Several functions can return negative value in case of error,
so their return type should be fixed as well as type of variables
to which this value is assigned.
The problem has been detected using proposed semantic patch
scripts/coccinelle/tests/assign_signed_to_unsigned.cocci [1].
[1]: http://permalink.gmane.org/gmane.linux.kernel/2046107
Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds hardware ECC support using the BCH encoder in the NFC IP.
The ECC encoder supports up to 32-bit correction by using 60 error
correction bytes. There is no sub-page ECC step, ECC is calculated
always across the whole page (up to 2k pages).
Limitations:
- HW ECC: Only 2K page with 64+ OOB.
- HW ECC: Only 24 and 32-bit error correction implemented.
Raw writes have been tested using the generic nand_write_page_raw
implementation. However, raw reads are currently not possible
because the controller need to know whether we are going to use
the ECC mode already at NAND_CMD_READ0 command time. At this point
we do not have the information whether it is a raw read or a
regular read at driver level...
Signed-off-by: Bill Pringlemeir <bpringlemeir@nbsps.com>
Signed-off-by: Stefan Agner <stefan@agner.ch>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
This driver supports Freescale NFC (NAND flash controller) found on
Vybrid (VF610), MPC5125, MCF54418 and Kinetis K70. The driver has
been tested using 8-bit and 16-bit NAND interface on the ARM based
Vybrid SoC VF500 and VF610 platform.
parameter page reading.
Limitations:
- Untested on MPC5125 and M54418.
- DMA and pipelining not used.
- 2K pages or less.
- No chip select, one NAND chip per controller.
- No hardware ECC.
Some paths have been hand-optimized and evaluated by measurements
made using mtd_speedtest.ko on a 100MB MTD partition.
Colibri VF50
eb write % eb read % page write % page read %
rel/opt 5175 11537 4560 11039
opt 5164 -0.21 11420 -1.01 4737 +3.88 10918 -1.10
none 5113 -1.20 11352 -1.60 4490 -1.54 10865 -1.58
Colibri VF61
eb write % eb read % page write % page read %
rel/opt 5766 13096 5459 12846
opt 5883 +2.03 13064 -0.24 5561 +1.87 12802 -0.34
none 5701 -1.13 12980 -0.89 5488 +0.53 12735 -0.86
rel = using readl_relaxed/writel_relaxed in optimized paths
opt = hand-optimized by combining multiple accesses into one read/write
The measurements have not been statistically verfied, hence use them
with care. The author came to the conclusion that using the relaxed
variants of readl/writel are not worth the additional code.
Signed-off-by: Bill Pringlemeir <bpringlemeir@nbsps.com>
Tested-by: Albert ARIBAUD <albert.aribaud@3adev.fr>
Signed-off-by: Stefan Agner <stefan@agner.ch>
Reviewed-by: Alexey Klimov <klimov.linux@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Aaron Conole says:
====================
af_unix: return data from multiple SKBs on recv() with MSG_PEEK flag
This patch set implements a bugfix for kernel.org bugzilla #12323, allowing
MSG_PEEK to return all queued data on the unix domain socket, not just the
data contained in a single SKB.
This is the v3 version of this patch, which includes a suggested modification
by Eric Dumazet to convert the unix_sk() conversion macro to a static inline
function. These patches are independent and can be applied separately.
This set was tested over a 24-hour period, utilizing a loop continually
executing the bugzilla issue attached python code. It was instrumented with
a pr_err_once() ([ 13.798683] unix: went there at least one time).
v2->v3:
- Added Eric Dumazet's suggestion for #define to static inline
- Fixed an issue calling unix_state_lock() with an invalid argument
v3->v4:
- Eliminated an XXX comment
- Changed from goto unlock to explicit unix_state_unlock() and break
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
AF_UNIX sockets now return multiple skbs from recv() when MSG_PEEK flag
is set.
This is referenced in kernel bugzilla #12323 @
https://bugzilla.kernel.org/show_bug.cgi?id=12323
As described both in the BZ and lkml thread @
http://lkml.org/lkml/2008/1/8/444 calling recv() with MSG_PEEK on an
AF_UNIX socket only reads a single skb, where the desired effect is
to return as much skb data has been queued, until hitting the recv
buffer size (whichever comes first).
The modified MSG_PEEK path will now move to the next skb in the tree
and jump to the again: label, rather than following the natural loop
structure. This requires duplicating some of the loop head actions.
This was tested using the python socketpair python code attached to
the bugzilla issue.
Signed-off-by: Aaron Conole <aconole@bytheb.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
As suggested by Eric Dumazet this change replaces the
#define with a static inline function to enjoy
complaints by the compiler when misusing the API.
Signed-off-by: Aaron Conole <aconole@bytheb.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
After a bit of poking around wondering why my 32-bit user-space can't
seem to send a proper ioctl(BLKPG) to an MTD on my 64-bit kernel
(ARM64), I noticed that struct blkpg_ioctl_arg is actually pretty
unsuitable for use in the ioctl() ABI, due to its use of raw pointers,
and its lack of alignment/packing restrictions (32-bit arch'es tend to
pack the 4 fields into 4 32-bit words, whereas 64-bit arch'es would add
padding after the third int, and make this 6 32-bit words).
Anyway, this means BLKPG deserves some special compat_ioctl handling. Do
the conversion in a small shim for MTD.
block/compat_ioctl.c already has compat support for the block subsystem,
but it does so by a re-marshalling data to/from user-space (see
compat_blkpg_ioctl()). Personally, I think this approach is cleaner.
Tested only on MTD, with an ARM32 user space on an ARM64 kernel.
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
This patch changes the bridge vlan implementation to use rhashtables
instead of bitmaps. The main motivation behind this change is that we
need extensible per-vlan structures (both per-port and global) so more
advanced features can be introduced and the vlan support can be
extended. I've tried to break this up but the moment net_port_vlans is
changed and the whole API goes away, thus this is a larger patch.
A few short goals of this patch are:
- Extensible per-vlan structs stored in rhashtables and a sorted list
- Keep user-visible behaviour (compressed vlans etc)
- Keep fastpath ingress/egress logic the same (optimizations to come
later)
Here's a brief list of some of the new features we'd like to introduce:
- per-vlan counters
- vlan ingress/egress mapping
- per-vlan igmp configuration
- vlan priorities
- avoid fdb entries replication (e.g. local fdb scaling issues)
The structure is kept single for both global and per-port entries so to
avoid code duplication where possible and also because we'll soon introduce
"port0 / aka bridge as port" which should simplify things further
(thanks to Vlad for the suggestion!).
Now we have per-vlan global rhashtable (bridge-wide) and per-vlan port
rhashtable, if an entry is added to a port it'll get a pointer to its
global context so it can be quickly accessed later. There's also a
sorted vlan list which is used for stable walks and some user-visible
behaviour such as the vlan ranges, also for error paths.
VLANs are stored in a "vlan group" which currently contains the
rhashtable, sorted vlan list and the number of "real" vlan entries.
A good side-effect of this change is that it resembles how hw keeps
per-vlan data.
One important note after this change is that if a VLAN is being looked up
in the bridge's rhashtable for filtering purposes (or to check if it's an
existing usable entry, not just a global context) then the new helper
br_vlan_should_use() needs to be used if the vlan is found. In case the
lookup is done only with a port's vlan group, then this check can be
skipped.
Things tested so far:
- basic vlan ingress/egress
- pvids
- untagged vlans
- undef CONFIG_BRIDGE_VLAN_FILTERING
- adding/deleting vlans in different scenarios (with/without global ctx,
while transmitting traffic, in ranges etc)
- loading/removing the module while having/adding/deleting vlans
- extracting bridge vlan information (user ABI), compressed requests
- adding/deleting fdbs on vlans
- bridge mac change, promisc mode
- default pvid change
- kmemleak ON during the whole time
Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes timeout issues seen on large NOR flash (e.g., 16MB
w25q128fw) when using ioctl(MEMERASE) with offset=0 and length=16M. The
input parameters matter because spi_nor_erase() uses a different code
path for full-chip erase, where we use the SPINOR_OP_CHIP_ERASE (0xc7)
opcode.
Fix: use a different timeout for full-chip erase than for other
commands.
While most operations can be expected to perform relatively similarly
across a variety of NOR flash types and sizes (and therefore might as
well use a similar timeout to keep things simple), full-chip erase is
unique, because the time it typically takes to complete:
(1) is much larger than most operations and
(2) scales with the size of the flash.
Let's base our timeout on the original comments stuck here -- that a 2MB
flash requires max 40s to erase.
Small survey of a few flash datasheets I have lying around:
Chip Size (MB) Max chip erase (seconds)
---- -------- ------------------------
w25q32fw 4 50
w25q64cv 8 30
w25q64fw 8 100
w25q128fw 16 200
s25fl128s 16 ~256
s25fl256s 32 ~512
From this data, it seems plenty sufficient to say we need to wait for
40 seconds for each 2MB of flash.
After this change, it might make some sense to decrease the timeout for
everything else, as even the most extreme operations (single block
erase?) shouldn't take more than a handful of seconds. But for safety,
let's leave it as-is. It's only an error case, after all, so we don't
exactly need to optimize it.
Signed-off-by: Furquan Shaikh <furquan@google.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
The change adds support of ARM PrimeCell PL176 MPMC.
Static memory configuration of PL175 MPMC is very similar to one found
on PL172 and PL175 controllers, so it is preferred to add its support
into the existing driver. The difference is that PL176 supports up to
10 slave ports (but only 4 of them may be connected to static memory
devices), AHB master bus width cab be 64-bit wide, also NAND devices
can be interfaced.
Similar to PL175 contoller, PL176 has no write buffer enable control
in static memory configuration register, the rest of static memory
configuration bits (with exception of NAND) is the same.
Signed-off-by: Vladimir Zapolskiy <vz@mleia.com>
Signed-off-by: Joachim Eastwood <manabian@gmail.com>
The change adds support of ARM PrimeCell PL175 MPMC, the controller is
found on NXP LPC32xx SoC.
PL175 MPMC is very similar to PL172 controller, so it is preferred to
add its support into the existing driver.
One of the differences between PL172 and PL175 is that the latter one
supports up to 6 AHB ports, but still only 4 AHB ports can be assigned
to a static memory device, also PL175 does not have write buffer
enable control in static memory configuration register.
Signed-off-by: Vladimir Zapolskiy <vz@mleia.com>
Signed-off-by: Joachim Eastwood <manabian@gmail.com>
According to PL172 TRM read of bits [7:6] of MPMCPeriphID3 is
undefined, so unmask them. Also the driver supports all currently
present revisions of PL172, this allows to alleviate requirements to
the revision version matched by the driver.
Signed-off-by: Vladimir Zapolskiy <vz@mleia.com>
Signed-off-by: Joachim Eastwood <manabian@gmail.com>
It is a 512KiB flash with 4 KiB erase sectors.
Signed-off-by: Yuan Yao <yao.yuan@freescale.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
The module pcmciamtd doesn't generate a mtd node for PRETEC 4MB SRAM
cards without the id and hash added to pcmciamtd.c
Tested on 3 different 4MB pretec sram cards.
Signed-off-by: Tom Englund <tomenglund26@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Only create the stacktrace trace option when CONFIG_STACKTRACE is
configured.
Cleaned up the ftrace_trace_stack() function call a little to allow better
encapsulation of the stacktrace trace flag.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Instead of only enabling the backlight (which seems to set it to max
brightness), just re-set the current backlight level, which also takes
care of enabling the backlight if necessary.
Port of radeon commit:
drm/radeon: Restore LCD backlight level on resume (>= R5xx)
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
When the function tracer is not compiled in, do not create the option files
for it.
Fix up both the sched_wakeup and irqsoff tracers to handle the change.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Gregory CLEMENT says:
====================
net: mvneta: Switch to per-CPU irq and make rxq_def useful
As stated in the first version: "this patchset reworks the Marvell
neta driver in order to really support its per-CPU interrupts, instead
of faking them as SPI, and allow the use of any RX queue instead of
the hardcoded RX queue 0 that we have currently."
Following the review which has been done, Maxime started adding the
CPU hotplug support. I continued his work a few weeks ago and here is
the result.
Since the 1st version the main change is this CPU hotplug support, in
order to validate it I powered up and down the CPUs while performing
iperf. I ran the tests during hours: the kernel didn't crash and the
network interfaces were still usable. Of course it impacted the
performance, but continuously power down and up the CPUs is not
something we usually do.
I also reorganized the series, the 3 first patches should go through
the irq subsystem, whereas the 4 others should go to the network
subsystem.
However, there is a runtime dependency between the two parts. Patch 5
depend on the patch 3 to be able to use the percpu irq.
Thanks,
Gregory
PS: Thanks to Willy who gave me some pointers on how to deal with the
NAPI.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Since the switch to per-CPU interrupts, we lost the ability to set which
CPU was going to receive our RX interrupt, which was now only the CPU on
which the mvneta_open function was run.
We can now assign our queues to their respective CPUs, and make sure only
this CPU is going to handle our traffic.
This also paves the road to be able to change that at runtime, and later on
to support RSS.
[gregory.clement@free-electrons.com]: hardened the CPU hotplug support.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The mvneta driver allows to change the default RX queue trough the rxq_def
kernel parameter.
However, the current code doesn't allow to have any value but 0. It is
actively checked for in the driver's probe because the drivers makes a
number of assumption and takes a number of shortcuts in order to just use
that RX queue.
Remove these limitations in order to be able to specify any available
queue.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that our interrupt controller is allowing us to use per-CPU interrupts,
actually use it in the mvneta driver.
This involves obviously reworking the driver to have a CPU-local NAPI
structure, and report for incoming packet using that structure.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The CPU_MAP register is duplicated for each CPUs at different addresses,
each instance being at a different address.
However, the code so far was using CONFIG_NR_CPUS to initialise the CPU_MAP
registers for each registers, while the SoCs embed at most 4 CPUs.
This is especially an issue with multi_v7_defconfig, where CONFIG_NR_CPUS
is currently set to 16, resulting in writes to registers that are not
CPU_MAP.
Fixes: c5aff18204 ("net: mvneta: driver for Marvell Armada 370/XP network unit")
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: <stable@vger.kernel.org> # v3.8+
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The MPIC driver currently has a list of interrupts to handle as per-cpu.
Since the timer, fabric and neta interrupts were the only per-cpu
interrupts in the system, we can now remove the switch and just check for
the hardware irq number to determine whether a given interrupt is per-cpu
or not.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some drivers might use the per-cpu interrupts and still might be built as a
module. Export request_percpu_irq an free_percpu_irq to these user, which
also make it consistent with enable/disable_percpu_irq that were exported.
Reported-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The documentation of request_percpu_irq is confusing and suggest that the
interrupt is not enabled at all, while it is actually enabled on the local
CPU.
Clarify that.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Read Denali hardware revision number and use it to
calculate max_banks, The encoding of max_banks changed
in Denali revision 5.1.
Signed-off-by: Graham Moore <grmoore@opensource.altera.com>
[Brian: parentheses around macro arg]
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
For some reason the name of the external interrupt controller
has changed name with r8a7795, so use "intc-ex" instead of "irqc"
as r8a7795 compat string to follow the friendly documentation.
Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Reviewed-by: horms@verge.net.au
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
Cc: jason@lakedaemon.net
Cc: Magnus Damm <magnus.damm@gmail.com>
Link: http://lkml.kernel.org/r/20150928094228.32552.83336.sendpatchset@little-apple
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
A read id operation followed by 0x00 reads the device ID while
a read id operation followed by 0x20 reads the possible ONFI identifier.
As the READID function did not propagate the second id parameter but had
a hard-coded call for 0x90 0x00, reading the ONFI identifier was not
possible and thus chips werde not detected (tested with
MT29F8G08ABABAWP)
Signed-off-by: Enrico Jorns <ejo@pengutronix.de>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
We can now kill a number of glue functions which were sitting between
the common tda998x code and the drm encoder/connector methods. This
results in slightly cleaner code.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Kill the redundant tda998x_priv2 structure now that its only member is
the struct tda998x_priv.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Remove the encoder pointer from struct tda998x_priv, moving the encoder
itself from struct tda998x_priv2 here.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Remove the DRM slave encoder compatibility from the TDA998x driver. We
now use the component helpers to manage the binding of DRM sub-drivers.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The network namespace is easiliy available in state->net so use it.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Don't make ip6_route_me_harder guess which network namespace
it is routing in, pass the network namespace in.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Don't make ip_route_me_harder guess which network namespace
it is routing in, pass the network namespace in.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
ip6t_SYNPROXY already does this and this is needed so that we have a
struct net that can be passed down into ip_route_me_harder, so
that ip_route_me_harder can stop guessing it's context.
Along the way pass snet into synproxy_send_client_synack as this
is the only caller of synprox_send_tcp that is not passed snet
already.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The network namespace is needed when routing a packet.
Stop making nf_afinfo.reroute guess which network namespace
is the proper namespace to route the packet in.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This is needed so struct net can be pushed down into
ip_route_me_harder.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Instead of only enabling the backlight (which seems to set it to max
brightness), just re-set the current backlight level, which also takes
care of enabling the backlight if necessary.
Only the radeon_atom_encoder_dpms_dig part tested on a Kaveri laptop,
the radeon_atom_encoder_dpms_avivo part is only compile tested.
Cc: stable@vger.kernel.org
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The "i" variable should be signed or it leads to a crash in the error
handling code.
Fixes: 1d263474c4 ('drm/amdgpu: unwind properly in amdgpu_cs_parser_init()')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
CPU hotplug handling for blk-mq (blk_mq_queue_reinit) acquires
all_q_mutex in blk_mq_queue_reinit_notify() and then removes sysfs
entries by blk_mq_sysfs_unregister(). Removing sysfs entry needs to
be blocked until the active reference of the kernfs_node to be zero.
On the other hand, reading blk_mq_hw_sysfs_cpu sysfs entry (e.g.
/sys/block/nullb0/mq/0/cpu_list) acquires all_q_mutex in
blk_mq_hw_sysfs_cpus_show().
If these happen at the same time, a deadlock can happen. Because one
can wait for the active reference to be zero with holding all_q_mutex,
and the other tries to acquire all_q_mutex with holding the active
reference.
The reason that all_q_mutex is acquired in blk_mq_hw_sysfs_cpus_show()
is to avoid reading an imcomplete hctx->cpumask. Since reading sysfs
entry for blk-mq needs to acquire q->sysfs_lock, we can avoid deadlock
and reading an imcomplete hctx->cpumask by protecting q->sysfs_lock
while hctx->cpumask is being updated.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Reviewed-by: Ming Lei <tom.leiming@gmail.com>
Cc: Ming Lei <tom.leiming@gmail.com>
Cc: Wanpeng Li <wanpeng.li@hotmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Notifier callbacks for CPU_ONLINE action can be run on the other CPU
than the CPU which was just onlined. So it is possible for the
process running on the just onlined CPU to insert request and run
hw queue before establishing new mapping which is done by
blk_mq_queue_reinit_notify().
This can cause a problem when the CPU has just been onlined first time
since the request queue was initialized. At this time ctx->index_hw
for the CPU, which is the index in hctx->ctxs[] for this ctx, is still
zero before blk_mq_queue_reinit_notify() is called by notifier
callbacks for CPU_ONLINE action.
For example, there is a single hw queue (hctx) and two CPU queues
(ctx0 for CPU0, and ctx1 for CPU1). Now CPU1 is just onlined and
a request is inserted into ctx1->rq_list and set bit0 in pending
bitmap as ctx1->index_hw is still zero.
And then while running hw queue, flush_busy_ctxs() finds bit0 is set
in pending bitmap and tries to retrieve requests in
hctx->ctxs[0]->rq_list. But htx->ctxs[0] is a pointer to ctx0, so the
request in ctx1->rq_list is ignored.
Fix it by ensuring that new mapping is established before onlined cpu
starts running.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Reviewed-by: Ming Lei <tom.leiming@gmail.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
CPU hotplug handling for blk-mq (blk_mq_queue_reinit) accesses
q->mq_usage_counter while freezing all request queues in all_q_list.
On the other hand, q->mq_usage_counter is deinitialized in
blk_mq_free_queue() before deleting the queue from all_q_list.
So if CPU hotplug event occurs in the window, percpu_ref_kill() is
called with q->mq_usage_counter which has already been marked dead,
and it triggers warning. Fix it by deleting the queue from all_q_list
earlier than destroying q->mq_usage_counter.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Reviewed-by: Ming Lei <tom.leiming@gmail.com>
Cc: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
CPU hotplug handling for blk-mq (blk_mq_queue_reinit) updates
q->mq_map by blk_mq_update_queue_map() for all request queues in
all_q_list. On the other hand, q->mq_map is released before deleting
the queue from all_q_list.
So if CPU hotplug event occurs in the window, invalid memory access
can happen. Fix it by releasing q->mq_map in blk_mq_release() to make
it happen latter than removal from all_q_list.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Suggested-by: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Ming Lei <tom.leiming@gmail.com>
Cc: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
There is a race between cpu hotplug handling and adding/deleting
gendisk for blk-mq, where both are trying to register and unregister
the same sysfs entries.
null_add_dev
--> blk_mq_init_queue
--> blk_mq_init_allocated_queue
--> add to 'all_q_list' (*)
--> add_disk
--> blk_register_queue
--> blk_mq_register_disk (++)
null_del_dev
--> del_gendisk
--> blk_unregister_queue
--> blk_mq_unregister_disk (--)
--> blk_cleanup_queue
--> blk_mq_free_queue
--> del from 'all_q_list' (*)
blk_mq_queue_reinit
--> blk_mq_sysfs_unregister (-)
--> blk_mq_sysfs_register (+)
While the request queue is added to 'all_q_list' (*),
blk_mq_queue_reinit() can be called for the queue anytime by CPU
hotplug callback. But blk_mq_sysfs_unregister (-) and
blk_mq_sysfs_register (+) in blk_mq_queue_reinit must not be called
before blk_mq_register_disk (++) and after blk_mq_unregister_disk (--)
is finished. Because '/sys/block/*/mq/' is not exists.
There has already been BLK_MQ_F_SYSFS_UP flag in hctx->flags which can
be used to track these sysfs stuff, but it is only fixing this issue
partially.
In order to fix it completely, we just need per-queue flag instead of
per-hctx flag with appropriate locking. So this introduces
q->mq_sysfs_init_done which is properly protected with all_q_mutex.
Also, we need to ensure that blk_mq_map_swqueue() is called with
all_q_mutex is held. Since hctx->nr_ctx is reset temporarily and
updated in blk_mq_map_swqueue(), so we should avoid
blk_mq_register_hctx() seeing the temporary hctx->nr_ctx value
in CPU hotplug handling or adding/deleting gendisk .
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Reviewed-by: Ming Lei <tom.leiming@gmail.com>
Cc: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
When unmapped hw queue is remapped after CPU topology is changed,
hctx->tags->cpumask has to be set after hctx->tags is setup in
blk_mq_map_swqueue(), otherwise it causes null pointer dereference.
Fixes: f26cdc8536 ("blk-mq: Shared tag enhancements")
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Keith Busch <keith.busch@intel.com>
Cc: Ming Lei <tom.leiming@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
