The following trace is seen when receiving a DHCP request over a flow-based
VXLAN tunnel. I believe this is caused by the metadata dst having a NULL
dev value and as a result dev_net(dev) is causing a NULL pointer dereference.
To resolve this I am replacing the check for skb_dst(skb)->dev with just
skb->dev. This makes sense as the callers of this function are usually in
the receive path and as such skb->dev should always be populated. In
addition other functions in the area where these are called are already
using dev_net(skb->dev) to determine the namespace the UDP packet belongs
in.
Fixes: 63058308cd55 ("udp: Add udp6_lib_lookup_skb and udp4_lib_lookup_skb") Signed-off-by: Alexander Duyck <[email protected]> Acked-by: Eric Dumazet <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Ingo Molnar [Fri, 13 May 2016 05:35:12 +0000 (07:35 +0200)]
Merge tag 'perf-urgent-for-mingo-20160512' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent
Pull perf/urgent fixes from Arnaldo Carvalho de Melo:
- Fallback to usermode-only counters when perf_event_paranoid > 1, which
is the case now (Arnaldo Carvalho de Melo)
- Do not reassign parg after collapse_tree() in libtraceevent, which
may cause tool crashes (Steven Rostedt)
- Fix the build on Fedora Rawhide, where readdir_r() is deprecated and
also wrt -Werror=unused-const-variable= + x86_32_regoffset_table on
!x86_64 (Arnaldo Carvalho de Melo)
- Fix the build on Ubuntu 12.04.5, where dwarf_getlocations() isn't
available, i.e. libdw-dev < 0.157 (Arnaldo Carvalho de Melo)
Linus Torvalds [Fri, 13 May 2016 01:44:24 +0000 (18:44 -0700)]
Merge branch 'akpm' (patches from Andrew)
Merge fixes from Andrew Morton:
"4 fixes"
* emailed patches from Andrew Morton <[email protected]>:
mm: thp: calculate the mapcount correctly for THP pages during WP faults
ksm: fix conflict between mmput and scan_get_next_rmap_item
ocfs2: fix posix_acl_create deadlock
ocfs2: revert using ocfs2_acl_chmod to avoid inode cluster lock hang
Andrea Arcangeli [Thu, 12 May 2016 22:42:25 +0000 (15:42 -0700)]
mm: thp: calculate the mapcount correctly for THP pages during WP faults
This will provide fully accuracy to the mapcount calculation in the
write protect faults, so page pinning will not get broken by false
positive copy-on-writes.
total_mapcount() isn't the right calculation needed in
reuse_swap_page(), so this introduces a page_trans_huge_mapcount()
that is effectively the full accurate return value for page_mapcount()
if dealing with Transparent Hugepages, however we only use the
page_trans_huge_mapcount() during COW faults where it strictly needed,
due to its higher runtime cost.
This also provide at practical zero cost the total_mapcount
information which is needed to know if we can still relocate the page
anon_vma to the local vma. If page_trans_huge_mapcount() returns 1 we
can reuse the page no matter if it's a pte or a pmd_trans_huge
triggering the fault, but we can only relocate the page anon_vma to
the local vma->anon_vma if we're sure it's only this "vma" mapping the
whole THP physical range.
Kirill A. Shutemov discovered the problem with moving the page
anon_vma to the local vma->anon_vma in a previous version of this
patch and another problem in the way page_move_anon_rmap() was called.
Andrew Morton discovered that CONFIG_SWAP=n wouldn't build in a
previous version, because reuse_swap_page must be a macro to call
page_trans_huge_mapcount from swap.h, so this uses a macro again
instead of an inline function. With this change at least it's a less
dangerous usage than it was before, because "page" is used only once
now, while with the previous code reuse_swap_page(page++) would have
called page_mapcount on page+1 and it would have increased page twice
instead of just once.
Dean Luick noticed an uninitialized variable that could result in a
rmap inefficiency for the non-THP case in a previous version.
Mike Marciniszyn said:
: Our RDMA tests are seeing an issue with memory locking that bisects to
: commit 61f5d698cc97 ("mm: re-enable THP")
:
: The test program registers two rather large MRs (512M) and RDMA
: writes data to a passive peer using the first and RDMA reads it back
: into the second MR and compares that data. The sizes are chosen randomly
: between 0 and 1024 bytes.
:
: The test will get through a few (<= 4 iterations) and then gets a
: compare error.
:
: Tracing indicates the kernel logical addresses associated with the individual
: pages at registration ARE correct , the data in the "RDMA read response only"
: packets ARE correct.
:
: The "corruption" occurs when the packet crosse two pages that are not physically
: contiguous. The second page reads back as zero in the program.
:
: It looks like the user VA at the point of the compare error no longer points to
: the same physical address as was registered.
:
: This patch totally resolves the issue!
Zhou Chengming [Thu, 12 May 2016 22:42:21 +0000 (15:42 -0700)]
ksm: fix conflict between mmput and scan_get_next_rmap_item
A concurrency issue about KSM in the function scan_get_next_rmap_item.
task A (ksmd): |task B (the mm's task):
|
mm = slot->mm; |
down_read(&mm->mmap_sem); |
|
... |
|
spin_lock(&ksm_mmlist_lock); |
|
ksm_scan.mm_slot go to the next slot; |
|
spin_unlock(&ksm_mmlist_lock); |
|mmput() ->
| ksm_exit():
|
|spin_lock(&ksm_mmlist_lock);
|if (mm_slot && ksm_scan.mm_slot != mm_slot) {
| if (!mm_slot->rmap_list) {
| easy_to_free = 1;
| ...
|
|if (easy_to_free) {
| mmdrop(mm);
| ...
|
|So this mm_struct may be freed in the mmput().
|
up_read(&mm->mmap_sem); |
As we can see above, the ksmd thread may access a mm_struct that already
been freed to the kmem_cache. Suppose a fork will get this mm_struct from
the kmem_cache, the ksmd thread then call up_read(&mm->mmap_sem), will
cause mmap_sem.count to become -1.
As suggested by Andrea Arcangeli, unmerge_and_remove_all_rmap_items has
the same SMP race condition, so fix it too. My prev fix in function
scan_get_next_rmap_item will introduce a different SMP race condition, so
just invert the up_read/spin_unlock order as Andrea Arcangeli said.
Junxiao Bi [Thu, 12 May 2016 22:42:18 +0000 (15:42 -0700)]
ocfs2: fix posix_acl_create deadlock
Commit 702e5bc68ad2 ("ocfs2: use generic posix ACL infrastructure")
refactored code to use posix_acl_create. The problem with this function
is that it is not mindful of the cluster wide inode lock making it
unsuitable for use with ocfs2 inode creation with ACLs. For example,
when used in ocfs2_mknod, this function can cause deadlock as follows.
The parent dir inode lock is taken when calling posix_acl_create ->
get_acl -> ocfs2_iop_get_acl which takes the inode lock again. This can
cause deadlock if there is a blocked remote lock request waiting for the
lock to be downconverted. And same deadlock happened in ocfs2_reflink.
This fix is to revert back using ocfs2_init_acl.
Junxiao Bi [Thu, 12 May 2016 22:42:15 +0000 (15:42 -0700)]
ocfs2: revert using ocfs2_acl_chmod to avoid inode cluster lock hang
Commit 743b5f1434f5 ("ocfs2: take inode lock in ocfs2_iop_set/get_acl()")
introduced this issue. ocfs2_setattr called by chmod command holds
cluster wide inode lock when calling posix_acl_chmod. This latter
function in turn calls ocfs2_iop_get_acl and ocfs2_iop_set_acl. These
two are also called directly from vfs layer for getfacl/setfacl commands
and therefore acquire the cluster wide inode lock. If a remote
conversion request comes after the first inode lock in ocfs2_setattr,
OCFS2_LOCK_BLOCKED will be set. And this will cause the second call to
inode lock from the ocfs2_iop_get_acl() to block indefinetly.
The deleted version of ocfs2_acl_chmod() calls __posix_acl_chmod() which
does not call back into the filesystem. Therefore, we restore
ocfs2_acl_chmod(), modify it slightly for locking as needed, and use that
instead.
Jon Paul Maloy [Wed, 11 May 2016 23:15:45 +0000 (19:15 -0400)]
tipc: eliminate risk of double link_up events
When an ACTIVATE or data packet is received in a link in state
ESTABLISHING, the link does not immediately change state to
ESTABLISHED, but does instead return a LINK_UP event to the caller,
which will execute the state change in a different lock context.
This non-atomic approach incurs a low risk that we may have two
LINK_UP events pending simultaneously for the same link, resulting
in the final part of the setup procedure being executed twice. The
only potential harm caused by this it that we may see two LINK_UP
events issued to subsribers of the topology server, something that
may cause confusion.
This commit eliminates this risk by checking if the link is already
up before proceeding with the second half of the setup.
Arnd Bergmann [Wed, 11 May 2016 20:13:23 +0000 (22:13 +0200)]
net: mvneta: bm: fix dependencies again
I tried to fix this before, but my previous fix was incomplete
and we can still get the same link error in randconfig builds
because of the way that Kconfig treats the
default y if MVNETA=y && MVNETA_BM_ENABLE
line that does not actually trigger when MVNETA_BM_ENABLE=m,
unlike I intended.
Changing the line to use MVNETA_BM_ENABLE!=n however has
the desired effect and hopefully makes all configurations
work as expected.
Signed-off-by: Arnd Bergmann <[email protected]> Fixes: 019ded3aa7c9 ("net: mvneta: bm: clarify dependencies") Acked-by: Gregory CLEMENT <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Linus Torvalds [Thu, 12 May 2016 19:55:42 +0000 (12:55 -0700)]
Merge tag 'sound-4.6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"This is a pretty boring pull request as you wish: including a few
small and trivial HD-audio and USB-audio quirks and a couple of small
regression fixes in HD-audio"
* tag 'sound-4.6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: usb-audio: Yet another Phoneix Audio device quirk
ALSA: hda - Fix regression on ATI HDMI audio
ALSA: hda - Fix subwoofer pin on ASUS N751 and N551
ALSA: hda - Fix broken reconfig
ALSA: hda - Fix white noise on Asus UX501VW headset
ALSA: usb-audio: Quirk for yet another Phoenix Audio devices (v2)
perf stat: Fallback to user only counters when perf_event_paranoid > 1
After 0161028b7c8a ("perf/core: Change the default paranoia level to 2")
'perf stat' fails for users without CAP_SYS_ADMIN, so just use
'perf_evsel__fallback()' to have the same behaviour as 'perf record',
i.e. set perf_event_attr.exclude_kernel to 1.
perf evsel: Handle EACCESS + perf_event_paranoid=2 in fallback()
Now with the default for the kernel.perf_event_paranoid sysctl being 2 [1]
we need to fall back to :u, i.e. to set perf_event_attr.exclude_kernel
to 1.
Before:
[acme@jouet linux]$ perf record usleep 1
Error:
You may not have permission to collect stats.
Consider tweaking /proc/sys/kernel/perf_event_paranoid,
which controls use of the performance events system by
unprivileged users (without CAP_SYS_ADMIN).
The current value is 2:
-1: Allow use of (almost) all events by all users
>= 0: Disallow raw tracepoint access by users without CAP_IOC_LOCK
>= 1: Disallow CPU event access by users without CAP_SYS_ADMIN
>= 2: Disallow kernel profiling by users without CAP_SYS_ADMIN
[acme@jouet linux]$
And if the user turns on verbose mode, an explanation will appear:
[acme@jouet linux]$ perf record -v usleep 1
Warning:
kernel.perf_event_paranoid=2, trying to fall back to excluding kernel samples
mmap size 528384B
[ perf record: Woken up 1 times to write data ]
Looking at the vmlinux_path (8 entries long)
Using /lib/modules/4.6.0-rc7+/build/vmlinux for symbols
[ perf record: Captured and wrote 0.016 MB perf.data (7 samples) ]
[acme@jouet linux]$
[1] 0161028b7c8a ("perf/core: Change the default paranoia level to 2")
perf evsel: Improve EPERM error handling in open_strerror()
We were showing a hardcoded default value for the kernel.perf_event_paranoid
sysctl, now that it became more paranoid (1 -> 2 [1]), this would need to be
updated, instead show the current value:
[acme@jouet linux]$ perf record ls
Error:
You may not have permission to collect stats.
Consider tweaking /proc/sys/kernel/perf_event_paranoid,
which controls use of the performance events system by
unprivileged users (without CAP_SYS_ADMIN).
The current value is 2:
-1: Allow use of (almost) all events by all users
>= 0: Disallow raw tracepoint access by users without CAP_IOC_LOCK
>= 1: Disallow CPU event access by users without CAP_SYS_ADMIN
>= 2: Disallow kernel profiling by users without CAP_SYS_ADMIN
[acme@jouet linux]$
[1] 0161028b7c8a ("perf/core: Change the default paranoia level to 2")
Linus Torvalds [Thu, 12 May 2016 18:23:08 +0000 (11:23 -0700)]
Merge tag 'pinctrl-v4.6-4' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl
Pull pinctrl fix from Linus Walleij:
"A single last pin control fix for v4.6. t's tagged for stable and
only hits a single driver with two added lines so should be safe.
Tested in linux-next.
- The pull up/down logic for the AT91 PIO4 controller was tilted: we
need to mask the reverse pull when unmasking a pull direction.
Setting both pull up & pull down is illegal and makes no sense"
* tag 'pinctrl-v4.6-4' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
pinctrl: at91-pio4: fix pull-up/down logic
Pablo Neira [Thu, 12 May 2016 15:16:31 +0000 (17:16 +0200)]
gtp: put back reference to netns when not required anymore
This patch fixes a netns leak.
Fixes: 93edb8c7f94f ("gtp: reload GTPv1 header after pskb_may_pull()") Reported-by: Cong Wang <[email protected]> Signed-off-by: Pablo Neira Ayuso <[email protected]> Signed-off-by: David S. Miller <[email protected]>
| If a down prepare callback fails, then DOWN_FAILED is invoked for all
| callbacks which have successfully executed DOWN_PREPARE.
|
| But, workqueue has actually two notifiers. One which handles
| UP/DOWN_FAILED/ONLINE and one which handles DOWN_PREPARE.
|
| Now look at the priorities of those callbacks:
|
| CPU_PRI_WORKQUEUE_UP = 5
| CPU_PRI_WORKQUEUE_DOWN = -5
|
| So the call order on DOWN_PREPARE is:
|
| CB 1
| CB ...
| CB workqueue_up() -> Ignores DOWN_PREPARE
| CB ...
| CB X ---> Fails
|
| So we call up to CB X with DOWN_FAILED
|
| CB 1
| CB ...
| CB workqueue_up() -> Handles DOWN_FAILED
| CB ...
| CB X-1
|
| So the problem is that the workqueue stuff handles DOWN_FAILED in the up
| callback, while it should do it in the down callback. Which is not a good idea
| either because it wants to be called early on rollback...
|
| Brilliant stuff, isn't it? The hotplug rework will solve this problem because
| the callbacks become symetric, but for the existing mess, we need some
| workaround in the workqueue code.
The boot CPU handles housekeeping duty(unbound timers, workqueues,
timekeeping, ...) on behalf of full dynticks CPUs. It must remain
online when nohz full is enabled. There is a priority set to every
notifier_blocks:
So tick_nohz_cpu_down callback failed when down prepare cpu 0, and
notifier_blocks behind tick_nohz_cpu_down will not be called any
more, which leads to workers are actually not unbound. Then hotplug
state machine will fallback to undo and online cpu 0 again. Workers
will be rebound unconditionally even if they are not unbound and
trigger the warning in this progress.
This patch fix it by catching !DISASSOCIATED to avoid rebind bound
workers.
David S. Miller [Thu, 12 May 2016 15:46:58 +0000 (11:46 -0400)]
Merge tag 'mac80211-next-for-davem-2016-05-12' of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next
Johannes Berg says:
====================
Some more work for 4.7, notably:
* completion and fixups of nla_put_64_64bit() work
* remove a/b/g/n from wext nickname to avoid confusion
with 11ac (which wouldn't even fit fully there due to
string length restrictions)
along with some other minor changes/cleanups.
====================
Arnd Bergmann [Thu, 12 May 2016 15:44:53 +0000 (17:44 +0200)]
Merge tag 'at91-fixes2' of git://git.kernel.org/pub/scm/linux/kernel/git/nferre/linux-at91 into fixes
Merge "Second AT91 fix PR for 4.6" from Nicolas Ferre:
- fix a regression on the clock subsystem while switching to syscon/regmap
due to a stricter check of the register map.
* tag 'at91-fixes2' of git://git.kernel.org/pub/scm/linux/kernel/git/nferre/linux-at91:
ARM: dts: at91: sam9x5: Fix the memory range assigned to the PMC
Felipe Balbi [Thu, 12 May 2016 09:34:38 +0000 (12:34 +0300)]
cgroup: fix compile warning
commit 4f41fc59620f ("cgroup, kernfs: make mountinfo
show properly scoped path for cgroup namespaces")
added the following compile warning:
kernel/cgroup.c: In function ‘cgroup_show_path’:
kernel/cgroup.c:1634:15: warning: unused variable ‘ret’ [-Wunused-variable]
int len = 0, ret = 0;
^
fix it.
Fixes: 4f41fc59620f ("cgroup, kernfs: make mountinfo show properly scoped path for cgroup namespaces") Signed-off-by: Felipe Balbi <[email protected]> Signed-off-by: Tejun Heo <[email protected]>
Steven Rostedt [Wed, 11 May 2016 19:09:36 +0000 (15:09 -0400)]
tools lib traceevent: Do not reassign parg after collapse_tree()
At the end of process_filter(), collapse_tree() was changed to update
the parg parameter, but the reassignment after the call wasn't removed.
What happens is that the "current_op" gets modified and freed and parg
is assigned to the new allocated argument. But after the call to
collapse_tree(), parg is assigned again to the just freed "current_op",
and this causes the tool to crash.
The current_op variable must also be assigned to NULL in case of error,
otherwise it will cause it to be free()ed twice.
perf dwarf: Guard !x86_64 definitions under #ifdef else clause
To fix the build on Fedora Rawhide (gcc 6.0.0 20160311 (Red Hat 6.0.0-0.17):
CC /tmp/build/perf/arch/x86/util/dwarf-regs.o
arch/x86/util/dwarf-regs.c:66:36: error: 'x86_32_regoffset_table' defined but not used [-Werror=unused-const-variable=]
static const struct pt_regs_offset x86_32_regoffset_table[] = {
^~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
perf tools: Use readdir() instead of deprecated readdir_r()
The readdir() function is thread safe as long as just one thread uses a
DIR, which is the case when parsing tracepoint event definitions, to
avoid breaking the build with glibc-2.23.90 (upcoming 2.24), use it
instead of readdir_r().
"However, in modern implementations (including the glibc implementation),
concurrent calls to readdir() that specify different directory streams
are thread-safe. In cases where multiple threads must read from the
same directory stream, using readdir() with external synchronization is
still preferable to the use of the deprecated readdir_r(3) function."
Noticed while building on a Fedora Rawhide docker container.
perf thread_map: Use readdir() instead of deprecated readdir_r()
The readdir() function is thread safe as long as just one thread uses a
DIR, which is the case in thread_map, so, to avoid breaking the build
with glibc-2.23.90 (upcoming 2.24), use it instead of readdir_r().
"However, in modern implementations (including the glibc implementation),
concurrent calls to readdir() that specify different directory streams
are thread-safe. In cases where multiple threads must read from the
same directory stream, using readdir() with external synchronization is
still preferable to the use of the deprecated readdir_r(3) function."
Noticed while building on a Fedora Rawhide docker container.
perf script: Use readdir() instead of deprecated readdir_r()
The readdir() function is thread safe as long as just one thread uses a
DIR, which is the case in 'perf script', so, to avoid breaking the build
with glibc-2.23.90 (upcoming 2.24), use it instead of readdir_r().
"However, in modern implementations (including the glibc implementation),
concurrent calls to readdir() that specify different directory streams
are thread-safe. In cases where multiple threads must read from the
same directory stream, using readdir() with external synchronization is
still preferable to the use of the deprecated readdir_r(3) function."
Noticed while building on a Fedora Rawhide docker container.
perf tools: Use readdir() instead of deprecated readdir_r()
The readdir() function is thread safe as long as just one thread uses a
DIR, which is the case when synthesizing events for pre-existing threads
by traversing /proc, so, to avoid breaking the build with glibc-2.23.90
(upcoming 2.24), use it instead of readdir_r().
"However, in modern implementations (including the glibc implementation),
concurrent calls to readdir() that specify different directory streams
are thread-safe. In cases where multiple threads must read from the
same directory stream, using readdir() with external synchronization is
still preferable to the use of the deprecated readdir_r(3) function."
Noticed while building on a Fedora Rawhide docker container.
CC /tmp/build/perf/util/event.o
util/event.c: In function '__event__synthesize_thread':
util/event.c:466:2: error: 'readdir_r' is deprecated [-Werror=deprecated-declarations]
while (!readdir_r(tasks, &dirent, &next) && next) {
^~~~~
In file included from /usr/include/features.h:368:0,
from /usr/include/stdint.h:25,
from /usr/lib/gcc/x86_64-redhat-linux/6.0.0/include/stdint.h:9,
from /git/linux/tools/include/linux/types.h:6,
from util/event.c:1:
/usr/include/dirent.h:189:12: note: declared here
perf/core: Disable the event on a truncated AUX record
When the PMU driver reports a truncated AUX record, it effectively means
that there is no more usable room in the event's AUX buffer (even though
there may still be some room, so that perf_aux_output_begin() doesn't take
action). At this point the consumer still has to be woken up and the event
has to be disabled, otherwise the event will just keep spinning between
perf_aux_output_begin() and perf_aux_output_end() until its context gets
unscheduled.
Again, for cpu-wide events this means never, so once in this condition,
they will be forever losing data.
Fix this by disabling the event and waking up the consumer in case of a
truncated AUX record.
perf/x86/intel/pt: Generate PMI in the STOP region as well
Currently, the PT driver always sets the PMI bit one region (page) before
the STOP region so that we can wake up the consumer before we run out of
room in the buffer and have to disable the event. However, we also need
an interrupt in the last output region, so that we actually get to disable
the event (if no more room from new data is available at that point),
otherwise hardware just quietly refuses to start, but the event is
scheduled in and we end up losing trace data till the event gets removed.
For a cpu-wide event it is even worse since there may not be any
re-scheduling at all and no chance for the ring buffer code to notice
that its buffer is filled up and the event needs to be disabled (so that
the consumer can re-enable it when it finishes reading the data out). In
other words, all the trace data will be lost after the buffer gets filled
up.
This patch makes PT also generate a PMI when the last output region is
full.
David Howells [Tue, 23 Feb 2016 11:03:12 +0000 (11:03 +0000)]
KEYS: Fix ASN.1 indefinite length object parsing
This fixes CVE-2016-0758.
In the ASN.1 decoder, when the length field of an ASN.1 value is extracted,
it isn't validated against the remaining amount of data before being added
to the cursor. With a sufficiently large size indicated, the check:
datalen - dp < 2
may then fail due to integer overflow.
Fix this by checking the length indicated against the amount of remaining
data in both places a definite length is determined.
Whilst we're at it, make the following changes:
(1) Check the maximum size of extended length does not exceed the capacity
of the variable it's being stored in (len) rather than the type that
variable is assumed to be (size_t).
(2) Compare the EOC tag to the symbolic constant ASN1_EOC rather than the
integer 0.
(3) To reduce confusion, move the initialisation of len outside of:
for (len = 0; n > 0; n--) {
since it doesn't have anything to do with the loop counter n.
Johannes Berg [Tue, 3 May 2016 13:58:00 +0000 (16:58 +0300)]
mac80211: allow software PS-Poll/U-APSD with AP_LINK_PS
When using RSS, frames might not be processed in the correct order,
and thus AP_LINK_PS must be used; most likely with firmware keeping
track of the powersave state, this is the case in iwlwifi now.
In this case, the driver can use ieee80211_sta_ps_transition() to
still have mac80211 manage powersave buffering. However, for U-APSD
and PS-Poll this isn't sufficient. If the device can't manage that
entirely on its own, mac80211's code should be used.
To allow this, export two functions: ieee80211_sta_uapsd_trigger()
and ieee80211_sta_pspoll().
Johannes Berg [Tue, 3 May 2016 13:52:04 +0000 (16:52 +0300)]
cfg80211: make wdev_list accessible to drivers
There's no harm in having drivers read the list, since they can
use RCU protection or RTNL locking; allow this to not require
each and every driver to also implement its own bookkeeping.
Sara Sharon [Tue, 3 May 2016 12:59:44 +0000 (15:59 +0300)]
mac80211: allow same PN for AMSDU sub-frames
Some hardware (iwlwifi an example) de-aggregate AMSDUs and copy the IV
as is to the generated MPDUs, so the same PN appears in multiple
packets without being a replay attack. Allow driver to explicitly
indicate that a frame is allowed to have the same PN as the previous
frame.
David Spinadel [Tue, 3 May 2016 13:05:02 +0000 (16:05 +0300)]
mac80211: remove disconnected APs from BSS table
In some cases, after a sudden AP disappearing and reconnection to
another AP in the same ESS, user space gets the old AP in scan
results (cached). User space may decide to roam to that old AP
which will cause a disconnection and longer recovery.
Remove APs that are probably out of range from BSS table.
hchrzani [Mon, 9 May 2016 07:36:59 +0000 (09:36 +0200)]
perf/x86/intel/uncore: Fix CHA registers configuration procedure for Knights Landing platform
CHA events in Knights Landing platform require programming filter registers properly.
Remote node, local node and NonNearMemCachable bits should be set to 1 at all times.
David S. Miller [Thu, 12 May 2016 04:04:08 +0000 (00:04 -0400)]
Merge branch 'qed-sriov'
Yuval Mintz says:
====================
qed*: Add SR-IOV support
This patch adds SR-IOV support to qed/qede drivers, adding a new PCI
device ID for a VF that is shared between all the various PFs that
support IOV.
This is quite a massive series - the first 7 parts of the series add
the infrastructure of supporting vfs in qed - mainly adding support in a
HW-based vf<->pf channel, as well as diverging all existing configuration
flows based on the pf/vf decision. I.e., while PF-originated requests
head directly to HW/FW, the VF requests first have to traverse to the PF
which will perform the configuration.
The 8th patch is the one that adds the support for the VF device in qede.
The remaining 6 patches each adds some user-based API support related to
VFs that can be used over the PF - forcing mac/vlan, changing speed, etc.
Dave,
Sorry in advance for the length of the series. Most of the bulk here is in
the infrastructure patches that have to go together [or at least, it makes
little sense to try splitting them up].
Please consider applying this to `net-next'.
Thanks,
Yuval
Changes from previous revision:
------------------------------
- V2 - Replace aligned_u64 with regular u64; This was possible as the
shared structures [between PF and VF] were already sufficiently
padded as-is in the API, making this redundant.
====================
Yuval Mintz [Wed, 11 May 2016 13:36:25 +0000 (16:36 +0300)]
qed*: Tx-switching configuration
Device should be configured by default to VEB once VFs are active.
This changes the configuration of both PFs' and VFs' vports into enabling
tx-switching once sriov is enabled.
Yuval Mintz [Wed, 11 May 2016 13:36:22 +0000 (16:36 +0300)]
qed*: IOV link control
This adds support in 2 ndo that allow PF to tweak the VF's view of the
link - `ndo_set_vf_link_state' to allow it a view independent of the PF's,
and `ndo_set_vf_rate' which would allow the PF to limit the VF speed.
Yuval Mintz [Wed, 11 May 2016 13:36:20 +0000 (16:36 +0300)]
qed*: Support PVID configuration
This adds support for PF control over the VF vlan configuration.
I.e., `ip link ... vf <x> vlan <vid>' should now be supported.
1. <vid> != 0 => VF receives [unknowingly] only traffic tagged by
<vid> and tags all outgoing traffic sent by VF with <vid>.
2. <vid> == 0 ==> Remove the pvid configuration, reverting to previous.
Yuval Mintz [Wed, 11 May 2016 13:36:19 +0000 (16:36 +0300)]
qede: Add VF support
Adding a PCI callback for `sriov_configure' and a new PCI device id for
the VF [+ Some minor changes to accomodate differences between PF and VF
at the qede].
Following this, VF creation should be possible and the entire subset of
existing PF functionality that's allow to VFs should be supported.
Yuval Mintz [Wed, 11 May 2016 13:36:18 +0000 (16:36 +0300)]
qed: Align TLVs
As the VF infrastructure is supposed to offer backward/forward
compatibility, the various types associated with VF<->PF communication
should be aligned across all various platforms that support IOV
on our family of adapters.
This adds a couple of currently missing values, specifically aligning
the enum for the various TLVs possible in the communication between them.
It then adds the PF implementation for some of those missing VF requests.
This support isn't really necessary for the Linux VF as those VFs aren't
requiring it [at least today], but are required by VFs running on other
OSes. LRO is an example of one such configuration.
Yuval Mintz [Wed, 11 May 2016 13:36:17 +0000 (16:36 +0300)]
qed: Bulletin and Link
Up to this point, VF and PF communication always originates from VF.
As a result, VF cannot be notified of any async changes, and specifically
cannot be informed of the current link state.
This introduces the bulletin board, the mechanism through which the PF
is going to communicate async notifications back to the VF. basically,
it's a well-defined structure agreed by both PF and VF which the VF would
continuously poll and into which the PF would DMA messages when needed.
[Bulletin board is actually allocated and communicated in previous patches
but never before used]
Based on the bulletin infrastructure, the VF can query its link status
and receive said async carrier changes.
Yuval Mintz [Wed, 11 May 2016 13:36:16 +0000 (16:36 +0300)]
qed: IOV l2 functionality
This adds sufficient changes to allow VFs l2-configuration flows to work.
While the fastpath of the VF and the PF are meant to be exactly the same,
the configuration of the VF is done by the PF.
This diverges all VF-related configuration flows that originate from a VF,
making them pass through the VF->PF channel and adding sufficient logic
on the PF side to support them.
Yuval Mintz [Wed, 11 May 2016 13:36:15 +0000 (16:36 +0300)]
qed: IOV configure and FLR
While previous patches have already added the necessary logic to probe
VFs as well as enabling them in the HW, this patch adds the ability to
support VF FLR & SRIOV disable.
It then wraps both flows together into the first IOV callback to be
provided to the protocol driver - `configure'. This would later to be used
to enable and disable SRIOV in the adapter.
Yuval Mintz [Wed, 11 May 2016 13:36:14 +0000 (16:36 +0300)]
qed: Introduce VFs
This adds the qed VFs for the first time -
The vfs are limited functions, with a very different PCI bar structure
[when compared with PFs] to better impose the related security demands
associated with them.
This patch includes the logic neccesary to allow VFs to successfully probe
[without actually adding the ability to enable iov].
This includes diverging all the flows that would occur as part of the pci
probe of the driver, preventing VF from accessing registers/memories it
can't and instead utilize the VF->PF channel to query the PF for needed
information.
Yuval Mintz [Wed, 11 May 2016 13:36:13 +0000 (16:36 +0300)]
qed: Add VF->PF channel infrastructure
Communication between VF and PF is based on a dedicated HW channel;
VF will prepare a messge, and by signaling the HW the PF would get a
notification of that message existance. The PF would then copy the
message, process it and DMA an answer back to the VF as a response.
The messages themselves are TLV-based - allowing easier backward/forward
compatibility.
This patch adds the infrastructure of the channel on the PF side -
starting with the arrival of the notification and ending with DMAing
the response back to the VF.
It also adds a dummy-response as reference, as it only lays the
groundwork of the communication; it doesn't really add support of any
actual messages.
Yuval Mintz [Wed, 11 May 2016 13:36:12 +0000 (16:36 +0300)]
qed: Add CONFIG_QED_SRIOV
Add support for a new Kconfig option for qed* driver which would allow
[eventually] the support in VFs.
This patch adds the necessary logic in the PF to learn about the possible
VFs it will have to support [Based on PCI configuration space and HW],
and prepare a database with an entry per-VF as infrastructure for future
interaction with said VFs.
Michael Chan [Tue, 10 May 2016 23:18:00 +0000 (19:18 -0400)]
bnxt_en: Add workaround to detect bad opaque in rx completion (part 2)
Add detection and recovery code when the hardware returned opaque value
does not match the expected consumer index. Once the issue is detected,
we skip the processing of all RX and LRO/GRO packets. These completion
entries are discarded without sending the SKB to the stack and without
producing new buffers. The function will be reset from a workqueue.
Michael Chan [Tue, 10 May 2016 23:17:59 +0000 (19:17 -0400)]
bnxt_en: Add workaround to detect bad opaque in rx completion (part 1)
There is a rare hardware bug that can cause a bad opaque value in the RX
or TPA completion. When this happens, the hardware may have used the
same buffer twice for 2 rx packets. In addition, the driver will also
crash later using the bad opaque as the index into the ring.
The rx opaque value is predictable and is always monotonically increasing.
The workaround is to keep track of the expected next opaque value and
compare it with the one returned by hardware during RX and TPA start
completions. If they miscompare, we will not process any more RX and
TPA completions and exit NAPI. We will then schedule a workqueue to
reset the function.
This patch adds the logic to keep track of the next rx consumer index.
Dan Carpenter [Tue, 10 May 2016 19:20:04 +0000 (22:20 +0300)]
qlcnic: potential NULL dereference in qlcnic_83xx_get_minidump_template()
If qlcnic_fw_cmd_get_minidump_temp() fails then "fw_dump->tmpl_hdr" is
NULL or possibly freed. It can lead to an oops later.
Fixes: d01a6d3c8ae1 ('qlcnic: Add support to enable capability to extend minidump for iSCSI') Signed-off-by: Dan Carpenter <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Dave Airlie [Thu, 12 May 2016 00:05:36 +0000 (10:05 +1000)]
Merge branch 'drm-fixes-4.6' of git://people.freedesktop.org/~agd5f/linux into drm-fixes
Two some radeon display fixes.
* 'drm-fixes-4.6' of git://people.freedesktop.org/~agd5f/linux:
drm/radeon: fix PLL sharing on DCE6.1 (v2)
drm/radeon: fix DP link training issue with second 4K monitor
Dave Airlie [Thu, 12 May 2016 00:05:06 +0000 (10:05 +1000)]
Merge tag 'drm-intel-fixes-2016-05-11' of git://anongit.freedesktop.org/drm-intel into drm-fixes
Misc intel fixes, reverting MST audio which was causing oops for now.
* tag 'drm-intel-fixes-2016-05-11' of git://anongit.freedesktop.org/drm-intel:
drm/i915: Bail out of pipe config compute loop on LPT
Revert "drm/i915: start adding dp mst audio"
drm/i915/bdw: Add missing delay during L3 SQC credit programming
drm/i915/lvds: separate border enable readout from panel fitter
drm/i915: Update CDCLK_FREQ register on BDW after changing cdclk frequency
David S. Miller [Thu, 12 May 2016 00:00:54 +0000 (20:00 -0400)]
Merge tag 'nfc-next-4.7-1' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/nfc-next
Samuel Ortiz says:
====================
NFC 4.7 pull request
This is the first NFC pull request for 4.7. With this one we
mainly have:
- Support for NXP's pn532 NFC chipset. The pn532 is based on the same
microcontroller as the pn533, but it talks to the host through i2c
instead of USB. By separating the pn533 driver into core and PHY
parts, we can not add the i2c layer and support the pn532 chipset.
- Support for NCI's loopback mode. This is a testing mode where each
packet received by the NFCC is sent back to the DH, allowing the
host to test that the controller can receive and send data.
- A few ACPI related fixes for the STMicro drivers, in order to match
the device tree naming scheme.
- A bunch of cleanups for the st-nci and the st21nfca STMicro drivers.
====================
CQE Compressing reduces PCI overhead by coalescing and compressing multiple CQEs into a
single merged CQE. Successful compressing improves message rate especially for small packet
traffic.
CQE Compressing in details:
Instead of writing full CQEs to memory, multiple almost identical CQEs are merged and compressed.
Information that is shared between the CQEs is written once, regardless of the number of
compressed CQEs. In addition, only the unique information (small amount of bytes compared to
full CQE size) is written per CQE.
CQE Compression Block:
This block contains multiple compressed CQEs. CQE Compression Block contains a single copy
of CQEs properties which are shared between all the compressed CQEs (called Title, see below)
and multiple mini CQEs (CQEs in compressed form).
Title:
The Title holds information which is shared between all the compressed CQEs in the CQE Compression
Block. In each Compression Block there is only a single Title regardless of the number
of compressed CQEs.
Mini CQE:
A CQE in compressed form that holds some data needed to extract a single full CQE, for example
8 Bytes instead of 64 Bytes.
The shared information between all compressed CQEs, which belong to the same CQE Compression
Block called Title, is written once, and only the unique information in each compressed
CQE, for example 8 bytes, is written per compressed CQE, called mini CQE.
Since CQE Compression can add overhead to the software (CPU),
it will be only enabled on "weak/slow" PCI slots, where it can actually help.
Saeed Mahameed [Tue, 10 May 2016 21:29:16 +0000 (00:29 +0300)]
net/mlx5e: Enable CQE compression when PCI is slower than link
We turn the feature ON, only for servers with PCI BW < MAX LINK BW, as it
helps reducing PCI pressure on weak PCI slots, but it adds some software
overhead.
Tariq Toukan [Tue, 10 May 2016 21:29:15 +0000 (00:29 +0300)]
net/mlx5e: Expand WQE stride when CQE compression is enabled
Make the MPWQE/Striding RQ default configuration dynamic and not
statically set at compile time. Now at driver load we set
stride size and num strides dynamically.
By default we use same values as before, but when CQE compression
is enabled, we set larger stride size to benefit from CQE
compression for larger packets.
Tariq Toukan [Tue, 10 May 2016 21:29:14 +0000 (00:29 +0300)]
net/mlx5e: CQE compression
CQE compression feature is meant to save PCIe bandwidth by
compressing few CQEs into smaller amount of bytes on PCIe.
CQE compression can be selectively enabled per CQ. By default
is disabled for now and will be enabled later on.
David S. Miller [Wed, 11 May 2016 23:36:29 +0000 (19:36 -0400)]
Merge branch 'more-dsa-probing'
Andrew Lunn says:
====================
More enabler patches for DSA probing
The complete set of patches for the reworked DSA probing is too big to
post as once. These subset contains some enablers which are easy to
review.
Eventually, the Marvell driver will instantiate its own internal MDIO
bus, rather than have the framework do it, thus allows devices on the
bus to be listed in the device tree. Initialize the main mutex as soon
as it is created, to avoid lifetime issues with the mdio bus.
A previous patch renamed all the DSA probe functions to make room for
a true device probe. However the recent merging of all the Marvell
switch drivers resulted in mv88e6xxx going back to the old probe
name. Rename it again, so we can have a driver probe function.
Add minimum support for the Marvell switch driver to probe as an MDIO
device, as well as an DSA driver. Later patches will then register
this device with the new DSA core framework.
Move the GPIO reset code out of the DSA code. Different drivers may
need different reset mechanisms, e.g. via a reset controller for
memory mapped devices. Don't clutter up the core with this. Let each
driver implement what it needs.
master_dev is no longer needed in the switch drivers, since they have
access to a device pointer from the probe function. Remove it.
Let the switch parse the eeprom length from its one device tree
node. This is required with the new binding when the central DSA
platform device no longer exists.
====================
Andrew Lunn [Tue, 10 May 2016 21:27:25 +0000 (23:27 +0200)]
dsa: mv88e6xxx: Handle eeprom-length property
A switch can export an attached EEPROM using the standard ethtool API.
However the switch itself cannot determine the size of the EEPROM, and
multiple sizes are allowed. Thus a device tree property is supported
to indicate the length of the EEPROM. Parse this property during
device probe, and implement a callback function to retrieve it.
Andrew Lunn [Tue, 10 May 2016 21:27:24 +0000 (23:27 +0200)]
dsa: Rename switch chip data to cd
The dsa_switch structure contains a dsa_chip_data member called pd.
However in the rest of the code, pd is used for dsa_platform_data.
This is confusing. Rename it cd, which is already often used in dsa.c
and slave.c for this data type.
Andrew Lunn [Tue, 10 May 2016 21:27:23 +0000 (23:27 +0200)]
dsa: Remove master_dev from switch structure
The switch drivers only use the master_dev member for dev_info()
messages. Now that the device is passed to the old style probe, and
new style drivers are probed as true linux drivers, this is no longer
needed.
Andrew Lunn [Tue, 10 May 2016 21:27:22 +0000 (23:27 +0200)]
dsa: Move gpio reset into switch driver
Resetting the switch is something the driver does, not the framework.
So move the parsing of this property into the driver.
There are no in kernel users of this property, so moving it does not
break anything. There is however a board which will make use of this
property making its way into the kernel.
Andrew Lunn [Tue, 10 May 2016 21:27:21 +0000 (23:27 +0200)]
dsa: Add mdio device support to Marvell switches
Allow Marvell switches to be mdio devices. Currently the driver just
allocate the private structure and detects what device is on the
bus. Later patches will make them register with the DSA framework.
Andrew Lunn [Tue, 10 May 2016 21:27:20 +0000 (23:27 +0200)]
dsa: mv88e6xxx: Rename probe function to fit the normal pattern
All other DSA drivers use _drv_ in there DSA probe function name, thus
allowing for a true linux driver probe function to use the
conventional name. Make mv88e6xxx fit this pattern.
David S. Miller [Wed, 11 May 2016 23:31:40 +0000 (19:31 -0400)]
Merge branch 'vrf-pktinfo'
David Ahern says:
====================
net: vrf: Fixup PKTINFO to return enslaved device index
Applications such as OSPF and BFD need the original ingress device not
the VRF device; the latter can be derived from the former. To that end
move the packet intercept from an rx handler that is invoked by
__netif_receive_skb_core to the ipv4 and ipv6 receive processing.
IPv6 already saves the skb_iif to the control buffer in ipv6_rcv. Since
the skb->dev has not been switched the cb has the enslaved device. Make
the same happen for IPv4 by adding the skb_iif to inet_skb_parm and set
it in ipv4 code after clearing the skb control buffer similar to IPv6.
From there the pktinfo can just pull it from cb with the PKTINFO_SKB_CB
cast.
====================
David Ahern [Tue, 10 May 2016 18:19:51 +0000 (11:19 -0700)]
net: original ingress device index in PKTINFO
Applications such as OSPF and BFD need the original ingress device not
the VRF device; the latter can be derived from the former. To that end
add the skb_iif to inet_skb_parm and set it in ipv4 code after clearing
the skb control buffer similar to IPv6. From there the pktinfo can just
pull it from cb with the PKTINFO_SKB_CB cast.
The previous patch moving the skb->dev change to L3 means nothing else
is needed for IPv6; it just works.
David Ahern [Tue, 10 May 2016 18:19:50 +0000 (11:19 -0700)]
net: l3mdev: Add hook in ip and ipv6
Currently the VRF driver uses the rx_handler to switch the skb device
to the VRF device. Switching the dev prior to the ip / ipv6 layer
means the VRF driver has to duplicate IP/IPv6 processing which adds
overhead and makes features such as retaining the ingress device index
more complicated than necessary.
This patch moves the hook to the L3 layer just after the first NF_HOOK
for PRE_ROUTING. This location makes exposing the original ingress device
trivial (next patch) and allows adding other NF_HOOKs to the VRF driver
in the future.
dev_queue_xmit_nit is exported so that the VRF driver can cycle the skb
with the switched device through the packet taps to maintain current
behavior (tcpdump can be used on either the vrf device or the enslaved
devices).
Linus Torvalds [Wed, 11 May 2016 20:17:12 +0000 (13:17 -0700)]
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"This is a couple of small fixes: one is a potential uninitialised
error variable in the alua code, potentially causing spurious failures
and the other is a problem caused by the conversion of SCSI to
hostwide tags which resulted in the qla1280 driver always failing in
host initialisation"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
qla1280: Don't allocate 512kb of host tags
scsi_dh_alua: uninitialized variable in alua_rtpg()
Pull networking fixes from David Miller:
"Hopefully the last round of fixes this release, fingers crossed :)
1) Initialize static nf_conntrack_locks_all_lock properly, from
Florian Westphal.
2) Need to cancel pending work when destroying IDLETIMER entries,
from Liping Zhang.
3) Fix TX param usage when sending TSO over iwlwifi devices, from
Emmanuel Grumbach.
4) NFACCT quota params not validated properly, from Phil Turnbull.
5) Resolve more glibc vs. kernel header conflicts, from Mikko
Tapeli.
6) Missing IRQ free in ravb_close(), from Geert Uytterhoeven.
7) Fix infoleak in x25, from Kangjie Lu.
8) Similarly in thunderx driver, from Heinrich Schuchardt.
9) tc_ife.h uapi header not exported properly, from Jamal Hadi Salim.
10) Don't reenable PHY interreupts if device is in polling mode, from
Shaohui Xie.
11) Packet scheduler actions late binding was not being handled
properly at all, from Jamal Hadi Salim.
12) Fix binding of conntrack entries to helpers in openvswitch, from
Joe Stringer"
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (21 commits)
gre: do not keep the GRE header around in collect medata mode
openvswitch: Fix cached ct with helper.
net sched: ife action fix late binding
net sched: skbedit action fix late binding
net sched: simple action fix late binding
net sched: mirred action fix late binding
net sched: ipt action fix late binding
net sched: vlan action fix late binding
net: phylib: fix interrupts re-enablement in phy_start
tcp: refresh skb timestamp at retransmit time
net: nps_enet: bug fix - handle lost tx interrupts
net: nps_enet: Tx handler synchronization
export tc ife uapi header
net: thunderx: avoid exposing kernel stack
net: fix a kernel infoleak in x25 module
ravb: Add missing free_irq() call to ravb_close()
uapi glibc compat: fix compile errors when glibc net/if.h included before linux/if.h
netfilter: nfnetlink_acct: validate NFACCT_QUOTA parameter
iwlwifi: mvm: don't override the rate with the AMSDU len
netfilter: IDLETIMER: fix race condition when destroy the target
...
Jiri Benc [Wed, 11 May 2016 13:53:57 +0000 (15:53 +0200)]
gre: do not keep the GRE header around in collect medata mode
For ipgre interface in collect metadata mode, it doesn't make sense for the
interface to be of ARPHRD_IPGRE type. The outer header of received packets
is not needed, as all the information from it is present in metadata_dst. We
already don't set ipgre_header_ops for collect metadata interfaces, which is
the only consumer of mac_header pointing to the outer IP header.
Just set the interface type to ARPHRD_NONE in collect metadata mode for
ipgre (not gretap, that still correctly stays ARPHRD_ETHER) and reset
mac_header.
Fixes: a64b04d86d14 ("gre: do not assign header_ops in collect metadata mode") Fixes: 2e15ea390e6f4 ("ip_gre: Add support to collect tunnel metadata.") Signed-off-by: Jiri Benc <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Joe Stringer [Wed, 11 May 2016 17:29:26 +0000 (10:29 -0700)]
openvswitch: Fix cached ct with helper.
When using conntrack helpers from OVS, a common configuration is to
perform a lookup without specifying a helper, then go through a
firewalling policy, only to decide to attach a helper afterwards.
In this case, the initial lookup will cause a ct entry to be attached to
the skb, then the later commit with helper should attach the helper and
confirm the connection. However, the helper attachment has been missing.
If the user has enabled automatic helper attachment, then this issue
will be masked as it will be applied in init_conntrack(). It is also
masked if the action is executed from ovs_packet_cmd_execute() as that
will construct a fresh skb.
This patch fixes the issue by making an explicit call to try to assign
the helper if there is a discrepancy between the action's helper and the
current skb->nfct.
Fixes: cae3a2627520 ("openvswitch: Allow attaching helpers to ct action") Signed-off-by: Joe Stringer <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Lawrence Brakmo [Wed, 11 May 2016 17:02:13 +0000 (10:02 -0700)]
tcp: replace cnt & rtt with struct in pkts_acked()
Replace 2 arguments (cnt and rtt) in the congestion control modules'
pkts_acked() function with a struct. This will allow adding more
information without having to modify existing congestion control
modules (tcp_nv in particular needs bytes in flight when packet
was sent).
As proposed by Neal Cardwell in his comments to the tcp_nv patch.
Mathias Krause [Tue, 10 May 2016 21:07:02 +0000 (23:07 +0200)]
x86/extable: ensure entries are swapped completely when sorting
The x86 exception table sorting was changed in commit 29934b0fb8ff
("x86/extable: use generic search and sort routines") to use the arch
independent code in lib/extable.c. However, the patch was mangled
somehow on its way into the kernel from the last version posted at [1].
The committed version kind of attempted to incorporate the changes of
commit 548acf19234d ("x86/mm: Expand the exception table logic to allow
new handling options") as in _completely_ _ignoring_ the x86 specific
'handler' member of struct exception_table_entry. This effectively
broke the sorting as entries will only partly be swapped now.
Fortunately, the x86 Kconfig selects BUILDTIME_EXTABLE_SORT, so the
exception table doesn't need to be sorted at runtime. However, in case
that ever changes, we better not break the exception table sorting just
because of that.
[ Ard Biesheuvel points out that BUILDTIME_EXTABLE_SORT applies to the
core image only, but we still rely on the sorting routines for modules
in that case - Linus ]
Fix this by providing a swap_ex_entry_fixup() macro that takes care of
the 'handler' member.
Linus Torvalds [Wed, 11 May 2016 17:21:16 +0000 (10:21 -0700)]
Merge tag 'spi-fix-v4.6-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi
Pull spi fixes from Mark Brown:
"A bunch of small driver specific fixes that have come up, none of them
remarkable in themselves. One fixes a regression introduced in the
merge window and another two are targetted at stable"
* tag 'spi-fix-v4.6-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: pxa2xx: Do not detect number of enabled chip selects on Intel SPT
spi: spi-ti-qspi: Handle truncated frames properly
spi: spi-ti-qspi: Fix FLEN and WLEN settings if bits_per_word is overridden
spi: omap2-mcspi: Undo broken fix for dma transfer of vmalloced buffer
spi: spi-fsl-dspi: Fix cs_change handling in message transfer
Linus Torvalds [Wed, 11 May 2016 17:11:44 +0000 (10:11 -0700)]
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM fixes from Paolo Bonzini:
"Two small x86 patches, improving "make kvmconfig" and fixing an
objtool warning for CONFIG_PROFILE_ALL_BRANCHES"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
kvmconfig: add more virtio drivers
x86/kvm: Add stack frame dependency to fastop() inline asm
Namhyung Kim [Tue, 10 May 2016 14:26:24 +0000 (11:26 -0300)]
perf diff: Fix duplicated output column
The commit b97511c5bc94 ("perf tools: Add overhead/overhead_children
keys defaults via string") moved initialization of column headers but it
missed to check the sort__mode. As 'perf diff' doesn't call
perf_hpp__init(), the setup_overhead() also should not be called.
Boris Brezillon [Wed, 11 May 2016 09:00:02 +0000 (11:00 +0200)]
ARM: dts: at91: sam9x5: Fix the memory range assigned to the PMC
The memory range assigned to the PMC (Power Management Controller) was
not including the PMC_PCR register which are used to control peripheral
clocks.
This was working fine thanks to the page granularity of ioremap(), but
started to fail when we switched to syscon/regmap, because regmap is
making sure that all accesses are falling into the reserved range.
Takashi Iwai [Wed, 11 May 2016 12:56:12 +0000 (14:56 +0200)]
ALSA: hda - Fix regression on ATI HDMI audio
The HDMI/DP audio output on ATI/AMD chips got broken due to the recent
restructuring of chmap. Fortunately, Daniel Exner could bisect, and
pointed the culprit commit [739ffee97ed5: ALSA: hda - Add hdmi chmap
verb programming ops to chmap object].
This commit moved some ops from hdmi_ops to chmap_ops, and reassigned
the ops in the embedded chmap object in hdmi_spec instead.
Unfortunately, the reassignment of these ops in patch_atihdmi() were
moved into an if block that is performed only for old chips. Thus, on
newer chips, the generic ops is still used, which doesn't work for
such ATI/AMD chips.
This patch addresses the regression, simply by moving the assignment
of chmap ops to the right place.
projects / linux.git / log