07 Oct, 2010
1 commit
-
The page-types utility still uses an out of date name for the
unpoison interface: debugfs:hwpoison/renew-pfn
This patch renames and fixes it.Signed-off-by: Naoya Horiguchi
Acked-by: Wu Fengguang
Signed-off-by: Andi Kleen
06 Aug, 2010
1 commit
-
Trivial typo fixes.
Signed-off-by: Tommi Rantala
Signed-off-by: Randy Dunlap
Signed-off-by: Linus Torvalds
16 Dec, 2009
6 commits
-
Signed-off-by: Wu Fengguang
Signed-off-by: Andi Kleen -
On a system with large amount of memory (256GB), invoking page-types can
take quite a long time, which is unreasonable considering the user only
wants a description of the flags:# time ./page-types -d 0x10
0x0000000000000010 ____D_____________________________ dirtyreal 0m34.285s
user 0m1.966s
sys 0m32.313sThis is because we still walk the entire address range.
Exiting early seems like a reasonble solution:
# time ./page-types -d 0x10
0x0000000000000010 ____D_____________________________ dirtyreal 0m0.007s
user 0m0.001s
sys 0m0.005sSigned-off-by: Alex Chiang
Cc: Andi Kleen
Cc: Haicheng Li
Acked-by: Wu Fengguang
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Align the output when page-type -h is invoked.
Signed-off-by: Alex Chiang
Acked-by: Wu Fengguang
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Teach page-types to describe page flags directly from the command line.
Why is this useful? For instance, if you're using memory hotplug and see
this in /var/log/messages:kernel: removing from LRU failed 3836dd0/1/1e00000000000010
It would be nice to decode those page flags without staring at the source.
Example usage and output:
# Documentation/vm/page-types -d 0x10
0x0000000000000010 ____D_____________________________ dirty# Documentation/vm/page-types -d anon
0x0000000000001000 ____________a_____________________ anonymous# Documentation/vm/page-types -d anon,0x10
0x0000000000001010 ____D_______a_____________________ dirty,anonymous[achiang@hp.com: documentation]
Signed-off-by: Alex Chiang
Signed-off-by: Wu Fengguang
Cc: Andi Kleen
Cc: Haicheng Li
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
If not signed, testing of the read() return value in this function
will not work.Signed-off-by: Roel Kluin
Cc: Wu Fengguang
Cc: Randy Dunlap
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Signed-off-by: Tommi Rantala
Cc: Randy Dunlap
Cc: Wu Fengguang
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
08 Dec, 2009
1 commit
-
Conflicts:
kernel/irq/chip.c
04 Dec, 2009
1 commit
-
That is "success", "unknown", "through", "performance", "[re|un]mapping"
, "access", "default", "reasonable", "[con]currently", "temperature"
, "channel", "[un]used", "application", "example","hierarchy", "therefore"
, "[over|under]flow", "contiguous", "threshold", "enough" and others.Signed-off-by: André Goddard Rosa
Signed-off-by: Jiri Kosina
16 Nov, 2009
1 commit
-
Nothing outside of Documentation/vm/page-types.c references
checked_open.Signed-off-by: Josh Triplett
08 Oct, 2009
8 commits
-
For hwpoison stress testing. The debugfs mount point is assumed to be
/debug/.Signed-off-by: Wu Fengguang
Cc: Andi Kleen
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Signed-off-by: Wu Fengguang
Cc: Andi Kleen
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Signed-off-by: Wu Fengguang
Cc: Andi Kleen
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Refactor the code to be more modular and easier to reuse.
Signed-off-by: Wu Fengguang
Cc: Andi Kleen
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This helps merge duplicate code (now and future) and outstand the main
logic.Signed-off-by: Wu Fengguang
Cc: Andi Kleen
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Signed-off-by: Wu Fengguang
Cc: Andi Kleen
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
It indicates to the system admin that processes mapping such pages may be
eating less physical memory than the reported numbers by legacy tools.Signed-off-by: Wu Fengguang
Cc: Hugh Dickins
Cc: Izik Eidus
Acked-by: Chris Wright
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This flag indicates a hardware detected memory corruption on the page.
Any future access of the page data may bring down the machine.Signed-off-by: Wu Fengguang
Cc: Andi Kleen
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
24 Sep, 2009
1 commit
-
Introduce "-p|--pid " for walking the process address space. The
default action is to walk raw memory PFNs.Both the virtual address and physical address of each present pages will
be listed:# ./tools/vm/page-types -lp $$ | head -3
voffset offset len flags
400 11bebe 1 __RU_lA____M______________________
402 11bebc 1 __RU_lA____M______________________Note that voffset/offset/len are now showed as hex numbers.
[akpm@linux-foundation.org: coding-style fixes]
Cc: Andi Kleen
Signed-off-by: Wu Fengguang
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
23 Sep, 2009
1 commit
-
Fix up -Wmissing-prototypes in compileable userspace code, mainly under
Documentation/.Signed-off-by: Ladinu Chandrasinghe
Signed-off-by: Trevor Keith
Cc: Sam Ravnborg
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
17 Jun, 2009
1 commit
-
Add page-types, a handy tool for querying page flags.
It will expand some of the overloaded flags:
PG_slob_free = PG_private
PG_slub_frozen = PG_active
PG_slub_debug = PG_error
PG_readahead = PG_reclaimand mask out obscure flags except in -raw mode:
PG_reserved
PG_mlocked
PG_mappedtodisk
PG_private
PG_private_2
PG_owner_priv_1
PG_arch_1
PG_uncached
PG_compound* for non hugeTLB pages[akpm@linux-foundation.org: fix warning]
Signed-off-by: Wu Fengguang
Cc: KOSAKI Motohiro
Cc: Andi Kleen
Cc: Matt Mackall
Cc: Alexey Dobriyan
Cc: Ingo Molnar
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds