g. blackfin).
Mike Travis [Fri, 16 Jan 2009 23:31:15 +0000 (15:31 -0800)]
cpufreq: use work_on_cpu in acpi-cpufreq.c for drv_read and drv_write
Impact: use new work_on_cpu function to reduce stack usage
Replace the saving of current->cpus_allowed and set_cpus_allowed_ptr() with
a work_on_cpu function for drv_read() and drv_write().
Basically converts do_drv_{read,write} into "work_on_cpu" functions that
are now called by drv_read and drv_write.
Note: This patch basically reverts 50c668d6 which reverted 7503bfba, now
that the work_on_cpu() function is more stable.
Signed-off-by: Mike Travis <travis@sgi.com> Acked-by: Rusty Russell <rusty@rustcorp.com.au> Tested-by: Dieter Ries <clip2@gmx.de> Tested-by: Maciej Rutecki <maciej.rutecki@gmail.com> Cc: Dave Jones <davej@redhat.com> Cc: <cpufreq@vger.kernel.org>
Rusty Russell [Fri, 16 Jan 2009 23:31:15 +0000 (15:31 -0800)]
work_on_cpu: Use our own workqueue.
Impact: remove potential clashes with generic kevent workqueue
Annoyingly, some places we want to use work_on_cpu are already in
workqueues. As per Ingo's suggestion, we create a different workqueue
for work_on_cpu.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Mike Travis <travis@sgi.com>
Rusty Russell [Fri, 16 Jan 2009 23:31:15 +0000 (15:31 -0800)]
work_on_cpu: don't try to get_online_cpus() in work_on_cpu.
Impact: remove potential circular lock dependency with cpu hotplug lock
This has caused more problems than it solved, with a pile of cpu
hotplug locking issues.
Followup patches will get_online_cpus() in callers that need it, but
if they don't do it they're no worse than before when they were using
set_cpus_allowed without locking.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Mike Travis <travis@sgi.com>
Tejun Heo [Fri, 16 Jan 2009 03:11:43 +0000 (12:11 +0900)]
x86_64: initialize this_cpu_off to __per_cpu_load
On x86_64, if get_per_cpu_var() is used before per cpu area is setup
(if lockdep is turned on, it happens), it needs this_cpu_off to point
to __per_cpu_load. Initialize accordingly.
I also cleaned up the x86-specific APIs and made the x86 code use
these new generic percpu primitives.
tj: * fixed generic percpu_sub() definition as Roel Kluin pointed out
* added percpu_and() for completeness's sake
* made generic percpu ops atomic against preemption
Tejun Heo [Tue, 13 Jan 2009 11:41:35 +0000 (20:41 +0900)]
x86: convert pda ops to wrappers around x86 percpu accessors
pda is now a percpu variable and there's no reason it can't use plain
x86 percpu accessors. Add x86_test_and_clear_bit_percpu() and replace
pda op implementations with wrappers around x86 percpu accessors.
Tejun Heo [Tue, 13 Jan 2009 11:41:35 +0000 (20:41 +0900)]
x86: make pda a percpu variable
[ Based on original patch from Christoph Lameter and Mike Travis. ]
As pda is now allocated in percpu area, it can easily be made a proper
percpu variable. Make it so by defining per cpu symbol from linker
script and declaring it in C code for SMP and simply defining it for
UP. This change cleans up code and brings SMP and UP closer a bit.
Tejun Heo [Tue, 13 Jan 2009 11:41:35 +0000 (20:41 +0900)]
x86: merge 64 and 32 SMP percpu handling
Now that pda is allocated as part of percpu, percpu doesn't need to be
accessed through pda. Unify x86_64 SMP percpu access with x86_32 SMP
one. Other than the segment register, operand size and the base of
percpu symbols, they behave identical now.
This patch replaces now unnecessary pda->data_offset with a dummy
field which is necessary to keep stack_canary at its place. This
patch also moves per_cpu_offset initialization out of init_gdt() into
setup_per_cpu_areas(). Note that this change also necessitates
explicit per_cpu_offset initializations in voyager_smp.c.
With this change, x86_OP_percpu()'s are as efficient on x86_64 as on
x86_32 and also x86_64 can use assembly PER_CPU macros.
Tejun Heo [Tue, 13 Jan 2009 11:41:35 +0000 (20:41 +0900)]
x86: fold pda into percpu area on SMP
[ Based on original patch from Christoph Lameter and Mike Travis. ]
Currently pdas and percpu areas are allocated separately. %gs points
to local pda and percpu area can be reached using pda->data_offset.
This patch folds pda into percpu area.
Due to strange gcc requirement, pda needs to be at the beginning of
the percpu area so that pda->stack_canary is at %gs:40. To achieve
this, a new percpu output section macro - PERCPU_VADDR_PREALLOC() - is
added and used to reserve pda sized chunk at the start of the percpu
area.
After this change, for boot cpu, %gs first points to pda in the
data.init area and later during setup_per_cpu_areas() gets updated to
point to the actual pda. This means that setup_per_cpu_areas() need
to reload %gs for CPU0 while clearing pda area for other cpus as cpu0
already has modified it when control reaches setup_per_cpu_areas().
This patch also removes now unnecessary get_local_pda() and its call
sites.
A lot of this patch is taken from Mike Travis' "x86_64: Fold pda into
per cpu area" patch.
Tejun Heo [Tue, 13 Jan 2009 11:41:35 +0000 (20:41 +0900)]
x86: use static _cpu_pda array
_cpu_pda array first uses statically allocated storage in data.init
and then switches to allocated bootmem to conserve space. However,
after folding pda area into percpu area, _cpu_pda array will be
removed completely. Drop the reallocation part to simplify the code
for soon-to-follow changes.
Tejun Heo [Tue, 13 Jan 2009 11:41:35 +0000 (20:41 +0900)]
x86: load pointer to pda into %gs while brining up a CPU
[ Based on original patch from Christoph Lameter and Mike Travis. ]
CPU startup code in head_64.S loaded address of a zero page into %gs
for temporary use till pda is loaded but address to the actual pda is
available at the point. Load the real address directly instead.
This will help unifying percpu and pda handling later on.
This patch is mostly taken from Mike Travis' "x86_64: Fold pda into
per cpu area" patch.
Tejun Heo [Tue, 13 Jan 2009 11:41:35 +0000 (20:41 +0900)]
x86: make percpu symbols zerobased on SMP
[ Based on original patch from Christoph Lameter and Mike Travis. ]
This patch makes percpu symbols zerobased on x86_64 SMP by adding
PERCPU_VADDR() to vmlinux.lds.h which helps setting explicit vaddr on
the percpu output section and using it in vmlinux_64.lds.S. A new
PHDR is added as existing ones cannot contain sections near address
zero. PERCPU_VADDR() also adds a new symbol __per_cpu_load which
always points to the vaddr of the loaded percpu data.init region.
The following adjustments have been made to accomodate the address
change.
* code to locate percpu gdt_page in head_64.S is updated to add the
load address to the gdt_page offset.
* __per_cpu_load is used in places where access to the init data area
is necessary.
* pda->data_offset is initialized soon after C code is entered as zero
value doesn't work anymore.
This patch is mostly taken from Mike Travis' "x86_64: Base percpu
variables at zero" patch.
Mike Travis [Thu, 15 Jan 2009 17:19:32 +0000 (09:19 -0800)]
x86: fix build warning when CONFIG_NUMA not defined.
Impact: fix build warning
The macro cpu_to_node did not reference it's argument, and instead
simply returned a 0. This causes a "unused variable" warning if
it's the only reference in a function (show_cache_disable).
arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c is still broken: there's no
allocation of the variable mask, so we pass in an uninitialized cmd.mask
field to drv_read(), which then passes it to the scheduler which then
crashes ...
Switch it over to the much simpler constant-cpumask-pointers approach.
Jan Beulich [Wed, 14 Jan 2009 12:27:35 +0000 (12:27 +0000)]
x86: fully honor "nolapic"
Impact: widen the effect of the 'nolapic' boot parameter
"nolapic" should not only suppress SMP and use of the LAPIC, but it
also ought to have the effect of disabling all IO-APIC related activity
as well as PCI MSI and HT-IRQs.
Signed-off-by: Jan Beulich <jbeulich@novell.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Mike Travis [Wed, 14 Jan 2009 23:43:54 +0000 (15:43 -0800)]
irq: update all arches for new irq_desc, fix
Impact: fix build errors
Since the SPARSE IRQS changes redefined how the kstat irqs are
organized, arch's must use the new accessor function:
kstat_incr_irqs_this_cpu(irq, DESC);
If CONFIG_SPARSE_IRQS is set, then DESC is a pointer to the
irq_desc which has a pointer to the kstat_irqs. If not, then
the .irqs field of struct kernel_stat is used instead.
Signed-off-by: Mike Travis <travis@sgi.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Suresh Siddha [Fri, 9 Jan 2009 22:35:20 +0000 (14:35 -0800)]
x86, pat: fix reserve_memtype() for legacy 1MB range
Thierry Vignaud reported:
> http://bugzilla.kernel.org/show_bug.cgi?id=12372
>
> On P4 with an SiS motherboard (video card is a SiS 651)
> X server fails to start with error:
> xf86MapVidMem: Could not mmap framebuffer (0x00000000,0x2000) (Invalid
> argument)
Here X is trying to map first 8KB of memory using /dev/mem. Existing
code treats first 0-4KB of memory as non-RAM and 4KB-8KB as RAM. Recent
code changes don't allow to map memory with different attributes
at the same time.
Fix this by treating the first 1MB legacy region as special and always
track the attribute requests with in this region using linear linked
list (and don't bother if the range is RAM or non-RAM or mixed)
Suresh Siddha [Wed, 14 Jan 2009 02:43:03 +0000 (18:43 -0800)]
x86: fix broken flush_tlb_others_ipi()
This commit broke flush_tlb_others_ipi() causing boot hangs on a
16 logical cpu system:
> commit 4595f9620cda8a1e973588e743cf5f8436dd20c6
> Author: Rusty Russell <rusty@rustcorp.com.au>
> Date: Sat Jan 10 21:58:09 2009 -0800
>
> x86: change flush_tlb_others to take a const struct cpumask
This change resulted in sending the invalidate tlb vector to the
sender itself causing the hang. flush_tlb_others_ipi() should exclude
the sender itself from the destination list.
Linus Torvalds [Tue, 13 Jan 2009 22:53:16 +0000 (14:53 -0800)]
Merge branch 'x86-pat-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'x86-pat-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86 PAT: remove CPA WARN_ON for zero pte
x86 PAT: return compatible mapping to remap_pfn_range callers
x86 PAT: change track_pfn_vma_new to take pgprot_t pointer param
x86 PAT: consolidate old memtype new memtype check into a function
x86 PAT: remove PFNMAP type on track_pfn_vma_new() error
Linus Torvalds [Tue, 13 Jan 2009 22:52:35 +0000 (14:52 -0800)]
Merge master.kernel.org:/home/rmk/linux-2.6-arm
* master.kernel.org:/home/rmk/linux-2.6-arm:
TWL4030: fix clk API usage
[ARM] 5364/1: allow flush_ioremap_region() to be used from modules
[ARM] w90x900: fix build errors and warnings
[ARM] i.MX add missing include
[ARM] i.MX: fix breakage from commit 278892736e99330195c8ae5861bcd9d791bbf19e
[ARM] i.MX: remove LCDC controller register definitions from imx-regs.h
Bernd Schmidt [Tue, 13 Jan 2009 21:14:48 +0000 (22:14 +0100)]
Fix timeouts in sys_pselect7
Since we (Analog Devices) updated our Blackfin kernel to 2.6.28, we've
seen occasional 5-second hangs from telnet. telnetd calls select with a
NULL timeout, but with the new kernel, the system call occasionally
returns 0, which causes telnet to call sleep (5). This did not happen
with earlier kernels.
The code in sys_pselect7 looks a bit strange, in particular the variable
"to" is initialized to NULL, then changed if a non-null timeout was
passed in, but not used further. It needs to be passed to
core_sys_select instead of &end_time.
Helge Deller [Tue, 13 Jan 2009 21:51:07 +0000 (22:51 +0100)]
fix early_serial_setup() regression
Commit b430428a188e8a434325e251d0704af4b88b4711 ("8250: Don't clobber
spinlocks.") introduced a regression on the parisc architecture, which
broke the handover to the serial port at boottime.
early_serial_setup() was changed to only copy a subset of the uart_port
fields, and sadly the "type" and "line" fields were forgotten and thus
the serial port was not initialized and could not be used for a
handover. This patch fixes this by copying the missing fields.
As this change to early_serial_setup() doesn't need an initialized
spinlock in the uart_port struct any longer, we can drop the spinlock
initialization in the superio driver.
Cc: David Daney <ddaney@caviumnetworks.com> Cc: Tomaso Paoletti <tpaoletti@caviumnetworks.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Acked-by: Kyle McMartin <kyle@mcmartin.ca> Cc: linux-parisc@vger.kernel.org Signed-off-by: Helge Deller <deller@gmx.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Russell King [Sat, 10 Jan 2009 10:40:42 +0000 (10:40 +0000)]
TWL4030: fix clk API usage
Always pass a struct device if one is available; and there's really
no reason for the processor specific stuff in this file if only
people would follow the API usage properly by using the struct device.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Impact: reduce scope of debug check - avoid warnings
The logic to find whether identity map exists or not using
high_memory or max_low_pfn_mapped/max_pfn_mapped are not complete
as the memory withing the range may not be mapped if there is a
unusable hole in e820.
Specifically, on my test system I started seeing these warnings with
tools like hwinfo, acpidump trying to map ACPI region.
We can add logic to check for this. But, there can also be other holes in
identity map when we have 1GB of aligned reserved space in e820.
This patch handles it by removing the WARN_ON and returning a specific
error value (EFAULT) to indicate that the address does not have any
identity mapping.
The code that tries to keep identity map in sync can ignore
this error, with other callers of cpa still getting error here.
x86 PAT: return compatible mapping to remap_pfn_range callers
Impact: avoid warning message, potentially solve 3D performance regression
Change x86 PAT code to return compatible memtype if the exact memtype that
was requested in remap_pfn_rage and friends is not available due to some
conflict.
This is done by returning the compatible type in pgprot parameter of
track_pfn_vma_new(), and the caller uses that memtype for page table.
Note that track_pfn_vma_copy() which is basically called during fork gets the
prot from existing page table and should not have any conflict. Hence we use
strict memtype check there and do not allow compatible memtypes.
x86 PAT: change track_pfn_vma_new to take pgprot_t pointer param
Impact: cleanup
Change the protection parameter for track_pfn_vma_new() into a pgprot_t pointer.
Subsequent patch changes the x86 PAT handling to return a compatible
memtype in pgprot_t, if what was requested cannot be allowed due to conflicts.
No fuctionality change in this patch.
x86 PAT: consolidate old memtype new memtype check into a function
Impact: cleanup
Move the new memtype old memtype allowed check to header so that is can be
shared by other users. Subsequent patch uses this in pat.c in remap_pfn_range()
code path. No functionality change in this patch.
Linus Torvalds [Tue, 13 Jan 2009 16:19:42 +0000 (08:19 -0800)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
IB/iser: Add dependency on INFINIBAND_ADDR_TRANS
IPoIB: Do not join broadcast group if interface is brought down
RDMA/nes: Fix for NIPQUAD removal
IPoIB: Fix loss of connectivity after bonding failover on both sides
IB/mlx4: Don't register IB device for adapters with no IB ports
mlx4_core: Fix warning from min()
IB/ehca: spin_lock_irqsave() takes an unsigned long
Randy Dunlap [Tue, 13 Jan 2009 03:30:41 +0000 (19:30 -0800)]
IB/iser: Add dependency on INFINIBAND_ADDR_TRANS
Fix ib_iser build to depend on INFINIBAND_ADDR_TRANS; if INET=y but
IPV6=n, then the RDMA CM is not built but INFINIBAND_ISER can be
enabled, leading to:
Yossi Etigin [Tue, 13 Jan 2009 03:28:42 +0000 (19:28 -0800)]
IPoIB: Do not join broadcast group if interface is brought down
Because the ipoib_workqueue is not flushed when ipoib interface is
brought down, ipoib_mcast_join() may trigger a join to the broadcast
group after priv->broadcast was set to NULL (during cleanup). This
will cause the system to be a member of the broadcast group when
interface is down. As a side effect, this breaks the optimization of
setting the Q_key only when joining the broadcast group.
Signed-off-by: Yossi Etigin <yosefe@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
projects / linux-beck.git / log