Ansuel Smith [Wed, 2 Feb 2022 00:03:27 +0000 (01:03 +0100)]
net: dsa: tag_qca: add support for handling mgmt and MIB Ethernet packet
Add connect/disconnect helper to assign private struct to the DSA switch.
Add support for Ethernet mgmt and MIB if the DSA driver provide an handler
to correctly parse and elaborate the data.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Wed, 2 Feb 2022 00:03:26 +0000 (01:03 +0100)]
net: dsa: tag_qca: add define for handling MIB packet
Add struct to correctly parse a mib Ethernet packet.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Wed, 2 Feb 2022 00:03:25 +0000 (01:03 +0100)]
net: dsa: tag_qca: add define for handling mgmt Ethernet packet
Add all the required define to prepare support for mgmt read/write in
Ethernet packet. Any packet of this type has to be dropped as the only
use of these special packet is receive ack for an mgmt write request or
receive data for an mgmt read request.
A struct is used that emulates the Ethernet header but is used for a
different purpose.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Wed, 2 Feb 2022 00:03:24 +0000 (01:03 +0100)]
net: dsa: tag_qca: enable promisc_on_master flag
Ethernet MDIO packets are non-standard and DSA master expects the first
6 octets to be the MAC DA. To address these kind of packet, enable
promisc_on_master flag for the tagger.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Wed, 2 Feb 2022 00:03:23 +0000 (01:03 +0100)]
net: dsa: tag_qca: move define to include linux/dsa
Move tag_qca define to include dir linux/dsa as the qca8k require access
to the tagger define to support in-band mdio read/write using ethernet
packet.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ansuel Smith [Wed, 2 Feb 2022 00:03:22 +0000 (01:03 +0100)]
net: dsa: tag_qca: convert to FIELD macro
Convert driver to FIELD macro to drop redundant define.
Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Vladimir Oltean [Wed, 2 Feb 2022 00:03:21 +0000 (01:03 +0100)]
net: dsa: replay master state events in dsa_tree_{setup,teardown}_master
In order for switch driver to be able to make simple and reliable use of
the master tracking operations, they must also be notified of the
initial state of the DSA master, not just of the changes. This is
because they might enable certain features only during the time when
they know that the DSA master is up and running.
Therefore, this change explicitly checks the state of the DSA master
under the same rtnl_mutex as we were holding during the
dsa_master_setup() and dsa_master_teardown() call. The idea being that
if the DSA master became operational in between the moment in which it
became a DSA master (dsa_master_setup set dev->dsa_ptr) and the moment
when we checked for the master being up, there is a chance that we
would emit a ->master_state_change() call with no actual state change.
We need to avoid that by serializing the concurrent netdevice event with
us. If the netdevice event started before, we force it to finish before
we begin, because we take rtnl_lock before making netdev_uses_dsa()
return true. So we also handle that early event and do nothing on it.
Similarly, if the dev_open() attempt is concurrent with us, it will
attempt to take the rtnl_mutex, but we're holding it. We'll see that
the master flag IFF_UP isn't set, then when we release the rtnl_mutex
we'll process the NETDEV_UP notifier.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Vladimir Oltean [Wed, 2 Feb 2022 00:03:20 +0000 (01:03 +0100)]
net: dsa: provide switch operations for tracking the master state
Certain drivers may need to send management traffic to the switch for
things like register access, FDB dump, etc, to accelerate what their
slow bus (SPI, I2C, MDIO) can already do.
Ethernet is faster (especially in bulk transactions) but is also more
unreliable, since the user may decide to bring the DSA master down (or
not bring it up), therefore severing the link between the host and the
attached switch.
Drivers needing Ethernet-based register access already should have
fallback logic to the slow bus if the Ethernet method fails, but that
fallback may be based on a timeout, and the I/O to the switch may slow
down to a halt if the master is down, because every Ethernet packet will
have to time out. The driver also doesn't have the option to turn off
Ethernet-based I/O momentarily, because it wouldn't know when to turn it
back on.
Which is where this change comes in. By tracking NETDEV_CHANGE,
NETDEV_UP and NETDEV_GOING_DOWN events on the DSA master, we should know
the exact interval of time during which this interface is reliably
available for traffic. Provide this information to switches so they can
use it as they wish.
An helper is added dsa_port_master_is_operational() to check if a master
port is operational.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Martin Habets [Mon, 31 Jan 2022 11:10:54 +0000 (11:10 +0000)]
sfc: The size of the RX recycle ring should be more flexible
Ideally the size would depend on the link speed, but the recycle
ring is created when the interface is brought up before the driver
knows the link speed. So size it for the maximum speed of a given NIC.
PowerPC is only supported on SFN7xxx and SFN8xxx NICs.
With this patch on a 40G NIC the number of calls to alloc_pages and
friends went down from about 18% to under 2%.
On a 10G NIC the number of calls to alloc_pages and friends went down
from about 15% to 0 (perf did not capture any calls during the 60
second test).
On a 100G NIC the number of calls to alloc_pages and friends went down
from about 23% to 4%.
Jakub Kicinski [Mon, 31 Jan 2022 23:33:57 +0000 (15:33 -0800)]
net: allow SO_MARK with CAP_NET_RAW via cmsg
There's not reason SO_MARK would be allowed via setsockopt()
and not via cmsg, let's keep the two consistent. See
commit 079925cce1d0 ("net: allow SO_MARK with CAP_NET_RAW")
for justification why NET_RAW -> SO_MARK is safe.
David S. Miller [Tue, 1 Feb 2022 14:18:44 +0000 (14:18 +0000)]
Merge branch 'lan966x-ptp'
Horatiu Vultur says:
====================
net: lan966x: Add PTP Hardward Clock support
Add support for PTP Hardware Clock (PHC) for lan966x. The switch supports
both PTP 1-step and 2-step modes.
v1->v2:
- fix commit messages
- reduce the scope of the lock ptp_lock inside the function
lan966x_ptp_hwtstamp_set
- the rx timestamping is always enabled for all packages
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Horatiu Vultur [Mon, 31 Jan 2022 10:01:21 +0000 (11:01 +0100)]
net: lan966x: Add support for ptp interrupts
When doing 2-step timestamping the HW will generate an interrupt when it
managed to timestamp a frame. It is the SW responsibility to read it
from the FIFO.
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Horatiu Vultur [Mon, 31 Jan 2022 10:01:20 +0000 (11:01 +0100)]
net: lan966x: Update extraction/injection for timestamping
Update both the extraction and injection to do timestamping of the
frames. The extraction is always doing the timestamping while for
injection is doing the timestamping only if it is configured.
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Horatiu Vultur [Mon, 31 Jan 2022 10:01:19 +0000 (11:01 +0100)]
net: lan966x: Implement SIOCSHWTSTAMP and SIOCGHWTSTAMP
Implement the ioctl callbacks SIOCSHWTSTAMP and SIOCGHWTSTAMP to allow
to configure the ports to enable/disable timestamping for TX. The RX
timestamping is always enabled. The HW is capable to run both 1-step
timestamping and 2-step timestamping.
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Horatiu Vultur [Mon, 31 Jan 2022 10:01:18 +0000 (11:01 +0100)]
net: lan966x: Add support for ptp clocks
The lan966x has 3 PHC. Enable each of them, for now all the
timestamping is happening on the first PHC.
Acked-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Guillaume Nault [Mon, 31 Jan 2022 15:42:06 +0000 (16:42 +0100)]
selftests: fib rule: Log test description
All callers of fib_rule6_test_match_n_redirect() and
fib_rule4_test_match_n_redirect() pass a third argument containing a
description of the test being run. Instead of ignoring this argument,
let's use it for logging instead of printing a truncated version of the
command.
Signed-off-by: Guillaume Nault <gnault@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Guillaume Nault [Mon, 31 Jan 2022 15:42:04 +0000 (16:42 +0100)]
selftests: fib rule: Drop erroneous TABLE variable
The fib_rule6_del_by_pref() and fib_rule4_del_by_pref() functions use
an uninitialised $TABLE variable. They should use $RTABLE instead.
This doesn't alter the result of the test, as it just makes the grep
command less specific (but since the script always uses the same table
number, that doesn't really matter).
Let's fix it anyway and, while there, specify the filtering parameters
directly in 'ip -X rule show' to avoid the extra grep command entirely.
Signed-off-by: Guillaume Nault <gnault@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 1 Feb 2022 12:53:18 +0000 (12:53 +0000)]
Merge branch '10GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next
-queue
Tony Nguyen says:
====================
10GbE Intel Wired LAN Driver Updates 2022-01-31
Alexander Lobakin says:
This is an interpolation of [0] to other Intel Ethernet drivers
(and is (re)based on its code).
The main aim is to keep XDP metadata not only in case with
build_skb(), but also when we do napi_alloc_skb() + memcpy().
All Intel drivers suffers from the same here:
- metadata gets lost on XDP_PASS in legacy-rx;
- excessive headroom allocation on XSK Rx to skbs;
- metadata gets lost on XSK Rx to skbs.
Those get especially actual in XDP Hints upcoming.
I couldn't have addressed the first one for all Intel drivers due to
that they don't reserve any headroom for now in legacy-rx mode even
with XDP enabled. This is hugely wrong, but requires quite a bunch
of work and a separate series. Luckily, ice doesn't suffer from
that.
igc has 1 and 3 already fixed in [0].
Hyeonggon Yoo [Sat, 29 Jan 2022 08:53:36 +0000 (08:53 +0000)]
net: ena: Do not waste napi skb cache
By profiling, discovered that ena device driver allocates skb by
build_skb() and frees by napi_skb_cache_put(). Because the driver
does not use napi skb cache in allocation path, napi skb cache is
periodically filled and flushed. This is waste of napi skb cache.
As ena_alloc_skb() is called only in napi, Use napi_build_skb()
and napi_alloc_skb() when allocating skb.
This patch was tested on aws a1.metal instance.
[ jwiedmann.dev@gmail.com: Use napi_alloc_skb() instead of
netdev_alloc_skb_ip_align() to keep things consistent. ]
qed: use msleep() in qed_mcp_cmd() and add qed_mcp_cmd_nosleep() for udelay.
Change qed_mcp_cmd() to use msleep() (by setting QED_MB_FLAG_CAN_SLEEP
flag) and add new nosleep() version of the api. These api are used to
issue cmds to management fw and the change affects how driver
behaves while waiting for a response/resource.
All sleepable callers of the existing api now use msleep() version. For
non-sleepable callers, the new nosleep() version is explicitly used.
For now, if the XDP prog returns XDP_PASS on XSK, the metadata
will be lost as it doesn't get copied to the skb.
Copy it along with the frame headers. Account its size on skb
allocation, and when copying just treat it as a part of the frame
and do a pull after to "move" it to the "reserved" zone.
net_prefetch() xdp->data_meta and align the copy size to speed-up
memcpy() a little and better match ixgbe_construct_skb().
Fixes: d0bcacd0a130 ("ixgbe: add AF_XDP zero-copy Rx support") Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com> Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
ixgbe: don't reserve excessive XDP_PACKET_HEADROOM on XSK Rx to skb
{__,}napi_alloc_skb() allocates and reserves additional NET_SKB_PAD
+ NET_IP_ALIGN for any skb.
OTOH, ixgbe_construct_skb_zc() currently allocates and reserves
additional `xdp->data - xdp->data_hard_start`, which is
XDP_PACKET_HEADROOM for XSK frames.
There's no need for that at all as the frame is post-XDP and will
go only to the networking stack core.
Pass the size of the actual data only to __napi_alloc_skb() and
don't reserve anything. This will give enough headroom for stack
processing.
Fixes: d0bcacd0a130 ("ixgbe: add AF_XDP zero-copy Rx support") Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
ixgbe: pass bi->xdp to ixgbe_construct_skb_zc() directly
To not dereference bi->xdp each time in ixgbe_construct_skb_zc(),
pass bi->xdp as an argument instead of bi. We can also call
xsk_buff_free() outside of the function as well as assign bi->xdp
to NULL, which seems to make it closer to its name.
Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Tested-by: Sandeep Penigalapati <sandeep.penigalapati@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
igc: don't reserve excessive XDP_PACKET_HEADROOM on XSK Rx to skb
{__,}napi_alloc_skb() allocates and reserves additional NET_SKB_PAD
+ NET_IP_ALIGN for any skb.
OTOH, igc_construct_skb_zc() currently allocates and reserves
additional `xdp->data_meta - xdp->data_hard_start`, which is about
XDP_PACKET_HEADROOM for XSK frames.
There's no need for that at all as the frame is post-XDP and will
go only to the networking stack core.
Pass the size of the actual data only (+ meta) to
__napi_alloc_skb() and don't reserve anything. This will give
enough headroom for stack processing.
Also, net_prefetch() xdp->data_meta and align the copy size to
speed-up memcpy() a little and better match igc_construct_skb().
Fixes: fc9df2a0b520 ("igc: Enable RX via AF_XDP zero-copy") Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Tested-by: Nechama Kraus <nechamax.kraus@linux.intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
For now, if the XDP prog returns XDP_PASS on XSK, the metadata will
be lost as it doesn't get copied to the skb.
Copy it along with the frame headers. Account its size on skb
allocation, and when copying just treat it as a part of the frame
and do a pull after to "move" it to the "reserved" zone.
net_prefetch() xdp->data_meta and align the copy size to speed-up
memcpy() a little and better match ice_construct_skb().
Fixes: 2d4238f55697 ("ice: Add support for AF_XDP") Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com> Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
ice: don't reserve excessive XDP_PACKET_HEADROOM on XSK Rx to skb
{__,}napi_alloc_skb() allocates and reserves additional NET_SKB_PAD
+ NET_IP_ALIGN for any skb.
OTOH, ice_construct_skb_zc() currently allocates and reserves
additional `xdp->data - xdp->data_hard_start`, which is
XDP_PACKET_HEADROOM for XSK frames.
There's no need for that at all as the frame is post-XDP and will
go only to the networking stack core.
Pass the size of the actual data only to __napi_alloc_skb() and
don't reserve anything. This will give enough headroom for stack
processing.
Fixes: 2d4238f55697 ("ice: Add support for AF_XDP") Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
ice: respect metadata in legacy-rx/ice_construct_skb()
In "legacy-rx" mode represented by ice_construct_skb(), we can
still use XDP (and XDP metadata), but after XDP_PASS the metadata
will be lost as it doesn't get copied to the skb.
Copy it along with the frame headers. Account its size on skb
allocation, and when copying just treat it as a part of the frame
and do a pull after to "move" it to the "reserved" zone.
Point net_prefetch() to xdp->data_meta instead of data. This won't
change anything when the meta is not here, but will save some cache
misses otherwise.
Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com> Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
For now, if the XDP prog returns XDP_PASS on XSK, the metadata will
be lost as it doesn't get copied to the skb.
Copy it along with the frame headers. Account its size on skb
allocation, and when copying just treat it as a part of the frame
and do a pull after to "move" it to the "reserved" zone.
net_prefetch() xdp->data_meta and align the copy size to speed-up
memcpy() a little and better match i40e_construct_skb().
Fixes: 0a714186d3c0 ("i40e: add AF_XDP zero-copy Rx support") Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com> Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
i40e: don't reserve excessive XDP_PACKET_HEADROOM on XSK Rx to skb
{__,}napi_alloc_skb() allocates and reserves additional NET_SKB_PAD
+ NET_IP_ALIGN for any skb.
OTOH, i40e_construct_skb_zc() currently allocates and reserves
additional `xdp->data - xdp->data_hard_start`, which is
XDP_PACKET_HEADROOM for XSK frames.
There's no need for that at all as the frame is post-XDP and will
go only to the networking stack core.
Pass the size of the actual data only to __napi_alloc_skb() and
don't reserve anything. This will give enough headroom for stack
processing.
Fixes: 0a714186d3c0 ("i40e: add AF_XDP zero-copy Rx support") Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Acked-by: Jesper Dangaard Brouer <brouer@redhat.com> Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
David S. Miller [Mon, 31 Jan 2022 15:08:20 +0000 (15:08 +0000)]
Merge branch 'smc-improvements'
Tony Lu says:
====================
net/smc: Improvements for TCP_CORK and sendfile()
Currently, SMC use default implement for syscall sendfile() [1], which
is wildly used in nginx and big data sences. Usually, applications use
sendfile() with TCP_CORK:
The above is an example of Nginx, when sendfile() on, Nginx first
enables TCP_CORK, write headers, the data will not be sent. Then call
sendfile(), it reads file and write to sndbuf. When TCP_CORK is cleared,
all pending data is sent out.
The performance of the default implement of sendfile is lower than when
it is off. After investigation, it shows two parts to improve:
- unnecessary lock contention of delayed work
- less data per send than when sendfile off
Patch #1 tries to reduce lock_sock() contention in smc_tx_work().
Patch #2 removes timed work for corking, and let applications control
it. See TCP_CORK [2] MSG_MORE [3].
Patch #3 adds MSG_SENDPAGE_NOTLAST for corking more data when
sendfile().
Test environments:
- CPU Intel Xeon Platinum 8 core, mem 32 GiB, nic Mellanox CX4
- socket sndbuf / rcvbuf: 16384 / 131072 bytes
- server: smc_run nginx
- client: smc_run ./wrk -c 100 -t 2 -d 30 http://192.168.100.1:8080/4k.html
- payload: 4KB local disk file
Items QPS
sendfile off 272477.10
sendfile on (orig) 223622.79
sendfile on (this) 395847.21
This benchmark shows +45.28% improvement compared with sendfile off, and
+77.02% compared with original sendfile implement.
Tony Lu [Sun, 30 Jan 2022 18:02:57 +0000 (02:02 +0800)]
net/smc: Cork when sendpage with MSG_SENDPAGE_NOTLAST flag
This introduces a new corked flag, MSG_SENDPAGE_NOTLAST, which is
involved in syscall sendfile() [1], it indicates this is not the last
page. So we can cork the data until the page is not specify this flag.
It has the same effect as MSG_MORE, but existed in sendfile() only.
This patch adds a option MSG_SENDPAGE_NOTLAST for corking data, try to
cork more data before sending when using sendfile(), which acts like
TCP's behaviour. Also, this reimplements the default sendpage to inform
that it is supported to some extent.
Tony Lu [Sun, 30 Jan 2022 18:02:56 +0000 (02:02 +0800)]
net/smc: Remove corked dealyed work
Based on the manual of TCP_CORK [1] and MSG_MORE [2], these two options
have the same effect. Applications can set these options and informs the
kernel to pend the data, and send them out only when the socket or
syscall does not specify this flag. In other words, there's no need to
send data out by a delayed work, which will queue a lot of work.
This removes corked delayed work with SMC_TX_CORK_DELAY (250ms), and the
applications control how/when to send them out. It improves the
performance for sendfile and throughput, and remove unnecessary race of
lock_sock(). This also unlocks the limitation of sndbuf, and try to fill
it up before sending.
Tony Lu [Sun, 30 Jan 2022 18:02:55 +0000 (02:02 +0800)]
net/smc: Send directly when TCP_CORK is cleared
According to the man page of TCP_CORK [1], if set, don't send out
partial frames. All queued partial frames are sent when option is
cleared again.
When applications call setsockopt to disable TCP_CORK, this call is
protected by lock_sock(), and tries to mod_delayed_work() to 0, in order
to send pending data right now. However, the delayed work smc_tx_work is
also protected by lock_sock(). There introduces lock contention for
sending data.
To fix it, send pending data directly which acts like TCP, without
lock_sock() protected in the context of setsockopt (already lock_sock()ed),
and cancel unnecessary dealyed work, which is protected by lock.
[1] https://linux.die.net/man/7/tcp
Signed-off-by: Tony Lu <tonylu@linux.alibaba.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 31 Jan 2022 15:05:25 +0000 (15:05 +0000)]
Merge branch 'hash-rethink'
Akhmat Karakotov says:
====================
Make hash rethink configurable
As it was shown in the report by Alexander Azimov, hash rethink at the
client-side may lead to connection timeout toward stateful anycast
services. Tom Herbert created a patchset to address this issue by applying
hash rethink only after a negative routing event (3RTOs) [1]. This change
also affects server-side behavior, which we found undesirable. This
patchset changes defaults in a way to make them safe: hash rethink at the
client-side is disabled and enabled at the server-side upon each RTO
event or in case of duplicate acknowledgments.
This patchset provides two options to change default behaviour. The hash
rethink may be disabled at the server-side by the new sysctl option.
Changes in the sysctl option don't affect default behavior at the
client-side.
Hash rethink can also be enabled/disabled with socket option or bpf
syscalls which ovewrite both default and sysctl settings. This socket
option is available on both client and server-side. This should provide
mechanics to enable hash rethink inside administrative domain, such as DC,
where hash rethink at the client-side can be desirable.
v2:
- Changed sysctl default to ENABLED in all patches. Reduced sysctl
and socket option size to u8. Fixed netns bug reported by kernel
test robot.
v3:
- Fixed bug with bad u8 comparison. Moved sk_txrehash to use less
bytes in struct. Added WRITE_ONCE() in setsockopt in and
READ_ONCE() in tcp_rtx_synack.
v4:
- Rebase and add documentation for sysctl option.
v5:
- Move sk_txrehash out of busy poll ifdef.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Akhmat Karakotov [Mon, 31 Jan 2022 13:31:25 +0000 (16:31 +0300)]
tcp: Change SYN ACK retransmit behaviour to account for rehash
Disabling rehash behavior did not affect SYN ACK retransmits because hash
was forcefully changed bypassing the sk_rethink_hash function. This patch
adds a condition which checks for rehash mode before resetting hash.
Signed-off-by: Akhmat Karakotov <hmukos@yandex-team.ru> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Akhmat Karakotov [Mon, 31 Jan 2022 13:31:24 +0000 (16:31 +0300)]
bpf: Add SO_TXREHASH setsockopt
Add bpf socket option to override rehash behaviour from userspace or from bpf.
Signed-off-by: Akhmat Karakotov <hmukos@yandex-team.ru> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Akhmat Karakotov [Mon, 31 Jan 2022 13:31:22 +0000 (16:31 +0300)]
txhash: Add socket option to control TX hash rethink behavior
Add the SO_TXREHASH socket option to control hash rethink behavior per socket.
When default mode is set, sockets disable rehash at initialization and use
sysctl option when entering listen state. setsockopt() overrides default
behavior.
Signed-off-by: Akhmat Karakotov <hmukos@yandex-team.ru> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Akhmat Karakotov [Mon, 31 Jan 2022 13:31:21 +0000 (16:31 +0300)]
txhash: Make rethinking txhash behavior configurable via sysctl
Add a per ns sysctl that controls the txhash rethink behavior:
net.core.txrehash. When enabled, the same behavior is retained,
when disabled, rethink is not performed. Sysctl is enabled by default.
Signed-off-by: Akhmat Karakotov <hmukos@yandex-team.ru> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Gerhard Engleder [Sun, 30 Jan 2022 09:54:22 +0000 (10:54 +0100)]
selftests/net: timestamping: Fix bind_phc check
timestamping checks socket options during initialisation. For the field
bind_phc of the socket option SO_TIMESTAMPING it expects the value -1 if
PHC is not bound. Actually the value of bind_phc is 0 if PHC is not
bound. This results in the following output:
This is fixed by setting default value and expected value of bind_phc to
0.
Fixes: 2214d7032479 ("selftests/net: timestamping: support binding PHC") Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 31 Jan 2022 11:42:13 +0000 (11:42 +0000)]
Merge branch 'renesas-dead-code'
Sergey Shtylyov says:
====================
Remove some dead code in the Renesas Ethernet drivers
Here are 2 patches against DaveM's 'net-next.git' repo. The Renesas drivers
call their ndo_stop() methods directly and they always return 0, making the
result checks pointless, hence remove them...
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Sergey Shtylyov [Sat, 29 Jan 2022 11:55:17 +0000 (14:55 +0300)]
sh_eth: sh_eth_close() always returns 0
sh_eth_close() always returns 0, hence the check in sh_eth_wol_restore()
is pointless (however we cannot change the prototype of sh_eth_close() as
it implements the driver's ndo_stop() method).
Found by Linux Verification Center (linuxtesting.org) with the SVACE static
analysis tool.
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru> Signed-off-by: David S. Miller <davem@davemloft.net>
Sergey Shtylyov [Sat, 29 Jan 2022 11:55:16 +0000 (14:55 +0300)]
ravb: ravb_close() always returns 0
ravb_close() always returns 0, hence the check in ravb_wol_restore() is
pointless (however, we cannot change the prototype of ravb_close() as it
implements the driver's ndo_stop() method).
Found by Linux Verification Center (linuxtesting.org) with the SVACE static
analysis tool.
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru> Signed-off-by: David S. Miller <davem@davemloft.net>
Wei Yongjun [Sat, 29 Jan 2022 01:27:02 +0000 (01:27 +0000)]
net/fsl: xgmac_mdio: fix return value check in xgmac_mdio_probe()
In case of error, the function devm_ioremap() returns NULL pointer
not ERR_PTR(). The IS_ERR() test in the return value check should
be replaced with NULL test.
Fixes: 1d14eb15dc2c ("net/fsl: xgmac_mdio: Use managed device resources") Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com> Reviewed-by: Tobias Waldekranz <tobias@waldekranz.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Fri, 28 Jan 2022 23:53:47 +0000 (16:53 -0700)]
ipv4: Make ip_idents_reserve static
ip_idents_reserve is only used in net/ipv4/route.c. Make it static
and remove the export.
Signed-off-by: David Ahern <dsahern@kernel.org> Cc: Eric Dumazet <edumazet@google.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Fri, 28 Jan 2022 20:41:42 +0000 (21:41 +0100)]
r8169: add rtl_disable_exit_l1()
Add rtl_disable_exit_l1() for ensuring that the chip doesn't
inadvertently exit ASPM L1 when being in a low-power mode.
The new function is called from rtl_prepare_power_down() which
has to be moved in the code to avoid a forward declaration.
According to Realtek OCP register 0xc0ac shadows ERI register 0xd4
on RTL8168 versions from RTL8168g. This allows to simplify the
code a little.
v2:
- call rtl_disable_exit_l1() also if DASH or WoL are enabled
Suggested-by: Chun-Hao Lin <hau@realtek.com> Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sergey Shtylyov [Fri, 28 Jan 2022 18:32:40 +0000 (21:32 +0300)]
phy: make phy_set_max_speed() *void*
After following the call tree of phy_set_max_speed(), it became clear
that this function never returns anything but 0, so we can change its
result type to *void* and drop the result checks from the three drivers
that actually bothered to do it...
Found by Linux Verification Center (linuxtesting.org) with the SVACE static
analysis tool.
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
The individual patches have all the details. This work was triggered
by recent work on a platform that took 16s (sic) to load the mv88e6xxx
module.
The first patch gets rid of most of that time by replacing a very long
delay with a tighter poll loop to wait for the busy bit to clear.
The second patch shaves off some more time by avoiding redundant
busy-bit-checks, saving 1 out of 4 MDIO operations for every register
read/write in the optimal case.
v1 -> v2:
- Make sure that we always poll the busy bit at least twice, in the
unlikely event that the first one is quick to query the hardware,
but is then scheduled out for a long time before the timeout is
checked.
v2 -> v3:
- Fallback to the longer sleeps after the initial two poll attempts.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Before this change, both the read and write callback would start out
by asserting that the chip's busy flag was cleared. However, both
callbacks also made sure to wait for the clearing of the busy bit
before returning - making the initial check superfluous. The only
time that would ever have an effect was if the busy bit was initially
set for some reason.
With that in mind, make sure to perform an initial check of the busy
bit, after which both read and write can rely the previous operation
to have waited for the bit to clear.
This cuts the number of operations on the underlying MDIO bus by 25%
Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
net: dsa: mv88e6xxx: Improve performance of busy bit polling
Avoid a long delay when a busy bit is still set and has to be polled
again.
Measurements on a system with 2 Opals (6097F) and one Agate (6352)
show that even with this much tighter loop, we have about a 50% chance
of the bit being cleared on the first poll, all other accesses see the
bit being cleared on the second poll.
On a standard MDIO bus running MDC at 2.5MHz, a single access with 32
bits of preamble plus 32 bits of data takes 64*(1/2.5MHz) = 25.6us.
This means that mv88e6xxx_smi_direct_wait took 26us + CPU overhead in
the fast scenario, but 26us + 1500us + 26us + CPU overhead in the slow
case - bringing the average close to 1ms.
With this change in place, the slow case is closer to 2*26us + CPU
overhead, with the average well below 100us - a 10x improvement.
This translates to real-world winnings. On a 3-chip 20-port system,
the modprobe time drops by 88%:
Before:
root@coronet:~# time modprobe mv88e6xxx
real 0m 15.99s
user 0m 0.00s
sys 0m 1.52s
After:
root@coronet:~# time modprobe mv88e6xxx
real 0m 2.21s
user 0m 0.00s
sys 0m 1.54s
Signed-off-by: Tobias Waldekranz <tobias@waldekranz.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Sun Shouxin [Fri, 28 Jan 2022 14:44:42 +0000 (09:44 -0500)]
net: bonding: Add support for IPV6 ns/na to balance-alb/balance-tlb mode
Since ipv6 neighbor solicitation and advertisement messages
isn't handled gracefully in bond6 driver, we can see packet
drop due to inconsistency between mac address in the option
message and source MAC .
Another examples is ipv6 neighbor solicitation and advertisement
messages from VM via tap attached to host bridge, the src mac
might be changed through balance-alb mode, but it is not synced
with Link-layer address in the option message.
The patch implements bond6's tx handle for ipv6 neighbor
solicitation and advertisement messages.
Suggested-by: Hu Yadi <huyd12@chinatelecom.cn> Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com> Signed-off-by: Sun Shouxin <sunshouxin@chinatelecom.cn> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sat, 29 Jan 2022 17:49:21 +0000 (17:49 +0000)]
Merge branch 'Cadence-ZyncMP-SGMII'
Robert Hancock says:
====================
Cadence MACB/GEM support for ZynqMP SGMII
Changes to allow SGMII mode to work properly in the GEM driver on the
Xilinx ZynqMP platform.
Changes since v3:
-more code formatting and error handling fixes
Changes since v2:
-fixed missing includes in DT binding example
-fixed phy_init and phy_power_on error handling/cleanup, moved
phy_power_on to open rather than probe
Changes since v1:
-changed order of controller reset and PHY init as per suggestion
-switched device reset to be optional
-updated bindings doc patch for switch to YAML
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Robert Hancock [Thu, 27 Jan 2022 16:37:36 +0000 (10:37 -0600)]
arm64: dts: zynqmp: Added GEM reset definitions
The Cadence GEM/MACB driver now utilizes the platform-level reset on the
ZynqMP platform. Add reset definitions to the ZynqMP platform device
tree to allow this to be used.
Signed-off-by: Robert Hancock <robert.hancock@calian.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Robert Hancock [Thu, 27 Jan 2022 16:37:35 +0000 (10:37 -0600)]
net: macb: Added ZynqMP-specific initialization
The GEM controllers on ZynqMP were missing some initialization steps which
are required in some cases when using SGMII mode, which uses the PS-GTR
transceivers managed by the phy-zynqmp driver.
The GEM core appears to need a hardware-level reset in order to work
properly in SGMII mode in cases where the GT reference clock was not
present at initial power-on. This can be done using a reset mapped to
the zynqmp-reset driver in the device tree.
Also, when in SGMII mode, the GEM driver needs to ensure the PHY is
initialized and powered on.
Signed-off-by: Robert Hancock <robert.hancock@calian.com> Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Robert Hancock [Thu, 27 Jan 2022 16:37:34 +0000 (10:37 -0600)]
dt-bindings: net: cdns,macb: added generic PHY and reset mappings for ZynqMP
Updated macb DT binding documentation to reflect the phy-names, phys,
resets, reset-names properties which are now used with ZynqMP GEM
devices, and added a ZynqMP-specific DT example.
Signed-off-by: Robert Hancock <robert.hancock@calian.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Fri, 28 Jan 2022 21:39:04 +0000 (13:39 -0800)]
Merge tag 'for-net-next-2022-01-28' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next
Luiz Augusto von Dentz says:
====================
bluetooth-next pull request for net-next:
- Add support for RTL8822C hci_ver 0x08
- Add support for RTL8852AE part 0bda:2852
- Fix WBS setting for Intel legacy ROM products
- Enable SCO over I2S ib mt7921s
- Increment management interface revision
* tag 'for-net-next-2022-01-28' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next: (30 commits)
Bluetooth: Increment management interface revision
Bluetooth: hci_sync: Fix queuing commands when HCI_UNREGISTER is set
Bluetooth: hci_h5: Add power reset via gpio in h5_btrtl_open
Bluetooth: btrtl: Add support for RTL8822C hci_ver 0x08
Bluetooth: hci_event: Fix HCI_EV_VENDOR max_len
Bluetooth: hci_core: Rate limit the logging of invalid SCO handle
Bluetooth: hci_event: Ignore multiple conn complete events
Bluetooth: msft: fix null pointer deref on msft_monitor_device_evt
Bluetooth: btmtksdio: mask out interrupt status
Bluetooth: btmtksdio: run sleep mode by default
Bluetooth: btmtksdio: lower log level in btmtksdio_runtime_[resume|suspend]()
Bluetooth: mt7921s: fix btmtksdio_[drv|fw]_pmctrl()
Bluetooth: mt7921s: fix bus hang with wrong privilege
Bluetooth: btmtksdio: refactor btmtksdio_runtime_[suspend|resume]()
Bluetooth: mt7921s: fix firmware coredump retrieve
Bluetooth: hci_serdev: call init_rwsem() before p->open()
Bluetooth: Remove kernel-doc style comment block
Bluetooth: btusb: Whitespace fixes for btusb_setup_csr()
Bluetooth: btusb: Add one more Bluetooth part for the Realtek RTL8852AE
Bluetooth: btintel: Fix WBS setting for Intel legacy ROM products
...
====================
Jisheng Zhang [Fri, 28 Jan 2022 14:52:13 +0000 (22:52 +0800)]
net: stmmac: dwmac-sun8i: make clk really gated during rpm suspended
Currently, the dwmac-sun8i's stmmaceth clk isn't disabled even if the
the device has been runtime suspended. The reason is the driver gets
the "stmmaceth" clk as tx_clk and enabling it during probe. But
there's no other usage of tx_clk except preparing and enabling, so
we can remove tx_clk and its usage then rely on the common routine
stmmac_probe_config_dt() to prepare and enable the stmmaceth clk
during driver initialization, and benefit from the runtime pm feature
after probed.
Signed-off-by: Jisheng Zhang <jszhang@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 28 Jan 2022 15:02:50 +0000 (15:02 +0000)]
Merge branch 'dsa-realtek-MDIO'
Luiz Angelo Daros de Luca says:
====================
net: dsa: realtek: MDIO interface and RTL8367S,RTL8367RB-VB
The old realtek-smi driver was linking subdrivers into a single
realtek-smi.ko After this series, each subdriver will be an independent
module required by either realtek-smi (platform driver) or the new
realtek-mdio (mdio driver). Both interface drivers (SMI or MDIO) are
independent, and they might even work side-by-side, although it will be
difficult to find such device. The subdriver can be individually
selected but only at buildtime, saving some storage space for custom
embedded systems.
Existing realtek-smi devices continue to work untouched during the
tests. The realtek-smi was moved into a realtek subdirectory, but it
normally does not break things.
I couldn't identify a fixed relation between port numbers (0..9) and
external interfaces (0..2), and I'm not sure if it is fixed for each
chip version or a device configuration. Until there is more info about
it, there is a new port property "realtek,ext-int" that can inform the
external interface.
The rtl8365mb might now handle multiple CPU ports and extint ports not
used as CPU ports. RTL8367S has an SGMII external interface, but my test
device (TP-Link Archer C5v4) uses only the second RGMII interface. We
need a test device with more external ports to test these features.
The driver still cannot handle SGMII ports.
RTL8367RB-VB support was added using information from Frank Wunderlich
<frank-w@public-files.de> but I didn't test it myself.
The rtl8365mb was tested with a MDIO-connected RTL8367S (TP-Link Acher
C5v4) and a SMI-connected RTL8365MB-VC switch (Asus RT-AC88U)
The rtl8366rb subdriver was not tested with this patch series, but it
was only slightly touched. It would be nice to test it, especially in an
MDIO-connected switch.
Best,
Luiz
Changelog:
v1-v2)
- formatting fixes
- dropped the rtl8365mb->rtl8367c rename
- other suggestions
v2-v3)
* realtek-mdio.c:
- cleanup realtek-mdio.c (BUG_ON, comments and includes)
- check devm_regmap_init return code
- removed realtek,rtl8366s string from realtek-mdio
* realtek-smi.c:
- removed void* type cast
* rtl8365mb.c:
- using macros to identify EXT interfaces
- rename some extra extport->extint cases
- allow extint as non cpu (not tested)
- allow multple cpu ports (not tested)
- dropped cpu info from struct rtl8365mb
* dropped dt-bindings changes (dealing outside this series)
* formatting issues fixed
v3-v4)
* fix cover message numbering 0/13 -> 0/11
* use static for realtek_mdio_read_reg
- Reported-by: kernel test robot <lkp@intel.com>
* use dsa_switch_for_each_cpu_port
* mention realtek_smi_{variant,ops} to realtek_{variant,ops}
in commit message
v5) sent again v4 branch. Sorry
v4-v6)
- added support for RTL8367RB-VB
- cleanup mdio_{read,write}, removing misterious START_OP, checking and
returning errors
- renamed priv->phy_id to priv->mdio_addr
- duplicated priv->ds_ops into ds_ops_{smi,mdio}. ds_ops_smi must not
set
phy_read or else both dsa and this driver might free slave_mii.
Dropped 401fd75c92f37
- Map port to extint using code instead of device-tree property. Added
comment
about port number, port description and external interfaces. Dropped
'realtek,ext-int' device-tree property
- Redacted the non-cpu ext port commit message, not highlighting the
possibility of using multiple CPU ports as it was just a byproduct.
- In a possible case of multiple cpu ports, use the first one as the
trap port.
Dropped 'realtek,trap-port' device-tree property
- Some formatting fixes
- BUG: rtl8365mb_phy_mode_supported was still checking for a cpu port
and not
an external interface
- BUG: fix trapdoor masking for port>7. Got a compiler error with a
bigger
constant value
- WARN: completed kdoc for rtl8366rb_drop_untagged()
- WARN: removed marks from incomplete kdoc
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Trap door number is a 4-bit number divided in two regions (3 and 1-bit).
Both values were not masked properly. This bug does not affect supported
devices as they use up to port 7 (ext2). It would only be a problem if
the driver becomes compatible with 10-port switches like RTL8370MB and
RTL8310SR.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
External interfaces can be configured, even if they are not CPU ports.
The first CPU port will also be the trap port (for receiving trapped
frames from the switch).
The CPU information was dropped from chip data as it was not used
outside setup. The only other place it was used is when it wrongly
checks for CPU port when it should check for extint.
The supported modes check now uses port type and not port usage.
As a byproduct, more than one CPU can be configured. although this
might not work well with DSA setups. Also, this driver is still only
blindly forwarding all traffic to CPU port(s).
This change was not tested in a device with multiple active external
interfaces ports.
realtek_priv->cpu_port is now only used by rtl8366rb.c
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: dsa: realtek: rtl8365mb: add RTL8367S support
Realtek's RTL8367S, a 5+2 port 10/100/1000M Ethernet switch.
It shares the same driver family (RTL8367C) with other models
as the RTL8365MB-VC. Its compatible string is "realtek,rtl8367s".
It was tested only with MDIO interface (realtek-mdio), although it might
work out-of-the-box with SMI interface (using realtek-smi).
This patch was based on an unpublished patch from Alvin Šipraga
<alsi@bang-olufsen.dk>.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com> Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com> Reviewed-by: Alvin Šipraga <alsi@bang-olufsen.dk> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: dsa: realtek: rtl8365mb: rename extport to extint
"extport" 0, 1, 2 was used to reference external ports id (ext0, ext1,
ext2). Meanwhile, port 0..9 is used as switch ports, including external
ports. "extport" was renamed to extint to make it clear it does not mean
the port number but the external interface number id.
The macros that map extint numbers to registers addresses now use inline
ifs instead of binary arithmetic.
Realtek uses in docs and drivers EXT_PORT0 (GMAC1) and EXT_PORT1
(GMAC2), with EXT_PORT0 being converted to ext_id == 1 and so on. It
might introduce some confusing while reading datasheets but it will not
be exposed to users.
"extint" was hardcoded to 1. However, some chips have multiple external
interfaces. It's not right to assume the CPU port uses extint 1 nor that
all extint are CPU ports. Now it came from a map between port number and
external interface id number.
This patch still does not allow multiple CPU ports nor extint as a non
CPU port.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: dsa: realtek: add new mdio interface for drivers
This driver is a mdio_driver instead of a platform driver (like
realtek-smi).
ds_ops was duplicated for smi and mdio usage as mdio interfaces uses
phy_{read,write} in ds_ops and the presence of phy_read is incompatible
with external slave_mii_bus allocation.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com> Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: dsa: realtek: remove direct calls to realtek-smi
Remove the only two direct calls from subdrivers to realtek-smi.
Now they are called from realtek_priv. Subdrivers can now be
linked independently from realtek-smi.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com> Reviewed-by: Alvin Šipraga <alsi@bang-olufsen.dk> Reviewed-by: Linus Walleij <linus.walleij@linaro.org> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: dsa: realtek: rename realtek_smi to realtek_priv
In preparation to adding other interfaces, the private data structure
was renamed to priv. Also, realtek_smi_variant and realtek_smi_ops
were renamed to realtek_variant and realtek_ops as those structs are
not SMI specific.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com> Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Alvin Šipraga <alsi@bang-olufsen.dk> Reviewed-by: Linus Walleij <linus.walleij@linaro.org> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
octeontx2-pf: Change receive buffer size using ethtool
ethtool rx-buf-len is for setting receive buffer size,
support setting it via ethtool -G parameter and getting
it via ethtool -g parameter.
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com> Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>
TCP ZC Rx requires data to be placed neatly into pages, separate
from the networking headers. This is not supported by most devices
so to make deployment easy this set adds a way for the driver to
report support for this feature thru ethtool.
The larger scope of configuring splitting headers and data, or DMA
scatter seems dauntingly broad, so this set focuses specifically
on the question "is this device usable with TCP ZC Rx?".
The aim is to avoid a litany of conditions on HW platforms, features,
and firmware versions in orchestration systems when the drivers can
easily tell their SG config.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Thu, 27 Jan 2022 18:42:59 +0000 (10:42 -0800)]
ethtool: add header/data split indication
For applications running on a mix of platforms it's useful
to have a clear indication whether host's NIC supports the
geometry requirements of TCP zero-copy. TCP zero-copy Rx
requires data to be neatly placed into memory pages.
Most NICs can't do that.
This patch is adding GET support only, since the NICs
I work with either always have the feature enabled or
enable it whenever MTU is set to jumbo. In other words
I don't need SET. But adding set should be trivial.
(The only note on SET is that we will likely want
the setting to be "sticky" and use 0 / `unknown`
to reset it back to driver default.)
Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
The reference clock output from the KSZ9477 and related Microchip
switch devices is not required on all board designs. Add a device
tree property to disable it for power and EMI reasons.
Changes since v3:
-rework some code for simplicity
Changes since v2:
-check for conflicting options in DT, added note in bindings doc
Changes since v1:
-added Acked-by on patch 1, rebase to net-next
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Robert Hancock [Thu, 27 Jan 2022 16:41:56 +0000 (10:41 -0600)]
net: dsa: microchip: Add property to disable reference clock
Add a new microchip,synclko-disable property which can be specified
to disable the reference clock output from the device if not required
by the board design.
Signed-off-by: Robert Hancock <robert.hancock@calian.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Robert Hancock [Thu, 27 Jan 2022 16:41:55 +0000 (10:41 -0600)]
net: dsa: microchip: Document property to disable reference clock
Document the new microchip,synclko-disable property which can be
specified to disable the reference clock output from the device if not
required by the board design.
Signed-off-by: Robert Hancock <robert.hancock@calian.com> Acked-by: Rob Herring <robh@kernel.org> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Lorenzo Bianconi [Thu, 27 Jan 2022 14:47:49 +0000 (15:47 +0100)]
net: mvneta: remove unnecessary if condition in mvneta_xdp_submit_frame
Get rid of unnecessary if check on tx_desc pointer in
mvneta_xdp_submit_frame routine since num_frames is always greater than
0 and tx_desc pointer is always initialized.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Convert sparx5 to use the mac_select_interface rather than using
phylink_set_pcs(). The intention here is to unify the approach for
PCS and eventually remove phylink_set_pcs().
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Fri, 28 Jan 2022 03:46:13 +0000 (19:46 -0800)]
Merge branch 'udp-ipv6-optimisations'
Pavel Begunkov says:
====================
udp/ipv6 optimisations
Shed some weight from udp/ipv6. Zerocopy benchmarks over dummy showed
~5% tx/s improvement, should be similar for small payload non-zc
cases.
The performance comes from killing 4 atomics and a couple of big struct
memcpy/memset. 1/10 removes a pair of atomics on dst refcounting for
cork->skb setup, 9/10 saves another pair on cork init. 5/10 and 8/10
kill extra 88B memset and memcpy respectively.
====================
Pavel Begunkov [Thu, 27 Jan 2022 00:36:31 +0000 (00:36 +0000)]
ipv6: partially inline ipv6_fixup_options
Inline a part of ipv6_fixup_options() to avoid extra overhead on
function call if opt is NULL.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>