Setting single memory pages at a time to wc takes a lot time in cache flush. To
reduce number of cache flush set_pages_array_wc and set_memory_array_wc can be
used to set multiple pages to WC with single cache flush.

This improves allocation performance for wc cached pages in drm/ttm.

CC: Suresh Siddha <suresh.b.siddha@intel.com>
CC: Venkatesh Pallipadi <venkatesh.pallipadi@gmail.com>
Signed-off-by: NPauli Nieminen <suokkos@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h

percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files.  percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.

percpu.h -> slab.h dependency is about to be removed.  Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability.  As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.

  http://userweb.kernel.org/~tj/misc/slabh-sweep.py

The script does the followings.

* Scan files for gfp and slab usages and update includes such that
  only the necessary includes are there.  ie. if only gfp is used,
  gfp.h, if slab is used, slab.h.

* When the script inserts a new include, it looks at the include
  blocks and try to put the new include such that its order conforms
  to its surrounding.  It's put in the include block which contains
  core kernel includes, in the same order that the rest are ordered -
  alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
  doesn't seem to be any matching order.

* If the script can't find a place to put a new include (mostly
  because the file doesn't have fitting include block), it prints out
  an error message indicating which .h file needs to be added to the
  file.

The conversion was done in the following steps.

1. The initial automatic conversion of all .c files updated slightly
   over 4000 files, deleting around 700 includes and adding ~480 gfp.h
   and ~3000 slab.h inclusions.  The script emitted errors for ~400
   files.

2. Each error was manually checked.  Some didn't need the inclusion,
   some needed manual addition while adding it to implementation .h or
   embedding .c file was more appropriate for others.  This step added
   inclusions to around 150 files.

3. The script was run again and the output was compared to the edits
   from #2 to make sure no file was left behind.

4. Several build tests were done and a couple of problems were fixed.
   e.g. lib/decompress_*.c used malloc/free() wrappers around slab
   APIs requiring slab.h to be added manually.

5. The script was run on all .h files but without automatically
   editing them as sprinkling gfp.h and slab.h inclusions around .h
   files could easily lead to inclusion dependency hell.  Most gfp.h
   inclusion directives were ignored as stuff from gfp.h was usually
   wildly available and often used in preprocessor macros.  Each
   slab.h inclusion directive was examined and added manually as
   necessary.

6. percpu.h was updated not to include slab.h.

7. Build test were done on the following configurations and failures
   were fixed.  CONFIG_GCOV_KERNEL was turned off for all tests (as my
   distributed build env didn't work with gcov compiles) and a few
   more options had to be turned off depending on archs to make things
   build (like ipr on powerpc/64 which failed due to missing writeq).

   * x86 and x86_64 UP and SMP allmodconfig and a custom test config.
   * powerpc and powerpc64 SMP allmodconfig
   * sparc and sparc64 SMP allmodconfig
   * ia64 SMP allmodconfig
   * s390 SMP allmodconfig
   * alpha SMP allmodconfig
   * um on x86_64 SMP allmodconfig

8. percpu.h modifications were reverted so that it could be applied as
   a separate patch and serve as bisection point.

Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: NTejun Heo <tj@kernel.org>
Guess-its-ok-by: NChristoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>

We currently enforce the !RW mapping for the kernel mapping that maps
holes between different text, rodata and data sections. However, kernel
identity mappings will have different RWX permissions to the pages mapping to
text and to the pages padding (which are freed) the text, rodata sections.
Hence kernel identity mappings will be broken to smaller pages. For 64-bit,
kernel text and kernel identity mappings are different, so we can enable
protection checks that come with CONFIG_DEBUG_RODATA, as well as retain 2MB
large page mappings for kernel text.

Konrad reported a boot failure with the Linux Xen paravirt guest because of
this. In this paravirt guest case, the kernel text mapping and the kernel
identity mapping share the same page-table pages. Thus forcing the !RW mapping
for some of the kernel mappings also cause the kernel identity mappings to be
read-only resulting in the boot failure. Linux Xen paravirt guest also
uses 4k mappings and don't use 2M mapping.

Fix this issue and retain large page performance advantage for native kernels
by not working hard and not enforcing !RW for the kernel text mapping,
if the current mapping is already using small page mapping.
Reported-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1266522700.2909.34.camel@sbs-t61.sc.intel.com>
Tested-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: stable@kernel.org	[2.6.32, 2.6.33]
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Make set_memory_x/set_memory_nx directly aware of if NX is supported
in the system or not, rather than requiring that every caller assesses
that support independently.
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Tim Starling <tstarling@wikimedia.org>
Cc: Hannes Eder <hannes@hanneseder.net>
LKML-Reference: <1258154897-6770-4-git-send-email-hpa@zytor.com>
Acked-by: NKees Cook <kees.cook@canonical.com>

On x86_64, kernel text mappings are mapped read-only with
CONFIG_DEBUG_RODATA. So use the kernel identity mapping instead
of the kernel text mapping to modify the kernel text.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Acked-by: NSteven Rostedt <rostedt@goodmis.org>
Tested-by: NSteven Rostedt <rostedt@goodmis.org>
LKML-Reference: <20091029024821.080941108@sbs-t61.sc.intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Steven Rostedt reported that we are unconditionally making the
kernel text mapping as read-only. i.e., if someone does cpa() to
the kernel text area for setting/clearing any page table
attribute, we unconditionally clear the read-write attribute for
the kernel text mapping that is set at compile time.

We should delay (to forbid the write attribute) and enforce only
after the kernel has mapped the text as read-only.
Reported-by: NSteven Rostedt <rostedt@goodmis.org>
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Acked-by: NSteven Rostedt <rostedt@goodmis.org>
Tested-by: NSteven Rostedt <rostedt@goodmis.org>
LKML-Reference: <20091029024820.996634347@sbs-t61.sc.intel.com>
[ marked kernel_set_to_readonly as __read_mostly ]
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The commit 74e08179
x86-64: align RODATA kernel section to 2MB with CONFIG_DEBUG_RODATA
prevents text sections from becoming read/write using set_memory_rw.

The dynamic ftrace changes all text pages to read/write just before
converting the calls to tracing to nops, and vice versa.

I orginally just added a flag to allow this transaction when ftrace
did the change, but I also found that when the CPA testing was running
it would remove the read/write as well, and ftrace does not do the text
conversion on boot up, and the CPA changes caused the dynamic tracer
to fail on self tests.

The current solution I have is to simply not to prevent
change_page_attr from setting the RW bit for kernel text pages.
Reported-by: NIngo Molnar <mingo@elte.hu>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

CONFIG_DEBUG_RODATA chops the large pages spanning boundaries of kernel
text/rodata/data to small 4KB pages as they are mapped with different
attributes (text as RO, RODATA as RO and NX etc).

On x86_64, preserve the large page mappings for kernel text/rodata/data
boundaries when CONFIG_DEBUG_RODATA is enabled. This is done by allowing the
RODATA section to be hugepage aligned and having same RWX attributes
for the 2MB page boundaries

Extra Memory pages padding the sections will be freed during the end of the boot
and the kernel identity mappings will have different RWX permissions compared to
the kernel text mappings.

Kernel identity mappings to these physical pages will be mapped with smaller
pages but large page mappings are still retained for kernel text,rodata,data
mappings.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20091014220254.190119924@sbs-t61.sc.intel.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Ever since we enabled GEM, the pre-9xx chipsets (particularly 865) have had
serious stability issues.  Back in May a wbinvd was added to the DRM to
work around much of the problem.  Some failure remained -- easily visible
by dragging a window around on an X -retro desktop, or by looking at bugzilla.

The chipset flush was on the right track -- hitting the right amount of
memory, and it appears to be the only way to flush on these chipsets, but the
flush page was mapped uncached.  As a result, the writes trying to clear the
writeback cache ended up bypassing the cache, and not flushing anything!  The
wbinvd would flush out other writeback data and often cause the data we wanted
to get flushed, but not always.  By removing the setting of the page to UC
and instead just clflushing the data we write to try to flush it, we get the
desired behavior with no wbinvd.

This exports clflush_cache_range(), which was laying around and happened to
basically match the code I was otherwise going to copy from the DRM.
Signed-off-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NBrice Goglin <Brice.Goglin@ens-lyon.org>
Cc: stable@kernel.org

Fix address passed to cpa_flush_range() when changing page
attributes from WB to UC. The address (*addr) is
modified by __change_page_attr_set_clr(). The result is that
the pages being flushed start at the _end_ of the changed range
instead of the beginning.

This should be considered for 2.6.30-stable and 2.6.31-stable.
Signed-off-by: NJack Steiner <steiner@sgi.com>
Acked-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>
Cc: Stable team <stable@kernel.org>

With x86 converted to embedding allocator, lpage doesn't have any user
left.  Kill it along with cpa handling code.
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Jan Beulich <JBeulich@novell.com>

The code was incorrectly reserving memtypes using the page
virtual address instead of the physical address. Furthermore,
the code was not ignoring highmem pages as it ought to.

( upstream does not pass in highmem pages yet - but upcoming
  graphics code will do it and there's no reason to not handle
  this properly in the CPA APIs.)

Fixes: http://bugzilla.kernel.org/show_bug.cgi?id=13884Signed-off-by: NThomas Hellstrom <thellstrom@vmware.com>
Acked-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Cc: <stable@kernel.org>
Cc: dri-devel@lists.sourceforge.net
Cc: venkatesh.pallipadi@intel.com
LKML-Reference: <1249284345-7654-1-git-send-email-thellstrom@vmware.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Changeset 3869c4aa
that went in after 2.6.30-rc1 was a seemingly small change to _set_memory_wc()
to make it complaint with SDM requirements. But, introduced a nasty bug, which
can result in crash and/or strange corruptions when set_memory_wc is used.
One such crash reported here
http://lkml.org/lkml/2009/7/30/94

Actually, that changeset introduced two bugs.
* change_page_attr_set() takes &addr as first argument and can the addr value
  might have changed on return, even for single page change_page_attr_set()
  call. That will make the second change_page_attr_set() in this routine
  operate on unrelated addr, that can eventually cause strange corruptions
  and bad page state crash.
* The second change_page_attr_set() call, before setting _PAGE_CACHE_WC, should
  clear the earlier _PAGE_CACHE_UC_MINUS, as otherwise cache attribute will not
  be WC (will be UC instead).

The patch below fixes both these problems. Sending a single patch to fix both
the problems, as the change is to the same line of code. The change to have a
addr_copy is not very clean. But, it is simpler than making more changes
through various routines in pageattr.c.

A huge thanks to Jerome for reporting this problem and providing a simple test
case that helped us root cause the problem.
Reported-by: NJerome Glisse <glisse@freedesktop.org>
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20090730214319.GA1889@linux-os.sc.intel.com>
Acked-by: NDave Airlie <airlied@redhat.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Generalize and move x86 setup_pcpu_lpage() into
pcpu_lpage_first_chunk().  setup_pcpu_lpage() now is a simple wrapper
around the generalized version.  Other than taking size parameters and
using arch supplied callbacks to allocate/free/map memory,
pcpu_lpage_first_chunk() is identical to the original implementation.

This simplifies arch code and will help converting more archs to
dynamic percpu allocator.

While at it, factor out pcpu_calc_fc_sizes() which is common to
pcpu_embed_first_chunk() and pcpu_lpage_first_chunk().

[ Impact: code reorganization and generalization ]
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>

lpage allocator aliases a PMD page for each cpu and returns whatever
is unused to the page allocator.  When the pageattr of the recycled
pages are changed, this makes the two aliases point to the overlapping
regions with different attributes which isn't allowed and known to
cause subtle data corruption in certain cases.

This can be handled in simliar manner to the x86_64 highmap alias.
pageattr code should detect if the target pages have PMD alias and
split the PMD alias and synchronize the attributes.

pcpur allocator is updated to keep the allocated PMD pages map sorted
in ascending address order and provide pcpu_lpage_remapped() function
which binary searches the array to determine whether the given address
is aliased and if so to which address.  pageattr is updated to use
pcpu_lpage_remapped() to detect the PMD alias and split it up as
necessary from cpa_process_alias().

Jan Beulich spotted the original problem and incorrect usage of vaddr
instead of laddr for lookup.

With this, lpage percpu allocator should work correctly.  Re-enable
it.

[ Impact: fix subtle lpage pageattr bug and re-enable lpage ]
Signed-off-by: NTejun Heo <tj@kernel.org>
Reported-by: NJan Beulich <JBeulich@novell.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Ingo Molnar <mingo@elte.hu>

Reorganize cpa_process_alias() so that new alias condition can be
added easily.

Jan Beulich spotted problem in the original cleanup thread which
incorrectly assumed the two existing conditions were mutially
exclusive.

[ Impact: code reorganization ]
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Jan Beulich <JBeulich@novell.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Ingo Molnar <mingo@elte.hu>

As these are allocated using the page allocator, we need to pass
__GFP_NOTRACK before we add page allocator support to kmemcheck.
Signed-off-by: NVegard Nossum <vegard.nossum@gmail.com>

Cleanup cpa_flush_array() to avoid back to back on_each_cpu() calls.

[ Impact: optimizes fix 0af48f42 ]
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

cpa_flush_array seems to prefer wbinvd() over clflush at 4M threshold.
clflush needs to be done on only one CPU as per instruction definition.
wbinvd() however, should be done on all CPUs.

[ Impact: fix missing flush which could cause data corruption ]
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

wbinvd is supported on all CPUs 486 or later. But,
pageattr.c is checking x86_model >= 4 before wbinvd(), which looks like
an oversight bug. It was first introduced at one place by changeset
d7c8f21a and got copied over to second
place in the same file later.

[ Impact: fix missing cache flush on early-model CPUs, potential data corruption ]
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

As per SDM, there should not be any aliasing of a WC with any cacheable
type across CPUs. That is if one CPU is changing the identity map
memtype to _WC, no other CPU at the time of this change should not have a
TLB for this page that carries a WB attribute. SDM suggests to make the
page not present. But for that we will have to handle any page faults
that can potentially happen due to these pages being not present.

Other way to deal with this without having any WB mapping is to change
the page first to UC and then to WC. This ensures that we meet the SDM
requirement of no cacheable alais to WC page. This also has same or
lower overhead than marking the page not present and making it present
later.
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20090409212708.797481000@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Handle faults and do proper cleanups in set_memory_*() functions. In
some cases, these functions were not doing proper free on failure paths.

With the changes to tracking memtype of RAM pages in struct page instead
of pat list, we do not need the changes in commits c5e147. This patch
reverts that change.
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20090409212708.653222000@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

To be free of aliasing due to races, set_memory_* interfaces should
follow ordering of reserving, changing memtype to UC/WC, changing
memtype back to WB followed by free.
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20090409212708.512280000@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: simplification, robustness

Make paravirt_lazy_mode() always return PARAVIRT_LAZY_NONE
when in an interrupt.  This prevents interrupt code from
accidentally inheriting an outer lazy state, and instead
does everything synchronously.  Outer batched operations
are left deferred.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>

Add new interfaces:

  set_pages_array_uc()
  set_pages_array_wb()

that can be used change the page attribute for a bunch of pages with
flush etc done once at the end of all the changes. These interfaces
are similar to existing set_memory_array_uc() and set_memory_array_wc().
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: arjan@infradead.org
Cc: eric@anholt.net
Cc: airlied@redhat.com
LKML-Reference: <20090319215358.901545000@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Add struct page array pointer to cpa struct and CPA_PAGES_ARRAY.

With that we can add change_page_attr_set_clr() a parameter to pass
struct page array pointer and that can be handled by the underlying
cpa code.

cpa_flush_array() is also changed to support both addr array or
struct page pointer array, depending on the flag.
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: arjan@infradead.org
Cc: eric@anholt.net
Cc: airlied@redhat.com
LKML-Reference: <20090319215358.758513000@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Change change_page_attr_set_clr() array parameter to a flag. This helps
following patches which adds an interface to change attr to uc/wb over a
set of pages referred by struct page.
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: arjan@infradead.org
Cc: eric@anholt.net
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: airlied@redhat.com
LKML-Reference: <20090319215358.611346000@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: new interface

Add a brk()-like allocator which effectively extends the bss in order
to allow very early code to do dynamic allocations.  This is better than
using statically allocated arrays for data in subsystems which may never
get used.

The space for brk allocations is in the bss ELF segment, so that the
space is mapped properly by the code which maps the kernel, and so
that bootloaders keep the space free rather than putting a ramdisk or
something into it.

The bss itself, delimited by __bss_stop, ends before the brk area
(__brk_base to __brk_limit).  The kernel text, data and bss is reserved
up to __bss_stop.

Any brk-allocated data is reserved separately just before the kernel
pagetable is built, as that code allocates from unreserved spaces
in the e820 map, potentially allocating from any unused brk memory.
Ultimately any unused memory in the brk area is used in the general
kernel memory pool.

Initially the brk space is set to 1MB, which is probably much larger
than any user needs (the largest current user is i386 head_32.S's code
to build the pagetables to map the kernel, which can get fairly large
with a big kernel image and no PSE support).  So long as the system
has sufficient memory for the bootloader to reserve the kernel+1MB brk,
there are no bad effects resulting from an over-large brk.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Impact: work around boot crash

Work around Intel Atom erratum AAH41 (probabilistically) - it's triggering
in the field.
Reported-by: NLinus Torvalds <torvalds@linux-foundation.org>
Tested-by: NKyle McMartin <kyle@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: future-proof the split_large_page() function

Linus noticed that split_large_page() is not safe wrt. the
PAT bit: it is bit 12 on the 1GB and 2MB page table level
(_PAGE_BIT_PAT_LARGE), and it is bit 7 on the 4K page
table level (_PAGE_BIT_PAT).

Currently it is not a problem because we never set
_PAGE_BIT_PAT_LARGE on any of the large-page mappings - but
should this happen in the future the split_large_page() would
silently lift bit 12 into the lowlevel 4K pte and would start
corrupting the physical page frame offset. Not fun.

So add a debug warning, to make sure if something ever sets
the PAT bit then this function gets updated too.

Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Steven Rostedt found a bug in where in his modified kernel
ftrace was unable to modify the kernel text, due to the PMD
itself having been marked read-only as well in
split_large_page().

The fix, suggested by Linus, is to not try to 'clone' the
reference protection of a huge-page, but to use the standard
(and permissive) page protection bits of KERNPG_TABLE.

The 'cloning' makes sense for the ptes but it's a confused and
incorrect concept at the page table level - because the
pagetable entry is a set of all ptes and hence cannot
'clone' any single protection attribute - the ptes can be any
mixture of protections.

With the permissive KERNPG_TABLE, even if the pte protections
get changed after this point (due to ftrace doing code-patching
or other similar activities like kprobes), the resulting combined
protections will still be correct and the pte's restrictive
(or permissive) protections will control it.

Also update the comment.

This bug was there for a long time but has not caused visible
problems before as it needs a rather large read-only area to
trigger. Steve possibly hacked his kernel with some really
large arrays or so. Anyway, the bug is definitely worth fixing.

[ Huang Ying also experienced problems in this area when writing
  the EFI code, but the real bug in split_large_page() was not
  realized back then. ]
Reported-by: NSteven Rostedt <rostedt@goodmis.org>
Reported-by: NHuang Ying <ying.huang@intel.com>
Acked-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: Flush the lazy MMU only once

Pending mmu updates only need to be flushed once to bring the
in-memory pagetable state up to date.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

Impact: fix race leading to crash under KVM and Xen

The CPA code may be called while we're in lazy mmu update mode - for
example, when using DEBUG_PAGE_ALLOC and doing a slab allocation
in an interrupt handler which interrupted a lazy mmu update.  In this
case, the in-memory pagetable state may be out of date due to pending
queued updates.  We need to flush any pending updates before inspecting
the page table.  Similarly, we must explicitly flush any modifications
CPA may have made (which comes down to flushing queued operations when
flushing the TLB).
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: NMarcelo Tosatti <mtosatti@redhat.com>
Cc: Stable Kernel <stable@kernel.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: fix sporadic slowdowns and warning messages

This patch fixes a performance issue reported by Linus on his
Nehalem system. While Linus reverted the PAT patch (commit
58dab916) which exposed the issue,
existing cpa() code can potentially still cause wrong(page attribute
corruption) behavior.

This patch also fixes the "WARNING: at arch/x86/mm/pageattr.c:560" that
various people reported.

In 64bit kernel, kernel identity mapping might have holes depending
on the available memory and how e820 reports the address range
covering the RAM, ACPI, PCI reserved regions. If there is a 2MB/1GB hole
in the address range that is not listed by e820 entries, kernel identity
mapping will have a corresponding hole in its 1-1 identity mapping.

If cpa() happens on the kernel identity mapping which falls into these holes,
existing code fails like this:

	__change_page_attr_set_clr()
		__change_page_attr()
			returns 0 because of if (!kpte). But doesn't
			set cpa->numpages and cpa->pfn.
		cpa_process_alias()
			uses uninitialized cpa->pfn (random value)
			which can potentially lead to changing the page
			attribute of kernel text/data, kernel identity
			mapping of RAM pages etc. oops!

This bug was easily exposed by another PAT patch which was doing
cpa() more often on kernel identity mapping holes (physical range between
max_low_pfn_mapped and 4GB), where in here it was setting the
cache disable attribute(PCD) for kernel identity mappings aswell.

Fix cpa() to handle the kernel identity mapping holes. Retain
the WARN() for cpa() calls to other not present address ranges
(kernel-text/data, ioremap() addresses)
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: <stable@kernel.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This reverts commit 58dab916, which
makes my Nehalem come to a nasty crawling almost-halt.  It looks like it
turns off caching of regular kernel RAM, with the understandable
slowdown of a few orders of magnitude as a result.
Acked-by: NIngo Molnar <mingo@elte.hu>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Peter Anvin <hpa@zytor.com>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

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.

[   27.400018] ------------[ cut here ]------------
[   27.400344] WARNING: at /home/venkip/src/linus/linux-2.6/arch/x86/mm/pageattr.c:560 __change_page_attr_set_clr+0xf3/0x8b8()
[   27.400821] Hardware name: X7DB8
[   27.401070] CPA: called for zero pte. vaddr = ffff8800cff6a000 cpa->vaddr = ffff8800cff6a000
[   27.401569] Modules linked in:
[   27.401882] Pid: 4913, comm: dmidecode Not tainted 2.6.28-05716-gfe0bdec6 #586
[   27.402141] Call Trace:
[   27.402488]  [<ffffffff80237c21>] warn_slowpath+0xd3/0x10f
[   27.402749]  [<ffffffff80274ade>] ? find_get_page+0xb3/0xc9
[   27.403028]  [<ffffffff80274a2b>] ? find_get_page+0x0/0xc9
[   27.403333]  [<ffffffff80226425>] __change_page_attr_set_clr+0xf3/0x8b8
[   27.403628]  [<ffffffff8028ec99>] ? __purge_vmap_area_lazy+0x192/0x1a1
[   27.403883]  [<ffffffff8028eb52>] ? __purge_vmap_area_lazy+0x4b/0x1a1
[   27.404172]  [<ffffffff80290268>] ? vm_unmap_aliases+0x1ab/0x1bb
[   27.404512]  [<ffffffff80290105>] ? vm_unmap_aliases+0x48/0x1bb
[   27.404766]  [<ffffffff80226d28>] change_page_attr_set_clr+0x13e/0x2e6
[   27.405026]  [<ffffffff80698fa7>] ? _spin_unlock+0x26/0x2a
[   27.405292]  [<ffffffff80227e6a>] ? reserve_memtype+0x19b/0x4e3
[   27.405590]  [<ffffffff80226ffd>] _set_memory_wb+0x22/0x24
[   27.405844]  [<ffffffff80225d28>] ioremap_change_attr+0x26/0x28
[   27.406097]  [<ffffffff80228355>] reserve_pfn_range+0x1a3/0x235
[   27.406427]  [<ffffffff80228430>] track_pfn_vma_new+0x49/0xb3
[   27.406686]  [<ffffffff80286c46>] remap_pfn_range+0x94/0x32c
[   27.406940]  [<ffffffff8022878d>] ? phys_mem_access_prot_allowed+0xb5/0x1a8
[   27.407209]  [<ffffffff803e9bf4>] mmap_mem+0x75/0x9d
[   27.407523]  [<ffffffff8028b3b4>] mmap_region+0x2cf/0x53e
[   27.407776]  [<ffffffff8028b8cc>] do_mmap_pgoff+0x2a9/0x30d
[   27.408034]  [<ffffffff8020f4a4>] sys_mmap+0x92/0xce
[   27.408339]  [<ffffffff8020b65b>] system_call_fastpath+0x16/0x1b
[   27.408614] ---[ end trace 4b16ad70c09a602d ]---
[   27.408871] dmidecode:4913 reserve_pfn_range ioremap_change_attr failed write-back for cff6a000-cff6b000

This is wih track_pfn_vma_new trying to keep identity map in sync.
The address cff6a000 is the ACPI region according to e820.

[    0.000000] BIOS-provided physical RAM map:
[    0.000000]  BIOS-e820: 0000000000000000 - 000000000009c000 (usable)
[    0.000000]  BIOS-e820: 000000000009c000 - 00000000000a0000 (reserved)
[    0.000000]  BIOS-e820: 00000000000cc000 - 00000000000d0000 (reserved)
[    0.000000]  BIOS-e820: 00000000000e4000 - 0000000000100000 (reserved)
[    0.000000]  BIOS-e820: 0000000000100000 - 00000000cff60000 (usable)
[    0.000000]  BIOS-e820: 00000000cff60000 - 00000000cff69000 (ACPI data)
[    0.000000]  BIOS-e820: 00000000cff69000 - 00000000cff80000 (ACPI NVS)
[    0.000000]  BIOS-e820: 00000000cff80000 - 00000000d0000000 (reserved)
[    0.000000]  BIOS-e820: 00000000e0000000 - 00000000f0000000 (reserved)
[    0.000000]  BIOS-e820: 00000000fec00000 - 00000000fec10000 (reserved)
[    0.000000]  BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved)
[    0.000000]  BIOS-e820: 00000000ff000000 - 0000000100000000 (reserved)
[    0.000000]  BIOS-e820: 0000000100000000 - 0000000230000000 (usable)

And is not mapped as per init_memory_mapping.

[    0.000000] init_memory_mapping: 0000000000000000-00000000cff60000
[    0.000000] init_memory_mapping: 0000000100000000-0000000230000000

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.
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: right-align /proc/meminfo consistent with other fields

When the split-LRU patches added Inactive(anon) and Inactive(file) lines
to /proc/meminfo, all counts were moved two columns rightwards to fit in.
Now move x86's DirectMap lines two columns rightwards to line up.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

and move it to fs/proc/meminfo.c while I'm at it.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>

Rewrite the vmap allocator to use rbtrees and lazy tlb flushing, and
provide a fast, scalable percpu frontend for small vmaps (requires a
slightly different API, though).

The biggest problem with vmap is actually vunmap.  Presently this requires
a global kernel TLB flush, which on most architectures is a broadcast IPI
to all CPUs to flush the cache.  This is all done under a global lock.  As
the number of CPUs increases, so will the number of vunmaps a scaled
workload will want to perform, and so will the cost of a global TLB flush.
 This gives terrible quadratic scalability characteristics.

Another problem is that the entire vmap subsystem works under a single
lock.  It is a rwlock, but it is actually taken for write in all the fast
paths, and the read locking would likely never be run concurrently anyway,
so it's just pointless.

This is a rewrite of vmap subsystem to solve those problems.  The existing
vmalloc API is implemented on top of the rewritten subsystem.

The TLB flushing problem is solved by using lazy TLB unmapping.  vmap
addresses do not have to be flushed immediately when they are vunmapped,
because the kernel will not reuse them again (would be a use-after-free)
until they are reallocated.  So the addresses aren't allocated again until
a subsequent TLB flush.  A single TLB flush then can flush multiple
vunmaps from each CPU.

XEN and PAT and such do not like deferred TLB flushing because they can't
always handle multiple aliasing virtual addresses to a physical address.
They now call vm_unmap_aliases() in order to flush any deferred mappings.
That call is very expensive (well, actually not a lot more expensive than
a single vunmap under the old scheme), however it should be OK if not
called too often.

The virtual memory extent information is stored in an rbtree rather than a
linked list to improve the algorithmic scalability.

There is a per-CPU allocator for small vmaps, which amortizes or avoids
global locking.

To use the per-CPU interface, the vm_map_ram / vm_unmap_ram interfaces
must be used in place of vmap and vunmap.  Vmalloc does not use these
interfaces at the moment, so it will not be quite so scalable (although it
will use lazy TLB flushing).

As a quick test of performance, I ran a test that loops in the kernel,
linearly mapping then touching then unmapping 4 pages.  Different numbers
of tests were run in parallel on an 4 core, 2 socket opteron.  Results are
in nanoseconds per map+touch+unmap.

threads           vanilla         vmap rewrite
1                 14700           2900
2                 33600           3000
4                 49500           2800
8                 70631           2900

So with a 8 cores, the rewritten version is already 25x faster.

In a slightly more realistic test (although with an older and less
scalable version of the patch), I ripped the not-very-good vunmap batching
code out of XFS, and implemented the large buffer mapping with vm_map_ram
and vm_unmap_ram...  along with a couple of other tricks, I was able to
speed up a large directory workload by 20x on a 64 CPU system.  I believe
vmap/vunmap is actually sped up a lot more than 20x on such a system, but
I'm running into other locks now.  vmap is pretty well blown off the
profiles.

Before:
1352059 total                                      0.1401
798784 _write_lock                              8320.6667 <- vmlist_lock
529313 default_idle                             1181.5022
 15242 smp_call_function                         15.8771  <- vmap tlb flushing
  2472 __get_vm_area_node                         1.9312  <- vmap
  1762 remove_vm_area                             4.5885  <- vunmap
   316 map_vm_area                                0.2297  <- vmap
   312 kfree                                      0.1950
   300 _spin_lock                                 3.1250
   252 sn_send_IPI_phys                           0.4375  <- tlb flushing
   238 vmap                                       0.8264  <- vmap
   216 find_lock_page                             0.5192
   196 find_next_bit                              0.3603
   136 sn2_send_IPI                               0.2024
   130 pio_phys_write_mmr                         2.0312
   118 unmap_kernel_range                         0.1229

After:
 78406 total                                      0.0081
 40053 default_idle                              89.4040
 33576 ia64_spinlock_contention                 349.7500
  1650 _spin_lock                                17.1875
   319 __reg_op                                   0.5538
   281 _atomic_dec_and_lock                       1.0977
   153 mutex_unlock                               1.5938
   123 iget_locked                                0.1671
   117 xfs_dir_lookup                             0.1662
   117 dput                                       0.1406
   114 xfs_iget_core                              0.0268
    92 xfs_da_hashname                            0.1917
    75 d_alloc                                    0.0670
    68 vmap_page_range                            0.0462 <- vmap
    58 kmem_cache_alloc                           0.0604
    57 memset                                     0.0540
    52 rb_next                                    0.1625
    50 __copy_user                                0.0208
    49 bitmap_find_free_region                    0.2188 <- vmap
    46 ia64_sn_udelay                             0.1106
    45 find_inode_fast                            0.1406
    42 memcmp                                     0.2188
    42 finish_task_switch                         0.1094
    42 __d_lookup                                 0.0410
    40 radix_tree_lookup_slot                     0.1250
    37 _spin_unlock_irqrestore                    0.3854
    36 xfs_bmapi                                  0.0050
    36 kmem_cache_free                            0.0256
    35 xfs_vn_getattr                             0.0322
    34 radix_tree_lookup                          0.1062
    33 __link_path_walk                           0.0035
    31 xfs_da_do_buf                              0.0091
    30 _xfs_buf_find                              0.0204
    28 find_get_page                              0.0875
    27 xfs_iread                                  0.0241
    27 __strncpy_from_user                        0.2812
    26 _xfs_buf_initialize                        0.0406
    24 _xfs_buf_lookup_pages                      0.0179
    24 vunmap_page_range                          0.0250 <- vunmap
    23 find_lock_page                             0.0799
    22 vm_map_ram                                 0.0087 <- vmap
    20 kfree                                      0.0125
    19 put_page                                   0.0330
    18 __kmalloc                                  0.0176
    17 xfs_da_node_lookup_int                     0.0086
    17 _read_lock                                 0.0885
    17 page_waitqueue                             0.0664

vmap has gone from being the top 5 on the profiles and flushing the crap
out of all TLBs, to using less than 1% of kernel time.

[akpm@linux-foundation.org: cleanups, section fix]
[akpm@linux-foundation.org: fix build on alpha]
Signed-off-by: NNick Piggin <npiggin@suse.de>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Do a global flush tlb after splitting the large page and before we do the
actual change page attribute in the PTE.

With out this, we violate the TLB application note, which says
    "The TLBs may contain both ordinary and large-page translations for
     a 4-KByte range of linear addresses. This may occur if software
     modifies the paging structures so that the page size used for the
     address range changes. If the two translations differ with respect
     to page frame or attributes (e.g., permissions), processor behavior
     is undefined and may be implementation-specific."

And also serialize cpa() (for !DEBUG_PAGEALLOC which uses large identity
mappings) using cpa_lock. So that we don't allow any other cpu, with stale
large tlb entries change the page attribute in parallel to some other cpu
splitting a large page entry along with changing the attribute.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: arjan@linux.intel.com
Cc: venkatesh.pallipadi@intel.com
Cc: jeremy@goop.org
Signed-off-by: NIngo Molnar <mingo@elte.hu>