Commit graph

932,869 commits

Author SHA1 Message Date
Mauro Carvalho Chehab
0f7ff39532 V4L/DVB (13825): ir-core: Don't OOPS if IR device props is not defined
As currently most drivers don't define ir_dev->props, we shouldn't assume
that this field is defined.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:24 -03:00
Igor M. Liplianin
cd79d33e16 V4L/DVB (13818): Add Prof 7500 DVB-S2 USB card
The card based on stv0903 demod, stb6100 tuner.

Signed-off-by: Igor M. Liplianin <liplianin@me.by>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:24 -03:00
Jean-Francois Moine
d41592a2a2 V4L/DVB (13815): gspca - sunplus: Add webcam 052b:1507.
Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:24 -03:00
Mauro Carvalho Chehab
971e8298de V4L/DVB (13680): ir: use unsigned long instead of enum
When preparing the linux-next patches, I got those errors:

include/media/ir-core.h:29: warning: left shift count >= width of type
In file included from include/media/ir-common.h:29,
                 from drivers/media/video/ir-kbd-i2c.c:50:
drivers/media/video/ir-kbd-i2c.c: In function ‘ir_probe’:
drivers/media/video/ir-kbd-i2c.c:324: warning: left shift count >= width of type

Unfortunately, enum is 32 bits on i386. As we define IR_TYPE_OTHER as 1<<63,
it won't work on non 64 bits arch.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:24 -03:00
Mauro Carvalho Chehab
3f831107ed V4L/DVB (13641): Properly update the driver representation for the protocol
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:24 -03:00
Mauro Carvalho Chehab
eecee32ac2 V4L/DVB (13639): ir-sysfs: Properly protect rc_tab changes with a lock
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:24 -03:00
Mauro Carvalho Chehab
d4b778d368 V4L/DVB (13638): ir-core: documment missed functions
While here, change ir_core_dev_number to be static

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:24 -03:00
Mauro Carvalho Chehab
950b0f5a0b V4L/DVB (13637): em28xx: allow changing keycode table protocol
Experimental patch to allow changing the IR protocol. Currently, it support
changing between RC-5 and NEC protocols.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:23 -03:00
Mauro Carvalho Chehab
09b01b90eb V4L/DVB (13636): ir-core: add method to change IR protocol
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:23 -03:00
Mauro Carvalho Chehab
53f870228d V4L/DVB (13635): ir-core: Implement protocol table type reading
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:23 -03:00
Mauro Carvalho Chehab
e93854da88 V4L/DVB (13634): ir-core: allow passing IR device parameters to ir-core
Adds an structure to ir_input_register to contain IR device characteristics,
like supported protocols and a callback to handle protocol event changes.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:23 -03:00
Mauro Carvalho Chehab
4714eda877 V4L/DVB (13633): ir-core: create a new class for remote controllers
Add sysfs skeleton to export remote controller information via
/sys/class/irrcv.

For now, the code doesn't do much. It just exports an attribute that
is meant to  report and control the IR protocol used by the keytable.
However, the callbacks for this new attribute weren't set yet.

Also, it lacks symlinks to the used event interface.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2010-02-26 15:10:23 -03:00
Maxim Levitsky
9fc51a37a8 mtd: common module for smartmedia/xD support
This small module implements few helpers that are usefull
for nand drivers for SmartMedia/xD card readers.

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 18:04:51 +00:00
Linus Torvalds
a4a47bc03f Lower USB storage settling delay to something more reasonable
The five-second delay can be rather annoying, and makes the system
appear much less responsive when you connect a USB drive.

It's also not entirely clear that it is needed - the settling delay has
at least historically been an issue on some Apple iPods, for example,
and some devices have been reported to need even more than the old 5s
delay.

But before we penalize them all, let's see how bad it really is.  Some
of the reasons for long delays seem to be actual historical kernel bugs
that should probably never have been papered over with a delay in the
first place (there's a Ubuntu bug report for 2.6.20 about a NULL pointer
dereference unless 'delay_use' is 8 or more, for example).

It also looks like some distros have already shipped with delay_use=0,
so the five second default may well be totally historical.

In other words: "Let's see if anybody screams".

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-02-26 10:03:22 -08:00
Maxim Levitsky
e0b58d0a70 mtd: nand: add ->badblockbits for minimum number of set bits in bad block byte
This can be used to protect against bitflips in that field, but now mostly
for smartmedia.

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 18:02:07 +00:00
Sage Weil
080af17e9c ceph: remove bogus mds forward warning
The must_resend flag is always true, not false.  In any case, we can
just ignore it anyway.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-02-26 10:02:03 -08:00
Maxim Levitsky
b64d39d8b0 mtd: nand: make reads using MTD_OOB_RAW affect only ECC validation
This changes the behavier of MTD_OOB_RAW. It used to read both OOB and 
data to the data buffer, however you would still need to specify the 
dummy oob buffer.

This is only used in one place, but makes it hard to read data+oob 
without ECC test, thus I removed that behavier, and fixed the user.

Now MTD_OOB_RAW behaves just like MTD_OOB_PLACE, but doesn't do ECC 
validation

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 18:00:08 +00:00
David Teigland
cf6620acc0 dlm: send reply before bast
When the lock master processes a successful operation (request,
convert, cancel, or unlock), it will process the effects of the
change before sending the reply for the operation.  The "effects"
of the operation are:

- blocking callbacks (basts) for any newly granted locks
- waiting or converting locks that can now be granted

The cast is queued on the local node when the reply from the lock
master is received.  This means that a lock holder can receive a
bast for a lock mode that is doesn't yet know has been granted.

Signed-off-by: David Teigland <teigland@redhat.com>
2010-02-26 11:57:37 -06:00
Maxim Levitsky
782ce79a45 mtd: nand: cleanup the nand_do_write_ops
nand_do_write_ops was broken in regard to writing several pages, each 
with its own oob.

Although nand_do_write_ops intends to allow such mode, it fails do do so
Probably this was never tested.

Also add missing checks for attempts to write at illegal offsets.

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 17:47:47 +00:00
Maxim Levitsky
9aca334e85 mtd: nand: make MTD_OOB_PLACE work correctly.
MTD_OOB_PLACE is supposed to read/write the raw oob data similiar to the 
MTD_OOB_RAW however due to a bug, currently it is not possible to read 
more data that is specified by the oob 'free' regions.

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 17:44:51 +00:00
Maxim Levitsky
49ef3c6ee1 mtd: nand: make suspend work if device is accessed by kernel threads.
Since all userspace threads are frozen at the time the nand_suspend is called,
they aren't inside any nand function.

We don't call try_to_freeze in nand ether. Thus the only user that can 
be inside the nand functions is an non freezeable kernel thread. Thus we 
can safely wait for it to finish.

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 17:43:21 +00:00
Maxim Levitsky
026ec57886 mtd: blktrans: allow FTL drivers to export sysfs attributes
This patch adds an ability to export sysfs attributes below
the block disk device.

This can be used to pass the udev an information about the FTL
and could include the vendor, serial, version, etc...

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 17:37:40 +00:00
Sage Weil
c99eb1c726 ceph: remove fragile __map_osds optimization
We used to try to avoid freeing and then reallocating the osd
struct.  This is a bit fragile due to potential interactions with
other references (beyond o_requests), and may be the cause of
this crash:

[120633.442358] BUG: unable to handle kernel NULL pointer dereference at (null)
[120633.443292] IP: [<ffffffff812549b6>] rb_erase+0x11d/0x277
[120633.443292] PGD f7ff3067 PUD f7f53067 PMD 0
[120633.443292] Oops: 0000 [#1] PREEMPT SMP
[120633.443292] last sysfs file: /sys/kernel/uevent_seqnum
[120633.443292] CPU 1
[120633.443292] Modules linked in: ceph fan ac battery psmouse ehci_hcd ide_pci_generic ohci_hcd thermal processor button
[120633.443292] Pid: 3023, comm: ceph-msgr/1 Not tainted 2.6.32-rc2 #12 H8SSL
[120633.443292] RIP: 0010:[<ffffffff812549b6>]  [<ffffffff812549b6>] rb_erase+0x11d/0x277
[120633.443292] RSP: 0018:ffff8800f7b13a50  EFLAGS: 00010246
[120633.443292] RAX: ffff880022907819 RBX: ffff880022907818 RCX: 0000000000000000
[120633.443292] RDX: ffff8800f7b13a80 RSI: ffff8800f587eb48 RDI: 0000000000000000
[120633.443292] RBP: ffff8800f7b13a60 R08: 0000000000000000 R09: 0000000000000004
[120633.443292] R10: 0000000000000000 R11: ffff8800c4441000 R12: ffff8800f587eb48
[120633.443292] R13: ffff8800f58eaa00 R14: ffff8800f413c000 R15: 0000000000000001
[120633.443292] FS:  00007fbef6e226e0(0000) GS:ffff880009200000(0000) knlGS:0000000000000000
[120633.443292] CS:  0010 DS: 0018 ES: 0018 CR0: 000000008005003b
[120633.443292] CR2: 0000000000000000 CR3: 00000000f7c53000 CR4: 00000000000006e0
[120633.443292] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[120633.443292] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[120633.443292] Process ceph-msgr/1 (pid: 3023, threadinfo ffff8800f7b12000, task ffff8800f5858b40)
[120633.443292] Stack:
[120633.443292]  ffff8800f413c000 ffff8800f587e9c0 ffff8800f7b13a80 ffffffffa0098a86
[120633.443292] <0> 00000000000006f1 0000000000000000 ffff8800f7b13af0 ffffffffa009959b
[120633.443292] <0> ffff8800f413c000 ffff880022a68400 ffff880022a68400 ffff8800f587e9c0
[120633.443292] Call Trace:
[120633.443292]  [<ffffffffa0098a86>] __remove_osd+0x4d/0xbc [ceph]
[120633.443292]  [<ffffffffa009959b>] __map_osds+0x199/0x4fa [ceph]
[120633.443292]  [<ffffffffa00999f4>] ? __send_request+0xf8/0x186 [ceph]
[120633.443292]  [<ffffffffa0099beb>] kick_requests+0x169/0x3cb [ceph]
[120633.443292]  [<ffffffffa009a8c1>] ceph_osdc_handle_map+0x370/0x522 [ceph]

Since we're probably screwed anyway if a small kmalloc is
failing, don't bother with trying to be clever here.

Signed-off-by: Sage Weil <sage@newdream.net>
2010-02-26 09:37:33 -08:00
Maxim Levitsky
75c0b84d41 mtd: call the remove notifiers before assuming it is in use
Now that mtd block common layer is prepared for proper hotplug support,
enable it here

Now all users of the mtd device have a chance to put the mtd device
when they are notified to do so, and they have to do so to make hotplug work.

[dwmw2: There's more work to be done to fix hotplug in the general case, but
        this is a reasonable start]

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 17:37:10 +00:00
David S. Miller
38bdbd8efc Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/kaber/nf-next-2.6 2010-02-26 09:31:09 -08:00
Pekka Enberg
6adad2d543 Merge branch 'kmemcheck/fixes' into kmemcheck-for-linus 2010-02-26 19:25:30 +02:00
Dmitry Monakhov
4c13dd3b48 failslab: add ability to filter slab caches
This patch allow to inject faults only for specific slabs.
In order to preserve default behavior cache filter is off by
default (all caches are faulty).

One may define specific set of slabs like this:
# mark skbuff_head_cache as faulty
echo 1 > /sys/kernel/slab/skbuff_head_cache/failslab
# Turn on cache filter (off by default)
echo 1 > /sys/kernel/debug/failslab/cache-filter
# Turn on fault injection
echo 1 > /sys/kernel/debug/failslab/times
echo 1 > /sys/kernel/debug/failslab/probability

Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Akinobu Mita <akinobu.mita@gmail.com>
Acked-by: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
2010-02-26 19:19:39 +02:00
Maxim Levitsky
298304f1a5 mtd: mtdblock: test return value of add_mtd_blktrans_dev, because if can fail
This prevents a memory leak

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 17:05:46 +00:00
Maxim Levitsky
048d871995 mtd: blktrans: Hotplug fixes
* Add locking where it was missing.

* Don't do a get_mtd_device in blktrans_open because it would lead to a
  deadlock; instead do that in add_mtd_blktrans_dev.

* Only free the mtd_blktrans_dev structure when the last user exits.

* Flush request queue on device removal.

* Track users, and call tr->release in del_mtd_blktrans_dev
  Due to that ->open and release aren't called more that once.

Now it is safe to call del_mtd_blktrans_dev while the device is still in use.

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 17:05:26 +00:00
Maxim Levitsky
a863862257 mtd: blktrans: remove mtd_blkcore_priv, switch to per device queue and thread
This is the biggest change. To make hotplug possible, and this layer 
clean, the mtd_blktrans_dev now contains everything for a single mtd 
block translation device. Also removed some very old leftovers.

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 17:04:03 +00:00
Maxim Levitsky
3bd456576f mtd: create unlocked versions of {get,put}_mtd_device
Use these only if you know that you already hold mtd_table_mutex

Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 17:03:15 +00:00
Anatolij Gustschin
bb315f749f mtd: nand: Add MPC5121 NAND Flash Controller driver
Adds NAND Flash Controller driver for MPC5121 Revision 2.
All device features, except hardware ECC and power management,
are supported.

Signed-off-by: Piotr Ziecik <kosmo@semihalf.com>
Signed-off-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 16:56:43 +00:00
Roel Kluin
1385858ee0 mtd: nand_bcm: fix hot spin and code duplication
In the branch where pagesize equalled NAND_DATA_ACCESS_SIZE, NumToRead
wasn't decremented in the `while (numToRead > 11)' loop.
Also the first and last while loops were duplicated in both branches.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Acked-by: Leo Chen <leochen@broadcom.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 16:56:35 +00:00
Jan Engelhardt
6b4ff2d767 netfilter: xtables: restore indentation
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
2010-02-26 17:53:31 +01:00
Vimal Singh
c3341d0ceb mtd: omap2 fix prefetch mode read issue
There is a bug in nand prefetch read routine, which comes into effect
only if nand device is a 16-bit device (as we have in zoom boards).
This bug is effective only with below combination of conditions:
1. nand deivce, in use, is a 16 bit device
2. nand driver supports 'subpage' read
3. SW ECC is in use

This was not seen old  kernel (ex: .23), because when, in early days,
we tested this (nand prefetch read in LDP boards) there was no
'subpage read' support.
Later when we had subpage read in (.27) kernel, we had hw ecc enabled
always in our internal tree. So, we missed this bug.

This patch fixes the issue.

Signed-off-by: Vimal Singh <vimalsingh@ti.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 16:51:23 +00:00
Alexey Dobriyan
a49c650371 netfilter: nfnetlink_log: fix silly refcount leak
Quick fix for memory/module refcount leak.
Reference count of listener instance never reaches 0.

Start/stop of ulogd2 is enough to trigger this bug!

Now, refcounting there looks very fishy in particular this code:

 	if (!try_module_get(THIS_MODULE)) {
		...

and creation of listener instance with refcount 2,
so it may very well be ripped and redone.  :-)

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
2010-02-26 17:48:40 +01:00
Vimal Singh
f35b6eda51 mtd: omap2: correct 'info' pointer in 'omap_nand_remove'
Removing OMAP NAND driver, when loaded as a module, gives error and
does not get success. This fixes this and makes driver loadable and
removable run time.

Signed-off-by: Vimal Singh <vimalsingh@ti.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 16:47:39 +00:00
Vimal Singh
bdaefc4162 mtd: omap2: fixing compilation warning
Fixing below warning in compilation:
drivers/mtd/nand/omap2.c: In function 'omap_write_buf_dma_pref':
drivers/mtd/nand/omap2.c:508: warning: passing argument 2 of
'omap_nand_dma_transfer' discards qualifiers from pointer target type

Signed-off-by: Vimal Singh <vimalsingh@ti.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 16:46:13 +00:00
Guillaume LECERF
f3e69c6584 mtd: move more manufacturers to the common cfi.h header file
Move MANUFACTURER_MACRONIX and MANUFACTURER_SST definitions to the
include/linux/mtd/cfi.h header file and rename them to CFI_MFR_MACRONIX and
CFI_MFR_SST.

All references in drivers/mtd/chips/cfi_cmdset_0002.c are updated to reflect
this.

Signed-off-by: Guillaume LECERF <glecerf@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 16:45:52 +00:00
Simon Horman
51f0bc7868 IPVS: ip_vs_lblcr: use list headA
Use list_head rather than a custom list implementation.

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Patrick McHardy <kaber@trash.net>
2010-02-26 17:45:14 +01:00
Peter Zijlstra
1dd2980d99 perf_event, amd: Fix spinlock initialization
Avoid kernels from exploding on AMD machines when they have any
lock debugging bits enabled.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-26 17:25:19 +01:00
Peter Zijlstra
24691ea964 perf_event: Fix preempt warning in perf_clock()
A recent commit introduced a preemption warning for
perf_clock(), use raw_smp_processor_id() to avoid this, it
really doesn't matter which cpu we use here.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1267198583.22519.684.camel@laptop>
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-26 17:25:00 +01:00
Tomi Valkeinen
5d68e0326b OMAP: DSS2: DSI: add error prints
Add error printing for dsi_vc_dcs_write() and dsi_vc_dcs_read().

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
2010-02-26 17:44:10 +02:00
David S. Miller
4385d580f2 perf tools: Flush maps on COMM events
Even though we don't register the counters until the child is right about
to exec(), we're still going to get at least a few events while the
fork()'d child is still executing 'perf' and in particular we're going to
get the MMAP events.

We can't distinguish the ones in the newly executed process because the
PID will be the same.

One way to solve this would be to have a PERF_RECORD_EXEC event, and when
this is seen 'perf' can flush it's map cache.  We can't use
PERF_RECORD_COMM since that's generated by other things, not just exec().

Actually, thinking about it some more, using PERF_RECORD_COMM might be a
good enough approximation.

Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1267196914-16238-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-26 16:28:45 +01:00
Eric Benard
66803762c1 mtd: mxc_nand: add RESET command support
mxc_nand driver must support the RESET Command in order to support
Micron NAND which need a reset before any other command.

Signed-off-by: Eric Benard <eric@eukrea.com>
Acked-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
2010-02-26 15:18:12 +00:00
Suresh Siddha
dd5feea14a sched: Fix SCHED_MC regression caused by change in sched cpu_power
On platforms like dual socket quad-core platform, the scheduler load
balancer is not detecting the load imbalances in certain scenarios. This
is leading to scenarios like where one socket is completely busy (with
all the 4 cores running with 4 tasks) and leaving another socket
completely idle. This causes performance issues as those 4 tasks share
the memory controller, last-level cache bandwidth etc. Also we won't be
taking advantage of turbo-mode as much as we would like, etc.

Some of the comparisons in the scheduler load balancing code are
comparing the "weighted cpu load that is scaled wrt sched_group's
cpu_power" with the "weighted average load per task that is not scaled
wrt sched_group's cpu_power". While this has probably been broken for a
longer time (for multi socket numa nodes etc), the problem got aggrevated
via this recent change:

 |
 |  commit f93e65c186
 |  Author: Peter Zijlstra <a.p.zijlstra@chello.nl>
 |  Date:   Tue Sep 1 10:34:32 2009 +0200
 |
 |	sched: Restore __cpu_power to a straight sum of power
 |

Also with this change, the sched group cpu power alone no longer reflects
the group capacity that is needed to implement MC, MT performance
(default) and power-savings (user-selectable) policies.

We need to use the computed group capacity (sgs.group_capacity, that is
computed using the SD_PREFER_SIBLING logic in update_sd_lb_stats()) to
find out if the group with the max load is above its capacity and how
much load to move etc.

Reported-by: Ma Ling <ling.ma@intel.com>
Initial-Analysis-by: Zhang, Yanmin <yanmin_zhang@linux.intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
[ -v2: build fix ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: <stable@kernel.org> # [2.6.32.x, 2.6.33.x]
LKML-Reference: <1266970432.11588.22.camel@sbs-t61.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-26 15:45:13 +01:00
Peter Zijlstra
f22f54f449 perf_events, x86: Split PMU definitions into separate files
Split amd,p6,intel into separate files so that we can easily deal with
CONFIG_CPU_SUP_* things, needed to make things build now that perf_event.c
relies on symbols from amd.c

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-26 15:44:04 +01:00
Arnaldo Carvalho de Melo
48fb4fdd6b perf annotate: Handle samples not at objdump output addr boundaries
Without this patch we get this for need_resched:

[root@mica ~]# perf annotate need_resched

------------------------------------------------
 Percent |      Source code & Disassembly of vmlinux
------------------------------------------------
         :
         :
         :      Disassembly of section .text:
         :
         :      ffffffff810095ed <need_resched>:
         :              return (state & TASK_INTERRUPTIBLE) || __fatal_signal_pending(p);
         :      }
         :
         :      static inline int need_resched(void)
         :      {
    0.00 :      ffffffff810095ed:       55                      push   %rbp
         :              return unlikely(test_thread_flag(TIF_NEED_RESCHED));
    0.00 :      ffffffff810095ee:       be 03 00 00 00          mov    $0x3,%esi
         :
         :      static inline struct thread_info *current_thread_info(void)
         :      {
         :              struct thread_info *ti;
         :              ti = (void *)(percpu_read_stable(kernel_stack) +
    0.00 :      ffffffff810095f3:       65 48 8b 3c 25 48 b5    mov    %gs:0xb548,%rdi
    0.00 :      ffffffff810095fa:       00 00
         :              return (state & TASK_INTERRUPTIBLE) || __fatal_signal_pending(p);
         :      }
         :
         :      static inline int need_resched(void)
         :      {
    0.00 :      ffffffff810095fc:       48 89 e5                mov    %rsp,%rbp
         :              return unlikely(test_thread_flag(TIF_NEED_RESCHED));
    0.00 :      ffffffff810095ff:       48 81 ef d8 1f 00 00    sub    $0x1fd8,%rdi
    0.00 :      ffffffff81009606:       e8 9d ff ff ff          callq  ffffffff810095a8 <test_ti_thread_flag>
         :      }
    0.00 :      ffffffff8100960b:       c9                      leaveq
    0.00 :      ffffffff8100960c:       85 c0                   test   %eax,%eax
    0.00 :      ffffffff8100960e:       0f 95 c0                setne  %al
    0.00 :      ffffffff81009611:       0f b6 c0                movzbl %al,%eax
         :      Disassembly of section .vsyscall_0:
         :      Disassembly of section .vsyscall_fn:
         :      Disassembly of section .vsyscall_1:
         :      Disassembly of section .vsyscall_2:
         :      Disassembly of section .init.text:
         :      Disassembly of section .altinstr_replacement:
         :      Disassembly of section .exit.text:
[root@mica ~]#

But from the 'perf report' result we know that there are hits
for need_resched on a 4 way machine mostly doing nothing, so
after adding code to show what is in each hist offset and
collapsing IP hits for what happens between objdump lines we
get, for the same perf.data file:

[root@mica ~]# perf annotate -v need_resched

------------------------------------------------
 Percent |      Source code & Disassembly of vmlinux
------------------------------------------------
         :
         :
         :      Disassembly of section .text:
         :
         :      ffffffff810095ed <need_resched>:
         :              return (state & TASK_INTERRUPTIBLE) || __fatal_signal_pending(p);
         :      }
         :
         :      static inline int need_resched(void)
         :      {
    0.00 :      ffffffff810095ed:       55                      push   %rbp
         :              return unlikely(test_thread_flag(TIF_NEED_RESCHED));
   52.78 :      ffffffff810095ee:       be 03 00 00 00          mov    $0x3,%esi
         :
         :      static inline struct thread_info *current_thread_info(void)
         :      {
         :              struct thread_info *ti;
         :              ti = (void *)(percpu_read_stable(kernel_stack) +
    0.00 :      ffffffff810095f3:       65 48 8b 3c 25 48 b5    mov    %gs:0xb548,%rdi
    0.00 :      ffffffff810095fa:       00 00
         :              return (state & TASK_INTERRUPTIBLE) || __fatal_signal_pending(p);
         :      }
         :
         :      static inline int need_resched(void)
         :      {
    0.00 :      ffffffff810095fc:       48 89 e5                mov    %rsp,%rbp
         :              return unlikely(test_thread_flag(TIF_NEED_RESCHED));
    9.72 :      ffffffff810095ff:       48 81 ef d8 1f 00 00    sub    $0x1fd8,%rdi
    0.00 :      ffffffff81009606:       e8 9d ff ff ff          callq  ffffffff810095a8 <test_ti_thread_flag>
         :      }
    0.00 :      ffffffff8100960b:       c9                      leaveq
    0.00 :      ffffffff8100960c:       85 c0                   test   %eax,%eax
   37.50 :      ffffffff8100960e:       0f 95 c0                setne  %al
    0.00 :      ffffffff81009611:       0f b6 c0                movzbl %al,%eax
         :      Disassembly of section .vsyscall_0:
         :      Disassembly of section .vsyscall_fn:
         :      Disassembly of section .vsyscall_1:
         :      Disassembly of section .vsyscall_2:
         :      Disassembly of section .init.text:
         :      Disassembly of section .altinstr_replacement:
         :      Disassembly of section .exit.text:
[root@mica ~]#

And now 'perf annotate -v', verbose mode, will show the hits per
precise IP, so that one can make sense of the attribution to
each objdumop line:

[root@mica ~]# perf annotate -v need_resched
Looking at the vmlinux_path (5 entries long)
Using /lib/modules/2.6.33-rc8-tip-00784-g3471df5-dirty/build/vmlinux
for symbols annotate_sym: filename=/lib/modules/2.6.33-rc8-tip-00784-g3471df5-dirty/build/vmlinux, sym=need_resched, start=0xffffffff810095ed, end=0xffffffff81009614

------------------------------------------------
 Percent |      Source code & Disassembly of vmlinux
------------------------------------------------
                ffffffff810095f1: 152
                ffffffff81009603: 28
                ffffffff8100960f: 55
                ffffffff81009610: 53
                          h->sum: 288
<SNIP same annotation>

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Miller <davem@davemloft.net>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1267194194-15670-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-26 15:42:49 +01:00
Robert Richter
cfc9c0b450 oprofile/x86: fix msr access to reserved counters
During switching virtual counters there is access to perfctr msrs. If
the counter is not available this fails due to an invalid
address. This patch fixes this.

Cc: stable@kernel.org
Signed-off-by: Robert Richter <robert.richter@amd.com>
2010-02-26 15:28:16 +01:00
Robert Richter
c17c8fbf34 oprofile/x86: use kzalloc() instead of kmalloc()
Cc: stable@kernel.org
Signed-off-by: Robert Richter <robert.richter@amd.com>
2010-02-26 15:20:03 +01:00