Impact: add better first percpu allocation for NUMA
On NUMA, embedding allocator can't be used as different units can't be
made to fall in the correct NUMA nodes. To use large page mapping,
each unit needs to be remapped. However, percpu areas are usually
much smaller than large page size and unused space hurts a lot as the
number of cpus grow. This allocator remaps large pages for each chunk
but gives back unused part to the bootmem allocator making the large
pages mapped twice.
This adds slightly to the TLB pressure but is much better than using
4k mappings while still being NUMA-friendly.
Ingo suggested that this would be the correct approach for NUMA.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Impact: add better first percpu allocation for !NUMA
On !NUMA, we can simply allocate contiguous memory and use it for the
first chunk without mapping it into vmalloc area. As the memory area
is covered by the large page physical memory mapping, it allows the
dynamic perpcu allocator to not add any TLB overhead for the static
percpu area and whatever falls into the first chunk and the
implementation is very simple too.
Signed-off-by: Tejun Heo <tj@kernel.org>
Impact: modularize percpu first chunk allocation
x86 is gonna have a few different strategies for the first chunk
allocation. Modularize it by separating out the current allocation
mechanism into pcpu_alloc_bootmem() and setup_pcpu_4k().
Signed-off-by: Tejun Heo <tj@kernel.org>
Impact: more latitude for first percpu chunk allocation
The first percpu chunk serves the kernel static percpu area and may or
may not contain extra room for further dynamic allocation.
Initialization of the first chunk needs to be done before normal
memory allocation service is up, so it has its own init path -
pcpu_setup_static().
It seems archs need more latitude while initializing the first chunk
for example to take advantage of large page mapping. This patch makes
the following changes to allow this.
* Define PERCPU_DYNAMIC_RESERVE to give arch hint about how much space
to reserve in the first chunk for further dynamic allocation.
* Rename pcpu_setup_static() to pcpu_setup_first_chunk().
* Make pcpu_setup_first_chunk() much more flexible by fetching page
pointer by callback and adding optional @unit_size, @free_size and
@base_addr arguments which allow archs to selectively part of chunk
initialization to their likings.
Signed-off-by: Tejun Heo <tj@kernel.org>
Impact: allow unit_size to be arbitrary multiple of PAGE_SIZE
In dynamic percpu allocator, there is no reason the unit size should
be power of two. Remove the restriction.
As non-power-of-two unit size means that empty chunks fall into the
same slot index as lightly occupied chunks which is bad for reclaming.
Reserve an extra slot for empty chunks.
Signed-off-by: Tejun Heo <tj@kernel.org>
Impact: minor change to populate_extra_pte() and addition of pmd flavor
Update populate_extra_pte() to return pointer to the pte_t for the
specified address and add populate_extra_pmd() which only populates
till the pmd and returns pointer to the pmd entry for the address.
For 64bit, pud/pmd/pte fill functions are separated out from
set_pte_vaddr[_pud]() and used for set_pte_vaddr[_pud]() and
populate_extra_{pte|pmd}().
Signed-off-by: Tejun Heo <tj@kernel.org>
Impact: allow larger alignment for early vmalloc area allocation
Some early vmalloc users might want larger alignment, for example, for
custom large page mapping. Add @align to vm_area_register_early().
While at it, drop docbook comment on non-existent @size.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Impact: cleanup and addition of missing interface wrapper
The interface functions in bootmem.h was ordered in not so orderly
manner. Reorder them such that
* functions allocating the same area group together -
ie. alloc_bootmem group and alloc_bootmem_low group.
* functions w/o node parameter come before the ones w/ node parameter.
* nopanic variants are immediately below their panicky counterparts.
While at it, add alloc_bootmem_pages_node_nopanic() which was missing.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Johannes Weiner <hannes@saeurebad.de>
Impact: cleaner and consistent bootmem wrapping
By setting CONFIG_HAVE_ARCH_BOOTMEM_NODE, archs can define
arch-specific wrappers for bootmem allocation. However, this is done
a bit strangely in that only the high level convenience macros can be
changed while lower level, but still exported, interface functions
can't be wrapped. This not only is messy but also leads to strange
situation where alloc_bootmem() does what the arch wants it to do but
the equivalent __alloc_bootmem() call doesn't although they should be
able to be used interchangeably.
This patch updates bootmem such that archs can override / wrap the
backend function - alloc_bootmem_core() instead of the highlevel
interface functions to allow simpler and consistent wrapping. Also,
HAVE_ARCH_BOOTMEM_NODE is renamed to HAVE_ARCH_BOOTMEM.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Johannes Weiner <hannes@saeurebad.de>
Impact: fix short allocation leading to memory corruption
While dropping rvalue wrapping macros around global parameters,
pcpu_chunk_struct_size was set incorrectly resulting in shorter page
pointer array. Fix it.
Signed-off-by: Tejun Heo <tj@kernel.org>
Andrew was concerned about the unit of variables named or have suffix
size. Every usage in percpu allocator is in bytes but make it super
clear by adding comments.
While at it, make pcpu_depopulate_chunk() take int @off and @size like
everyone else.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Impact: use new dynamic allocator, unified access to static/dynamic
percpu memory
Convert to the new dynamic percpu allocator.
* implement populate_extra_pte() for both 32 and 64
* update setup_per_cpu_areas() to use pcpu_setup_static()
* define __addr_to_pcpu_ptr() and __pcpu_ptr_to_addr()
* define config HAVE_DYNAMIC_PER_CPU_AREA
Signed-off-by: Tejun Heo <tj@kernel.org>
Impact: new scalable dynamic percpu allocator which allows dynamic
percpu areas to be accessed the same way as static ones
Implement scalable dynamic percpu allocator which can be used for both
static and dynamic percpu areas. This will allow static and dynamic
areas to share faster direct access methods. This feature is optional
and enabled only when CONFIG_HAVE_DYNAMIC_PER_CPU_AREA is defined by
arch. Please read comment on top of mm/percpu.c for details.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Impact: two more public map/unmap functions
Implement map_kernel_range_noflush() and unmap_kernel_range_noflush().
These functions respectively map and unmap address range in kernel VM
area but doesn't do any vcache or tlb flushing. These will be used by
new percpu allocator.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Impact: allow multiple early vm areas
There are places where kernel VM area needs to be allocated before
vmalloc is initialized. This is done by allocating static vm_struct,
initializing several fields and linking it to vmlist and later vmalloc
initialization picking up these from vmlist. This is currently done
manually and if there's more than one such areas, there's no defined
way to arbitrate who gets which address.
This patch implements vm_area_register_early(), which takes vm_area
struct with flags and size initialized, assigns address to it and puts
it on the vmlist. This way, multiple early vm areas can determine
which addresses they should use. The only current user - alpha mm
init - is converted to use it.
Signed-off-by: Tejun Heo <tj@kernel.org>
Impact: kill unused functions
percpu_alloc() and its friends never saw much action. It was supposed
to replace the cpu-mask unaware __alloc_percpu() but it never happened
and in fact __percpu_alloc_mask() itself never really grew proper
up/down handling interface either (no exported interface for
populate/depopulate).
percpu allocation is about to go through major reimplementation and
there's no reason to carry this unused interface around. Replace it
with __alloc_percpu() and free_percpu().
Signed-off-by: Tejun Heo <tj@kernel.org>
This prepares for a real __alloc_percpu, by adding an alignment argument.
Only one place uses __alloc_percpu directly, and that's for a string.
tj: af_inet also uses __alloc_percpu(), update it.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Jens Axboe <axboe@kernel.dk>
Impact: cleanup
There are two allocated per-cpu accessor macros with almost identical
spelling. The original and far more popular is per_cpu_ptr (44
files), so change over the other 4 files.
tj: kill percpu_ptr() and update UP too
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: mingo@redhat.com
Cc: lenb@kernel.org
Cc: cpufreq@vger.kernel.org
Signed-off-by: Tejun Heo <tj@kernel.org>
Impact: proper vcache flush on unmap_kernel_range()
flush_cache_vunmap() should be called before pages are unmapped. Add
a call to it in unmap_kernel_range().
Signed-off-by: Tejun Heo <tj@kernel.org>
Impact: economize memory for large NR_CPUS
percpu data is setup earlier than irq, we can use percpu data
to economize memory.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Impact: Fixes warning
Fix uv.h struct usage:
arch/x86/include/asm/uv/uv.h:16: warning: 'struct mm_struct' declared inside parameter list
arch/x86/include/asm/uv/uv.h:16: warning: its scope is only this definition or declaration, which is probably not what you want
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Impact: cleanup
With the recent changes in the 32-bit code to make system calls which
use struct pt_regs take a pointer, sys_rt_sigreturn() have become
identical between 32 and 64 bits, and both are empty wrappers around
do_rt_sigreturn(). Remove both wrappers and rename both to
sys_rt_sigreturn().
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Some syscalls need to access the pt_regs structure, either to copy
user register state or to modifiy it. This patch adds stubs to load
the address of the pt_regs struct into the %eax register, and changes
the syscalls to take the pointer as an argument instead of relying on
the assumption that the pt_regs structure overlaps the function
arguments.
Drop the use of regparm(1) due to concern about gcc bugs, and to move
in the direction of the eventual removal of regparm(0) for asmlinkage.
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Fix the ia64 build error that occurs in the linux-next tree by introducing
an ia64 version of uv.h.
Additionally, clean up the usage of is_uv_system().
Signed-off-by: Dean Nelson <dcn@sgi.com>
Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Now that no functions rely on struct pt_regs being passed by value,
various "no stack protector" annotations can be dropped.
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Some syscalls need to access the pt_regs structure, either to copy
user register state or to modifiy it. This patch adds stubs to load
the address of the pt_regs struct into the %eax register, and changes
the syscalls to regparm(1) to receive the pt_regs pointer as the
first argument.
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The generic exception handler (error_code) passes in the pt_regs
pointer and the error code (unused in this case). The commit
"x86: fix math_emu register frame access" changed this to pass by
value, which doesn't work correctly with stack protector enabled.
Change it back to use the pt_regs pointer.
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Stackprotector builds were failing if CROSS_COMPILER was more than
a single world (such as when distcc was used) - because the check
scripts used $1 instead of $*.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix x86_32 stack protector
Brian Gerst found out that %gs was being initialized to stack_canary
instead of stack_canary - 20, which basically gave the same canary
value for all threads. Fixing this also exposed the following bugs.
* cpu_idle() didn't call boot_init_stack_canary()
* stack canary switching in switch_to() was being done too late making
the initial run of a new thread use the old stack canary value.
Fix all of them and while at it update comment in cpu_idle() about
calling boot_init_stack_canary().
Reported-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: stack protector for x86_32
Implement stack protector for x86_32. GDT entry 28 is used for it.
It's set to point to stack_canary-20 and have the length of 24 bytes.
CONFIG_CC_STACKPROTECTOR turns off CONFIG_X86_32_LAZY_GS and sets %gs
to the stack canary segment on entry. As %gs is otherwise unused by
the kernel, the canary can be anywhere. It's defined as a percpu
variable.
x86_32 exception handlers take register frame on stack directly as
struct pt_regs. With -fstack-protector turned on, gcc copies the
whole structure after the stack canary and (of course) doesn't copy
back on return thus losing all changed. For now, -fno-stack-protector
is added to all files which contain those functions. We definitely
need something better.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: pt_regs changed, lazy gs handling made optional, add slight
overhead to SAVE_ALL, simplifies error_code path a bit
On x86_32, %gs hasn't been used by kernel and handled lazily. pt_regs
doesn't have place for it and gs is saved/loaded only when necessary.
In preparation for stack protector support, this patch makes lazy %gs
handling optional by doing the followings.
* Add CONFIG_X86_32_LAZY_GS and place for gs in pt_regs.
* Save and restore %gs along with other registers in entry_32.S unless
LAZY_GS. Note that this unfortunately adds "pushl $0" on SAVE_ALL
even when LAZY_GS. However, it adds no overhead to common exit path
and simplifies entry path with error code.
* Define different user_gs accessors depending on LAZY_GS and add
lazy_save_gs() and lazy_load_gs() which are noop if !LAZY_GS. The
lazy_*_gs() ops are used to save, load and clear %gs lazily.
* Define ELF_CORE_COPY_KERNEL_REGS() which always read %gs directly.
xen and lguest changes need to be verified.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
On x86_32, %gs is handled lazily. It's not saved and restored on
kernel entry/exit but only when necessary which usually is during task
switch but there are few other places. Currently, it's done by
calling savesegment() and loadsegment() explicitly. Define
get_user_gs(), set_user_gs() and task_user_gs() and use them instead.
While at it, clean up register access macros in signal.c.
This cleans up code a bit and will help future changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Use .macro instead of cpp #define where approriate. This cleans up
code and will ease future changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: no default -fno-stack-protector if stackp is enabled, cleanup
Stackprotector make rules had the following problems.
* cc support test and warning are scattered across makefile and
kernel/panic.c.
* -fno-stack-protector was always added regardless of configuration.
Update such that cc support test and warning are contained in makefile
and -fno-stack-protector is added iff stackp is turned off. While at
it, prepare for 32bit support.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: misc udpate
* wrap content with CONFIG_CC_STACK_PROTECTOR so that other arch files
can include it directly
* add missing includes
This will help future changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
ELF core dump is used for both user land core dump and kernel crash
dump. Depending on architecture, register might need to be accessed
differently for userland and kernel. Allow architectures to define
ELF_CORE_COPY_KERNEL_REGS() and use different operation for kernel
register dump.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
do_device_not_available() is the handler for #NM and it declares that
it takes a unsigned long and calls math_emu(), which takes a long
argument and surprisingly expects the stack frame starting at the zero
argument would match struct math_emu_info, which isn't true regardless
of configuration in the current code.
This patch makes do_device_not_available() take struct pt_regs like
other exception handlers and initialize struct math_emu_info with
pointer to it and pass pointer to the math_emu_info to math_emulate()
like normal C functions do. This way, unless gcc makes a copy of
struct pt_regs in do_device_not_available(), the register frame is
correctly accessed regardless of kernel configuration or compiler
used.
This doesn't fix all math_emu problems but it at least gets it
somewhat working.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
* Come on, struct info? s/struct info/struct math_emu_info/
* Use struct pt_regs and kernel_vm86_regs instead of defining its own
register frame structure.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: dump the correct %gs into a.out core dump
aout_dump_thread() read %gs but didn't include it in core dump. Fix
it.
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Commit 6194ba6ff6 ("x86: don't special-case
pmd allocations as much") made changes to the way we handle pmd allocations,
and while doing that it dropped a call to paravirt_release_pd on the
pgd page from the pgd_dtor code path.
As a result of this missing release, the hypervisor is now unaware of the
pgd page being freed, and as a result it ends up tracking this page as a
page table page.
After this the guest may start using the same page for other purposes, and
depending on what use the page is put to, it may result in various performance
and/or functional issues ( hangs, reboots).
Since this release is only required for VMI, I now release the pgd page from
the (vmi)_pgd_free hook.
Signed-off-by: Alok N Kataria <akataria@vmware.com>
Acked-by: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: <stable@kernel.org>
Impact: find right nr_irqs_gsi on some systems.
One test-system has gap between gsi's:
[ 0.000000] ACPI: IOAPIC (id[0x04] address[0xfec00000] gsi_base[0])
[ 0.000000] IOAPIC[0]: apic_id 4, version 0, address 0xfec00000, GSI 0-23
[ 0.000000] ACPI: IOAPIC (id[0x05] address[0xfeafd000] gsi_base[48])
[ 0.000000] IOAPIC[1]: apic_id 5, version 0, address 0xfeafd000, GSI 48-54
[ 0.000000] ACPI: IOAPIC (id[0x06] address[0xfeafc000] gsi_base[56])
[ 0.000000] IOAPIC[2]: apic_id 6, version 0, address 0xfeafc000, GSI 56-62
...
[ 0.000000] nr_irqs_gsi: 38
So nr_irqs_gsi is not right. some irq for MSI will overwrite with io_apic.
need to get that with acpi_probe_gsi when acpi io_apic is used
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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>
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.
Reported-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Acked-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: bug fix
IA-64 needs to put percpu data in the seperate section even on UP.
Fixes regression caused by "percpu: refactor percpu.h"
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Acked-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
