The following two commits fixed a problem that x86 ioremap() doesn't handle
physical address higher than 32-bit properly in X86_32 PAE mode.
ffa71f33a8 (x86, ioremap: Fix incorrect
physical address handling in PAE mode)
35be1b716a (x86, ioremap: Fix normal
ram range check)
But these fixes are not enough, since pat_pagerange_is_ram() in PAT code
also has a same problem. This patch fixes it.
Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Reviewed-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
LKML-Reference: <4C47DDCF.80300@jp.fujitsu.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
subtree_max_end that was recently added to struct memtype was not getting
properly initialized resulting in
WARNING: kmemcheck: Caught 64-bit read from uninitialized memory
in memtype_rb_augment_cb()
reported here
https://bugzilla.kernel.org/show_bug.cgi?id=16092
This change fixes the problem.
Reported-by: Christian Casteyde <casteyde.christian@free.fr>
Tested-by: Christian Casteyde <casteyde.christian@free.fr>
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
LKML-Reference: <1276217101-11515-1-git-send-email-venki@google.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Reserve_memtype will allocate memory for new memtype, but
in free_memtype, after the memtype erased from rbtree, the
memory is not freed.
Changes since V1:
make rbt_memtype_erase return erased memtype so that
it can be freed in free_memtype.
[ hpa: not for -stable: 2.6.34 and earlier not affected ]
Signed-off-by: Xiaotian Feng <dfeng@redhat.com>
LKML-Reference: <1274838670-8731-1-git-send-email-dfeng@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Jack Steiner <steiner@sgi.com>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
While testing an application using the xpmem (out of kernel) driver, we
noticed a significant page fault rate reduction of x86_64 with respect
to ia64. For one test running with 32 cpus, one thread per cpu, it
took 01:08 for each of the threads to vm_insert_pfn 2GB worth of pages.
For the same test running on 256 cpus, one thread per cpu, it took 14:48
to vm_insert_pfn 2 GB worth of pages.
The slowdown was tracked to lookup_memtype which acquires the
spinlock memtype_lock. This heavily contended lock was slowing down
vm_insert_pfn().
With the cmpxchg on page->flags method, both the 32 cpu and 256 cpu
cases take approx 00:01.3 seconds to complete.
Signed-off-by: Robin Holt <holt@sgi.com>
LKML-Reference: <20100423153627.751194346@gulag1.americas.sgi.com>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@gmail.com>
Cc: Rafael Wysocki <rjw@novell.com>
Reviewed-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
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: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Move pat backend to fully rbtree based implementation from the existing
rbtree and linked list hybrid.
New rbtree based solution uses interval trees (augmented rbtrees) in
order to store the PAT ranges. The new code seprates out the pat backend
to pat_rbtree.c file, making is cleaner. The change also makes the PAT
lookup, reserve and free operations more optimal, as we don't have to
traverse linear linked list of few tens of entries in normal case.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
LKML-Reference: <20100210232607.GB11465@linux-os.sc.intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Minor changes in pat.c to cleanup code and make it smoother to introduce
bigger rbtree only change in the following patch. The changes are cleaup
only and should not have any functional impact.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
LKML-Reference: <20100210195909.792781000@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
While Linux provided an O_SYNC flag basically since day 1, it took until
Linux 2.4.0-test12pre2 to actually get it implemented for filesystems,
since that day we had generic_osync_around with only minor changes and the
great "For now, when the user asks for O_SYNC, we'll actually give
O_DSYNC" comment. This patch intends to actually give us real O_SYNC
semantics in addition to the O_DSYNC semantics. After Jan's O_SYNC
patches which are required before this patch it's actually surprisingly
simple, we just need to figure out when to set the datasync flag to
vfs_fsync_range and when not.
This patch renames the existing O_SYNC flag to O_DSYNC while keeping it's
numerical value to keep binary compatibility, and adds a new real O_SYNC
flag. To guarantee backwards compatiblity it is defined as expanding to
both the O_DSYNC and the new additional binary flag (__O_SYNC) to make
sure we are backwards-compatible when compiled against the new headers.
This also means that all places that don't care about the differences can
just check O_DSYNC and get the right behaviour for O_SYNC, too - only
places that actuall care need to check __O_SYNC in addition. Drivers and
network filesystems have been updated in a fail safe way to always do the
full sync magic if O_DSYNC is set. The few places setting O_SYNC for
lower layers are kept that way for now to stay failsafe.
We enforce that O_DSYNC is set when __O_SYNC is set early in the open path
to make sure we always get these sane options.
Note that parisc really screwed up their headers as they already define a
O_DSYNC that has always been a no-op. We try to repair it by using it for
the new O_DSYNC and redefinining O_SYNC to send both the traditional
O_SYNC numerical value _and_ the O_DSYNC one.
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Grant Grundler <grundler@parisc-linux.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andreas Dilger <adilger@sun.com>
Acked-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Acked-by: Kyle McMartin <kyle@mcmartin.ca>
Acked-by: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jan Kara <jack@suse.cz>
If pat is disabled (boot with nopat), there's no need to create
debugfs for it, it's empty all the time.
Signed-off-by: Xiaotian Feng <dfeng@redhat.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
LKML-Reference: <1259236428-16329-1-git-send-email-dfeng@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
- Change is_untracked_pat_range() to return bool.
- Clean up the initialization of is_untracked_pat_range() -- by default,
we simply point it at is_ISA_range() directly.
- Move is_untracked_pat_range to the end of struct x86_platform, since
it is the newest field.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Jack Steiner <steiner@sgi.com>
LKML-Reference: <20091119202341.GA4420@sgi.com>
is_untracked_pat_range() -- like its components, is_ISA_range() and
is_GRU_range(), takes a normal semiclosed interval (>=, <) whereas the
PAT code called it as if it took a closed range (>=, <=). Fix.
Although this is a bug, I believe it is non-manifest, simply because
none of the callers will call this with non-page-aligned addresses.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20091119202341.GA4420@sgi.com>
GRU space is always mapped as WB in the page table. There is
no need to track the mappings in the PAT. This also eliminates
the "freeing invalid memtype" messages when the GRU space is
unmapped.
Signed-off-by: Jack Steiner <steiner@sgi.com>
LKML-Reference: <20091119202341.GA4420@sgi.com>
[ v2: fix build failure ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Commit:
b6ff32d: x86, PAT: Consolidate code in pat_x_mtrr_type() and reserve_memtype()
consolidated code in pat_x_mtrr_type() and reserve_memtype(),
which removed the special case (req_type is -1) for the
PAT-enabled part.
We should also change comments and the PAT-disabled part.
Signed-off-by: Xiaotian Feng <dfeng@redhat.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
LKML-Reference: <1257844987-7906-1-git-send-email-dfeng@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
On modern systems, the kernel prints the message
x86 PAT enabled: cpu 0, old 0x7040600070406, new 0x7010600070106
once for every CPU.
This gets kind of ridiculous on huge systems; for example, on a
64-thread system I was lucky enough to get:
dmesg| grep 'PAT enabled' | wc
64 704 5174
There is already a BUG() if non-boot CPUs have PAT capabilities
that don't match the boot CPU, so just print the message on the
boot CPU. (I kept the print after the wrmsrl() that enables PAT,
so that the log output continues to mean that the system survived
enabling PAT on the boot CPU)
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
LKML-Reference: <adavdj92sso.fsf@cisco.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Recent enhancement of rb-tree based lookup exposed a bug with the lookup
mechanism in the reserve_memtype() which ensures that there are no conflicting
memtype requests for the memory range.
memtype_rb_search() returns an entry which has a start address <= new start
address. And from here we traverse the linear linked list to check if there
any conflicts with the existing mappings. As the rbtree is based on the
start address of the memory range, it is quite possible that we have several
overlapped mappings whose start address is much less than new requested start
but the end is >= new requested end. This results in conflicting memtype
mappings.
Same bug exists with the old code which uses cached_entry from where
we traverse the linear linked list. But the new rb-tree code exposes this
bug fairly easily.
For now, don't use the memtype_rb_search() and always start the search from
the head of linear linked list in reserve_memtype(). Linear linked list
for most of the systems grow's to few 10's of entries(as we track memory type
of RAM pages using struct page). So we should be ok for now.
We still retain the rbtree and use it to speed up free_memtype() which
doesn't have the same bug(as we know what exactly we are searching for
in free_memtype).
Also use list_for_each_entry_from() in free_memtype() so that we start
the search from rb-tree lookup result.
Reported-by: Markus Trippelsdorf <markus@trippelsdorf.de>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
LKML-Reference: <1253136483.4119.12.camel@sbs-t61.sc.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Add sanity check for remap_pfn_range of RAM regions using
lookup_memtype(). Previously, we did not have anyway to get the type of
RAM memory regions as they were tracked using a single bit in
page_struct (WB, nonWB). Now we can get the actual type from page struct
(WB, WC, UC_MINUS) and make sure the requester gets that type.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Lookup the reserved memtype during vm_insert_pfn and use that memtype
for the new mapping. This takes care or handling of vm_insert_pfn()
interface in track_pfn_vma*/untrack_pfn_vma.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Add a new routine lookup_memtype() to get the current memtype based on
the PAT reserves and frees.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Change reserve_ram_pages_type and free_ram_pages_type to use 2 page
flags to track UC_MINUS, WC, WB and default types. Previous RAM tracking
just tracked WB or NonWB, which was not complete and did not allow
tracking of RAM fully and there was no way to get the actual type
reserved by looking at the page flags.
We use the memtype_lock spinlock for atomicity in dealing with
memtype tracking in struct page.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
PAT memtype tracking uses a linear link list to keep track of IO
(non-RAM) regions and their memtypes. The code used a last_accessed
pointer as a cache to speedup the lookup. As per discussions with
H. Peter Anvin a while back, having a rbtree here will avoid bad
performances in pathological cases where we may end up with huge
linked list. This may not add any noticable performance speedup
in normal case as the number of entires in PAT memtype list tend
to be ~20-30 range. The patch removes the "cached_entry" logic
as with rbtree we have more generic way of speeding up the lookup.
With this patch, we use rbtree to do the quick lookup. We still use
linked list as the memtype range tracked can be of different sizes
and can overlap in different ways. We also keep track of usage counts
with linked list.
Example:
Multiple ioremaps with different sizes
uncached-minus @ 0xfffff00000-0xfffff04000
uncached-minus @ 0xfffff02000-0xfffff03000
And one userlevel mmap and the thread forks a new process
uncached-minus @ 0xbf453000-0xbf454000
uncached-minus @ 0xbf453000-0xbf454000
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Add new routines to request memtype for IO regions. This will currently
be a backend for io_mapping_* routines. But, it can also be made available
to drivers directly in future, in case it is needed.
reserve interface reserves the memory, makes sure we have a compatible
memory type available and keeps the identity map in sync when needed.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Make reserve_memtype internally take care of pat disabled case and fallback
to default return values.
Remove the specific pat_disabled checks in track_* routines.
Change kernel_map_sync_memtype to sync identity map even when
pat_disabled.
This change ensures that, even for pat_disabled case, we take care of
keeping identity map in sync. Before this patch, in pat disabled case,
ioremap() keeps the identity maps in sync and other APIs like pci and
/dev/mem mmap don't, which is not a very consistent behavior.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Max Vozeler reported:
> Bug 13877 - bogl-term broken with CONFIG_X86_PAT=y, works with =n
>
> strace of bogl-term:
> 814 mmap2(NULL, 65536, PROT_READ|PROT_WRITE, MAP_SHARED, 4, 0)
> = -1 EAGAIN (Resource temporarily unavailable)
> 814 write(2, "bogl: mmaping /dev/fb0: Resource temporarily unavailable\n",
> 57) = 57
PAT code maps the ISA memory range as WB in the PAT attribute, so that
fixed range MTRR registers define the actual memory type (UC/WC/WT etc).
But the upper level is_new_memtype_allowed() API checks are failing,
as the request here is for UC and the return tracked type is WB (Tracked type is
WB as MTRR type for this legacy range potentially will be different for each
4k page).
Fix is_new_memtype_allowed() by always succeeding the ISA address range
checks, as the null PAT (WB) and def MTRR fixed range register settings
satisfy the memory type needs of the applications that map the ISA address
range.
Reported-and-Tested-by: Max Vozeler <xam@debian.org>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
This change resolves the problem of too many single page entries
in pat_memtype_list and "freeing invalid memtype" errors with i915,
reported here:
http://marc.info/?l=linux-kernel&m=123845244713183&w=2
Remove page level granularity track and untrack of vm_insert_pfn.
memtype tracking at page granularity does not scale and cleaner
approach would be for the driver to request a type for a bigger
IO address range or PCI io memory range for that device, either at
mmap time or driver init time and just use that type during
vm_insert_pfn.
This patch just removes the track/untrack of vm_insert_pfn. That
means we will be in same state as 2.6.28, with respect to these APIs.
Newer APIs for the drivers to request a memtype for a bigger region
is coming soon.
[ Impact: fix Xorg startup warnings and hangs ]
Reported-by: Arkadiusz Miskiewicz <a.miskiewicz@gmail.com>
Tested-by: Arkadiusz Miskiewicz <a.miskiewicz@gmail.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
LKML-Reference: <20090408223716.GC3493@linux-os.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
pat_disable cannot be __cpuinit anymore because it's called from pat_init
and the callchain looks like this:
pat_disable [cpuinit] <- pat_init <- generic_set_all <-
ipi_handler <- set_mtrr <- (other non init/cpuinit functions)
WARNING: arch/x86/mm/built-in.o(.text+0x449e): Section mismatch in reference
from the function pat_init() to the function .cpuinit.text:pat_disable()
The function pat_init() references
the function __cpuinit pat_disable().
This is often because pat_init lacks a __cpuinit
annotation or the annotation of pat_disable is wrong.
Non CONFIG_X86_PAT version of pat_disable is static inline, so this version
can be static too (and there are no callers outside of this file).
Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
LKML-Reference: <49DFB055.6070405@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
/dev/mem mmap code was doing memtype reserve/free for a while now.
Recently we added memtype tracking in remap_pfn_range, and /dev/mem mmap
uses it indirectly. So, we don't need seperate tracking in /dev/mem code
any more. That means another ~100 lines of code removed :-).
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
LKML-Reference: <20090409212709.085210000@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix pat_x_mtrr_type() to use UC_MINUS when the mtrr type return UC. This
is to be consistent with ioremap() and ioremap_nocache() which uses
UC_MINUS.
Consolidate the code such that reserve_memtype() also uses
pat_x_mtrr_type() when the caller doesn't specify any special attribute
(non WB attribute).
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
LKML-Reference: <20090409212708.939936000@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix false positive PAT warnings - also fix VirtalBox hang
Use of vma->vm_pgoff to identify the pfnmaps that are fully
mapped at mmap time is broken. vm_pgoff is set by generic mmap
code even for cases where drivers are setting up the mappings
at the fault time.
The problem was originally reported here:
http://marc.info/?l=linux-kernel&m=123383810628583&w=2
Change is_linear_pfn_mapping logic to overload VM_INSERTPAGE
flag along with VM_PFNMAP to mean full PFNMAP setup at mmap
time.
Problem also tracked at:
http://bugzilla.kernel.org/show_bug.cgi?id=12800
Reported-by: Thomas Hellstrom <thellstrom@vmware.com>
Tested-by: Frans Pop <elendil@planet.nl>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha>@intel.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: "ebiederm@xmission.com" <ebiederm@xmission.com>
Cc: <stable@kernel.org> # only for 2.6.29.1, not .28
LKML-Reference: <20090313004527.GA7176@linux-os.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Add a function to check and keep identity maps in sync, when changing
any memory type. One of the follow on patches will also use this
routine.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: Eric Anholt <eric@anholt.net>
Cc: Keith Packard <keithp@keithp.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Jeff Mahoney reported:
> With Suse's hwinfo tool, on -tip:
> WARNING: at arch/x86/mm/pat.c:637 reserve_pfn_range+0x5b/0x26d()
reserve_pfn_range() is not tracking the memory range below 1MB
as non-RAM and as such is inconsistent with similar checks in
reserve_memtype() and free_memtype()
Rename the pagerange_is_ram() to pat_pagerange_is_ram() and add the
"track legacy 1MB region as non RAM" condition.
And also, fix reserve_pfn_range() to return -EINVAL, when the pfn
range is RAM. This is to be consistent with this API design.
Reported-and-tested-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: Cleanup
When PAT was originally introduced, it was handled specially for a few
reasons:
- PAT bugs are hard to track down, so we wanted to maintain a
whitelist of CPUs.
- The i386 and x86-64 CPUID code was not yet unified.
Both of these are now obsolete, so handle PAT like any other features,
including ordinary feature blacklisting due to known bugs.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Beschorner Daniel reported:
> hwinfo problem since 2.6.28, showing this in the oops:
> Corrupted page table at address 7fd04de3ec00
Also, PaX Team reported a regression with this commit:
> commit 9542ada803
> Author: Suresh Siddha <suresh.b.siddha@intel.com>
> Date: Wed Sep 24 08:53:33 2008 -0700
>
> x86: track memtype for RAM in page struct
This commit breaks mapping any RAM page through /dev/mem, as the
reserve_memtype() was not initializing the return attribute type and as such
corrupting the PTE entry that was setup with the return attribute type.
Because of this bug, application mapping this RAM page through /dev/mem
will die with "Corrupted page table at address xxxx" message in the kernel
log and also the kernel identity mapping which maps the underlying RAM
page gets converted to UC.
Fix this by initializing the return attribute type before calling
reserve_ram_pages_type()
Reported-by: PaX Team <pageexec@freemail.hu>
Reported-and-tested-by: Beschorner Daniel <Daniel.Beschorner@facton.com>
Tested-and-Acked-by: PaX Team <pageexec@freemail.hu>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This reverts commit 58dab916df, 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: Ingo 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: Linus Torvalds <torvalds@linux-foundation.org>
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)
Reported-and-tested-by: Thierry Vignaud <tvignaud@mandriva.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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-gfe0bdec #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: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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.
This patch fixes the bug reported here:
http://marc.info/?l=linux-kernel&m=123108883716357&w=2
Specifically the error message:
X:5010 map pfn expected mapping type write-back for d0000000-d0101000,
got write-combining
Should go away.
Reported-and-bisected-by: Kevin Winchester <kjwinchester@gmail.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup, fix warning
This warning:
arch/x86/mm/pat.c: In function track_pfn_vma_copy:
arch/x86/mm/pat.c:701: warning: passing argument 5 of follow_phys from incompatible pointer type
Triggers because physical addresses are resource_size_t, not u64.
This really matters when calling an interface like follow_phys() which
takes a pointer to a physical address -- although on x86, being
littleendian, it would generally work anyway as long as the memory region
wasn't completely uninitialized.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: Cleanup - removes a new function in favor of a recently modified older one.
Replace follow_pfnmap_pte in pat code with follow_phys. follow_phys lso
returns protection eliminating the need of pte_pgprot call. Using follow_phys
also eliminates the need for pte_pa.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: New mm functionality.
Add pgprot_writecombine. pgprot_writecombine will be aliased to
pgprot_noncached when not supported by the architecture.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: New mm functionality.
Hookup remap_pfn_range and vm_insert_pfn and corresponding copy and free
routines with reserve and free tracking.
reserve and free here only takes care of non RAM region mapping. For RAM
region, driver should use set_memory_[uc|wc|wb] to set the cache type and
then setup the mapping for user pte. We can bypass below
reserve/free in that case.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Impact: allow /dev/mem mmaps on non-PAT CPUs/platforms
Fix mmap to /dev/mem when CONFIG_X86_PAT is off and CONFIG_STRICT_DEVMEM is
off
mmap to /dev/mem on kernel memory has been failing since the
introduction of PAT (CONFIG_STRICT_DEVMEM=n case). Seems like
the check to avoid cache aliasing with PAT is kicking in even
when PAT is disabled. The bug seems to have crept in 2.6.26.
This patch makes sure that the mmap to regular
kernel memory succeeds if CONFIG_STRICT_DEVMEM=n and
PAT is disabled, and the checks to avoid cache aliasing
still happens if PAT is enabled.
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Tested-by: Tim Sirianni <tim@scalemp.com>
Cc: <stable@kernel.org>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Track the memtype for RAM pages in page struct instead of using the
memtype list. This avoids the explosion in the number of entries in
memtype list (of the order of 20,000 with AGP) and makes the PAT
tracking simpler.
We are using PG_arch_1 bit in page->flags.
We still use the memtype list for non RAM pages.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
All kernel mappings like ioremap(), etc uses UC_MINUS as the type. /dev/mem
mappings with /dev/mem being opened with O_SYNC however was using UC,
resulting in a conflict with /dev/mem mmap failing. This seems to be
affecting some apps (one being flashrom) which are using O_SYNC and which were
working before.
Switch /dev/mem with O_SYNC also to UC_MINUS.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Rene Herman reported significant Xorg startup/shutdown slowdown due
to PAT. It turns out that the memtype list has thousands of entries.
Add cached_entry to list add routine, in order to speed up the
lookup for sequential reserve_memtype calls.
Reported-by: Rene Herman <rene.herman@keyaccess.nl>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Recently I've run a kernel w/o PAT support and wondered why there was
a file "x86/pat_memtype_list" in debugfs. Of course it's empty if PAT
is disabled ... so just get rid of this interface if PAT is disabled.
Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
From: Arjan van de Ven <arjan@infradead.org>
Date: Sat, 19 Jul 2008 15:47:17 -0700
CONFIG_NONPROMISC_DEVMEM was a rather confusing name - but renaming it
to CONFIG_PROMISC_DEVMEM causes problems on architectures that do not
support this feature; this patch renames it to CONFIG_STRICT_DEVMEM,
so that architectures can opt-in into it.
( the polarity of the option is still the same as it was originally; it
needs to be for now to not break architectures that don't have the
infastructure yet to support this feature)
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: "V.Radhakrishnan" <rk@atr-labs.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
Yasuaki Ishimatsu