- 13 May, 2021 12 commits
-
-
Linus Lüssing authored
A multicast router for IPv4 does not imply that the same host also is a multicast router for IPv6 and vice versa. To reduce multicast traffic when a host is only a multicast router for one of these two protocol families, keep router state for IPv4 and IPv6 separately. Similar to how querier state is kept separately. For backwards compatibility for netlink and switchdev notifications these two will still only notify if a port switched from either no IPv4/IPv6 multicast router to any IPv4/IPv6 multicast router or the other way round. However a full netlink MDB router dump will now also include a multicast router timeout for both IPv4 and IPv6. Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Linus Lüssing authored
In preparation for the upcoming split of multicast router state into their IPv4 and IPv6 variants split router port deletion and notification into two functions. When we disable a port for instance later we want to only send one notification to switchdev and netlink for compatibility and want to avoid sending one for IPv4 and one for IPv6. For that the split is needed. Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Linus Lüssing authored
In preparation for the upcoming split of multicast router state into their IPv4 and IPv6 variants move the protocol specific router list and timer access to ip4 wrapper functions. Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Linus Lüssing authored
In preparation for the upcoming split of multicast router state into their IPv4 and IPv6 variants move the protocol specific timer access to an ip4 wrapper function. Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Linus Lüssing authored
In preparation for the upcoming split of multicast router state into their IPv4 and IPv6 variants make br_multicast_is_router() protocol family aware. Note that for now br_ip6_multicast_is_router() uses the currently still common ip4_mc_router_timer for now. It will be renamed to ip6_mc_router_timer later when the split is performed. While at it also renames the "1" and "2" constants in br_multicast_is_router() to the MDB_RTR_TYPE_TEMP_QUERY and MDB_RTR_TYPE_PERM enums. Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Linus Lüssing authored
In preparation for the upcoming split of multicast router state into their IPv4 and IPv6 variants and as the br_multicast_mark_router() will be split for that remove the select querier wrapper and instead add ip4 and ip6 variants for br_multicast_query_received(). Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Linus Lüssing authored
In preparation for the upcoming split of multicast router state into their IPv4 and IPv6 variants and to avoid IPv6 #ifdef clutter later add some inline functions for the protocol specific parts in the mdb router netlink code. Also the we need iterate over the port instead of router list to be able put one router port entry with both the IPv4 and IPv6 multicast router info later. Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Linus Lüssing authored
In preparation for the upcoming split of multicast router state into their IPv4 and IPv6 variants and to avoid IPv6 #ifdef clutter later add two wrapper functions for router node retrieval in the payload forwarding code. Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Linus Lüssing authored
In preparation for the upcoming split of multicast router state into their IPv4 and IPv6 variants, rename the affected variable to the IPv4 version first to avoid some renames in later commits. Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Sebastian Andrzej Siewior authored
__napi_schedule_irqoff() is an optimized version of __napi_schedule() which can be used where it is known that interrupts are disabled, e.g. in interrupt-handlers, spin_lock_irq() sections or hrtimer callbacks. On PREEMPT_RT enabled kernels this assumptions is not true. Force- threaded interrupt handlers and spinlocks are not disabling interrupts and the NAPI hrtimer callback is forced into softirq context which runs with interrupts enabled as well. Chasing all usage sites of __napi_schedule_irqoff() is a whack-a-mole game so make __napi_schedule_irqoff() invoke __napi_schedule() for PREEMPT_RT kernels. The callers of ____napi_schedule() in the networking core have been audited and are correct on PREEMPT_RT kernels as well. Reported-by: Juri Lelli <juri.lelli@redhat.com> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Juri Lelli <juri.lelli@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Johannes Berg authored
In the alx driver, all locking depended on the RTNL, but that causes issues with ipconfig ("ip=..." command line) because that waits for the netdev to have a carrier while holding the RTNL, but the alx workers etc. require RTNL, so the carrier won't be set until the RTNL is dropped and can be acquired by alx workers. This causes long delays at boot, as reported by Nikolai Zhubr. Really the only sensible thing to do here is to not use the RTNL for everything, but instead have fine-grained locking for just the driver. Do that, it's not that hard. Reported-by: Nikolai Zhubr <zhubr.2@gmail.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Yannick Vignon authored
Even though the taprio qdisc is designed for multiqueue devices, all the queues still point to the same top-level taprio qdisc. This works and is probably required for software taprio, but at least with offload taprio, it has an undesirable side effect: because the whole qdisc is run when a packet has to be sent, it allows packets in a best-effort class to be processed in the context of a task sending higher priority traffic. If there are packets left in the qdisc after that first run, the NET_TX softirq is raised and gets executed immediately in the same process context. As with any other softirq, it runs up to 10 times and for up to 2ms, during which the calling process is waiting for the sendmsg call (or similar) to return. In my use case, that calling process is a real-time task scheduled to send a packet every 2ms, so the long sendmsg calls are leading to missed timeslots. By attaching each netdev queue to its own qdisc, as it is done with the "classic" mq qdisc, each traffic class can be processed independently without touching the other classes. A high-priority process can then send packets without getting stuck in the sendmsg call anymore. Signed-off-by: Yannick Vignon <yannick.vignon@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
- 12 May, 2021 2 commits
-
-
Jim Ma authored
In function tls_sw_splice_read, before call tls_sw_advance_skb it checks likely(!(flags & MSG_PEEK)), while MSG_PEEK is used for recvmsg, splice supports SPLICE_F_NONBLOCK, SPLICE_F_MOVE, SPLICE_F_MORE, should remove this checking. Signed-off-by: Jim Ma <majinjing3@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Zou Wei authored
This module's remove path calls del_timer(). However, that function does not wait until the timer handler finishes. This means that the timer handler may still be running after the driver's remove function has finished, which would result in a use-after-free. Fix by calling del_timer_sync(), which makes sure the timer handler has finished, and unable to re-schedule itself. Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Zou Wei <zou_wei@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
- 11 May, 2021 5 commits
-
-
Guenter Roeck authored
The rcu_head pointer passed to taprio_free_sched_cb is never NULL. That means that the result of container_of() operations on it is also never NULL, even though rcu_head is the first element of the structure embedding it. On top of that, it is misleading to perform a NULL check on the result of container_of() because the position of the contained element could change, which would make the check invalid. Remove the unnecessary NULL check. This change was made automatically with the following Coccinelle script. @@ type t; identifier v; statement s; @@ <+... ( t v = container_of(...); | v = container_of(...); ) ... when != v - if (\( !v \| v == NULL \) ) s ...+> Signed-off-by: Guenter Roeck <linux@roeck-us.net> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Loic Poulain authored
The WWAN framework provides a unified way to handle WWAN/modems and its control port(s). It has initially been introduced to support MHI/PCI modems, offering the same control protocols as the USB variants such as MBIM, QMI, AT... The WWAN framework exposes these control protocols as character devices, similarly to cdc-wdm, but in a bus agnostic fashion. This change adds registration of the USB modem cdc-wdm control endpoints to the WWAN framework as standard control ports (wwanXpY...). Exposing cdc-wdm through WWAN framework normally maintains backward compatibility, e.g: $ qmicli --device-open-qmi -d /dev/wwan0p1QMI --dms-get-ids instead of $ qmicli --device-open-qmi -d /dev/cdc-wdm0 --dms-get-ids However, some tools may rely on cdc-wdm driver/device name for device detection. It is then safer to keep the 'legacy' cdc-wdm character device to prevent any breakage. This is handled in this change by API mutual exclusion, only one access method can be used at a time, either cdc-wdm chardev or WWAN API. Note that unknown channel types (other than MBIM, AT or MBIM) are not registered to the WWAN framework. Signed-off-by: Loic Poulain <loic.poulain@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Loic Poulain authored
Some devices may have ports with unknown type/protocol which need to be tagged (though not supported by WWAN core). This will be the case for cdc-wdm based drivers. Signed-off-by: Loic Poulain <loic.poulain@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Zou Wei authored
This module's remove path calls del_timer(). However, that function does not wait until the timer handler finishes. This means that the timer handler may still be running after the driver's remove function has finished, which would result in a use-after-free. Fix by calling del_timer_sync(), which makes sure the timer handler has finished, and unable to re-schedule itself. Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Zou Wei <zou_wei@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Zou Wei authored
This module's remove path calls del_timer(). However, that function does not wait until the timer handler finishes. This means that the timer handler may still be running after the driver's remove function has finished, which would result in a use-after-free. Fix by calling del_timer_sync(), which makes sure the timer handler has finished, and unable to re-schedule itself. Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Zou Wei <zou_wei@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
- 10 May, 2021 21 commits
-
-
David S. Miller authored
Matteo Croce says: ==================== mvpp2: resolve two warnings Just two small changes to suppress two warnings. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
-
Matteo Croce authored
Remove some unreachable code, so to suppress this warning: drivers/net/ethernet/marvell/mvpp2/mvpp2_prs.c: In function ‘mvpp2_prs_tcam_first_free’: drivers/net/ethernet/marvell/mvpp2/mvpp2_prs.c:397:10: warning: comparison is always false due to limited range of data type [-Wtype-limits] 397 | if (end >= MVPP2_PRS_TCAM_SRAM_SIZE) | ^~ Fixes: 3f518509 ("ethernet: Add new driver for Marvell Armada 375 network unit") Signed-off-by: Matteo Croce <mcroce@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Matteo Croce authored
mvpp2_run_xdp() has an unused parameter rxq, remove it. Fixes: 07dd0a7a ("mvpp2: add basic XDP support") Signed-off-by: Matteo Croce <mcroce@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
David S. Miller authored
Andy Shevchenko says: ==================== net: pch_gbe: fix and a few cleanups The series provides one fix (patch 1) for GPIO to be able to wait for the GPIO driver to appear. This is separated from the conversion to the GPIO descriptors (patch 2) in order to have a possibility for backporting. Patches 3 and 4 fix minor warnings from Sparse while moving to a new APIs. Patch 5 is MODULE_VERSION() clean up. Tested on Intel Minnowboard (v1). Since v3: - rebased on top of v5.13-rc1 - added Tested-by (Flavio) - added Reported-by to certain changes (LKP) Since v2: - added a few cleanups on top of the fix ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
-
Andy Shevchenko authored
Remove MODULE_VERSION(), as it doesn't seem to serve any practical purpose. For in-tree drivers, the kernel version matters. The code received lots of changes, but module version remained constant, since the driver landed in mainline. So, this version doesn't seem have any practical meaning anymore. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Flavio Suligoi <f.suligoi@asem.it> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Andy Shevchenko authored
Sparse is not happy about handling of strict types in pch_ptp_match(): .../pch_gbe_main.c:158:33: warning: incorrect type in argument 2 (different base types) .../pch_gbe_main.c:158:33: expected unsigned short [usertype] uid_hi .../pch_gbe_main.c:158:33: got restricted __be16 [usertype] .../pch_gbe_main.c:158:45: warning: incorrect type in argument 3 (different base types) .../pch_gbe_main.c:158:45: expected unsigned int [usertype] uid_lo .../pch_gbe_main.c:158:45: got restricted __be32 [usertype] .../pch_gbe_main.c:158:56: warning: incorrect type in argument 4 (different base types) .../pch_gbe_main.c:158:56: expected unsigned short [usertype] seqid .../pch_gbe_main.c:158:56: got restricted __be16 [usertype] Fix that by switching to use proper accessors to BE data. Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Flavio Suligoi <f.suligoi@asem.it> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Andy Shevchenko authored
Use readx_poll_timeout_atomic() instead of open coded variants. While at it, add __iomem attribute to the parameter of pch_gbe_wait_clr_bit(). This in particular will fix a lot of warnings detected by Sparse, e.g. .../pch_gbe_main.c:308:26: warning: incorrect type in argument 1 (different address spaces) .../pch_gbe_main.c:308:26: expected void const [noderef] __iomem * .../pch_gbe_main.c:308:26: got void *reg Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Flavio Suligoi <f.suligoi@asem.it> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Andy Shevchenko authored
This switches the PCH GBE driver to use GPIO descriptors. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Flavio Suligoi <f.suligoi@asem.it> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Andy Shevchenko authored
If GPIO controller is not available yet we need to defer the probe of GBE until provider will become available. While here, drop GPIOF_EXPORT because it's deprecated and may not be available. Fixes: f1a26fdf ("pch_gbe: Add MinnowBoard support") Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Tested-by: Flavio Suligoi <f.suligoi@asem.it> Signed-off-by: David S. Miller <davem@davemloft.net>
-
David S. Miller authored
Radu Pirea says: ==================== Add PTP support for TJA1103 This is the PTP support for TJA1103. The RX timestamp is found in the reserved2 field of the PTP package. The TX timestamp has to be read from the phy registers. Reading of the timestamp works with interrupts or with polling(that starts when .nxp_c45_txtstamp is called). The implementation of .adjtime is done by read modify write because there is no way to atomically add/subtract a constant from the clock value. I've moved scaled_ppm_to_ppb function from ptp_clock.c to ptp_clock_kernel.h in order to be able to build the driver without PTP_1588_CLOCK=y. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
-
Radu Pirea (NXP OSS) authored
Add mii_timestamper interface and register a ptp clock. The package timestamping can work with or without interrupts. RX timestamps are received in the reserved field of the PTP package. TX timestamps are read via MDIO from a set of registers. Signed-off-by: Radu Pirea (NXP OSS) <radu-nicolae.pirea@oss.nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Radu Pirea (NXP OSS) authored
Make scaled_ppm_to_ppb static inline to be able to build drivers that use this function even with PTP_1588_CLOCK disabled. Signed-off-by: Radu Pirea (NXP OSS) <radu-nicolae.pirea@oss.nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Andy Shevchenko authored
Custom isprint() definition may collide with one form ctype.h. In order to avoid this, replace it with a functional analogue which is isascii() && isprint() in this case. First appearance of the code is in the commit 636b3843 ("Import 2.3.43"). Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Zhen Lei authored
The statement of the last "if (xxx)" branch is the same as the "else" branch. Delete it to simplify code. No functional change. Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Andy Shevchenko authored
device_property_read_u32() may return different error codes. Propagate it to the caller. Signed-off-by: Andy Shevchenko <andy.shevchenko@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Andy Shevchenko authored
Replace open coded variants of devm_clk_get_optional(). Signed-off-by: Andy Shevchenko <andy.shevchenko@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Andy Shevchenko authored
Use the device_get_match_data() helper instead of open coding. Signed-off-by: Andy Shevchenko <andy.shevchenko@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Andy Shevchenko authored
In each iteration fwnode_for_each_available_child_node() bumps a reference counting of a loop variable followed by dropping in on a next iteration, Since in error case the loop is broken, we have to drop a reference count by ourselves. Do it for port_fwnode in error case during ->probe(). Fixes: 24812221 ("net: mvpp2: use device_*/fwnode_* APIs instead of of_*") Cc: Marcin Wojtas <mw@semihalf.com> Signed-off-by: Andy Shevchenko <andy.shevchenko@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Cong Wang authored
When we call af_ops->set_link_af() we hold a RCU read lock as we retrieve af_ops from the RCU protected list, but this is unnecessary because we already hold RTNL lock, which is the writer lock for protecting rtnl_af_ops, so it is safer than RCU read lock. Similar for af_ops->validate_link_af(). This was not a problem until we begin to take mutex lock down the path of ->set_link_af() in __ipv6_dev_mc_dec() recently. We can just drop the RCU read lock there and assert RTNL lock. Reported-and-tested-by: syzbot+7d941e89dd48bcf42573@syzkaller.appspotmail.com Fixes: 63ed8de4 ("mld: add mc_lock for protecting per-interface mld data") Tested-by: Taehee Yoo <ap420073@gmail.com> Signed-off-by: Cong Wang <cong.wang@bytedance.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
David S. Miller authored
Stefan Wahren says: ==================== net: qca_spi: Improve sync handling This small patch series contains some improvements of the sync handling. This was discovered while the QCA7000 was doing SLAC (Signal Level Attenuation Characterization). ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
-
Stefan Wahren authored
In order to identify significant signature issues add a new stat counter, which increases on bad signature values that causes a sync loss. Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-