Julian Wiedmann [Tue, 7 Dec 2021 09:04:48 +0000 (10:04 +0100)]
s390/qeth: simplify qeth_receive_skb()
Now that the OSN code is gone, we don't need the second switch statement
in the RX path. And getting rid of its (unreachable) default case is a
nice simplification.
Also don't pass in the full HW header, all we still need is a flag to
indicate whether the skb can use CSO. This we can already obtain during
the first peek at the HW header.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com> Reviewed-by: Alexandra Winter <wintera@linux.ibm.com> Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com> Signed-off-by: Alexandra Winter <wintera@linux.ibm.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Kees Cook [Tue, 7 Dec 2021 06:32:17 +0000 (22:32 -0800)]
hv_sock: Extract hvs_send_data() helper that takes only header
When building under -Warray-bounds, the compiler is especially
conservative when faced with casts from a smaller object to a larger
object. While this has found many real bugs, there are some cases that
are currently false positives (like here). With this as one of the last
few instances of the warning in the kernel before -Warray-bounds can be
enabled globally, rearrange the functions so that there is a header-only
version of hvs_send_data(). Silences this warning:
net/vmw_vsock/hyperv_transport.c: In function 'hvs_shutdown_lock_held.constprop':
net/vmw_vsock/hyperv_transport.c:231:32: warning: array subscript 'struct hvs_send_buf[0]' is partly outside array bounds of 'struct vmpipe_proto_header[1]' [-Warray-bounds]
231 | send_buf->hdr.pkt_type = 1;
| ~~~~~~~~~~~~~~~~~~~~~~~^~~
net/vmw_vsock/hyperv_transport.c:465:36: note: while referencing 'hdr'
465 | struct vmpipe_proto_header hdr;
| ^~~
This change results in no executable instruction differences.
Yihao Han [Tue, 7 Dec 2021 06:44:18 +0000 (22:44 -0800)]
net: dsa: felix: use kmemdup() to replace kmalloc + memcpy
Fix following coccicheck warning:
/drivers/net/dsa/ocelot/felix_vsc9959.c:1627:13-20:
WARNING opportunity for kmemdup
/drivers/net/dsa/ocelot/felix_vsc9959.c:1506:16-23:
WARNING opportunity for kmemdup
====================
prepare ocelot for external interface control
This patch set is derived from an attempt to include external control
for a VSC751[1234] chip via SPI. That patch set has grown large and is
getting unwieldy for reviewers and the developers... me.
I'm breaking out the changes from that patch set. Some are trivial
net: dsa: ocelot: remove unnecessary pci_bar variables
net: dsa: ocelot: felix: Remove requirement for PCS in felix devices
some are required for SPI
net: dsa: ocelot: felix: add interface for custom regmaps
and some are just to expose code to be shared
net: mscc: ocelot: split register definitions to a separate file
The entirety of this patch set should have essentially no impact on the
system performance.
====================
Colin Foster [Tue, 7 Dec 2021 17:00:28 +0000 (09:00 -0800)]
net: dsa: ocelot: felix: Remove requirement for PCS in felix devices
Existing felix devices all have an initialized pcs array. Future devices
might not, so running a NULL check on the array before dereferencing it
will allow those future drivers to not crash at this point
Signed-off-by: Colin Foster <colin.foster@in-advantage.com> Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The pci_bar variables for the switch and imdio don't make sense for the
generic felix driver. Moving them to felix_vsc9959 to limit scope and
simplify the felix_info struct.
Signed-off-by: Colin Foster <colin.foster@in-advantage.com> Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jakub Kicinski [Wed, 8 Dec 2021 05:01:16 +0000 (21:01 -0800)]
Merge tag 'wireless-drivers-next-2021-12-07' of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers-next
Kalle Valo says:
====================
wireless-drivers-next patches for v5.17
First set of patches for v5.17. The biggest change is the iwlmei
driver for Intel's AMT devices. Also now WCN6855 support in ath11k
should be usable.
Major changes:
ath10k
* fetch (pre-)calibration data via nvmem subsystem
ath11k
* enable 802.11 power save mode in station mode for qca6390 and wcn6855
* trace log support
* proper board file detection for WCN6855 based on PCI ids
* BSS color change support
rtw88
* add debugfs file to force lowest basic rate
* add quirk to disable PCI ASPM on HP 250 G7 Notebook PC
mwifiex
* add quirk to disable deep sleep with certain hardware revision in
Surface Book 2 devices
iwlwifi
* add iwlmei driver for co-operating with Intel's Active Management
Technology (AMT) devices
* tag 'wireless-drivers-next-2021-12-07' of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers-next: (87 commits)
iwlwifi: mei: fix linking when tracing is not enabled
rtlwifi: rtl8192de: Style clean-ups
mwl8k: Use named struct for memcpy() region
intersil: Use struct_group() for memcpy() region
libertas_tf: Use struct_group() for memcpy() region
libertas: Use struct_group() for memcpy() region
wlcore: no need to initialise statics to false
rsi: Fix out-of-bounds read in rsi_read_pkt()
rsi: Fix use-after-free in rsi_rx_done_handler()
brcmfmac: Configure keep-alive packet on suspend
wilc1000: remove '-Wunused-but-set-variable' warning in chip_wakeup()
iwlwifi: mvm: read the rfkill state and feed it to iwlmei
iwlwifi: mvm: add vendor commands needed for iwlmei
iwlwifi: integrate with iwlmei
iwlwifi: mei: add debugfs hooks
iwlwifi: mei: add the driver to allow cooperation with CSME
mei: bus: add client dma interface
mwifiex: Ignore BTCOEX events from the 88W8897 firmware
mwifiex: Ensure the version string from the firmware is 0-terminated
mwifiex: Add quirk to disable deep sleep with certain hardware revision
...
====================
====================
net: second round of netdevice refcount tracking
The most interesting part of this series is probably
("inet: add net device refcount tracker to struct fib_nh_common")
but only future reports will confirm this guess.
====================
====================
mptcp: New features for MPTCP sockets and netlink PM
This collection of patches adds MPTCP socket support for a few socket
options, ioctls, and one ancillary data type (specifics for each are
listed below). There's also a patch modifying the netlink MPTCP path
manager API to allow setting the backup flag on a configured interface
using the endpoint ID instead of the full IP address.
Patches 1 & 2: TCP_INQ cmsg and selftests.
Patches 2 & 3: SIOCINQ, OUTQ, and OUTQNSD ioctls and selftests.
Patch 5: Change backup flag using endpoint ID.
Patches 6 & 7: IP_TOS socket option and selftests.
Patches 8-10: TCP_CORK and TCP_NODELAY socket options. Includes a tcp
change to expose __tcp_sock_set_cork() and __tcp_sock_set_nodelay() for
use by MPTCP.
====================
Maxim Galaganov [Fri, 3 Dec 2021 22:35:41 +0000 (14:35 -0800)]
mptcp: support TCP_CORK and TCP_NODELAY
First, add cork and nodelay fields to the mptcp_sock structure
so they can be used in sync_socket_options(), and fill them on setsockopt
while holding the msk socket lock.
Then, on setsockopt set proper tcp_sk(ssk)->nonagle values for subflows
by calling __tcp_sock_set_cork() or __tcp_sock_set_nodelay() on the ssk
while holding the ssk socket lock.
tcp_push_pending_frames() will be invoked on the ssk if a cork was cleared
or nodelay was set. Also set MPTCP_PUSH_PENDING bit by calling
mptcp_check_and_set_pending(). This will lead to __mptcp_push_pending()
being called inside mptcp_release_cb() with new tcp_sk(ssk)->nonagle.
Also add getsockopt support for TCP_CORK and TCP_NODELAY.
Acked-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net> Signed-off-by: Maxim Galaganov <max@internet.ru> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Maxim Galaganov [Fri, 3 Dec 2021 22:35:40 +0000 (14:35 -0800)]
mptcp: expose mptcp_check_and_set_pending
Expose the mptcp_check_and_set_pending() function for use inside MPTCP
sockopt code. The next patch will call it when TCP_CORK is cleared or
TCP_NODELAY is set on the MPTCP socket in order to push pending data
from mptcp_release_cb().
Acked-by: Paolo Abeni <pabeni@redhat.com> Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net> Signed-off-by: Maxim Galaganov <max@internet.ru> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Maxim Galaganov [Fri, 3 Dec 2021 22:35:39 +0000 (14:35 -0800)]
tcp: expose __tcp_sock_set_cork and __tcp_sock_set_nodelay
Expose __tcp_sock_set_cork() and __tcp_sock_set_nodelay() for use in
MPTCP setsockopt code -- namely for syncing MPTCP socket options with
subflows inside sync_socket_options() while already holding the subflow
socket lock.
Acked-by: Paolo Abeni <pabeni@redhat.com> Acked-by: Matthieu Baerts <matthieu.baerts@tessares.net> Signed-off-by: Maxim Galaganov <max@internet.ru> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Florian Westphal [Fri, 3 Dec 2021 22:35:35 +0000 (14:35 -0800)]
selftests: mptcp: add inq test case
client & server use a unix socket connection to communicate
outside of the mptcp connection.
This allows the consumer to know in advance how many bytes have been
(or will be) sent by the peer.
This allows stricter checks on the bytecounts reported by TCP_INQ cmsg.
Suggested-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Florian Westphal [Fri, 3 Dec 2021 22:35:33 +0000 (14:35 -0800)]
selftests: mptcp: add TCP_INQ support
Do checks on the returned inq counter.
Fail on:
1. Huge value (> 1 kbyte, test case files are 1 kb)
2. last hint larger than returned bytes when read was short
3. erronenous indication of EOF.
3) happens when a hint of X bytes reads X-1 on next call
but next recvmsg returns more data (instead of EOF).
Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
If we do not transfer dst->dev_tracker to the new device,
we will get warnings from ref_tracker_dir_exit() when odev
is finally dismantled.
Fixes: 9038c320001d ("net: dst: add net device refcount tracking to dst_entry") Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: David Ahern <dsahern@kernel.org> Link: https://lore.kernel.org/r/20211207055603.1926372-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Tony Lu [Fri, 3 Dec 2021 11:33:31 +0000 (12:33 +0100)]
net/smc: Clear memory when release and reuse buffer
Currently, buffers are cleared when smc connections are created and
buffers are reused. This slows down the speed of establishing new
connections. In most cases, the applications want to establish
connections as quickly as possible.
This patch moves memset() from connection creation path to release and
buffer unuse path, this trades off between speed of establishing and
release.
Test environments:
- CPU Intel Xeon Platinum 8 core, mem 32 GiB, nic Mellanox CX4
- socket sndbuf / rcvbuf: 16384 / 131072 bytes
- w/o first round, 5 rounds, avg, 100 conns batch per round
- smc_buf_create() use bpftrace kprobe, introduces extra latency
Latency benchmarks for smc_buf_create():
w/o patch : 19040.0 ns
w/ patch : 1932.6 ns
ratio : 10.2% (-89.8%)
Latency benchmarks for socket create and connect:
w/o patch : 143.3 us
w/ patch : 102.2 us
ratio : 71.3% (-28.7%)
The latency of establishing connections is reduced by 28.7%.
net: prestera: replace zero-length array with flexible-array member
One-element and zero-length arrays are deprecated and should be
replaced with flexible-array members:
https://www.kernel.org/doc/html/latest/process/deprecated.html#zero-length-and-one-element-arrays
Replace zero-length array with flexible-array member and make use
of the struct_size() helper.
Arnd Bergmann [Sat, 4 Dec 2021 17:40:25 +0000 (18:40 +0100)]
net: wwan: iosm: select CONFIG_RELAY
The iosm driver started using relayfs, but is missing the Kconfig
logic to ensure it's built into the kernel:
x86_64-linux-ld: drivers/net/wwan/iosm/iosm_ipc_trace.o: in function `ipc_trace_create_buf_file_handler':
iosm_ipc_trace.c:(.text+0x16): undefined reference to `relay_file_operations'
x86_64-linux-ld: drivers/net/wwan/iosm/iosm_ipc_trace.o: in function `ipc_trace_subbuf_start_handler':
iosm_ipc_trace.c:(.text+0x31): undefined reference to `relay_buf_full'
x86_64-linux-ld: drivers/net/wwan/iosm/iosm_ipc_trace.o: in function `ipc_trace_ctrl_file_write':
iosm_ipc_trace.c:(.text+0xd5): undefined reference to `relay_flush'
x86_64-linux-ld: drivers/net/wwan/iosm/iosm_ipc_trace.o: in function `ipc_trace_port_rx':
Fixes: 00ef32565b9b ("net: wwan: iosm: device trace collection using relayfs") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Sergey Ryazanov <ryazanov.s.a@gmail.com> Reviewed-by: M Chetan Kumar <m.chetan.kumar@linux.intel.com> Link: https://lore.kernel.org/r/20211204174033.950528-1-arnd@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Eric Dumazet [Sat, 4 Dec 2021 04:53:56 +0000 (20:53 -0800)]
net: fix recent csum changes
Vladimir reported csum issues after my recent change in skb_postpull_rcsum()
Issue here is the following:
initial skb->csum is the csum of
[part to be pulled][rest of packet]
Old code:
skb->csum = csum_sub(skb->csum, csum_partial(pull, pull_length, 0));
New code:
skb->csum = ~csum_partial(pull, pull_length, ~skb->csum);
This is broken if the csum of [pulled part]
happens to be equal to skb->csum, because end
result of skb->csum is 0 in new code, instead
of being 0xffffffff
Two first patches add a generic infrastructure, that will be used
to get tracking of refcount increments/decrements.
The general idea is to be able to precisely pair each decrement with
a corresponding prior increment. Both share a cookie, basically
a pointer to private data storing stack traces.
The third patch adds dev_hold_track() and dev_put_track() helpers
(CONFIG_NET_DEV_REFCNT_TRACKER)
Then a series of 20 patches converts some dev_hold()/dev_put()
pairs to new hepers : dev_hold_track() and dev_put_track().
Hopefully this will be used by developpers and syzbot to
root cause bugs that cause netdevice dismantles freezes.
With CONFIG_PCPU_DEV_REFCNT=n option, we were able to detect
some class of bugs, but too late (when too many dev_put()
were happening).
Another series will be sent after this one is merged.
v3: moved NET_DEV_REFCNT_TRACKER to net/Kconfig.debug
added "depends on DEBUG_KERNEL && STACKTRACE_SUPPORT"
to hopefully get rid of kbuild reports for ARCH=nios2
Reworded patch 3 changelog.
Added missing htmldocs (Jakub)
v2: added four additional patches,
added netdev_tracker_alloc() and netdev_tracker_free()
addressed build error (kernel bots),
use GFP_ATOMIC in test_ref_tracker_timer_func()
====================
Eric Dumazet [Sun, 5 Dec 2021 04:21:57 +0000 (20:21 -0800)]
net: add net device refcount tracker infrastructure
net device are refcounted. Over the years we had numerous bugs
caused by imbalanced dev_hold() and dev_put() calls.
The general idea is to be able to precisely pair each decrement with
a corresponding prior increment. Both share a cookie, basically
a pointer to private data storing stack traces.
This patch adds dev_hold_track() and dev_put_track().
To use these helpers, each data structure owning a refcount
should also use a "netdevice_tracker" to pair the hold and put.
It can be hard to track where references are taken and released.
In networking, we have annoying issues at device or netns dismantles,
and we had various proposals to ease root causing them.
This patch adds new infrastructure pairing refcount increases
and decreases. This will self document code, because programmers
will have to associate increments/decrements.
This is controled by CONFIG_REF_TRACKER which can be selected
by users of this feature.
This adds both cpu and memory costs, and thus should probably be
used with care.
Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Manish Chopra [Thu, 2 Dec 2021 21:01:57 +0000 (13:01 -0800)]
qed*: esl priv flag support through ethtool
ESL(Enhanced System Lockdown) was designed to lock PCI adapter firmware
images and prevent changes to critical non-volatile configuration data
so that uncontrolled, malicious or unintentional modification to the
adapters are avoided, ensuring it's operational state. Once this feature is
enabled, the device is locked, rejecting any modification to non-volatile
images. Once unlocked, the protection is off such that firmware and
non-volatile configurations may be altered.
Driver just reflects the capability and status of this through
the ethtool private flag.
Manish Chopra [Thu, 2 Dec 2021 21:01:56 +0000 (13:01 -0800)]
qed*: enhance tx timeout debug info
This patch add some new qed APIs to query status block
info and report various data to MFW on tx timeout event
Along with that it enhances qede to dump more debug logs
(not just specific to the queue which was reported by stack)
on tx timeout which includes various other basic metadata about
all tx queues and other info (like status block etc.)
Dan Carpenter [Fri, 3 Dec 2021 09:55:31 +0000 (12:55 +0300)]
net: lan966x: fix a IS_ERR() vs NULL check in lan966x_create_targets()
The devm_ioremap() function does not return error pointers. It returns
NULL.
Fixes: db8bcaad5393 ("net: lan966x: add the basic lan966x driver") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yang Yingliang [Fri, 3 Dec 2021 07:04:18 +0000 (15:04 +0800)]
net: prestera: acl: fix return value check in prestera_acl_rule_entry_find()
rhashtable_lookup_fast() returns NULL pointer not ERR_PTR().
Return rhashtable_lookup_fast() directly to fix this.
Fixes: 47327e198d42 ("net: prestera: acl: migrate to new vTCAM api") Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andy Shevchenko [Thu, 2 Dec 2021 21:00:29 +0000 (23:00 +0200)]
net: dsa: vsc73xxx: Get rid of duplicate of_node assignment
GPIO library does copy the of_node from the parent device of
the GPIO chip, there is no need to repeat this in the individual
drivers. Remove assignment here.
For the details one may look into the of_gpio_dev_init() implementation.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Chris Mi [Wed, 1 Dec 2021 13:31:53 +0000 (15:31 +0200)]
net/sched: act_ct: Offload only ASSURED connections
Short-lived connections increase the insertion rate requirements,
fill the offload table and provide very limited offload value since
they process a very small amount of packets. The ct ASSURED flag is
designed to filter short-lived connections for early expiration.
Offload connections when they are ESTABLISHED and ASSURED.
Signed-off-by: Chris Mi <cmi@nvidia.com> Reviewed-by: Oz Shlomo <ozsh@nvidia.com> Reviewed-by: Roi Dayan <roid@nvidia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jie Wang [Fri, 3 Dec 2021 09:20:59 +0000 (17:20 +0800)]
net: hns3: fix hns3 driver header file not self-contained issue
The hns3 driver header file uses the structure of other files, but does
not include corresponding file, which causes a check warning that the
header file is not self-contained.
Therefore, the required header file is included in the header file, and
the structure declaration is added to the header file to avoid cyclic
dependency of the header file.
Signed-off-by: Jie Wang <wangjie125@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Hao Chen [Fri, 3 Dec 2021 09:20:58 +0000 (17:20 +0800)]
net: hns3: replace one tab with space in for statement
Replace one tab with space between symbol ')' and '{' in for statement of
function hclge_map_tqp().
Signed-off-by: Hao Chen <chenhao288@hisilicon.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Hao Chen [Fri, 3 Dec 2021 09:20:57 +0000 (17:20 +0800)]
net: hns3: remove rebundant line for hclge_dbg_dump_tm_pg()
Return value judgment should follow the function call, so remove line
between them.
Signed-off-by: Hao Chen <chenhao288@hisilicon.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Hao Chen [Fri, 3 Dec 2021 09:20:56 +0000 (17:20 +0800)]
net: hns3: add comments for hclge_dbg_fill_content()
When we use hclge_dbg_fill_content() to fill contents with
specific format according to struct hclge_dbg_item *items,
it may cause content cover due to unreasonable items.
So add comments to explain how to avoid it.
Signed-off-by: Hao Chen <chenhao288@hisilicon.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Hao Chen [Fri, 3 Dec 2021 09:20:55 +0000 (17:20 +0800)]
net: hns3: add void before function which don't receive ret
Add void before function which don't receive ret to improve code
readability.
Signed-off-by: Hao Chen <chenhao288@hisilicon.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Hao Chen [Fri, 3 Dec 2021 09:20:54 +0000 (17:20 +0800)]
net: hns3: align return value type of atomic_read() with its output
Change output value type of atomic_read() from %u to %d.
Signed-off-by: Hao Chen <chenhao288@hisilicon.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Hao Chen [Fri, 3 Dec 2021 09:20:52 +0000 (17:20 +0800)]
net: hns3: Align type of some variables with their print type
The c language has a set of implicit type conversions, when
two variables perform bitwise or arithmetic operations.
For example, variable A (type u16/u8) -1, its output is int type variable.
u16/u8 will convert to int type implicitly before it does arithmetic
operations. So, change 1 to unsigned type.
Signed-off-by: Hao Chen <chenhao288@hisilicon.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Guangbin Huang [Fri, 3 Dec 2021 09:20:50 +0000 (17:20 +0800)]
net: hns3: refactor function hclge_set_vlan_filter_hw
Function hclge_set_vlan_filter_hw() is a bit too long, so add a new
function hclge_need_update_port_vlan() to simplify code and improve
code readability.
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yufeng Mo [Fri, 3 Dec 2021 09:20:49 +0000 (17:20 +0800)]
net: hns3: optimize function hclge_cfg_common_loopback()
hclge_cfg_common_loopback() is a bit too long, so
encapsulate hclge_cfg_common_loopback_cmd_send() and
hclge_cfg_common_loopback_wait() two functions to
improve readability.
Signed-off-by: Yufeng Mo <moyufeng@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Fri, 3 Dec 2021 03:00:17 +0000 (19:00 -0800)]
Merge tag 'mlx5-updates-2021-12-02' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
Saeed Mahameed says:
====================
mlx5-updates-2021-12-02
Misc updates to mlx5 driver
1) Various code cleanups
2) Error path handling fixes of latest features
3) Print more information on pci error handling
4) Dynamically resize flow counters query buffer
====================
The flow counters bulk query buffer is allocated once during
mlx5_fc_init_stats(). For PFs and VFs this buffer usually takes a little
more than 512KB of memory, which is aligned to the next power of 2, to
1MB. For SFs, this buffer is reduced and takes around 128 Bytes.
The buffer size determines the maximum number of flow counters that
can be queried at a time. Thus, having a bigger buffer can improve
performance for users that need to query many flow counters.
There are cases that don't use many flow counters and don't need a big
buffer (e.g. SFs, VFs). Since this size is critical with large scale,
in these cases the buffer size should be reduced.
In order to reduce memory consumption while maintaining query
performance, change the query buffer's allocation scheme to the
following:
- First allocate the buffer with small initial size.
- If the number of counters surpasses the initial size, resize the
buffer to the maximum size.
The buffer only grows and isn't shrank, because users with many flow
counters don't care about the buffer size and we don't want to add
resize overhead if the current number of counters drops.
This solution is preferable to the current one, which is less accurate
and only addresses SFs.
Roi Dayan [Tue, 9 Nov 2021 12:07:49 +0000 (14:07 +0200)]
net/mlx5e: TC, Set flow attr ip_version earlier
Setting flow attr ip_version is not related to parsing tc flow actions.
It needs to be set after parsing flower matches which changes the spec.
So move it outside parse_tc_fdb_actions() and set it in
__mlx5e_add_fdb_flow().
Signed-off-by: Roi Dayan <roid@nvidia.com> Reviewed-by: Oz Shlomo <ozsh@nvidia.com> Reviewed-by: Maor Dickman <maord@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
net/mlx5e: Hide function mlx5e_num_channels_changed
No calls for mlx5e_num_channels_changed() out of en_main.c,
turn it static and remove from header.
Keep the wrapper function mlx5e_num_channels_changed_ctx exposed.
Dan Carpenter [Sat, 27 Nov 2021 14:19:53 +0000 (17:19 +0300)]
net/mlx5: SF, silence an uninitialized variable warning
This code sometimes calls mlx5_sf_hw_table_hwc_init() when "ext_base_id"
is uninitialized. It's not used on that path, but it generates a static
checker warning to pass uninitialized variables to another function.
It may also generate runtime UBSan warnings depending on if the
mlx5_sf_hw_table_hwc_init() function is inlined or not.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Arnd Bergmann [Mon, 8 Nov 2021 11:10:33 +0000 (12:10 +0100)]
mlx5: fix mlx5i_grp_sw_update_stats() stack usage
The mlx5e_sw_stats structure has grown to the point of triggering
a warning when put on the stack of a function:
mlx5/core/ipoib/ipoib.c: In function 'mlx5i_grp_sw_update_stats':
mlx5/core/ipoib/ipoib.c:136:1: error: the frame size of 1028 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
In this case, only five of the structure members are actually set,
so it's sufficient to have those as separate local variables.
As en_rep.c uses 'struct rtnl_link_stats64' for this, just use
the same one here for consistency.
projects / users / dwmw2 / linux.git / log