ath10k: fix rfc1042 header retrieval in QCA4019 with eth decap mode
Chipset from QCA99X0 onwards (QCA99X0, QCA9984, QCA4019 & future)
rx_hdr_status is not padded to align in 4-byte boundary. Define a
new hw_params field to handle different alignment behaviour between
different hw. This patch fixes improper retrieval of rfc1042 header
with QCA4019. This patch along with "ath10k: Properly remove padding
from the start of rx payload" will fix traffic failure in ethernet
decap mode for QCA4019.
Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qti.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Larry Finger [Wed, 21 Sep 2016 01:19:29 +0000 (21:19 -0400)]
rtl8xxxu: Stop log spam from each successful interrupt
As soon as debugging is turned on, the logs are filled with messages
reporting the interrupt status. As this quantity is usually zero, this
output is not needed. In fact, there will be a report if the status is
not zero, thus the debug line in question could probably be deleted.
Rather than taking that action, I have changed it to only be printed
when the newly added RTL8XXXU_DEBUG_INTERRUPT bit is set in the debug
mask.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Remove last_tx_page argument from the llt_init() function. The
rtl8xxxu_fileops structure contains the correct TX_TOTAL_PAGE_NUM
value for the device, and rtl8xxxu_auto_llt_table() doesn't need to
know the value in the first place.
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Cathy Luo [Tue, 20 Sep 2016 15:19:03 +0000 (20:49 +0530)]
mwifiex: fix race condition causing tx timeout
It's been observed that in a corner case mwifiex_usb_tx_complete()
gets called before we exit from mwifiex_usb_host_to_card() after
submitting the urb. 'data_sent' flag remains set in this case. It
blocks further Tx packets and triggers watchdog timeout.
The problem is fixed by setting data_sent and port_block flag at
correct place.
Signed-off-by: Cathy Luo <cluo@marvell.com> Signed-off-by: Shengzhen Li <szli@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
In mwifiex_usb_tx_complete(), we are updating port->block_status first
and then freeing the skb attached to that URB. We may end up attaching
new skb to URB in a corner case and same will be freed. This results in
the kernel crash. The problem is solved by changing the sequence.
Signed-off-by: Cathy Luo <cluo@marvell.com> Signed-off-by: Shengzhen Li <szli@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
The 802.11 standard only has four valid auth type configurations of which
our firmware only supports two, ie. Open System and Shared Key. Simplify
the mapping falling back to automatic for other types specified by
user-space.
Reviewed-by: Arend Van Spriel <arend.vanspriel@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Signed-off-by: Hante Meuleman <hante.meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
brcmfmac: fix out of bound access on clearing wowl wake indicator
Clearing the wowl wakeindicator happens with a rather odd
construction where the string "clear" is used to set the iovar
wowl_wakeind. This was implemented incorrectly as it caused an
out of bound access. Use an intermediate variable of correct
length and copy string in that. Problem was found using coverity.
Reviewed-by: Arend Van Spriel <arend.vanspriel@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Signed-off-by: Hante Meuleman <hante.meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
When IPv6 address is to be cleared there is a possible out of
bound access. But also the clearing of the last entry and the
adjustment of total number of stored IPv6 addresses is not
updated. This patch fixes that bug. Bug was found using coverity.
Reviewed-by: Arend Van Spriel <arend.vanspriel@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Signed-off-by: Hante Meuleman <hante.meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
in the function brcmf_bus_start() in the exception handling a
check is made to dermine whether ifp is null, though this is not
possible. Removing the unnessary check.
Reviewed-by: Arend Van Spriel <arend.vanspriel@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Signed-off-by: Hante Meuleman <hante.meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
brcmfmac: remove worker from .ndo_set_mac_address() callback
As it turns out there is no need to use a worker for the callback
because it is not called from atomic context.
Reported-by: Dan Williams <dcbw@redhat.com> Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
brcmfmac: initialize variable in brcmf_sdiod_regrl()
In case of an error the variable returned is uninitialized. The caller
will probably check the error code before using it, but better assure
it is set to zero.
Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
brcmfmac: rework pointer trickery in brcmf_proto_bcdc_query_dcmd()
The variable info is assigned to point to bcdc->msg[1], which is the
same as pointing to bcdc->buf. As that is what we want to access
make it clear by fixing the assignment. This also avoid out-of-bounds
errors from static analyzers are bcdc->msg[1] is not in the structure
definition.
Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
802.11d is not always supported by firmware anymore. Currently the
AP configuration of 11d will cause an abort if the ioctl set is
failing. This behavior is not correct and the error should be
ignored.
Reviewed-by: Arend Van Spriel <arend.vanspriel@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Signed-off-by: Hante Meuleman <hante.meuleman@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Previously device used to start using IGTK key as Tx key as soon as it
gets downloaded in add_key(). This patch implements set_default_mgmt_key
handler. We will update Tx key ID in set_default_mgmt_key().
Signed-off-by: Ganapathi Bhat <gbhat@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Tony Lindgren [Sat, 17 Sep 2016 16:06:32 +0000 (09:06 -0700)]
wlcore: Fix config firmware loading issues
Booting multiple wl12xx and wl18xx devices using the same rootfs is
a pain. You currently have to symlink the right nvs file depending
on the wl12xx type.
For example, with wl1271-nvs.bin being a symlink to wl127x-nvs.bin
by default and trying to bring up a wl128x based device:
wlcore: ERROR nvs size is not as expected: 1113 != 912
wlcore: ERROR NVS file is needed during boot
wlcore: ERROR NVS file is needed during boot
wlcore: ERROR firmware boot failed despite 3 retries
Note that wl18xx uses a separate config firmware wl18xx-conf.bin
that can be generated with tools using the following two git repos:
So let's not configure the nvs file for wl18xx as it's not needed
AFAIK. If it turns out that we also need the nvs file for wl18xx,
we can just add it to the config firmware data for wl18xx.
Let's fix the issue by using the chip specific config firmware
data, and make sure we produce understandable warnings if something
is missing.
Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Kalle Valo [Mon, 26 Sep 2016 11:10:31 +0000 (14:10 +0300)]
Merge tag 'iwlwifi-next-for-kalle-2016-09-19-2' of git://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next
* added support for MU-MIMO sniffer
* added support for RRM by scan
* added support for packet injection
* migrate to devm memory allocation handling
* some fixes, mostly in DQA and new HW support
* other generic cleanups
Sara Sharon [Wed, 31 Aug 2016 16:03:01 +0000 (19:03 +0300)]
iwlwifi: mvm: set HCMD_NAME for PHY_DB as well
Currently it is logged as UNKNOWN.
Also, 0x6c seems to be the permanent ID for this command, remove
incorrect comment and uncomment the command from the commands
list.
Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Ido Yariv [Tue, 23 Aug 2016 18:44:59 +0000 (14:44 -0400)]
iwlwifi: mvm: Add mem debugfs entry
In order to access cached/paged memory, there are a couple of firmware
commands (one for UMAC and one for LMAC) that let the host access memory
and registers indirectly. Since this is done by the firmware on behalf
of the host, even if memory is paged out or cached, the host will
retrieve the memory as the firmware sees it (paged out memory will get
paged in).
Export this mechanism via a debugfs entry for both read and write
access.
WARNING: This mechanism has no protections at all. Invalid addresses may
crash or hang the firmware. Writing to arbitrary memory also comes with
no guarantees.
Johannes Berg [Wed, 24 Aug 2016 08:05:40 +0000 (10:05 +0200)]
iwlwifi: mvm: move AP-specific code to right function
There's no need for the common MAC context function to have an
if on AP mode, the values can be overridden in the AP-specific
function later. Clean that up by adding the full command as a
new parameter to the AP-specific function, and doing it there.
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
When we get an unexpected Block Ack Request frame, the BAID from
the hardware will be invalid, and we'll pass it to mac80211 for
further handling (sending a delBA action frame.)
Add a comment explaining that, in case anyone looks in the future.
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Implement support for RRM by adding an option to configure the scan
dwell time and reporting scan start time and BSS detection time, and
Advertise support for these features.
Sara Sharon [Mon, 8 Aug 2016 10:07:01 +0000 (13:07 +0300)]
iwlwifi: mvm: support BAR in reorder buffer
On default queue we will not receive frame release notification,
but the BAR itself.
Upon receiving the BAR driver should look at the NSSN and adjust
window accordingly.
Fixes: b915c10174fb ("iwlwifi: mvm: add reorder buffer per queue") Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Sara Sharon [Mon, 8 Aug 2016 08:51:24 +0000 (11:51 +0300)]
iwlwifi: mvm: fix DQA AP mode station assumption
There was recently a Full AP mode feature added where hostap adds
the station at auth stage, and not assoc.
However, when running with legacy hostapd, we get the auth response
before station was added, and tx_skb_non_sta fails to allocate
a queue, resulting in a complete failure of association.
Take care of this situation as well.
Add a warning when no valid queue is returned at all and make mvm
drop the packet instead of passing it on.
Refactor the function a bit while at it.
Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Haim Dreyfuss [Tue, 2 Aug 2016 12:28:23 +0000 (15:28 +0300)]
iwlwifi: check for valid ethernet address provided by OEM
In 9000 family products we added an option to let the OEM fuse the
mac address via registers. If these registers are zeroed we use the OTP
address instead. Make sure that the address provided by the OEM is valid
and, if not, fall back to the OTP address as well.
Fixes: commit 17c867bfe89b ("iwlwifi: add support for getting HW address from CSR") Signed-off-by: Haim Dreyfuss <haim.dreyfuss@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Johannes Berg [Thu, 4 Aug 2016 06:57:59 +0000 (08:57 +0200)]
iwlwifi: mvm: compare full command ID
When comparing command IDs, the group should be taken
into account so the same command/notification from a
different group doesn't trigger anything unexpected.
Fix this by comparing to the wide ID.
Colin Ian King [Fri, 16 Sep 2016 09:37:31 +0000 (10:37 +0100)]
mwifiex: fix null pointer deference when adapter is null
If adapter is null the error exit path in mwifiex_shutdown_sw is
to down the semaphore sem and print some debug via mwifiex_dbg.
However, passing a NULL adapter to mwifiex_dbg causes a null
pointer deference when accessing adapter->dev. This fix checks
for a null adapter at the start of the function and to exit
without the need to up the semaphore and we also skip the debug
to avoid the null pointer dereference.
Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Bob Copeland [Wed, 14 Sep 2016 12:42:36 +0000 (08:42 -0400)]
mwifiex: fix error handling in mwifiex_create_custom_regdomain
smatch reports:
sta_cmdresp.c:1053 mwifiex_create_custom_regdomain() warn: possible memory leak of 'regd'
Indeed, mwifiex_create_custom_regdomain() returns NULL in the
case that channel is missing in the TLV without freeing regd.
Moreover, some other error paths in this function return ERR_PTR
values which are assigned without checking to the regd field in
the mwifiex_adapter struct. The latter is only null-checked where
used.
Fix by freeing regd in the error path, and only update
priv->adapter->regd if the returned pointer is valid.
Cc: Amitkumar Karwar <akarwar@marvell.com> Cc: Nishant Sarmukadam <nishants@marvell.com> Signed-off-by: Bob Copeland <me@bobcopeland.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Kalle Valo [Sat, 17 Sep 2016 15:10:33 +0000 (18:10 +0300)]
Merge tag 'iwlwifi-next-for-kalle-2016-09-15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next
* work for new hardware support continues
* dynamic queue allocation stabilization
* improvements in the MSIx code
* multiqueue support work continues
* new firmware version support
* general cleanups and improvements
samples/bpf: add comprehensive ipip, ipip6, ip6ip6 test
the test creates 3 namespaces with veth connected via bridge.
First two namespaces simulate two different hosts with the same
IPv4 and IPv6 addresses configured on the tunnel interface and they
communicate with outside world via standard tunnels.
Third namespace creates collect_md tunnel that is driven by BPF
program which selects different remote host (either first or
second namespace) based on tcp dest port number while tcp dst
ip is the same.
This scenario is rough approximation of load balancer use case.
The tests check both traditional tunnel configuration and collect_md mode.
Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Similar to gre, vxlan, geneve tunnels allow IPIP6 and IP6IP6 tunnels
to operate in 'collect metadata' mode.
Unlike ipv4 code here it's possible to reuse ip6_tnl_xmit() function
for both collect_md and traditional tunnels.
bpf_skb_[gs]et_tunnel_key() helpers and ovs (in the future) are the users.
Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Thomas Graf <tgraf@suug.ch> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Similar to gre, vxlan, geneve tunnels allow IPIP tunnels to
operate in 'collect metadata' mode.
bpf_skb_[gs]et_tunnel_key() helpers can make use of it right away.
ovs can use it as well in the future (once appropriate ovs-vport
abstractions and user apis are added).
Note that just like in other tunnels we cannot cache the dst,
since tunnel_info metadata can be different for every packet.
Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Thomas Graf <tgraf@suug.ch> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Julia Lawall [Thu, 15 Sep 2016 20:23:26 +0000 (22:23 +0200)]
l2tp: constify net_device_ops structures
Check for net_device_ops structures that are only stored in the netdev_ops
field of a net_device structure. This field is declared const, so
net_device_ops structures that have this property can be declared as const
also.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
Julia Lawall [Thu, 15 Sep 2016 20:23:25 +0000 (22:23 +0200)]
dwc_eth_qos: constify net_device_ops structures
Check for net_device_ops structures that are only stored in the netdev_ops
field of a net_device structure. This field is declared const, so
net_device_ops structures that have this property can be declared as const
also.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
@bad@
position p != {r.p,ok.p};
identifier r.i;
struct net_device_ops e;
@@
e@i@p
@depends on !bad disable optional_qualifier@
identifier r.i;
@@
static
+const
struct net_device_ops i = { ... };
// </smpl>
The result of size on this file before the change is:
text data bss dec hex filename
21623 1316 40 22979 59c3
drivers/net/ethernet/synopsys/dwc_eth_qos.o
and after the change it is:
text data bss dec hex filename
22199 724 40 22963 59b3
drivers/net/ethernet/synopsys/dwc_eth_qos.o
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
Julia Lawall [Thu, 15 Sep 2016 20:23:24 +0000 (22:23 +0200)]
hisilicon: constify net_device_ops structures
Check for net_device_ops structures that are only stored in the netdev_ops
field of a net_device structure. This field is declared const, so
net_device_ops structures that have this property can be declared as const
also.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
Eric Dumazet [Thu, 15 Sep 2016 16:33:02 +0000 (09:33 -0700)]
tcp: prepare skbs for better sack shifting
With large BDP TCP flows and lossy networks, it is very important
to keep a low number of skbs in the write queue.
RACK and SACK processing can perform a linear scan of it.
We should avoid putting any payload in skb->head, so that SACK
shifting can be done if needed.
With this patch, we allow to pack ~0.5 MB per skb instead of
the 64KB initially cooked at tcp_sendmsg() time.
This gives a reduction of number of skbs in write queue by eight.
tcp_rack_detect_loss() likes this.
We still allow payload in skb->head for first skb put in the queue,
to not impact RPC workloads.
Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Yuchung Cheng <ycheng@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sat, 17 Sep 2016 13:53:29 +0000 (09:53 -0400)]
Merge tag 'wireless-drivers-next-for-davem-2016-09-15' of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers-next
Kalle Valo says:
====================
wireless-drivers-next patches for 4.9
Major changes:
iwlwifi
* preparation for new a000 HW continues
* some DQA improvements
* add support for GMAC
* add support for 9460, 9270 and 9170 series
mwifiex
* support random MAC address for scanning
* add HT aggregation support for adhoc mode
* add custom regulatory domain support
* add manufacturing mode support via nl80211 testmode interface
bcma
* support BCM53573 series of wireless SoCs
bitfield.h
* add FIELD_PREP() and FIELD_GET() macros
mt7601u
* convert to use the new bitfield.h macros
brcmfmac
* add support for bcm4339 chip with modalias sdio:c00v02D0d4339
ath10k
* add nl80211 testmode support for 10.4 firmware
* hide kernel addresses from logs using %pK format specifier
* implement NAPI support
* enable peer stats by default
ath9k
* use ieee80211_tx_status_noskb where possible
wil6210
* extract firmware capabilities from the firmware file
ath6kl
* enable firmware crash dumps on the AR6004
ath-current is also merged to fix a conflict in ath10k.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sat, 17 Sep 2016 13:51:48 +0000 (09:51 -0400)]
Merge branch 'mlx5e-order-0'
Tariq Toukan says:
====================
mlx5e Order-0 pages for Striding RQ
In this series, we refactor our Striding RQ receive-flow to always use
fragmented WQEs (Work Queue Elements) using order-0 pages, omitting the
flow that allocates and splits high-order pages which would fragment
and deplete high-order pages in the system.
The first patch gives a slight degradation, but opens the opportunity
to using a simple page-cache mechanism of a fair size.
The page-cache, implemented in patch 3, not only closes the performance
gap but even gives a gain.
In patch 2 we re-organize the code to better manage the calls for
alloc/de-alloc pages in the RX flow.
Series generated against net-next commit: bed806cb266e "Merge branch 'mlxsw-ethtool'"
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
net/mlx5e: Implement RX mapped page cache for page recycle
Instead of reallocating and mapping pages for RX data-path,
recycle already used pages in a per ring cache.
Performance tests:
The following results were measured on a freshly booted system,
giving optimal baseline performance, as high-order pages are yet to
be fragmented and depleted.
We ran pktgen single-stream benchmarks, with iptables-raw-drop:
Single stride, 64 bytes:
* 4,739,057 - baseline
* 4,749,550 - order0 no cache
* 4,786,899 - order0 with cache
1% gain
Larger packets, no page cross, 1024 bytes:
* 3,982,361 - baseline
* 3,845,682 - order0 no cache
* 4,127,852 - order0 with cache
3.7% gain
Larger packets, every 3rd packet crosses a page, 1500 bytes:
* 3,731,189 - baseline
* 3,579,414 - order0 no cache
* 3,931,708 - order0 with cache
5.4% gain
Signed-off-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Manage the allocation and deallocation of mapped RX pages only
through dedicated API functions.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net/mlx5e: Single flow order-0 pages for Striding RQ
To improve the memory consumption scheme, we omit the flow that
demands and splits high-order pages in Striding RQ, and stay
with a single Striding RQ flow that uses order-0 pages.
Moving to fragmented memory allows the use of larger MPWQEs,
which reduces the number of UMR posts and filler CQEs.
Moving to a single flow allows several optimizations that improve
performance, especially in production servers where we would
anyway fallback to order-0 allocations:
- inline functions that were called via function pointers.
- improve the UMR post process.
This patch alone is expected to give a slight performance reduction.
However, the new memory scheme gives the possibility to use a page-cache
of a fair size, that doesn't inflate the memory footprint, which will
dramatically fix the reduction and even give a performance gain.
Performance tests:
The following results were measured on a freshly booted system,
giving optimal baseline performance, as high-order pages are yet to
be fragmented and depleted.
We ran pktgen single-stream benchmarks, with iptables-raw-drop:
Single stride, 64 bytes:
* 4,739,057 - baseline
* 4,749,550 - this patch
no reduction
Larger packets, no page cross, 1024 bytes:
* 3,982,361 - baseline
* 3,845,682 - this patch
3.5% reduction
Larger packets, every 3rd packet crosses a page, 1500 bytes:
* 3,731,189 - baseline
* 3,579,414 - this patch
4% reduction
Fixes: 461017cb006a ("net/mlx5e: Support RX multi-packet WQE (Striding RQ)") Fixes: bc77b240b3c5 ("net/mlx5e: Add fragmented memory support for RX multi packet WQE") Signed-off-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Howells [Sat, 17 Sep 2016 06:26:01 +0000 (07:26 +0100)]
rxrpc: Make IPv6 support conditional on CONFIG_IPV6
Add CONFIG_AF_RXRPC_IPV6 and make the IPv6 support code conditional on it.
This is then made conditional on CONFIG_IPV6.
Without this, the following can be seen:
net/built-in.o: In function `rxrpc_init_peer':
>> peer_object.c:(.text+0x18c3c8): undefined reference to `ip6_route_output_flags'
Reported-by: kbuild test robot <fengguang.wu@intel.com> Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 16 Sep 2016 08:31:56 +0000 (04:31 -0400)]
Merge branch 'QCA8K'
John Crispin says:
====================
net-next: dsa: add QCA8K support
This series is based on the AR8xxx series posted by Matthieu Olivari in may
2015. The following changes were made since then
* fixed the nitpicks from the previous review
* updated to latest API
* turned it into an mdio device
* added callbacks for fdb, bridge offloading, stp, eee, port status
* fixed several minor issues to the port setup and arp learning
* changed the namespacing as this driver to qca8k
The driver has so far only been tested on qca8337/N. It should work on other QCA
switches such as the qca8327 with minor changes.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
John Crispin [Thu, 15 Sep 2016 14:26:41 +0000 (16:26 +0200)]
net-next: dsa: add new driver for qca8xxx family
This patch contains initial support for the QCA8337 switch. It
will detect a QCA8337 switch, if present and declared in the DT.
Each port will be represented through a standalone net_device interface,
as for other DSA switches. CPU can communicate with any of the ports by
setting an IP@ on ethN interface. Most of the extra callbacks of the DSA
subsystem are already supported, such as bridge offloading, stp, fdb.
Signed-off-by: John Crispin <john@phrozen.org> Signed-off-by: David S. Miller <davem@davemloft.net>
John Crispin [Thu, 15 Sep 2016 14:26:40 +0000 (16:26 +0200)]
net-next: dsa: add Qualcomm tag RX/TX handler
Add support for the 2-bytes Qualcomm tag that gigabit switches such as
the QCA8337/N might insert when receiving packets, or that we need
to insert while targeting specific switch ports. The tag is inserted
directly behind the ethernet header.
Reviewed-by: Andrew Lunn <andrew@lunn.ch> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: John Crispin <john@phrozen.org> Signed-off-by: David S. Miller <davem@davemloft.net>
John Crispin [Thu, 15 Sep 2016 14:26:39 +0000 (16:26 +0200)]
Documentation: devicetree: add qca8k binding
Add device-tree binding for ar8xxx switch families.
Cc: devicetree@vger.kernel.org Signed-off-by: John Crispin <john@phrozen.org> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
The driver core clears the driver data to NULL after device_release
or on probe failure. Thus, it is not needed to manually clear the
device driver data to NULL.
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: dsa: b53: Remove unused including <linux/version.h>
Remove including <linux/version.h> that don't need it.
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Acked-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
drivers/net/dsa/bcm_sf2.c:963:19: warning:
symbol 'bcm_sf2_io_ops' was not declared. Should it be static?
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Acked-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes: 9f5afeae5152 ("tcp: use an RB tree for ooo receive queue") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Yuchung Cheng <ycheng@google.com> Cc: Yaogong Wang <wygivan@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 16 Sep 2016 06:23:06 +0000 (02:23 -0400)]
Merge branch 'mediatek-reset-flow'
Sean Wang says:
====================
mediatek: add enhancement into the existing reset flow
Current driver only resets DMA used by descriptor rings which
can't guarantee it can recover all various kinds of fatal
errors, so the patch
1) tries to reset the underlying hardware resource from scratch on
Mediatek SoC required for ethernet running.
2) refactors code in order to the reusability of existing code.
3) considers handling for race condition between the reset flow and
callbacks registered into core driver called about hardware accessing.
4) introduces power domain usage to hardware setup which leads to have
cleanly and completely restore to the state as the initial.
Changes since v1:
- fix the build error with module built causing undefined symbol for
pinctrl_bind_pins, so using pinctrl_select_state instead accomplishes
the pin mux setup during the reset process.
====================
Reviewed-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sean Wang [Wed, 14 Sep 2016 15:13:20 +0000 (23:13 +0800)]
net: ethernet: mediatek: add more resets for internal ethernet circuit block
struct mtk_eth has already contained struct regmap ethsys pointer
to the address range of the internal circuit reset, so we reuse it
to reset more internal blocks on ethernet hardware such as packet
processing engine (PPE) and frame engine (FE) instead of rstc which
deals with FE only.
Signed-off-by: Sean Wang <sean.wang@mediatek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sean Wang [Wed, 14 Sep 2016 15:13:19 +0000 (23:13 +0800)]
net: ethernet: mediatek: add the whole ethernet reset into the reset process
1) original driver only resets DMA used by descriptor rings
which can't guarantee it can recover all various kinds of fatal
errors, so the patch tries to reset the underlying hardware
resource from scratch on Mediatek SoC required for ethernet
running, including power, pin mux control, clock and internal
circuits on the ethernet in order to restore into the initial
state which the rebooted machine gives.
2) add state variable inside structure mtk_eth to help distinguish
mtk_hw_init is called between the initialization during boot time
or re-initialization during the reset process.
3) add ge_mode variable inside structure mtk_mac for restoring
the interface mode of the current setup for the target MAC.
4) remove __init attribute from mtk_hw_init definition
Signed-off-by: Sean Wang <sean.wang@mediatek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sean Wang [Wed, 14 Sep 2016 15:13:18 +0000 (23:13 +0800)]
net: ethernet: mediatek: add controlling power domain the ethernet belongs to
introduce power domain control which the digital circuit of
the ethernet belongs to inside the flow of hardware initialization
and deinitialization which helps the entire ethernet hardware block
could restart cleanly and completely as being back to the initial
state when the whole machine reboot.
Signed-off-by: Sean Wang <sean.wang@mediatek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This cleans up the error path inside mtk_hw_init call, causing it able
to exit appropriately when something fails and also includes refactoring
mtk_cleanup call to make the partial logic reusable on the error path.
Signed-off-by: Sean Wang <sean.wang@mediatek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sean Wang [Wed, 14 Sep 2016 15:13:15 +0000 (23:13 +0800)]
net: ethernet: mediatek: refactoring mtk_hw_init to be reused
the existing mtk_hw_init includes hardware and software
initialization inside so that it is slightly hard to reuse
them for the process of the reset recovery, so some splitting
is made here for keeping hardware initializing relevant thing
and the else such as IRQ registration and MDIO initialization
what are all about to the interface of core driver moved to the
other proper place because they have no needs to register IRQ and
re-initialize structure again during the reset process.
Signed-off-by: Sean Wang <sean.wang@mediatek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Johannes Berg [Wed, 3 Aug 2016 11:52:56 +0000 (13:52 +0200)]
iwlwifi: mvm: tighten BAID range check
As pointed out by smatch, checking the BAID for just >= INVALID
is a bad idea since only 32 (IWL_MAX_BAID) actually exist. Check
the range for that and print invalid ones in the warning.
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Johannes Berg [Wed, 3 Aug 2016 07:34:39 +0000 (09:34 +0200)]
iwlwifi: mvm: make RSS RX more robust
If the firmware ever decides to send any new/more notifications
to the RSS queues, the driver would currently try to interpret
those as REPLY_RX_MPDU_CMD and, if the notification was small,
access invalid memory.
Prevent that by checking for REPLY_RX_MPDU_CMD explicitly which
allows ignoring unexpected notifications.
Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
iwlwifi: pcie: replace possible_cpus() with online_cpus() in MSIX mode
In MSIX mode the number of irq depends on the number of
possible cpus existing on the host.
This cause to bug in case there are offline cores.
Take into account only the online CPUs instead.
Also save it in temporary variable.
Fixes: commit 2e5d4a8f61dc ("iwlwifi: pcie: Add new configuration to enable MSIX") Signed-off-by: Haim Dreyfuss <haim.dreyfuss@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
projects / linux-beck.git / log