Brian Norris [Fri, 13 Jan 2017 23:35:38 +0000 (15:35 -0800)]
mwifiex: pcie: read FROMDEVICE DMA-able memory with READ_ONCE()
In mwifiex_delay_for_sleep_cookie(), we're looping and waiting for the
PCIe endpoint to write a magic value back to memory, to signal that it
has finished going to sleep. We're not letting the compiler know that
this might change underneath our feet though. Let's do that, for good
hygiene.
I'm not aware of this fixing any concrete problems. I also give no
guarantee that this loop is actually correct in any other way, but at
least this looks like an improvement to me.
Signed-off-by: Brian Norris <briannorris@chromium.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Brian Norris [Fri, 13 Jan 2017 23:35:37 +0000 (15:35 -0800)]
mwifiex: pcie: don't loop/retry interrupt status checks
The following sequence occurs when using IEEE power-save on 8997:
(a) driver sees SLEEP event
(b) driver issues SLEEP CONFIRM
(c) driver recevies CMD interrupt; within the interrupt processing loop,
we do (d) and (e):
(d) wait for FW sleep cookie (and often time out; it takes a while), FW
is putting card into low power mode
(e) re-check PCIE_HOST_INT_STATUS register; quit loop with 0 value
But at (e), no one actually signaled an interrupt (i.e., we didn't check
adapter->int_status). And what's more, because the card is going to
sleep, this register read appears to take a very long time in some cases
-- 3 milliseconds in my case!
Now, I propose that (e) is completely unnecessary. If there were any
additional interrupts signaled after the start of this loop, then the
interrupt handler would have set adapter->int_status to non-zero and
queued more work for the main loop -- and we'd catch it on the next
iteration of the main loop.
So this patch drops all the looping/re-reading of PCIE_HOST_INT_STATUS,
which avoids the problematic (and slow) register read in step (e).
Incidentally, this is a very similar issue to the one fixed in commit ec815dd2a5f1 ("mwifiex: prevent register accesses after host is
sleeping"), except that the register read is just very slow instead of
fatal in this case.
Tested on 8997 in both MSI and (though not technically supported at the
moment) MSI-X mode.
Signed-off-by: Brian Norris <briannorris@chromium.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Brian Norris [Fri, 13 Jan 2017 23:35:36 +0000 (15:35 -0800)]
mwifiex: pcie: use posted write to wake up firmware
Depending on system factors (e.g., the PCIe link PM state), the first
read to wake up the Wifi firmware can take a long time. There is no
reason to use a (blocking, non-posted) read at this point, so let's just
use a write instead. Write vs. read doesn't matter functionality-wise --
it's just a dummy operation. But let's make sure to re-write with the
correct "ready" signature, since we check for that in other parts of the
driver.
This has been shown to decrease the time spent blocking in this function
on RK3399.
Signed-off-by: Brian Norris <briannorris@chromium.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Felix Fietkau [Mon, 16 Jan 2017 03:13:01 +0000 (04:13 +0100)]
rt2x00: rt2800lib: fix rf id for RT3352
Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Daniel Golle <daniel@makrotopia.org> Acked-by: Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
rt2800: remove warning on bcn_num != rt2x00dev->intf_beaconing
Since rt2800pci update beacon settings asynchronously from
tbtt tasklet, without beacon_skb_mutex protection, number of
currently active beacons entries can be different than
number pointed by rt2x00dev->intf_beaconing. Remove warning
about that inconsistency.
Reported-by: evaxige@qq.com Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Rafał Miłecki [Sat, 7 Jan 2017 22:43:45 +0000 (23:43 +0100)]
brcmfmac: make brcmf_of_probe more generic
We may want to use Open Firmware for other devices than just SDIO ones.
In future we may want to support more Broadcom properties so there is
really no reason for such limitation.
Call brcmf_of_probe for all kind of devices & move extra conditions to
the body of that funcion.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Pan Bian [Sat, 3 Dec 2016 10:27:37 +0000 (18:27 +0800)]
libertas: fix improper return value
Function lbs_cmd_802_11_sleep_params() always return 0, even if the call
to lbs_cmd_with_response() fails. In this case, the parameter @sp will
keep uninitialized. Because the return value is 0, its caller (say
lbs_sleepparams_read()) will not detect the error, and will copy the
uninitialized stack memory to user sapce, resulting in stack information
leak. To avoid the bug, this patch returns variable ret (which takes
the return value of lbs_cmd_with_response()) instead of 0.
Larry Finger [Tue, 10 Jan 2017 16:15:26 +0000 (10:15 -0600)]
rtlwifi: rtl8192de: Remove a pointless goto
In commit c93ac39da0064 ("rtlwifi: Remove some redundant code), a goto
statement was inadvertently left in the code.
Fixes: c93ac39da0064 ("rtlwifi: Remove some redundant code) Reported-by: kbuild test robot <fengguang.wu@intel.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Similar to commit fcd2042e8d36 ("mwifiex: printk() overflow with 32-byte
SSIDs"), we failed to account for the existence of 32-char SSIDs in our
debugfs code. Unlike in that case though, we zeroed out the containing
struct first, and I'm pretty sure we're guaranteed to have some padding
after the 'ssid.ssid' and 'ssid.ssid_len' fields (the struct is 33 bytes
long).
Fixes: 5e6e3a92b9a4 ("wireless: mwifiex: initial commit for Marvell mwifiex driver") Signed-off-by: Brian Norris <briannorris@chromium.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Rafał Miłecki [Sat, 7 Jan 2017 20:36:05 +0000 (21:36 +0100)]
brcmfmac: setup wiphy bands after registering it first
During bands setup we disable all channels that firmware doesn't support
in the current regulatory setup. If we do this before wiphy_register
it will result in copying set flags (including IEEE80211_CHAN_DISABLED)
to the orig_flags which is supposed to be persistent. We don't want this
as regulatory change may result in enabling some channels. We shouldn't
mess with orig_flags then (by changing them or ignoring them) so it's
better to just take care of their proper values.
This patch cleanups code a bit (by taking orig_flags more seriously) and
allows further improvements like disabling really unavailable channels.
We will need that e.g. if some frequencies should be disabled for good
due to hardware setup (design).
Signed-off-by: Rafał Miłecki <rafal@milecki.pl> Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Rafał Miłecki [Sat, 7 Jan 2017 20:36:04 +0000 (21:36 +0100)]
brcmfmac: don't preset all channels as disabled
During init we take care of regulatory stuff by disabling all
unavailable channels (see brcmf_construct_chaninfo) so this predisabling
them is not really required (and this patch won't change any behavior).
It will on the other hand allow more detailed runtime control over
channels which is the main reason for this change.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Enable RTS frame retry fall-back and limit number of RTS retries to 7
what is default number of retries for small frames. As RTS/CTS is used
for TXOP protection, those settings prevent posting lots of RTS
frames when remote station do not response with CTS at the moment. After
sending 7 RTS's the HW will start back-off mechanism and after it will
start posing RTS again to get access to the medium.
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
We do not have option to set per frame retry count. We have only global
TX_RTY_CFG registers which specify the number or retries. Set setting
of that register to value that correspond rate control algorithm number
of frame post (number of retries + 1), which is 3 for aggregated frames.
This should help with big amount of retries on bad conditions, hence
mitigate buffer-bloat like problems.
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
When medium is busy or frames have to be resend, it takes time to send
the frames and get TX status from hardware. For some really bad medium
conditions it can take seconds. Patch change TX status timeout to give
HW more time to provide it, however 500ms is not enough for bad
conditions. In the future this timeout should be removed and replaced
with proper watchdog mechanism.
Increase flush timeout accordingly as well.
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
On rt2800usb, if we do not get TX status from HW, we assume frames were
posted and after entry->last_action timeout, we forcibly provide TX
status to mac80211. So it's not possible to detect hardware TX hung
based on the timeout. Additionally TXRQ_PCNT tells on number of frames
in the Packet Buffer (buffer between bus interface and chip MAC
subsystem), which can be non zero on normal conditions. To check HW hung
we will need provide some different mechanism, for now remove watchdog
as current implementation is wrong and not useful.
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Rafał Miłecki [Wed, 4 Jan 2017 11:09:41 +0000 (12:09 +0100)]
brcmfmac: avoid writing channel out of allocated array
Our code was assigning number of channels to the index variable by
default. If firmware reported channel we didn't predict this would
result in using that initial index value and writing out of array. This
never happened so far (we got a complete list of supported channels) but
it means possible memory corruption so we should handle it anyway.
This patch simply detects unexpected channel and ignores it.
As we don't try to create new entry now, it's also safe to drop hw_value
and center_freq assignment. For known channels we have these set anyway.
I decided to fix this issue by assigning NULL or a target channel to the
channel variable. This was one of possible ways, I prefefred this one as
it also avoids using channel[index] over and over.
Fixes: 58de92d2f95e ("brcmfmac: use static superset of channels for wiphy bands") Signed-off-by: Rafał Miłecki <rafal@milecki.pl> Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Arnd Bergmann [Fri, 13 Jan 2017 15:35:03 +0000 (16:35 +0100)]
mwifiex: fix uninitialized variable access in pcie_remove
Checking the firmware status from PCIe register only works
if the register is available, otherwise we end up with
random behavior:
drivers/net/wireless/marvell/mwifiex/pcie.c: In function 'mwifiex_pcie_remove':
drivers/net/wireless/marvell/mwifiex/pcie.c:585:5: error: 'fw_status' may be used uninitialized in this function [-Werror=maybe-uninitialized]
This makes sure we treat the absence of the register as a failure.
Fixes: 045f0c1b5e26 ("mwifiex: get rid of global user_rmmod flag") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Xinming Hu [Wed, 14 Dec 2016 14:10:49 +0000 (19:40 +0530)]
mwifiex: sdio card reset enhancement
Commit b4336a282db8 ("mwifiex: sdio: reset adapter using mmc_hw_reset")
introduces a simple sdio card reset solution based on card remove and
re-probe. This solution has proved to be vulnerable, as card and
adapter structures are not protected, concurrent access will result in
kernel panic issues.
Let's reuse PCIe FLR's functions for SDIO reset to avoid freeing and
reallocating adapter and card structures.
Signed-off-by: Xinming Hu <huxm@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Xinming Hu [Wed, 14 Dec 2016 14:10:48 +0000 (19:40 +0530)]
mwifiex: cleanup in PCIe flr code path
adapter and card variables don't get freed during PCIe function level
reset. "adapter->ext_scan" variable need not be re-initialized.
fw_name and tx_buf_size initialization is moved to pcie specific code
so that mwifiex_reinit_sw() can be used by SDIO.
Signed-off-by: Xinming Hu <huxm@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Xinming Hu [Wed, 14 Dec 2016 14:10:47 +0000 (19:40 +0530)]
mwifiex: get rid of mwifiex_do_flr wrapper
This patch gets rid of mwifiex_do_flr. We will call
mwifiex_shutdown_sw() and mwifiex_reinit_sw() directly.
These two general purpose functions will be useful for
sdio card reset handler.
Signed-off-by: Xinming Hu <huxm@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
After user_rmmod global flag removal, *_init_module() and
*_cleanup_module() have become just a wrapper functions.
We will get rid of them with the help of module_*_driver() macros.
For pcie, existing ".init_if" handler has same name as what
module_pcie_driver() macro will create. Let's rename it to
avoid conflict.
Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Xinming Hu [Wed, 30 Nov 2016 14:52:17 +0000 (20:22 +0530)]
mwifiex: get rid of global user_rmmod flag
bus.remove() callback function is called when user removes this module
from kernel space or ejects the card from the slot. The driver handles
these 2 cases differently. Few commands (FUNC_SHUTDOWN etc.) are sent to
the firmware only for module unload case.
The variable 'user_rmmod' is used to distinguish between these two
scenarios.
This patch checks hardware status and get rid of global variable
user_rmmod.
Signed-off-by: Xinming Hu <huxm@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Ganapathi Bhat [Wed, 16 Nov 2016 13:09:09 +0000 (18:39 +0530)]
mwifiex: move pcie_work and related variables inside card
Currently pcie_work and related variables are global. It may create
problem while supporting multiple devices simultaneously. Let's move
it inside card structure so that separate instance will be created/
cancelled in init/teardown threads of each connected devices.
Signed-off-by: Ganapathi Bhat <gbhat@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Xinming Hu [Wed, 16 Nov 2016 13:09:08 +0000 (18:39 +0530)]
mwifiex: wait firmware dump complete during card remove process
Wait for firmware dump complete in card remove function.
For sdio interface, there are two diffenrent cases,
card reset trigger sdio_work and firmware dump trigger sdio_work.
Do code rearrangement for distinguish between these two cases.
Signed-off-by: Xinming Hu <huxm@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Reviewed-by: Brian Norris <briannorris@chromium.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Xinming Hu [Wed, 16 Nov 2016 13:09:05 +0000 (18:39 +0530)]
mwifiex: don't wait for main_process in shutdown_drv
main_process is not expected to be running when shutdown_drv function
is called. currently we wait for main_process completion in the
function.
Actually the caller has already made sure main_process is completed by
performing below actions.
(1) disable interrupts in if_ops->disable_int.
(2) set adapter->surprise_removed = true, main_process wont be queued.
(3) mwifiex_terminate_workqueue(adapter), wait for workqueue to be
completed.
This patch removes redundant wait code and takes care of related
cleanup.
Signed-off-by: Xinming Hu <huxm@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Sowmini Varadhan [Tue, 10 Jan 2017 15:47:15 +0000 (07:47 -0800)]
packet: pdiag_put_ring() should return TX_RING info for TPACKET_V3
Commit 7f953ab2ba46 ("af_packet: TX_RING support for TPACKET_V3")
now makes it possible to use TX_RING with TPACKET_V3, so make the
the relevant information available via 'ss -e -a --packet'
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tobias Klauser [Tue, 10 Jan 2017 14:02:16 +0000 (15:02 +0100)]
bpf: Make unnecessarily global functions static
Make the functions __local_list_pop_free(), __local_list_pop_pending(),
bpf_common_lru_populate() and bpf_percpu_lru_populate() static as they
are not used outide of bpf_lru_list.c
This fixes the following GCC warnings when building with 'W=1':
kernel/bpf/bpf_lru_list.c:363:22: warning: no previous prototype for ‘__local_list_pop_free’ [-Wmissing-prototypes]
kernel/bpf/bpf_lru_list.c:376:22: warning: no previous prototype for ‘__local_list_pop_pending’ [-Wmissing-prototypes]
kernel/bpf/bpf_lru_list.c:560:6: warning: no previous prototype for ‘bpf_common_lru_populate’ [-Wmissing-prototypes]
kernel/bpf/bpf_lru_list.c:577:6: warning: no previous prototype for ‘bpf_percpu_lru_populate’ [-Wmissing-prototypes]
Cc: Martin KaFai Lau <kafai@fb.com> Signed-off-by: Tobias Klauser <tklauser@distanz.ch> Acked-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Martin KaFai Lau <kafai@fb.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tobias Klauser [Tue, 10 Jan 2017 14:02:07 +0000 (15:02 +0100)]
bpf: Remove unused but set variable in __bpf_lru_list_shrink_inactive()
Remove the unused but set variable 'first_node' in
__bpf_lru_list_shrink_inactive() to fix the following GCC warning when
building with 'W=1':
kernel/bpf/bpf_lru_list.c:216:41: warning: variable ‘first_node’ set but not used [-Wunused-but-set-variable]
Cc: Martin KaFai Lau <kafai@fb.com> Signed-off-by: Tobias Klauser <tklauser@distanz.ch> Acked-by: Martin KaFai Lau <kafai@fb.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Mahesh Bandewar [Mon, 9 Jan 2017 23:05:54 +0000 (15:05 -0800)]
ipvlan: improvise dev_id generation logic in IPvlan
The patch 009146d117b ("ipvlan: assign unique dev-id for each slave
device.") used ida_simple_get() to generate dev_ids assigned to the
slave devices. However (Eric has pointed out that) there is a shortcoming
with that approach as it always uses the first available ID. This
becomes a problem when a slave gets deleted and a new slave gets added.
The ID gets reassigned causing the new slave to get the same link-local
address. This side-effect is undesirable.
This patch adds a per-port variable that keeps track of the IDs
assigned and used as the stat-base for the IDR api. This base will be
wrapped around when it reaches the MAX (0xFFFE) value possibly on a
busy system where slaves are added and deleted routinely.
Fixes: 009146d117b ("ipvlan: assign unique dev-id for each slave device.") Signed-off-by: Mahesh Bandewar <maheshb@google.com> CC: Eric Dumazet <edumazet@google.com> CC: David Miller <davem@davemloft.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Bert Kenward [Tue, 10 Jan 2017 16:24:31 +0000 (16:24 +0000)]
sfc: stop setting dev_port
Setting dev_port changes the device names allocated by systemd. Any devices
with a dev_port >0 will (in default distro configurations) have a suffix of
"d<port-number>" appended.
This is not something done by other drivers, and causes confusion for users.
Fixes: 8be41320f346 ("sfc: Add code to export port_num in netdev->dev_port") Signed-off-by: Edward Cree <ecree@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Bert Kenward [Tue, 10 Jan 2017 16:23:56 +0000 (16:23 +0000)]
sfc: implement ndo_get_phys_port_name
Output is of the form p<port-number>.
Note that the port numbers don't necessarily map one-to-one to physical
cages, partly because of 4x10G port modes on QSFP+ and partly because
of hw/fw implementation details.
Signed-off-by: Edward Cree <ecree@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Timur Tabi [Mon, 9 Jan 2017 18:03:12 +0000 (12:03 -0600)]
net: qcom/emac: add ethtool support
Add support for some ethtool methods: get/set link settings, get/set
message level, get statistics, get link status, get ring params, get
pause params, and restart autonegotiation.
The code to collect the hardware statistics is moved into its own
function so that it can be used by "get statistics" method.
Signed-off-by: Timur Tabi <timur@codeaurora.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Vivien Didelot [Mon, 9 Jan 2017 21:49:26 +0000 (16:49 -0500)]
net: dsa: select NET_SWITCHDEV
The support for DSA Ethernet switch chips depends on TCP/IP networking,
thus explicit that HAVE_NET_DSA depends on INET.
DSA uses SWITCHDEV, thus select it instead of depending on it.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Tested-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 9 Jan 2017 22:09:31 +0000 (17:09 -0500)]
Merge tag 'mlx5-4kuar-for-4.11' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux
Saeed Mahameed says:
====================
mlx5 4K UAR
The following series of patches optimizes the usage of the UAR area which is
contained within the BAR 0-1. Previous versions of the firmware and the driver
assumed each system page contains a single UAR. This patch set will query the
firmware for a new capability that if published, means that the firmware can
support UARs of fixed 4K regardless of system page size. In the case of
powerpc, where page size equals 64KB, this means we can utilize 16 UARs per
system page. Since user space processes by default consume eight UARs per
context this means that with this change a process will need a single system
page to fulfill that requirement and in fact make use of more UARs which is
better in terms of performance.
In addition to optimizing user-space processes, we introduce an allocator
that can be used by kernel consumers to allocate blue flame registers
(which are areas within a UAR that are used to write doorbells). This provides
further optimization on using the UAR area since the Ethernet driver makes
use of a single blue flame register per system page and now it will use two
blue flame registers per 4K.
The series also makes changes to naming conventions and now the terms used in
the driver code match the terms used in the PRM (programmers reference manual).
Thus, what used to be called UUAR (micro UAR) is now called BFREG (blue flame
register).
In order to support compatibility between different versions of
library/driver/firmware, the library has now means to notify the kernel driver
that it supports the new scheme and the kernel can notify the library if it
supports this extension. So mixed versions of libraries can run concurrently
without any issues.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Mon, 9 Jan 2017 18:29:27 +0000 (10:29 -0800)]
tcp: make TCP_INFO more consistent
tcp_get_info() has to lock the socket, so lets lock it
for an extended critical section, so that various fields
have consistent values.
This solves an annoying issue that some applications
reported when multiple counters are updated during one
particular rx/rx event, and TCP_INFO was called from
another cpu.
Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
since ARG_PTR_TO_STACK is no longer just pointer to stack
rename it to ARG_PTR_TO_MEM and adjust comment.
Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, helpers that read and write from/to the stack can do so using
a pair of arguments of type ARG_PTR_TO_STACK and ARG_CONST_STACK_SIZE.
ARG_CONST_STACK_SIZE accepts a constant register of type CONST_IMM, so
that the verifier can safely check the memory access. However, requiring
the argument to be a constant can be limiting in some circumstances.
Since the current logic keeps track of the minimum and maximum value of
a register throughout the simulated execution, ARG_CONST_STACK_SIZE can
be changed to also accept an UNKNOWN_VALUE register in case its
boundaries have been set and the range doesn't cause invalid memory
accesses.
One common situation when this is useful:
int len;
char buf[BUFSIZE]; /* BUFSIZE is 128 */
if (some_condition)
len = 42;
else
len = 84;
some_helper(..., buf, len & (BUFSIZE - 1));
The compiler can often decide to assign the constant values 42 or 48
into a variable on the stack, instead of keeping it in a register. When
the variable is then read back from stack into the register in order to
be passed to the helper, the verifier will not be able to recognize the
register as constant (the verifier is not currently tracking all
constant writes into memory), and the program won't be valid.
However, by allowing the helper to accept an UNKNOWN_VALUE register,
this program will work because the bitwise AND operation will set the
range of possible values for the UNKNOWN_VALUE register to [0, BUFSIZE),
so the verifier can guarantee the helper call will be safe (assuming the
argument is of type ARG_CONST_STACK_SIZE_OR_ZERO, otherwise one more
check against 0 would be needed). Custom ranges can be set not only with
ALU operations, but also by explicitly comparing the UNKNOWN_VALUE
register with constants.
Another very common example happens when intercepting system call
arguments and accessing user-provided data of variable size using
bpf_probe_read(). One can load at runtime the user-provided length in an
UNKNOWN_VALUE register, and then read that exact amount of data up to a
compile-time determined limit in order to fit into the proper local
storage allocated on the stack, without having to guess a suboptimal
access size at compile time.
Also, in case the helpers accepting the UNKNOWN_VALUE register operate
in raw mode, disable the raw mode so that the program is required to
initialize all memory, since there is no guarantee the helper will fill
it completely, leaving possibilities for data leak (just relevant when
the memory used by the helper is the stack, not when using a pointer to
map element value or packet). In other words, ARG_PTR_TO_RAW_STACK will
be treated as ARG_PTR_TO_STACK.
Signed-off-by: Gianluca Borello <g.borello@gmail.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
commit 484611357c19 ("bpf: allow access into map value arrays")
introduces the ability to do pointer math inside a map element value via
the PTR_TO_MAP_VALUE_ADJ register type.
The current support doesn't handle the case where a PTR_TO_MAP_VALUE_ADJ
is spilled into the stack, limiting several use cases, especially when
generating bpf code from a compiler.
Handle this case by explicitly enabling the register type
PTR_TO_MAP_VALUE_ADJ to be spilled. Also, make sure that min_value and
max_value are reset just for BPF_LDX operations that don't result in a
restore of a spilled register from stack.
Signed-off-by: Gianluca Borello <g.borello@gmail.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Enable helpers to directly access a map element value by passing a
register type PTR_TO_MAP_VALUE (or PTR_TO_MAP_VALUE_ADJ) to helper
arguments ARG_PTR_TO_STACK or ARG_PTR_TO_RAW_STACK.
This enables several use cases. For example, a typical tracing program
might want to capture pathnames passed to sys_open() with:
/* consume data.pathname, for example via
* bpf_trace_printk() or bpf_perf_event_output()
*/
}
Such a program could easily hit the stack limit in case PATHLEN needs to
be large or more local variables need to exist, both of which are quite
common scenarios. Allowing direct helper access to map element values,
one could do:
SEC("kprobe/sys_open")
int bpf_sys_open(struct pt_regs *ctx)
{
int id = 0;
struct trace_data *p = bpf_map_lookup_elem(&scratch_map, &id);
if (!p)
return;
bpf_probe_read(p->pathname, sizeof(p->pathname), ctx->di);
/* consume p->pathname, for example via
* bpf_trace_printk() or bpf_perf_event_output()
*/
}
And wouldn't risk exhausting the stack.
Code changes are loosely modeled after commit 6841de8b0d03 ("bpf: allow
helpers access the packet directly"). Unlike with PTR_TO_PACKET, these
changes just work with ARG_PTR_TO_STACK and ARG_PTR_TO_RAW_STACK (not
ARG_PTR_TO_MAP_KEY, ARG_PTR_TO_MAP_VALUE, ...): adding those would be
trivial, but since there is not currently a use case for that, it's
reasonable to limit the set of changes.
Also, add new tests to make sure accesses to map element values from
helpers never go out of boundary, even when adjusted.
Signed-off-by: Gianluca Borello <g.borello@gmail.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Move the logic to check memory accesses to a PTR_TO_MAP_VALUE_ADJ from
check_mem_access() to a separate helper check_map_access_adj(). This
enables to use those checks in other parts of the verifier as well,
where boundaries on PTR_TO_MAP_VALUE_ADJ might need to be checked, for
example when checking helper function arguments. The same thing is
already happening for other types such as PTR_TO_PACKET and its
check_packet_access() helper.
The code has been copied verbatim, with the only difference of removing
the "off += reg->max_value" statement and moving the sum into the call
statement to check_map_access(), as that was only needed due to the
earlier common check_map_access() call.
Signed-off-by: Gianluca Borello <g.borello@gmail.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
here is now V4 of the SMC-R patches having processed your feedback from end
of November. The most important change is the replacement of sysfs by a
generic netlink solution in patch 04. And I tried to get rid of the __packed
attributes. There are still a few usages left due to SMC-R protocol defined
structures.
V4 changes:
The order of patches 03 and 04 for pnet table management and SMC IB-client
establishing has been exchanged, since pnet table management is now built on
top of smc_ib_devices.
Patch 01: Use EXPORT_SYMBOL_GPL().
Patch 02: Define "use_fallback" as bool.
Get rid of useless smc_sock fields clearing in smc_sock_alloc(),
since sk_alloc() clears out the memory.
Patch 03: Postpone smc_ib_remember_port_attr() call till ib_device is
mentioned in the pnet table.
Patch 04: Replace sysfs-usage by a generic netlink approach for pnet table
configuration.
Change layout of pnet table entries to reference net_device and
ib_device instead of dealing with names of net_devices and
ib_devices.
Patch 05: Adapt "use_fallback" usages to new type bool.
Get rid of useless smc_sock fields clearing in smc_sock_alloc()
Avoid __packed where possible.
Check if clc responses are not too big.
Patch 09: Postpone smc_setup_per_ibdev till the first connection with this
ib_device is really created.
Patch 11: Get rid of __packed usage.
V3 changes:
Patch 05: Remove unneeded DEFINE_WAIT
Patch 06: Improve synchronization of link group creation
Patch 07: Rename peer_rmbe_len into peer_rmbe_size to be more consistent
Patch 09: Avoid calls of ib_get_memory_region with IB_ACCESS_LOCAL_WRITE,
use new default local_dma_lkey from protection domain as lkey
instead.
Remove no longer needed function smc_ib_dereg_memory_region().
Patch 14: Switch to state ACTIVE only if still in state INIT.
Return 0 for recvmsg invoked in a socket closing state.
Allow getname call in state APPCLOSEWAIT1
Do not trigger destruction of a socket-in-error queued in accept
queue.
During cleanup of accept queue, make sure sockets are destructed,
and sockets in fallback mode are handled appropriately.
When freeing sndbufs/rmbs, remove them from their list and free
the entry.
Use add_wait_queue() and remove_wait_queue() in close wait
functions.
If actively closing a socket in state for PEERFINCLOSEWAIT, keep
this state.
If passively closing a socket while bytes are to be received, move
to state APPCLOSEWAIT1.
If actively aborting a socket, skip sending the close_abort flag,
since RDMA communication is no longer possible.
When terminating a link group, do not schedule link group freeing a
2nd time, since already done when unregistering the last remaining
connection.
Patch 15: Introduce smc_diag module for monitoring SMC protocol sockets.
This replaces the old patch 0015 dealing with procfs.
V2 changes:
Patch 0002: Add SMC versions for family key strings in net/core/sock.c.
Patch 0006: initialize rb_tree.
Patch 0007: Get rid of unneeded use of xchg() in smc_sndbuf_unuse() and
smc_rmb_unuse().
Patch 0008: Correct error checking logic for ib_function calls.
Define struct smc_link field wr_tx_id as atomic_long_t.
Use "do_div" instead of "%" to be architecture-independent.
Patch 0009: Correct error checking logic for ib_function calls.
Patch 0011: Remove xchg() calls in cursor handling. Use atomic64_t for cursor
overlays on 64-bit architectures. If not available, use plain u64
and add locking for cursor reading and writing.
Implement smc_curs_add() without modulo operator "%".
Patch 0012: Remove xchg() calls in cursor handling.
Implement smc_tx_rdma_writes() without module operator "%".
Patch 0013: Remove xchg() calls in cursor handling.
Patch 0014: Return type bool in smc_wr_tx_has_pending().
Remove unneeded semicolon in smc_close_shutdown_write().
Call smc_close_active() in non-fallback case only.
Get rid of duplicate schedule of sock_put_work().
Take nested sock_lock in smc_listen_work().
Start close stream_wait in case of prepared sends only.
Patch 0015: Remove unneeded socket ref_count in smc_proc_seq_show().
Take lock before list_empty check in smc_proc_sock_list_del().
These patches are the initial part of the implementation of the
"Shared Memory Communications-RDMA" (SMC-R) protocol as defined in
RFC7609 [1]. While SMC-R does not aim to replace TCP,
it taps a wealth of existing data center TCP socket applications
to become more efficient without the need for rewriting them.
SMC-R uses RDMA over Converged Ethernet (RoCE) to save CPU consumption.
For instance, when running 10 parallel connections with uperf, we measured
a decrease of 60% in CPU consumption with SMC-R compared to TCP/IP
(with throughput and latency comparable;
measured on x86_64 with the same RoCE card and port).
SMC-R does not require an RDMA communication manager (RDMA CM).
SMC-R inherits TCP qualities such as reliable connections, host-based
firewall packet filtering (on connection establishment) and unmodified
application of communication encryption such as TLS (transport layer
security) or SSL (secure sockets layer). Since original TCP is used to
establish SMC-R connections, load balancers and packet inspection based
on TCP/IP connection establishment continue to work for SMC-R.
On the other hand, using SMC-R implies:
- either involving a preload library when invoking the unchanged TCP-application
or slightly modifying the source by simply changing the socket family in
the socket() call
- accepting extra overhead and latency in connection establishment due to
SMC Connection Layer Control (CLC) handshake
- explicit coupling of RoCE ports with Ethernet ports
- not routable as currently built on RoCE V1
- bypassing of packet-based networking features
- filtering (netfilter)
- sniffing (libpcap, packet sockets, (E)BPF)
- traffic control (scheduling, shaping)
- bypassing of IP-header based socket options
- bypassing of memory buffer (pressure) management
- unusable together with IPsec
Overview of the SMC-R Protocol described in informational RFC 7609
SMC-R is an open protocol that provides RDMA capabilities over RoCE
transparently for applications exploiting TCP sockets.
A new socket protocol family PF_SMC is introduced.
There are no changes required to applications using the sockets API for TCP
stream sockets other than the specification of the new socket family AF_SMC.
Unmodified applications can be used by means of a dynamic preload shared
library which rewrites the socket API call
socket(AF_INET, SOCK_STREAM, IPPROTO_TCP) into
socket(AF_SMC, SOCK_STREAM, IPPROTO_TCP).
SMC-R re-uses the address family AF_INET for all addressing purposes around
struct sockaddr.
A link group is determined by an ordered peer pair of TCP client and TCP server
(IP addresses and subnet). Reversed client server roles cause an own link group.
A link is a logical point-to-point connection based on an
infiniband reliable connected queue pair (RC-QP) between two RoCE ports
(MACs and GIDs) of a peer pair.
A link group can have 1..8 links for failover and load balancing.
This initial Linux implementation always has 1 link per link group.
Each link group on a peer can have 1..255 remote memory buffers (RMBs).
If more RMBs are needed, a peer can open another link group
(this initial Linux implementation) or fall back to TCP.
Each RMB has its own particular size and its own (R)DMA mapping and credentials
(rtoken consisting of rkey and RDMA "virtual address").
This initial Linux implementation uses physically contiguous memory for RMBs
but we are working towards scattered memory because of memory fragmentation.
Each RMB has 1..255 RMB elements (RMBEs) of equal size
to provide multiplexing of connections within an RMB.
An RMBE is the RDMA Write destination organized as wrapping ring buffer
for data transmit of a particular connection in one direction
(duplex by means of mirror symmetry as with TCP).
This initial Linux implementation always has 1 RMBE per RMB
and thus an individual RMB for each connection.
SMC-R connection establishment with subsequent data transfer:
CLIENT SERVER
TCP three-way handshake:
regular TCP SYN
-------------------------------------------------------->
regular TCP SYN ACK
<--------------------------------------------------------
regular TCP ACK
-------------------------------------------------------->
SMC Connection Layer Control (CLC) handshake
exchanges RDMA credentials between peers:
via above TCP connection: SMC CLC Proposal
-------------------------------------------------------->
via above TCP connection: SMC CLC Accept
<--------------------------------------------------------
via above TCP connection: SMC CLC Confirm
-------------------------------------------------------->
SMC Link Layer Control (LLC) (only once per link, i.e. 1st conn. of link group):
RoCE RC-QP: SMC LLC Confirm Link
<========================================================
RoCE RC-QP: SMC LLC Confirm Link response
========================================================>
SMC data transmission (incl. SMC Connection Data Control (CDC) message):
RoCE RC-QP: RDMA Write
========================================================>
RoCE RC-QP: SMC CDC message (flow control)
========================================================>
...
RoCE RC-QP: RDMA Write
<========================================================
RoCE RC-QP: SMC CDC message (flow control)
<========================================================
...
Data flow within an established connection:
+----------------------------------------------------------------------------
| SENDER
| sendmsg()
| |
| | produces into sndbuf [sender's process context]
| v
| +--------+
| | sndbuf | [ring buffer]
| +--------+
| |
| | consumes from sndbuf and produces into receiver's RMBE [any context]
| | by sending RDMA Write followed by SMC CDC message over RoCE RC-QP
| |
+----|-----------------------------------------------------------------------
|
+----|-----------------------------------------------------------------------
| v RECEIVER
| +------+
| | RMBE | [ring buffer, can have size different from sender's sndbuf]
| | | [RMBE represents rcvbuf, no further de-coupling as on sender side]
| +------+
| |
| | consumes from RMBE [receiver's process context]
| v
| recvmsg()
+----------------------------------------------------------------------------
Flow control ("cursor" updates) by means of SMC CDC messages:
SENDER RECEIVER
sends updates via CDC-------------+ sends updates via CDC
on consuming from sndbuf | on consuming from RMBE
and producing into RMBE | by means of recvmsg()
| |
| |
+-----------------------------------|------------+
| |
+--v-------------------------+ +--v-----------------------+
| receiver's consumer cursor | | sender's producer cursor----+
+----------------|-----------+ +--------------------------+ |
| |
| receiver's RMBE |
| +--------------------------+ |
| | | |
+--------------------------------+ | |
| | | |
| v | |
| +------------| |
|-------------+////////////| |
|//RDMA data written by////| |
|////sender that is////////| |
|/available to be consumed/| |
|///////// +---------------| |
|----------+^ | |
| | | |
| +-----------------+
| |
+--------------------------+
Sending updates of the producer cursor is immediate for low latency;
something like Nagle's algorithm (absence of TCP_NODELAY) is optional and
currently not part of this initial Linux implementation.
Sending updates of the consumer cursor is conditional to avoid the
silly window syndrome.
Normal connection termination:
Normal connection termination starts transitioning from socket state
ACTIVE via either "Active Close" or "Passive Close".
shutdown rdwr +-----------------+
or close, +-------------->| INIT / CLOSED |<-------------+
send PeerCon|nClosed +-----------------+ | PeerConnClosed
| | | received
| connection | established |
| V |
+----------------+ +-----------------+ +----------------+
|AppFinCloseWait | | ACTIVE | |PeerFinCloseWait|
+----------------+ +-----------------+ +----------------+
| | | |
| Active Close: | |Passive Close: |
| close or | |PeerConnClosed or |
| shutdown wr or| |PeerDoneWriting |
| shutdown rdwr | |received |
| V V |
PeerConnClo|sed +--------------+ +-------------+ | close or
received +--<----|PeerCloseWait1| |AppCloseWait1|--->----+ shutdown rdwr,
| +--------------+ +-------------+ | send
| PeerDoneWri|ting | shutdown wr, | PeerConnClosed
| received | send Pee|rDoneWriting |
| V V |
| +--------------+ +-------------+ |
+--<----|PeerCloseWait2| |AppCloseWait2|--->----+
+--------------+ +-------------+
In state CLOSED, the socket can be destructed only, once the application has
issued a close().
A PNET table in sysfs provides the mapping between network device names and
RoCE Infiniband device names for the transparent switch of data communication.
A PNET table can contain an arbitrary number of PNETIDs.
Each PNETID contains exactly one (Ethernet) network device name
and one or more RoCE Infiniband device names.
Each device name can only exist in at most one PNETID (no overlapping).
This initial Linux implementation allows at most one RoCE Infiniband device
name per PNETID.
After a new TCP connection is established, the network device name
used for egress traffic with the TCP connection's local source IP address
is used as key to lookup the unique PNETID, and the RoCE Infiniband device
of this PNETID is used to switch data communication from TCP to RDMA
during SMC CLC handshake.
Problem determination:
A protocol dissector is available with upstream wireshark for formatting
SMC-R related RoCE LAN traffic.
[https://code.wireshark.org/review/gitweb?p=wireshark.git;a=blob;f=epan/dissectors/packet-smcr.c]
We are working on enhancing the Linux implementation to cover:
- Improve default socket closing asynchronicity
- Address corner cases with many parallel connections
- Tracing
- Integrated load balancing and fail-over within a link group
- Splice and sendpage support
- IPv6 addressing support
- Keepalive, Cork
- Namespaces support
- Urgent data
- More socket options
- Diagnostics
- Statistics support
- SNMP support
Ursula Braun [Mon, 9 Jan 2017 15:55:19 +0000 (16:55 +0100)]
smc: work request (WR) base for use by LLC and CDC
The base containers for RDMA transport are work requests and completion
queue entries processed through Infiniband verbs:
* allocate and initialize these areas
* map these areas to DMA
* implement the basic communication consisting of work request posting
and receival of completion queue events
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ursula Braun [Mon, 9 Jan 2017 15:55:18 +0000 (16:55 +0100)]
smc: remote memory buffers (RMBs)
* allocate data RMB memory for sending and receiving
* size depends on the maximum socket send and receive buffers
* allocated RMBs are kept during life time of the owning link group
* map the allocated RMBs to DMA
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Thomas Richter [Mon, 9 Jan 2017 15:55:15 +0000 (16:55 +0100)]
smc: establish pnet table management
Connection creation with SMC-R starts through an internal
TCP-connection. The Ethernet interface for this TCP-connection is not
restricted to the Ethernet interface of a RoCE device. Any existing
Ethernet interface belonging to the same physical net can be used, as
long as there is a defined relation between the Ethernet interface and
some RoCE devices. This relation is defined with the help of an
identification string called "Physical Net ID" or short "pnet ID".
Information about defined pnet IDs and their related Ethernet
interfaces and RoCE devices is stored in the SMC-R pnet table.
A pnet table entry consists of the identifying pnet ID and the
associated network and IB device.
This patch adds pnet table configuration support using the
generic netlink message interface referring to network and IB device
by their names. Commands exist to add, delete, and display pnet table
entries, and to flush or display the entire pnet table.
There are cross-checks to verify whether the ethernet interfaces
or infiniband devices really exist in the system. If either device
is not available, the pnet ID entry is not created.
Loss of network devices and IB devices is also monitored;
a pnet ID entry is removed when an associated network or
IB device is removed.
Signed-off-by: Thomas Richter <tmricht@linux.vnet.ibm.com> Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ursula Braun [Mon, 9 Jan 2017 15:55:13 +0000 (16:55 +0100)]
smc: establish new socket family
* enable smc module loading and unloading
* register new socket family
* basic smc socket creation and deletion
* use backing TCP socket to run CLC (Connection Layer Control)
handshake of SMC protocol
* Setup for infiniband traffic is implemented in follow-on patches.
For now fallback to TCP socket is always used.
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com> Reviewed-by: Utz Bacher <utz.bacher@de.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ursula Braun [Mon, 9 Jan 2017 15:55:12 +0000 (16:55 +0100)]
net: introduce keepalive function in struct proto
Direct call of tcp_set_keepalive() function from protocol-agnostic
sock_setsockopt() function in net/core/sock.c violates network
layering. And newly introduced protocol (SMC-R) will need its own
keepalive function. Therefore, add "keepalive" function pointer
to "struct proto", and call it from sock_setsockopt() via this pointer.
Signed-off-by: Ursula Braun <ubraun@linux.vnet.ibm.com> Reviewed-by: Utz Bacher <utz.bacher@de.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 9 Jan 2017 20:55:18 +0000 (15:55 -0500)]
Merge branch 'sh_eth-wol'
Niklas Söderlund says:
====================
sh_eth: add wake-on-lan support via magic packet
This series adds support for Wake-on-Lan using Magic Packet for a few
models of the sh_eth driver. Patch 1/6 fix a naming error, patch 2/6
adds generic support to control and support WoL while patches 3/6 - 6/6
enable different models.
Based ontop of net-next master.
Changes since v2.
- Fix bookkeeping for "active_count" and "event_count" reported in
/sys/kernel/debug/wakeup_sources. Thanks Geert for noticing this.
- Add new patch 1/6 which corrects the name of ECMR_MPDE bit, suggested
by Sergei.
- s/sh7743/sh7734/ in patch 5/6. Thanks Geert for spotting this.
- Spelling improvements suggested by Sergei and Geert.
- Add Tested-by to 3/6 and 4/6.
Changes since v1.
- Split generic WoL functionality and device enablement to different
patches.
- Enable more devices then Gen2 after feedback from Geert and
datasheets.
- Do not set mdp->irq_enabled = false and remove specific MagicPacket
interrupt clearing, instead let sh_eth_error() clear the interrupt as
for other EMAC interrupts, thanks Sergei for the suggestion.
- Use the original return logic in sh_eth_resume().
- Moved sh_eth_private variable *clk to top of data structure to avoid
possible gaps due to alignment restrictions.
- Make wol_enabled in sh_eth_private part of the already existing
bitfield instead of a bool.
- Do not initiate mdp->wol_enabled to 0, the struct is kzalloc'ed so
it's already set to 0.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Geert Uytterhoeven reported WoL worked on his Armadillo board.
Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Tested-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Tested-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: David S. Miller <davem@davemloft.net>
sh_eth: add generic wake-on-lan support via magic packet
Add generic functionality to support Wake-on-LAN using MagicPacket which
are supported by at least a few versions of sh_eth. Only add
functionality for WoL, no specific sh_eth versions are marked to support
WoL yet.
WoL is enabled in the suspend callback by setting MagicPacket detection
and disabling all interrupts expect MagicPacket. In the resume path the
driver needs to reset the hardware to rearm the WoL logic, this prevents
the driver from simply restoring the registers and to take advantage of
that sh_eth was not suspended to reduce resume time. To reset the
hardware the driver closes and reopens the device just like it would do
in a normal suspend/resume scenario without WoL enabled, but it both
closes and opens the device in the resume callback since the device
needs to be open for WoL to work.
One quirk needed for WoL is that the module clock needs to be prevented
from being switched off by Runtime PM. To keep the clock alive the
suspend callback need to call clk_enable() directly to increase the
usage count of the clock. Then when Runtime PM decreases the clock usage
count it won't reach 0 and be switched off.
Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Signed-off-by: David S. Miller <davem@davemloft.net>
This bit was wrongly named due to a typo, Sergei checked the SH7734/63
manuals and this bit should be named MPDE.
Suggested-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Signed-off-by: David S. Miller <davem@davemloft.net>
This patchset is optimizing the ICMP-reply code path, for ICMP packets
that gets rate limited. A remote party can easily trigger this code
path by sending packets to port number with no listening service.
Generally the patchset moves the sysctl_icmp_msgs_per_sec ratelimit
checking to earlier in the code path and removes an allocation.
Use-case: The specific case I experienced this being a bottleneck is,
sending UDP packets to a port with no listener, which obviously result
in kernel replying with ICMP Destination Unreachable (type:3), Port
Unreachable (code:3), which cause the bottleneck.
After Eric and Paolo optimized the UDP socket code, the kernels PPS
processing capabilities is lower for no-listen ports, than normal UDP
sockets. This is bad for capacity planning when restarting a service.
UDP no-listen benchmark 8xCPUs using pktgen_sample04_many_flows.sh:
Baseline: 6.6 Mpps
Patch: 14.7 Mpps
Driver mlx5 at 50Gbit/s.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
net: for rate-limited ICMP replies save one atomic operation
It is possible to avoid the atomic operation in icmp{v6,}_xmit_lock,
by checking the sysctl_icmp_msgs_per_sec ratelimit before these calls,
as pointed out by Eric Dumazet, but the BH disabled state must be correct.
The icmp_global_allow() call states it must be called with BH
disabled. This protection was given by the calls icmp_xmit_lock and
icmpv6_xmit_lock. Thus, split out local_bh_disable/enable from these
functions and maintain it explicitly at callers.
Suggested-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: reduce cycles spend on ICMP replies that gets rate limited
This patch split the global and per (inet)peer ICMP-reply limiter
code, and moves the global limit check to earlier in the packet
processing path. Thus, avoid spending cycles on ICMP replies that
gets limited/suppressed anyhow.
The global ICMP rate limiter icmp_global_allow() is a good solution,
it just happens too late in the process. The kernel goes through the
full route lookup (return path) for the ICMP message, before taking
the rate limit decision of not sending the ICMP reply.
Details: The kernels global rate limiter for ICMP messages got added
in commit 4cdf507d5452 ("icmp: add a global rate limitation"). It is
a token bucket limiter with a global lock. It brilliantly avoids
locking congestion by only updating when 20ms (HZ/50) were elapsed. It
can then avoids taking lock when credit is exhausted (when under
pressure) and time constraint for refill is not yet meet.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Revert "icmp: avoid allocating large struct on stack"
This reverts commit 9a99d4a50cb8 ("icmp: avoid allocating large struct
on stack"), because struct icmp_bxm no really a large struct, and
allocating and free of this small 112 bytes hurts performance.
Fixes: 9a99d4a50cb8 ("icmp: avoid allocating large struct on stack") Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
These patches provide some tracepoints for AFS and fix a potential leak by
adding refcounting to the afs_call struct.
The patches are:
(1) Add some tracepoints for logging incoming calls and monitoring
notifications from AF_RXRPC and data reception.
(2) Get rid of afs_wait_mode as it didn't turn out to be as useful as
initially expected. It can be brought back later if needed. This
clears some stuff out that I don't then need to fix up in (4).
(3) Allow listen(..., 0) to be used to disable listening. This makes
shutting down the AFS cache manager server in the kernel much easier
and the accounting simpler as we can then be sure that (a) all
preallocated afs_call structs are relesed and (b) no new incoming
calls are going to be started.
For the moment, listening cannot be reenabled.
(4) Add refcounting to the afs_call struct to fix a potential multiple
release detected by static checking and add a tracepoint to follow the
lifecycle of afs_call objects.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
net: dsa: Encapsulate legacy switch drivers into dsa_switch_driver
In preparation for making struct dsa_switch_ops const, encapsulate it
within a dsa_switch_driver which has a list pointer and a pointer to
dsa_switch_ops. This allows us to take the list_head pointer out of
dsa_switch_ops, which is written to by {un,}register_switch_driver.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 9 Jan 2017 20:40:57 +0000 (15:40 -0500)]
Merge branch 'sh_eth-csum'
Sergei Shtylyov says:
====================
sh_eth: "intgelligent checksum" related cleanups
Here's a set of 2 patches against DaveM's 'net.git' repo, as they are based
on a couple patches merged there recently; however, the patches are destined
for 'net-next.git' (once 'net.git' gets merged there next time). I'm cleaning
up the "intelligent checksum" related code (however, the driver only disables
this feature for now, theres's no proper offload supprt yet).
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Sergei Shtylyov [Fri, 6 Jan 2017 21:02:52 +0000 (00:02 +0300)]
sh_eth: get rid of 'sh_eth_cpu_data::shift_rd0'
After checking all the available manuals, I have enough information to
conclude that the 'shift_rd0' flag is only relevant for the Ether cores
supporting so called "intelligent checksum" (and hence having CSMR) which
is indicated by the 'hw_crc' flag. Since all the relevant SoCs now have
both these flags set, we can at last get rid of the former flag...
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Zefir Kurtisi [Fri, 6 Jan 2017 11:14:48 +0000 (12:14 +0100)]
phy state machine: failsafe leave invalid RUNNING state
While in RUNNING state, phy_state_machine() checks for link changes by
comparing phydev->link before and after calling phy_read_status().
This works as long as it is guaranteed that phydev->link is never
changed outside the phy_state_machine().
If in some setups this happens, it causes the state machine to miss
a link loss and remain RUNNING despite phydev->link being 0.
This has been observed running a dsa setup with a process continuously
polling the link states over ethtool each second (SNMPD RFC-1213
agent). Disconnecting the link on a phy followed by a ETHTOOL_GSET
causes dsa_slave_get_settings() / dsa_slave_get_link_ksettings() to
call phy_read_status() and with that modify the link status - and
with that bricking the phy state machine.
This patch adds a fail-safe check while in RUNNING, which causes to
move to CHANGELINK when the link is gone and we are still RUNNING.
Signed-off-by: Zefir Kurtisi <zefir.kurtisi@neratec.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
1) Fix dumping of nft_quota entries, from Pablo Neira Ayuso.
2) Fix out of bounds access in nf_tables discovered by KASAN, from
Florian Westphal.
3) Fix IRQ enabling in dp83867 driver, from Grygorii Strashko.
4) Fix unicast filtering in be2net driver, from Ivan Vecera.
5) tg3_get_stats64() can race with driver close and ethtool
reconfigurations, fix from Michael Chan.
6) Fix error handling when pass limit is reached in bpf code gen on
x86. From Daniel Borkmann.
7) Don't clobber switch ops and use proper MDIO nested reads and writes
in bcm_sf2 driver, from Florian Fainelli.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (21 commits)
net: dsa: bcm_sf2: Utilize nested MDIO read/write
net: dsa: bcm_sf2: Do not clobber b53_switch_ops
net: stmmac: fix maxmtu assignment to be within valid range
bpf: change back to orig prog on too many passes
tg3: Fix race condition in tg3_get_stats64().
be2net: fix unicast list filling
be2net: fix accesses to unicast list
netlabel: add CALIPSO to the list of built-in protocols
vti6: fix device register to report IFLA_INFO_KIND
net: phy: dp83867: fix irq generation
amd-xgbe: Fix IRQ processing when running in single IRQ mode
sh_eth: R8A7740 supports packet shecksumming
sh_eth: fix EESIPR values for SH77{34|63}
r8169: fix the typo in the comment
nl80211: fix sched scan netlink socket owner destruction
bridge: netfilter: Fix dropping packets that moving through bridge interface
netfilter: ipt_CLUSTERIP: check duplicate config when initializing
netfilter: nft_payload: mangle ckecksum if NFT_PAYLOAD_L4CSUM_PSEUDOHDR is set
netfilter: nf_tables: fix oob access
netfilter: nft_queue: use raw_smp_processor_id()
...
David S. Miller [Mon, 9 Jan 2017 19:54:30 +0000 (14:54 -0500)]
Merge branch 'dwmac-dwc-qos-eth'
Joao Pinto says:
====================
adding new glue driver dwmac-dwc-qos-eth
This patch set contains the porting of the synopsys/dwc_eth_qos.c driver
to the stmmac structure. This operation resulted in the creation of a new
platform glue driver called dwmac-dwc-qos-eth which was based in the
dwc_eth_qos as is.
dwmac-dwc-qos-eth inherited dwc_eth_qos DT bindings, to assure that current
and old users can continue to use it as before. We can see this driver as
being deprecated, since all new development will be done in stmmac.
Please check each patch for implementation details.
====================
Tested-by: Niklas Cassel <niklas.cassel@axis.com> Reviewed-by: Lars Persson <larper@axis.com> Acked-by: Alexandre TORGUE <alexandre.torgue@st.com> Signed-off-by: David S. Miller <davem@davemloft.net>
jpinto [Mon, 9 Jan 2017 12:35:10 +0000 (12:35 +0000)]
stmmac: adding new glue driver dwmac-dwc-qos-eth
This patch adds a new glue driver called dwmac-dwc-qos-eth which
was based in the dwc_eth_qos as is. To assure retro-compatibility a slight
tweak was also added to stmmac_platform.
Signed-off-by: Joao Pinto <jpinto@synopsys.com> Tested-by: Niklas Cassel <niklas.cassel@axis.com> Reviewed-by: Lars Persson <larper@axis.com> Acked-by: Alexandre TORGUE <alexandre.torgue@st.com> Signed-off-by: David S. Miller <davem@davemloft.net>
jpinto [Mon, 9 Jan 2017 12:35:09 +0000 (12:35 +0000)]
stmmac: move stmmac_clk, pclk, clk_ptp_ref and stmmac_rst to platform structure
This patch moves stmmac_clk, pclk, clk_ptp_ref and stmmac_rst to the
plat_stmmacenet_data structure. It also moves these platform variables
initialization to stmmac_platform. This was done for two reasons:
a) If PCI is used, platform related code is being executed in stmmac_main
resulting in warnings that have no sense and conceptually was not right
b) stmmac as a synopsys reference ethernet driver stack will be hosting
more and more drivers to its structure like synopsys/dwc_eth_qos.c.
These drivers have their own DT bindings that are not compatible with
stmmac's. One of the most important are the clock names, and so they need
to be parsed in the glue logic and initialized there, and that is the main
reason why the clocks were passed to the platform structure.
Signed-off-by: Joao Pinto <jpinto@synopsys.com> Tested-by: Niklas Cassel <niklas.cassel@axis.com> Reviewed-by: Lars Persson <larper@axis.com> Acked-by: Alexandre TORGUE <alexandre.torgue@st.com> Signed-off-by: David S. Miller <davem@davemloft.net>
jpinto [Mon, 9 Jan 2017 12:35:08 +0000 (12:35 +0000)]
stmmac: adding DT parameter for LPI tx clock gating
This patch adds a new parameter to the stmmac DT: snps,en-tx-lpi-clockgating.
It was ported from synopsys/dwc_eth_qos.c and it is useful if lpi tx clock
gating is needed by stmmac users also.
Signed-off-by: Joao Pinto <jpinto@synopsys.com> Tested-by: Niklas Cassel <niklas.cassel@axis.com> Reviewed-by: Lars Persson <larper@axis.com> Acked-by: Alexandre TORGUE <alexandre.torgue@st.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tobias Regnery [Mon, 9 Jan 2017 12:15:55 +0000 (13:15 +0100)]
alx: add feature flag for rx checksumming
The code to handle rx checksumming was in the driver since its introduction
but for reasons unknown the feature flag was left out. Now it is possible
to enable this feature with ethtool.
Tested on my AR8161 ethernet card, there are no regressions observed in
netperf if this feature is enabled.
Signed-off-by: Tobias Regnery <tobias.regnery@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
projects / mirror_ubuntu-bionic-kernel.git / log