James Nunez [Mon, 7 Mar 2016 23:10:21 +0000 (18:10 -0500)]
staging: lustre: Correct missing newline
Several error messages are missing newline characters
at the end of the message. Newlines are added where
necessary and other minor corrections; no punctuation
at the end of an error message, add a return code to
the end of error messages, device name at the beginning,
etc.
There are just a couple of places where newlines are
removed and this is only in LDLM_DEBUG_NOLOCK. The definition
of LDLM_DEBUG_NOLOCK already has a newline in it and
resulted in double newlines printed.
Signed-off-by: James Nunez <james.a.nunez@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4871
Reviewed-on: http://review.whamcloud.com/10000 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Cliff White <cliff.white@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: lustre: fix 'data race condition' issue in framework.c
Fix 'data race condition' defects found by Coverity version 6.5.0:
Data race condition (MISSING_LOCK)
Accessing variable without holding lock. Elsewhere,
this variable is accessed with lock held.
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2744
Reviewed-on: http://review.whamcloud.com/6568 Reviewed-by: Bob Glossman <bob.glossman@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix 'NULL pointer dereference' defects found by Coverity version
6.5.3:
Dereference after null check (FORWARD_NULL)
For instance, Passing null pointer to a function which dereferences
it.
Dereference before null check (REVERSE_INULL)
Null-checking variable suggests that it may be null, but it has
already been dereferenced on all paths leading to the check.
Dereference null return value (NULL_RETURNS)
The following fixes for the LNet layer are broken out of patch
http://review.whamcloud.com/4720.
staging: lustre: fix 'data race condition' issue in conrpc.c
Fix 'data race condition' defects found by Coverity version 6.5.0:
Data race condition (MISSING_LOCK)
Accessing variable without holding lock. Elsewhere,
this variable is accessed with lock held.
Liang Zhen [Mon, 7 Mar 2016 23:10:16 +0000 (18:10 -0500)]
staging: lustre: LNet drop rule implementation
This is implementation of LNet Drop Rule, which can randomly drop
LNet messages at specified rate.
LNet Drop Rule can only be applied to receive side of message. User
can add drop_rule either on end point of cluster (client/server) or
on LNet routers.
Here are lctl command to control LNet Drop Rules:
- net_drop_add -s SRC_NID -d DEST_NID --rate VALUE
drop 1/@VALUE of messages from @SRC_NID to @DEST_NID
- net_drop_del -s SRC_NID -d DEST_NID
remove all drop rules from @SRC_NID to @DEST_NID
- net_drop_list
list all drop rules on current node
Examples:
- lctl net_drop_add -s *@o2ib0 -d 192.168.1.102@tcp 1000
add new drop rule, it will drop 1/1000 messages from network o2ib0
to 192.168.1.102@tcp
- lctl net_drop_add -s 10.8.6.123@o2ib1 -d * 500
add new drop rule, it will drop 1/500 messages from 10.8.6.123@o2ib1
to all nodes
Signed-off-by: Liang Zhen <liang.zhen@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5435
Reviewed-on: http://review.whamcloud.com/11314 Reviewed-by: Bobi Jam <bobijam@gmail.com> Reviewed-by: Amir Shehata <amir.shehata@intel.com> Reviewed-by: Johann Lombardi <johann.lombardi@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Arnd Bergmann [Wed, 2 Mar 2016 19:06:46 +0000 (20:06 +0100)]
isdn: i4l: move active-isdn drivers to staging
The icn, act2000 and pcbit drivers are all for very old hardware,
and it is highly unlikely that anyone is actually still using them
on modern kernels, if at all.
All three drivers apparently are for hardware that predates PCI
being the common connector, as they are ISA-only and active
PCI ISDN cards were widely available in the 1990s.
Looking through the git logs, it I cannot find any indication of a
patch to any of these drivers that has been tested on real hardware,
only cleanups or global API changes.
Arnd Bergmann [Wed, 2 Mar 2016 19:06:45 +0000 (20:06 +0100)]
isdn: icn: remove a #warning
The icn driver currently produces an unconditional #warning whenever
we build it, introduced by Karsten Keil back in 2003:
#warning TODO test headroom or use skb->nb to flag ACK
Karsten's original commit (from BitKeeper) contains this description:
- here are lot of bugs left, so ISDN is not stable yet but
I think it's really time to fix it, even if it need some cycles
to get it right (normally I'm only send patches if it works 100% for
me).
- I add some additional #warnings to address places which need fixing
(I hope that some of the other ISDN developer jump in)
Apparently this has not happened, and it is unlikely that it ever will,
given that the driver doesn't seem to work. No substantial bug fixes
other than janitorial cleanups have happened in the driver since then,
and I see no indication that anyone who patched it had the hardware.
We should probably either remove the driver, or remove all of i4l,
but for now, this shuts up the distracting #warning by turning it
into a comment.
Ian Abbott [Wed, 2 Mar 2016 14:27:38 +0000 (14:27 +0000)]
staging: comedi: comedi.h: Fix comment for struct comedi_krange
The kernel-doc comment for `struct comedi_krange` refers to the macro
constant `RF_external`. It should be `RF_EXTERNAL`, so fix it. It also
documents the value of the constant as `(1 << 8)`, but the macro now
expands to the hexadecimal constant `0x100`, so use that as the
documented value.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Ian Abbott [Wed, 2 Mar 2016 14:27:37 +0000 (14:27 +0000)]
staging: comedi: comedi.h: Do not use BIT macro
The "comedi.h" file is part of the user API for COMEDI devices, and is
intended to be migrated to "include/uapi/linux". The `BIT` macro from
"include/linux/bitops.h" should not be used there.
Replace the use of the `BIT` macro with hexadecimal constants of the
same value. The `BIT` macro replaced expressions of the form `(1 << N)`
in this file originally, but reverting back to that form would encourage
patches changing them back to use the `BIT` macro.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Reviewed-by: H Hartley Sweeten <hsweeten@visionengravers.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Bhumika Goyal [Sun, 28 Feb 2016 07:58:47 +0000 (13:28 +0530)]
Staging: comedi: s626.c: Remove unused functions
The functions s626_get_clk_mult, s626_get_clk_mult, s626_get_enc_mode,
s626_set_index_pol are not used anywhere in the kernel so they can be
removed. This also cleans the code. Grepped to find the occurences.
alloc_and_init_dma_members does not affect the interrupt status and is
only called by auto_attach, which also does not affect the interrupt
status. auto_attach() also contains a call to comedi_alloc_devpriv()
which calls kzalloc with GFP_KERNEL flag. Thus, there seems to be no
danger that dma_alloc_coherent can be called with interrupts turned
off, and GFP_KERNEL can be used.
Bhumika Goyal [Tue, 1 Mar 2016 19:43:22 +0000 (01:13 +0530)]
Staging: gdm724x: Replace random_ether_addr with eth_random_addr
The macro random_ether_addr is calling the function eth_random_addr.
Therefore, the call to random_ether_addr can be replaced with
eth_random_addr.
Done using coccinelle:
Bhumika Goyal [Fri, 26 Feb 2016 10:04:32 +0000 (15:34 +0530)]
Staging: gdm724x: Use min instead of ternary operator
This patch replaces ternary operator with macro min as it shorter and
thus increases code readability. Macro min return the minimum of the
two compared values.
Made a semantic patch for changes:
@@
type T;
T x;
T y;
@@
(
- x < y ? x : y
+ min(x,y)
|
- x > y ? x : y
+ max(x,y)
)
With concurrency managed workqueues, use of dedicated workqueues
can be replaced by using system_wq. Drop usb_tx_wq and usb_rx_wq
by using system_wq.
Since there are multiple work items per udev but different udevs
do not need to be ordered, increase of concurrency level by
switching to system_wq should not break anything.
cancel_work_sync() is used to ensure that work is not pending or
executing on any CPU.
Lastly, since all devices are suspended, which shutdowns the work
items before the driver can be unregistered, it is guaranteed
that no work item is pending or executing by the time exit path
runs.
Dan Carpenter [Mon, 22 Feb 2016 19:33:09 +0000 (22:33 +0300)]
Staging: gdm72xx: underflow bug in gdm_wimax_ioctl_get_data()
"size" here should be unsigned, otherwise we might end up trying to copy
negative bytes in gdm_wimax_ioctl_get_data() resulting in an information
leak.
Reported-by: Alan Cox <gnomes@lxorguk.ukuu.org.uk> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Dan Carpenter [Mon, 22 Feb 2016 19:31:31 +0000 (22:31 +0300)]
Staging: gdm72xx: silence underflow warning in netlink_send()
The value of "group" comes from "idx" in __gdm_wimax_event_send():
if (sscanf(e->dev->name, "wm%d", &idx) == 1)
Smatch marks sscanf values as user controlled. It's supposed to be a
number in 0-30 range. We cap the upper bound but allow negatives. Fix
this by making it type u16 instead.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Dan Carpenter [Mon, 22 Feb 2016 19:30:46 +0000 (22:30 +0300)]
staging: gdm72xx: underflow in netlink_rcv_cb()
If nlh->nlmsg_len is less than ND_IFINDEX_LEN we end up trying to memcpy
a negative size. I also re-ordered slighty the condition to make it
more uniform.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Alan [Mon, 15 Feb 2016 18:58:10 +0000 (18:58 +0000)]
rtl871x: avoid running off end of buffer
If 32 bytes of non zero are passed in pdata->pointer then the mac_pton
function will run off the end of the buffer. Make sure we always have a
terminated string kernel side.
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: speakup: Remove unnecessary test in if condition
Remove unnecessary test on synth->alive since it has already been
tested previously.
This fixes the following smatch warning:
drivers/staging/speakup/synth.c:182 spk_synth_is_alive_restart() warn:
we tested 'synth->alive' before and it was 'false'
James Simmons [Thu, 3 Mar 2016 16:00:58 +0000 (11:00 -0500)]
staging: lustre: handle complex strings in cfs_str2num_check
Originally the function cfs_str2num_check used simple_strtoul
but has been updated to kstrtoul. The string passed into
cfs_str2num_check can be a very complex, for example we could
have 10.37.202.[59-61]. When simple_strtoul was used the first
number until we hit a non-digit character could be extracted
but testing showed that kstrtoul will not return any value if
it detects any non-digit character. Because of this change in
behavior a different approach is needed to handle these types
of complex strings. The use of sscanf was investigated to see
if it could be used to extract numbers from the passed in
string but unlike its glibc counterpart the kernel version
also just reported a error with no results if a non-digit value
in the string was encountered. Another possible approach would
be to use __parse_int directly but that class of functions is
not exported by the kernel. So the approach in this patch is
to scan the string passed in for the first non-digit character
and replace that character with a '\0' so kstrtoul can be used.
Once completed the original character is restored. We also
restore a original behavior that was removed to return 0 when
we encounter any non digit character before the nob count.
Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Liang Zhen [Wed, 2 Mar 2016 23:53:30 +0000 (18:53 -0500)]
staging: lustre: do less intense allocating retry for ko2iblnd
ko2iblnd may retry too frequent for growing pools, all schedulers
are spinning if another thread is in progress of allocating a new
pool and can't finish right away because of high system load.
Liang Zhen [Wed, 2 Mar 2016 23:53:29 +0000 (18:53 -0500)]
staging: lustre: avoid intensive reconnecting for ko2iblnd
When there is a connection race between two nodes and one side
of the connection is rejected by the other side. o2iblnd will
reconnect immediately, this is going to generate a lot of
trashes if:
- race winner is slow and can't send out connecting request
in short time.
- remote side leaves a cmid in TIMEWAIT state, which will reject
future connection requests
To resolve this problem, this patch changed the reconnection
behave: reconnection is submitted by connd only if a zombie
connection is being destroyed and there is a pending
reconnection request for the corresponding peer.
Also, after a few rejections, reconnection will have a time
interval between each attempt.
Liang Zhen [Wed, 2 Mar 2016 23:53:28 +0000 (18:53 -0500)]
staging: lustre: check wr_id returned by ib_poll_cq
If ib_poll_cq returned +ve without initialising ib_wc::wr_id (bug
in driver), then o2iblnd will run into unpredictable situation
because ib_wc::wr_id may refer to stale tx/rx pointer in stack.
It indicates bug in HCA driver if this happened, ko2iblnd should
output console error then close current connection.
Doug Oucharek [Wed, 2 Mar 2016 23:53:27 +0000 (18:53 -0500)]
staging: lustre: Change connect peer failed cleanup order
A race condition has been found where connd is cleaning up failed
connections, the peer ref counter goes to zero, but we stil have
a connecting counter > 0.
One possible race is when we are retrying a connection by
calling kiblnd_connect_peer() which itself fails and decrements
the peer ref counter and gets swapped out before it can decrement
the connecting counter. connd swaps in and cleans up the
connection where it sees a peer ref counter of 1 and a connecting
counter of 1. This will trigger the assert seen in LU-7210 when
it decrements the peer counter.
The solution: be sure to decrement the connecting counter
before decrementing the peer counter in the peer connect
failure path.
Liang Zhen [Wed, 2 Mar 2016 23:53:26 +0000 (18:53 -0500)]
staging: lustre: take extra refcount in kiblnd_connreq_done
refcount taken by cmid is not reliable after kiblnd_connreq_done
released the glock because this connection is visible to other
threads, another thread can find and close this connection right
after kiblnd_connreq_done released the glock, if kiblnd_cm_callback
for RDMA_CM_EVENT_DISCONNECTED is called, it can release the
connection refcount taken by cmid. It means the connection could be
destroyed before kiblnd_connreq_done() finish operations on it.
Amir Shehata [Wed, 2 Mar 2016 23:53:25 +0000 (18:53 -0500)]
staging: lustre: make ko2iblnd connect parameters persistent
Store map-on-demand and peertx credits in the peer, since the peer
is persistent. Also made sure that when assigning the parameters
received on the connection to the peer structure through create,
that if another peer is added before grabbing the lock we assign
these parameters to it as well.
Bhumika Goyal [Wed, 2 Mar 2016 19:16:37 +0000 (00:46 +0530)]
Staging: lustre: lnet: Remove and rename struct typedefs
Remove typedefs from structures stt_timer_t and lst_ping_data_t as typedef
for a structure type is not preferred. The suffix '_t' and the typedefs
are removed using coccinelle.
Script 1:
//Drop typedefs and '_t'
@r1@
type T;
@@
typedef struct { ... } T;
@script:python c1@
T2;
T << r1.T;
@@
if T[-2:] =="_t":
coccinelle.T2 = T[:-2];
print T
else:
coccinelle.T2=T;
staging: lustre: llite: Remove unnecessary test in if condition
Remove unnecessary test on `rc` variable since it has already been
tested previously.
Fixes following smatch warning:
drivers/staging/lustre/lustre/llite/llite_lib.c:1331 ll_setattr_raw()
warn: we tested 'rc' before and it was 'false'
James Simmons [Wed, 2 Mar 2016 22:01:59 +0000 (17:01 -0500)]
staging: lustre: reverse LNet and infinband header order
LNet is an abstraction built on top of network APIs such
as infiniband or TCP/IP layer. Since LNet is dependent
on these other layers we should ensure LNet headers
should always come after the infiniband header since the
infiniband headers can influence the LNet header
definitions.
Dmitry Eremin [Wed, 2 Mar 2016 22:01:58 +0000 (17:01 -0500)]
staging: lustre: fix framework.c issues found by Klocwork Insight tool
The functions sfw_test_buffers() and sfw_unload_test() from LNet
selftest both assume sfw_test_instance_t being passed in is never
null. This is corrected here.
Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4629
Reviewed-on: http://review.whamcloud.com/9386 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Isaac Huang <he.huang@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Dmitry Eremin [Wed, 2 Mar 2016 22:01:57 +0000 (17:01 -0500)]
staging: lustre: fix conctl.c issues found by Klocwork Insight tool
The function lst_test_add_ioctl is always copying lstio_tes_param
from userland even if the user doesn't send this data to LNet
selftest. Only consider lstio_tes_param data if lstio_tes_param_len
is not zero.
Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4629
Reviewed-on: http://review.whamcloud.com/9386 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Isaac Huang <he.huang@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Dmitry Eremin [Wed, 2 Mar 2016 22:01:56 +0000 (17:01 -0500)]
staging: lustre: fix api-ni.c issues found by Klocwork Insight tool
Pointer 'ni' checked for NULL at line 1569 may be passed to
function and may be dereferenced there by passing argument 1 to
function 'lnet_ni_notify_locked' at line 1621.
Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4629
Reviewed-on: http://review.whamcloud.com/9386 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Isaac Huang <he.huang@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
James Simmons [Wed, 2 Mar 2016 22:01:54 +0000 (17:01 -0500)]
staging: lustre: bind socklnd peers to a specific CPT
Currently the socklnd driver doesn't support
CPT affinity for its peers. Binding peers to
a specific CPT and memory allocated to the
NUMA node belonging to the CPT should give a
performance boost.
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-7245
Reviewed-on: http://review.whamcloud.com/16710 Reviewed-by: Olaf Weber <olaf@sgi.com> Reviewed-by: Amir Shehata <amir.shehata@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Jian Yu [Wed, 2 Mar 2016 22:01:53 +0000 (17:01 -0500)]
staging: lustre: replace direct LNet HZ access with kernel APIs
On some customers' systems, the kernel was compiled with HZ defined
to 100, instead of 1000. This improves performance for HPC applications.
However, to use these systems with Lustre, customers have to re-build
Lustre for the kernel because Lustre directly uses the defined constant
HZ.
Since kernel 2.6.21, some non-HZ dependent timing APIs become non-
inline functions, which can be used in Lustre codes to replace the
direct HZ access.
These kernel APIs include:
jiffies_to_msecs()
jiffies_to_usecs()
jiffies_to_timespec()
msecs_to_jiffies()
usecs_to_jiffies()
timespec_to_jiffies()
And here are some samples of the replacement:
HZ -> msecs_to_jiffies(MSEC_PER_SEC)
n * HZ -> msecs_to_jiffies(n * MSEC_PER_SEC)
HZ / n -> msecs_to_jiffies(MSEC_PER_SEC / n)
n / HZ -> jiffies_to_msecs(n) / MSEC_PER_SEC
n / HZ * 1000 -> jiffies_to_msecs(n)
This patch replaces the direct HZ access in lnet module.
John L. Hammond [Wed, 2 Mar 2016 22:01:52 +0000 (17:01 -0500)]
staging: lustre: set task state before scheduling in lnet_sock_accept
In the original code change when libcfs_sock_accept() was made
into lnet_sock_accept() a call to set_current_state(TASK_INTERRUPTIBLE)
got dropped which was restored. For upstream this is an
optimization of calling init_waitqueue_entry() only if
accept() return -EAGAIN. Also we can remove setting the
task to TASK_RUNNING that is not needed.
Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6407
Reviewed-on: http://review.whamcloud.com/14265 Reviewed-by: James Simmons <uja.ornl@yahoo.com> Reviewed-by: Amir Shehata <amir.shehata@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: lustre: fix 'copy into fixed size buffer' errors
Fix 'copy into fixed size buffer' defects found by Coverity
version 6.0.3:
Copy into fixed size buffer (STRING_OVERFLOW)
The fixed-size string might be overrun by copying without
checking the length.
Olaf Weber [Wed, 2 Mar 2016 22:01:49 +0000 (17:01 -0500)]
staging: lustre: Use after free in lnet_ptl_match_delay()
In lnet_ptl_match_delay() we check msg->msg_rx_delayed to see whether
the message has been added to the delay queue. But this check is done
after lnet_ptl_unlock() and lnet_res_unlock(), and the message can be
processed and freed before the check.
Replace the check with checking rc against LNET_MATCHMD_NONE, which
is how the callers of lnet_ptl_match_delay() know whether the message
was added to the delay queue. To make this work we reset rc in the
loop when there was no match and the message hasn't been delayed. In
addition reorganize the code and add comments to clarify the logic.
In lnet_ptl_match_md() a similar msg->msg_rx_delayed is replaced for
the same reason.
Li Xi [Wed, 2 Mar 2016 22:01:48 +0000 (17:01 -0500)]
staging: lustre: remove annoying message in parse_nidrange
When setting TBF rules of jobid, parse_nidrange() prints warning
messages. However, this is unnecessary and annoying since paring
a TBF rule will always try to parse the jobid like a nid.
Signed-off-by: Li Xi <lixi@ddn.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-7647
Reviewed-on: http://review.whamcloud.com/17916 Reviewed-by: Emoly Liu <emoly.liu@intel.com> Reviewed-by: Bobi Jam <bobijam@hotmail.com> Reviewed-by: Lai Siyao <lai.siyao@intel.com> Reviewed-by: James Simmons <uja.ornl@yahoo.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Amir Shehata [Wed, 2 Mar 2016 22:01:46 +0000 (17:01 -0500)]
staging: lustre: Ignore hops if not explicitly set
Since the # of hops is not a mandatory parameter the LU-6060
patch will cause problems to already existing systems since it
changes the behavior by which a route is determined down.
To fix this case the # of hops now defaults to LNET_UNDEFINED_HOPS
if no hop count is specified.
LNET_UNDEFINED_HOPS is defined to ((__u32)-1). When it's printed as
%d, it displays as -1.
__u32 is used through out the call stack for hop count to explicitly
define the size of the hop count and to avoid any sizing issues when
passing data to and from the kernel.
To keep existing behavior both lnet_compare_routes() and LNetDist()
will treat undefined hop count as hop count 1.
When executing the logic in lnet_parse_rc_info() there is no
longer an assumption that the default hop count is 1. If
the hop count is 1 then it must've been explicitly set by
the user.
Liang Zhen [Wed, 2 Mar 2016 22:01:45 +0000 (17:01 -0500)]
staging: lustre: recv could access freed message
When lnet_parse_put calls lnet_ptl_match_md, this function can attach
current message on the delayed list if there is no match. It means
this message can be taken over and freed by another thread who is
posting new MD, then it is not safe for caller of lnet_parse_put to
check this message again.
This patch fixes this issue by adding a local variable "ready_delay"
to store corresponding status of lnet_msg, so lnet doesn't need to
check the message again if lnet_ptl_match_md returned MATCH_NONE for
it.
Liang Zhen [Wed, 2 Mar 2016 22:01:44 +0000 (17:01 -0500)]
staging: lustre: set downis to 1 if there's no NI for remote net
lnet_route_t::lr_downis is marked as zero even if there is no NI to
target network, this is wrong and breaks logic of ARF. This patch
fixes this problem.
Oleg Drokin [Wed, 2 Mar 2016 13:25:39 +0000 (08:25 -0500)]
staging/lustre: Unbreak ll_prep_inode iget result check
commit 020ecc6f3229 ("staging: lustre: llite: Remove IS_ERR tests")
managed to break one of the converted checks by losing a dereference
causing the condition to be wrong and hence leading
to a crash later on in case of actual error.
IS_ERR_OR_NULL(*inode) got converted into !inode.
staging: wilc1000: wilc_wlan: Remove wrapper function
Remove is_tcp_ack_filter() which only returns the value of boolean
variable `enabled` by replacing the only call site with the variable
`enabled` itself.
Variables pcgroup_encrypt_val,pccipher_group,pcwpa_version,
hold different values at different stages of the execution of
connect(), however they are not being used anywhere.
Hence, the unused variables have been removed.
This was done using Coccinelle.
@@ type T; identifier i; constant c; @@
-T i;
<... when != i
-i = c;
...>
Alison Schofield [Fri, 26 Feb 2016 07:03:42 +0000 (23:03 -0800)]
staging: wilc1000: remove code for HOST_IF_MSG_SET_MAC_ADDRESS
The function that sent this message id was previously removed.
Finish the cleanup by removing the call to the message handler,
the message handler, and the #define.
Alison Schofield [Fri, 26 Feb 2016 07:02:59 +0000 (23:02 -0800)]
staging: wilc1000: remove code for HOST_IF_MSG_Q_IDLE
The function that sent this message id was previously removed.
Finish the cleanup by removing the call to the message handler,
the message handler, and the #define.
Alison Schofield [Fri, 26 Feb 2016 07:02:17 +0000 (23:02 -0800)]
staging: wilc1000: remove code for HOST_IF_MSG_FLUSH_CONNECT
The function that sent this message id was previously removed.
Finish the cleanup by removing the call to the message handler,
the message handler, and the #define.
Alison Schofield [Fri, 26 Feb 2016 07:01:13 +0000 (23:01 -0800)]
staging: wilc1000: remove code for HOST_IF_GET_LINKSPEED
The function that sent this message id was previously removed.
Finish the cleanup by removing the call to the message handler,
the message handler, the #define, and the link_speed var.
Alison Schofield [Fri, 26 Feb 2016 07:00:42 +0000 (23:00 -0800)]
staging: wilc1000: remove code for HOST_IF_MSG_GET_CHNL
The function that sent this message id was previously removed.
Finish the cleanup by removing the call to the message handler,
the message handler, the #define, and the channel var.
Bhumika Goyal [Sun, 28 Feb 2016 19:44:33 +0000 (01:14 +0530)]
Staging: lustre: Remove print statement on function failure
The memory allocation functions generates a call stack containing
all the context information on failure, so print statements can be
removed on failure of these functions. Also remove
unwanted {} around if block after removal of these statements.
Done using coccinelle:
M. Vefa Bicakci [Sun, 28 Feb 2016 14:26:22 +0000 (09:26 -0500)]
staging: lustre: Use __user for a pointer to a user space address
This commit corrects two sparse warnings caused by the lack of a __user
annotation for the third argument of the libcfs_ioctl_handle function.
module.c:165:68: warning: incorrect type in argument 1
(different address spaces)
module.c:165:68: expected void [noderef] <asn:1>*arg
module.c:165:68: got void *arg
module.c:209:47: warning: incorrect type in argument 3
(different address spaces)
module.c:209:47: expected void *arg
module.c:209:47: got void [noderef] <asn:1>*arg
The need to have the __user annotation is supported by the fact that
libcfs_ioctl_handle passes its third argument to a helper function
(libcfs_ioctl_popdata) which also has a __user annotation for its
corresponding argument.
Signed-off-by: M. Vefa Bicakci <m.v.b@runbox.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
M. Vefa Bicakci [Sun, 28 Feb 2016 14:26:21 +0000 (09:26 -0500)]
staging: lustre: Use const static file_operations struct
This commit corrects the following sparse warning regarding a
file_operations structure being non-static and constifies the
structure in question as well.
warning: symbol 'lprocfs_stats_seq_fops' was not declared.
Should it be static?
Signed-off-by: M. Vefa Bicakci <m.v.b@runbox.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: lustre: lnet: Use list_for_each_entry_safe
Doubly linked lists which are iterated using list_empty
and list_entry macros have been replaced with list_for_each_entry_safe macro.
This makes the iteration simpler and more readable.
This patch replaces the while loop containing list_empty and list_entry
with list_for_each_entry_safe.
This was done with Coccinelle.
@@
expression E1;
identifier I1, I2;
type T;
iterator name list_for_each_entry_safe;
@@
T *I1;
+ T *tmp;
...
- while (list_empty(&E1) == 0)
+ list_for_each_entry_safe (I1, tmp, &E1, I2)
{
...when != T *I1;
- I1 = list_entry(E1.next, T, I2);
...
}
Andreas Dilger [Fri, 26 Feb 2016 16:36:06 +0000 (11:36 -0500)]
staging: lustre: update comment for lnet_lib_init/exit
The documentation about the return values for lnet_lib_init
and lnet_lib_exit was in the old style format. Bring it in
sync with the rest of the LNet core. Broken out of patch 16787.
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6204
Reviewed-on: http://review.whamcloud.com/16787 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com> Reviewed-by: Frank Zago <fzago@cray.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Andreas Dilger [Fri, 26 Feb 2016 16:36:04 +0000 (11:36 -0500)]
staging: lustre: update the MODULE_DESCRIPTION for all lustre modules
Fixup the MODULE_DESCRIPTION for several lustre modules. Some wrongly
place the version in the string or they are not descriptive enough.
Broken out of patch http://review.whamcloud.com/16787.
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6204
Reviewed-on: http://review.whamcloud.com/16787 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com> Reviewed-by: Frank Zago <fzago@cray.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
James Simmons [Fri, 26 Feb 2016 16:36:03 +0000 (11:36 -0500)]
staging: lustre: add missing MODULE_AUTHOR for LNet selftest module
For several lustre modules the MODULE_VERSION has the wrong value,
located in the wrong place in the source code, or completely missing.
This patch brings it up to date.
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6204
Reviewed-on: http://review.whamcloud.com/16729 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Frank Zago <fzago@cray.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / linux-beck.git / log