Clarify the usage of mtrr_lookup() in PAT code, and to make PAT code
resilient to mtrr lookup problems.
Specifically, pat_x_mtrr_type() is restructured to highlight, under what
conditions we look for mtrr hint. pat_x_mtrr_type() uses a default type
when there are any errors in mtrr lookup (still maintaining the pat
consistency). And, reserve_memtype() highlights its usage ot mtrr_lookup
for request type of '-1' and also defaults in a sane way on any mtrr
lookup failure.
pat.c looks at mtrr type of a range to get a hint on what mapping type
to request when user/API: (1) hasn't specified any type (/dev/mem
mapping) and we do not want to take performance hit by always mapping
UC_MINUS. This will be the case for /dev/mem mappings used to map BIOS
area or ACPI region which are WB'able. In this case, as long as MTRR is
not WB, PAT will request UC_MINUS for such mappings.
(2) user/API requests WB mapping while in reality MTRR may have UC or
WC. In this case, PAT can map as WB (without checking MTRR) and still
effective type will be UC or WC. But, a subsequent request to map same
region as UC or WC may fail, as the region will get trackked as WB in
PAT list. Looking at MTRR hint helps us to track based on effective type
rather than what user requested. Again, here mtrr_lookup is only used as
hint and we fallback to WB mapping (as requested by user) as default.
In both cases, after using the mtrr hint, we still go through the
memtype list to make sure there are no inconsistencies among multiple
users.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Tested-by: Rufus & Azrael <rufus-azrael@numericable.fr>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Changed the call to find_e820_area_size to pass u64 instead of unsigned long.
Signed-off-by: Kevin Winchester <kjwinchester@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
OGAWA Hirofumi and Fede have reported rare pmd_ERROR messages:
mm/memory.c:127: bad pmd ffff810000207xxx(9090909090909090).
Initialization's cleanup_highmap was leaving alignment filler
behind in the pmd for MODULES_VADDR: when vmalloc's guard page
would occupy a new page table, it's not allocated, and then
module unload's vfree hits the bad 9090 pmd entry left over.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
this way 32-bit is more similar to 64-bit, and smarter e820 and numa.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
when 1/3 user/kernel split is used, and less memory is installed, or if
we have a big hole below 4g, max_low_pfn is still using 3g-128m
try to go down from max_low_pfn until we get it. otherwise will panic.
need to make 32-bit code to use register_e820_active_regions ... later.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
on 32-bit in head_32.S after initial page table is done, we get initial
max_pfn_mapped, and then kernel_physical_mapping_init will give us
a final one.
We need to use that to make sure find_e820_area will get valid addresses
for boot_map and for NODE_DATA(0) on numa32.
XEN PV and lguest may need to assign max_pfn_mapped too.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
we don't need to call memory_present that early.
numa and sparse will call memory_present later and might
even fail, it will call memory_present for the full range.
also for sparse it will call alloc_bootmem ... before we set up bootmem.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
... otherwise alloc_remap will not get node_mem_map from kva area, and
alloc_node_mem_map has to alloc_bootmem_node to get mem_map.
It will use two low address copies ...
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
so every element will represent 64M instead of 256M.
AMD opteron could have HW memory hole remapping, so could have
[0, 8g + 64M) on node0. Reduce element size to 64M to keep that on node 0
Later we need to use find_e820_area() to allocate memory_node_map like
on 64-bit. But need to move memory_present out of populate_mem_map...
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
on Summit it's possible to have:
CONFIG_ACPI_SRAT=y
CONFIG_HAVE_ARCH_PARSE_SRAT=y
in which case acpi.h defines the acpi_numa_slit_init() and
acpi_numa_processor_affinity_init() methods as a macro.
reserve early numa kva, so it will not clash with new RAMDISK
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
(Updated with a common max-stack-used checker that knows about
the canary, as suggested by Joe Perches)
Use a canary at the end of the stack to clearly indicate
at oops time whether the stack has ever overflowed.
This is a very simple implementation with a couple of
drawbacks:
1) a thread may legitimately use exactly up to the last
word on the stack
-- but the chances of doing this and then oopsing later seem slim
2) it's possible that the stack usage isn't dense enough
that the canary location could get skipped over
-- but the worst that happens is that we don't flag the overrun
-- though this happens fairly often in my testing :(
With the code in place, an intentionally-bloated stack oops might
do:
BUG: unable to handle kernel paging request at ffff8103f84cc680
IP: [<ffffffff810253df>] update_curr+0x9a/0xa8
PGD 8063 PUD 0
Thread overran stack or stack corrupted
Oops: 0000 [1] SMP
CPU 0
...
... unless the stack overrun is so bad that it corrupts some other
thread.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Remove three extern declarations for routines
that don't exist. Fix a typo in a comment.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
memory_add_physaddr_to_nid() is only used in the
CONFIG_MEMORY_HOTPLUG_SPARSE || CONFIG_ACPI_HOTPLUG_MEMORY case.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
free_initrd_mem needs a prototype, which is in linux/initrd.h
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
sparse mutters:
arch/x86/mm/k8topology_64.c:108:7: warning: symbol 'nodeid' shadows an earlier one
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
k8_scan_nodes is global and needs a prototype. Add the header file
which contains it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
arch/x86/mm/pat.c: In function 'phys_mem_access_prot_allowed':
arch/x86/mm/pat.c:526: warning: passing argument 2 of 'constant_test_bit' from incompatible pointer type
arch/x86/mm/pat.c:526: warning: passing argument 2 of 'variable_test_bit' from incompatible pointer type
arch/x86/mm/pat.c:527: warning: passing argument 2 of 'constant_test_bit' from incompatible pointer type
arch/x86/mm/pat.c:527: warning: passing argument 2 of 'variable_test_bit' from incompatible pointer type
arch/x86/mm/pat.c:528: warning: passing argument 2 of 'constant_test_bit' from incompatible pointer type
arch/x86/mm/pat.c:528: warning: passing argument 2 of 'variable_test_bit' from incompatible pointer type
arch/x86/mm/pat.c:529: warning: passing argument 2 of 'constant_test_bit' from incompatible pointer type
arch/x86/mm/pat.c:529: warning: passing argument 2 of 'variable_test_bit' from incompatible pointer type
Don't open-code test_bit() on a __u32
Cc: Andrea Arcangeli <andrea@qumranet.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
mmiotrace_ioremap() expects to receive the original unaligned map phys address
and size. Also fix {un,}register_kmmio_probe() to deal properly with
unaligned size.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
From 8979ee55cb6a429c4edd72ebec2244b849f6a79a Mon Sep 17 00:00:00 2001
From: Pekka Paalanen <pq@iki.fi>
Date: Sat, 12 Apr 2008 00:18:57 +0300
Mmiotrace is not reliable with multiple CPUs and may
miss events. Drop to single CPU when mmiotrace is activated.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix gcc printk format warnings:
next-20080415/arch/x86/mm/mmio-mod.c: In function 'print_pte':
next-20080415/arch/x86/mm/mmio-mod.c:154: warning: format '%lx' expects type 'long unsigned int', but argument 3 has type 'pteval_t'
next-20080415/arch/x86/mm/mmio-mod.c:154: warning: format '%lx' expects type 'long unsigned int', but argument 4 has type 'pteval_t'
next-20080415/arch/x86/mm/mmio-mod.c: At top level:
next-20080415/arch/x86/mm/mmio-mod.c:403: warning: 'downed_cpus' defined but not used
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This had become a no-op.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Kconfig.debug, Makefile and testmmiotrace.c style fixes.
Use real mutex instead of mutex.
Fix failure path in register probe func.
kmmio: RCU read-locked over single stepping.
Generate mapping id's.
Make mmio-mod.c built-in and rewrite its locking.
Add debugfs file to enable/disable mmiotracing.
kmmio: use irqsave spinlocks.
Lots of cleanups in mmio-mod.c
Marker file moved from /proc into debugfs.
Call mmiotrace entrypoints directly from ioremap.c.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
kmmio.c handles the list of mmio probes with callbacks, list of traced
pages, and attaching into the page fault handler and die notifier. It
arms, traps and disarms the given pages, this is the core of mmiotrace.
mmio-mod.c is a user interface, hooking into ioremap functions and
registering the mmio probes. It also decodes the required information
from trapped mmio accesses via the pre and post callbacks in each probe.
Currently, hooking into ioremap functions works by redefining the symbols
of the target (binary) kernel module, so that it calls the traced
versions of the functions.
The most notable changes done since the last discussion are:
- kmmio.c is a built-in, not part of the module
- direct call from fault.c to kmmio.c, removing all dynamic hooks
- prepare for unregistering probes at any time
- make kmmio re-initializable and accessible to more than one user
- rewrite kmmio locking to remove all spinlocks from page fault path
Can I abuse call_rcu() like I do in kmmio.c:unregister_kmmio_probe()
or is there a better way?
The function called via call_rcu() itself calls call_rcu() again,
will this work or break? There I need a second grace period for RCU
after the first grace period for page faults.
Mmiotrace itself (mmio-mod.c) is still a module, I am going to attack
that next. At some point I will start looking into how to make mmiotrace
a tracer component of ftrace (thanks for the hint, Ingo). Ftrace should
make the user space part of mmiotracing as simple as
'cat /debug/trace/mmio > dump.txt'.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The custom page fault handler list is replaced with a single function
pointer. All related functions and variables are renamed for
mmiotrace.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: pq@iki.fi
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Use lookup_address() from pageattr.c instead of doing the same
manually. Also had to EXPORT_SYMBOL_GPL(lookup_address) to make this
work for modules. This also fixes "undefined symbol 'init_mm'"
compile error for x86_32.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Without CONFIG_DYNAMIC_FTRACE, mark_rodata_ro() would mark a wrong
number of pages as no-execute. The bug was introduced in the patch
"ftrace: dont write protect kernel text". The symptom was machine reboot
after a CPU hotplug.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Acked-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Provides kernel modules a way to register custom page fault handlers.
On every page fault this will call a list of registered functions. The
functions may handle the fault and force do_page_fault() to return
immediately.
This functionality is similar to the now removed page fault notifiers.
Custom page fault handlers are used by debugging and reverse engineering
tools. Mmiotrace is one such tool and a patch to add it into the tree
will follow.
The custom page fault handlers are called earlier in do_page_fault()
than the page fault notifiers were.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Dynamic ftrace cant work when the kernel has its text write protected.
This patch keeps the kernel from being write protected when
dynamic ftrace is in place.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
When allocating the pgdat's for numa nodes on x86_32 we attempt to place
them in the numa remap space for that node. However should the node not
have any remap space allocated (such as due to having non-ram pages in
the remap location in the node) then we will incorrectly place the pgdat
at zero. Check we have remap available, falling back to node 0 memory
where we do not.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Recent kernels have been panic'ing trying to allocate memory early in boot,
in __alloc_pages:
BUG: unable to handle kernel paging request at 00001568
IP: [<c10407b6>] __alloc_pages+0x33/0x2cc
*pdpt = 00000000013a5001 *pde = 0000000000000000
Oops: 0000 [#1] SMP
Modules linked in:
Pid: 1, comm: swapper Not tainted (2.6.25 #78)
EIP: 0060:[<c10407b6>] EFLAGS: 00010246 CPU: 0
EIP is at __alloc_pages+0x33/0x2cc
EAX: 00001564 EBX: 000412d0 ECX: 00001564 EDX: 000005c3
ESI: f78012a0 EDI: 00000001 EBP: 00001564 ESP: f7871e50
DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
Process swapper (pid: 1, ti=f7870000 task=f786f670 task.ti=f7870000)
Stack: 00000000 f786f670 00000010 00000000 0000b700 000412d0 f78012a0 00000001
00000000 c105b64d 00000000 000412d0 f78012a0 f7803120 00000000 c105c1c5
00000010 f7803144 000412d0 00000001 f7803130 f7803120 f78012a0 00000001
Call Trace:
[<c105b64d>] kmem_getpages+0x94/0x129
[<c105c1c5>] cache_grow+0x8f/0x123
[<c105c689>] ____cache_alloc_node+0xb9/0xe4
[<c105c999>] kmem_cache_alloc_node+0x92/0xd2
[<c1018929>] build_sched_domains+0x536/0x70d
[<c100b63c>] do_flush_tlb_all+0x0/0x3f
[<c100b63c>] do_flush_tlb_all+0x0/0x3f
[<c10572d6>] interleave_nodes+0x23/0x5a
[<c105c44f>] alternate_node_alloc+0x43/0x5b
[<c1018b47>] arch_init_sched_domains+0x46/0x51
[<c136e85e>] kernel_init+0x0/0x82
[<c137ac19>] sched_init_smp+0x10/0xbb
[<c136e8a1>] kernel_init+0x43/0x82
[<c10035cf>] kernel_thread_helper+0x7/0x10
Debugging this showed that the NODE_DATA() for nodes other than node 0
were all NULL. Tracing this back showed that the NODE_DATA() pointers
were being initialised to each nodes remap space. However under
SPARSEMEM remap is disabled which leads to the pgdat's being placed
incorrectly at kernel virtual address 0. Leading to the panic when
attempting to allocate memory from these nodes.
Numa remap was disabled in the commit below. This occured while fixing
problems triggered when attempting to boot x86_32 NUMA SPARSEMEM kernels
on non-numa hardware.
x86: make NUMA work on 32-bit
commit 1b000a5dbe
The real problem is believed to be related to other alignment issues in
the regions blocked out from the bootmem allocator for small memory
systems, and has been fixed separately. Therefore re-enable remap for
SPARSMEM, which fixes pgdat allocation issues. Testing confirms that
SPARSMEM NUMA kernels will boot correctly with this part of the change
reverted.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
pat_disable() is __init, which means it goes away after booting is complete.
Unfortunately it is used by the hotplug code if the machine is not
pat-capable, causing a crash.
Fix by marking pat_disable() as __cpuinit.
Signed-off-by: Avi Kivity <avi@qumranet.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
