Split a function out of __zerocopy_sg_from_iter() that only cares about
the traditional path with refcounted pages and doesn't need to know
about ->sg_from_iter. A preparation patch, we'll improve on the function
later.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
skb_zcopy_set() does nothing if there is already a ubuf_info associated
with an skb, and since ->link_skb should have set it several lines above
the check here essentially does nothing and can be removed. It's also
safer this way, because even if the callback is faulty we'll
have it set.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
phy_suspend() checks the WoL status, and then dereferences
phydrv->flags if (and only if) we decided that WoL has been enabled
on either the PHY or the netdev.
We then check whether phydrv was NULL, but we've potentially already
dereferenced the pointer.
If phydrv is NULL, then phy_ethtool_get_wol() will return an error
and leave wol.wolopts set to zero. However, if netdev->wol_enabled
is true, then we would dereference a NULL pointer.
Checking the PHY drivers, the only place that phydev->wol_enabled is
checked by them is in their suspend/resume callbacks and nowhere else
(which is correct, because phylib only updates this in phy_suspend()).
So, move the NULL pointer check earlier to avoid a NULL pointer
dereference. Leave the check for phydrv->suspend in place as a driver
may populate the .resume method but not the .suspend method.
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/E1sN8tn-00GDCZ-Jj@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----
iQFHBAABCgAxFiEEUEC6huC2BN0pvD5fKDiiPnotvG8FAmZ/8VUTHG1rbEBwZW5n
dXRyb25peC5kZQAKCRAoOKI+ei28b35qB/44WrOFDefCGNuERFq5/8Kq0FGoDvoa
ie1PJXjfGbhn5A53pNpB7jOyDnr6DwWVTdF6AMZObHGpqjRT03LuflWqeC9+JIC1
X6pIAeZVToBMTi8SKTGvK0Fw13dZuxc/QqlHp3S6cd8s+9ETTDnIBN3iQx4vqPTc
+G0iJUNC49zf9D4F6MsJTYGP8bmzeexTcjOwj+i2oiZvNN/vSrWs+cjC9l8K3Y0a
b1FL7rv4l+TcUWrSIxbqNUUZBrzA5uV0JFUOokhC3MvLKw+61r3L9DMUUBZBIB+Q
aFgJyCwX+ju/j1c76lJT2U3EnzNLxego7fLzha5BcftzCk3974Hvr/3y
=Ykzx
-----END PGP SIGNATURE-----
Merge tag 'linux-can-next-for-6.11-20240629' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next
Marc Kleine-Budde says:
====================
pull-request: can-next 2024-06-29
Geert Uytterhoeven contributes 3 patches with small improvements and
cleanups for the rcar_canfd driver.
A patch by Christophe JAILLET constifies the struct m_can_ops in the
m_can driver to reduce the code size.
The last 9 patches are by me an work around erratum DS80000789E 6 of
mcp2518fd.
* tag 'linux-can-next-for-6.11-20240629' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next:
can: mcp251xfd: tef: update workaround for erratum DS80000789E 6 of mcp2518fd
can: mcp251xfd: tef: prepare to workaround broken TEF FIFO tail index erratum
can: mcp251xfd: rx: add workaround for erratum DS80000789E 6 of mcp2518fd
can: mcp251xfd: rx: prepare to workaround broken RX FIFO head index erratum
can: mcp251xfd: mcp251xfd_handle_rxif_ring_uinc(): factor out in separate function
can: mcp251xfd: clarify the meaning of timestamp
can: mcp251xfd: move mcp251xfd_timestamp_start()/stop() into mcp251xfd_chip_start/stop()
can: mcp251xfd: update errata references
can: mcp251xfd: properly indent labels
can: gs_usb: add VID/PID for Xylanta SAINT3 product family
can: m_can: Constify struct m_can_ops
can: rcar_canfd: Remove superfluous parentheses in address calculations
can: rcar_canfd: Improve printing of global operational state
can: rcar_canfd: Simplify clock handling
====================
Link: https://patch.msgid.link/20240629114017.1080160-1-mkl@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
syzbot reported a general protection fault caused by a null pointer
dereference in coalesce_fill_reply(). The issue occurs when req_base->dev
is null, leading to an invalid memory access.
This panic occurs if dumping coalesce when no device name is specified.
Fixes: f750dfe825 ("ethtool: provide customized dim profile management")
Reported-by: syzbot+e77327e34cdc8c36b7d3@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=e77327e34cdc8c36b7d3
Signed-off-by: Heng Qi <hengqi@linux.alibaba.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Tony nguyen says:
====================
Intel Wired LAN Driver Updates 2024-06-28 (MAINTAINERS, ice)
This series contains updates to MAINTAINERS file and ice driver.
Jesse replaces himself with Przemek in the maintainers file.
Karthik Sundaravel adds support for VF get/set MAC address via devlink.
Eric checks for errors from ice_vsi_rebuild() during queue
reconfiguration.
Paul adjusts FW API version check for E830 devices.
Piotr adds differentiation of unload type when shutting down AdminQ.
Przemek changes ice_adapter initialization to occur once per physical
card.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Michael Chan says:
====================
bnxt_en: PTP updates for net-next
The first 5 patches implement the PTP feature on the new BCM5760X
chips. The main new hardware feature is the new TX timestamp
completion which enables the driver to retrieve the TX timestamp
in NAPI without deferring to the PTP worker.
The last 5 patches increase the number of TX PTP packets in-flight
from 1 to 4 on the older BCM5750X chips. On these older chips, we
need to call firmware in the PTP worker to retrieve the timestamp.
We use an arry to keep track of the in-flight TX PTP packets.
v2: Patch #2: Fix the unwind of txr->is_ts_pkt when bnxt_start_xmit() aborts.
Patch #4: Set the SKBTX_IN_PROGRESS flag for timestamp packets.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that we require the spinlock to protect ptp->txts_prod, change
ptp->tx_avail to non-atomic and protect it under the same spinlock.
Add a new helper function bnxt_ptp_get_txts_prod() to decrement
ptp->tx_avail under spinlock and return the producer.
Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Start accepting up to 4 TX TS requests on BCM5750X (P5) chips.
These PTP TX packets will be queued in the ptp->txts_req[] array
waiting for the TX timestamp to complete. The entries in the
array will be managed by a producer and consumer index. The
producer index is updated under spinlock since multiple TX rings
can try to send PTP packets at the same time.
Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change the function bnxt_stamp_tx_skb() to return 0 for suceess
or -EAGAIN if the timestamp is still pending in firmware. The
calling PTP aux worker will reschedule based on the return code.
Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the current 5750X PTP code paths, there is always at most one TX
SKB requested for timestamp and we won't accept another one until we
have retrieved the timestamp or it has timed out. Remove the
unnecessary check in bnxt_get_tx_ts_p5() for a pending SKB and change
the function to void.
Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On the older 5750X (P5) chips, we currently support only 1 TX PTP
packet in-flight waiting for the timestamp. Refactor the
datastructures to prepare to support up to 4 TX PTP packets.
Combine all fields required for PTP TX timestamp query into one
structure. An array of this structure will be added in follow-on
patches to support multiple outstanding TX timestamps.
Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
BCM5760X firmware will advertise direct 64-bit PHC registers access
for the driver from BAR0.
Make the necessary changes in handling HWRM_PORT_MAC_PTP_QCFG's
response and PHC register mapping for 5760X chips.
Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The new BCM5760X chips will return the timestamp of TX packets in a
new completion. Add logic in __bnxt_poll_work() to handle this
completion type to retrieve the timestamp. This feature eliminates
the limit on the number of in-flight PTP TX packets.
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The driver's current logic will always free all the TX SKBs up to
txr->tx_hw_cons within NAPI. In the next patches, we'll be adding
logic to handle TX timestamp completion and we may need to hold
some remaining TX SKBs if we don't have the timestamp completions
yet.
Modify __bnxt_poll_work_done() to clear each event bit separately to
allow bnapi->tx_int() to decide whether to clear BNXT_TX_CMP_EVENT or
not. bnapi->tx_int() will not clear BNXT_TX_CMP_EVENT if some TX
SKBs are held waiting for TX timestamps. Note that legacy chips will
never hold any SKBs this way. The SKB is always deferred to the PTP
worker slow path to retrieve the timestamp from firmware. On the new
P7 chips, the timestamp is returned by the hardware directly and we
can retrieve it directly from NAPI.
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove the unused is_gso field and add the is_ts_pkt field to struct
bnxt_sw_tx_bd. This field will mark the TX BD that has requested
HW TX timestamp. The field needs to be cleared if the timestamp packet
is later aborted. This field will be useful when processing the
new TX timestamp completion from the hardware in the next patches.
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The new BCM5760X chips will generate this new TX timestamp completion
when a TX packet's timestamp has been taken right before transmission.
The driver logic to retrieve the timestamp will be added in the next
few patches.
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Octeontx2 hardware uses Near Data Cache(NDC) block to cache
contexts in it so that access to LLC/DRAM can be avoided.
It is recommended in HRM to sync the NDC contents before
releasing/resetting LF resources. Hence implement NDC_SYNC
mailbox and sync contexts during driver teardown.
Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Call phylink_ethtool_ksettings_get() for get_link_ksettings method and
ethtool_op_get_link() for get_link method.
Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
-----BEGIN PGP SIGNATURE-----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=0BtN
-----END PGP SIGNATURE-----
Merge tag 'nf-next-24-06-28' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next into main
Pablo Neira Ayuso says:
====================
Netfilter/IPVS updates for net-next
The following patchset contains Netfilter/IPVS updates for net-next:
Patch #1 to #11 to shrink memory consumption for transaction objects:
struct nft_trans_chain { /* size: 120 (-32), cachelines: 2, members: 10 */
struct nft_trans_elem { /* size: 72 (-40), cachelines: 2, members: 4 */
struct nft_trans_flowtable { /* size: 80 (-48), cachelines: 2, members: 5 */
struct nft_trans_obj { /* size: 72 (-40), cachelines: 2, members: 4 */
struct nft_trans_rule { /* size: 80 (-32), cachelines: 2, members: 6 */
struct nft_trans_set { /* size: 96 (-24), cachelines: 2, members: 8 */
struct nft_trans_table { /* size: 56 (-40), cachelines: 1, members: 2 */
struct nft_trans_elem can now be allocated from kmalloc-96 instead of
kmalloc-128 slab.
Series from Florian Westphal. For the record, I have mangled patch #1
to add nft_trans_container_*() and use if for every transaction object.
I have also added BUILD_BUG_ON to ensure struct nft_trans always comes
at the beginning of the container transaction object. And few minor
cleanups, any new bugs are of my own.
Patch #12 simplify check for SCTP GSO in IPVS, from Ismael Luceno.
Patch #13 nf_conncount key length remains in the u32 bound, from Yunjian Wang.
Patch #14 removes unnecessary check for CTA_TIMEOUT_L3PROTO when setting
default conntrack timeouts via nfnetlink_cttimeout API, from
Lin Ma.
Patch #15 updates NFT_SECMARK_CTX_MAXLEN to 4096, SELinux could use
larger secctx names than the existing 256 bytes length.
Patch #16 adds a selftest to exercise nfnetlink_queue listeners leaving
nfnetlink_queue, from Florian Westphal.
Patch #17 increases hitcount from 255 to 65535 in xt_recent, from Phil Sutter.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski says:
====================
tcp_metrics: add netlink protocol spec in YAML
Add a netlink protocol spec for the tcp_metrics generic netlink family.
First patch adjusts the uAPI header guards to make it easier to build
tools/ with non-system headers.
v1: https://lore.kernel.org/all/20240626201133.2572487-1-kuba@kernel.org
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a protocol spec for tcp_metrics, so that it's accessible via YNL.
Useful at the very least for testing fixes.
In this episode of "10,000 ways to complicate netlink" the metric
nest has defines which are off by 1. iproute2 does:
struct rtattr *m[TCP_METRIC_MAX + 1 + 1];
parse_rtattr_nested(m, TCP_METRIC_MAX + 1, a);
for (i = 0; i < TCP_METRIC_MAX + 1; i++) {
// ...
attr = m[i + 1];
This is too weird to support in YNL, add a new set of defines
with _correct_ values to the official kernel header.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
tcp_metrics' header lacks the customary _UAPI in the header guard.
This makes YNL build rules work less seamlessly.
We can easily fix that on YNL side, but this could also be
problematic if we ever needed to create a kernel-only tcp_metrics.h.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Realtek RTL8211F Ethernet PHY supports 3 LED pins which are used to
indicate link status and activity. Add minimal LED controller driver
supporting the most common uses with the 'netdev' trigger.
Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Edward Cree says:
====================
ethtool: track custom RSS contexts in the core
Make the core responsible for tracking the set of custom RSS contexts,
their IDs, indirection tables, hash keys, and hash functions; this
lets us get rid of duplicative code in drivers, and will allow us to
support netlink dumps later.
This series only moves the sfc EF10 & EF100 driver over to the new API;
other drivers (mvpp2, octeontx2, mlx5, sfc/siena, bnxt_en) can be converted
afterwards and the legacy API removed.
====================
Link: https://patch.msgid.link/cover.1719502239.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The core now always satisfies 'ethtool -x context nonzero' from its own
tracking, so our lookup code for that case is never called. Remove it.
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/b426fcc416dedc8f203e52eebef6891eccebe4c1.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
On 'ethtool -x' with rss_context != 0, instead of calling the driver to
read the RSS settings for the context, just get the settings from the
rss_ctx xarray, and return them to the user with no driver involvement.
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/2d0190fa29638f307ea720f882ebd41f6f867694.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The core is now responsible for allocating IDs and a memory region for
us to store our state (struct efx_rss_context_priv), so we no longer
need efx_alloc_rss_context_entry() and friends.
Since the contexts are now maintained by the core, use the core's lock
(net_dev->ethtool->rss_lock), rather than our own mutex (efx->rss_lock),
to serialise access against changes; and remove the now-unused
efx->rss_lock from struct efx_nic.
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/150274740ea8cc137fef5502541ce573d32fb319.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
While this is not needed to serialise the ethtool entry points (which
are all under RTNL), drivers may have cause to asynchronously access
dev->ethtool->rss_ctx; taking dev->ethtool->rss_lock allows them to
do this safely without needing to take the RTNL.
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/7f9c15eb7525bf87af62c275dde3a8570ee8bf0a.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add a new API to create/modify/remove RSS contexts, that passes in the
newly-chosen context ID (not as a pointer) rather than leaving the
driver to choose it on create. Also pass in the ctx, allowing drivers
to easily use its private data area to store their hardware-specific
state.
Keep the existing .set_rxfh API for now as a fallback, but deprecate it
for custom contexts (rss_context != 0).
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/45f1fe61df2163c091ec394c9f52000c8b16cc3b.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Since drivers are still choosing the context IDs, we have to force the
XArray to use the ID they've chosen rather than picking one ourselves,
and handle the case where they give us an ID that's already in use.
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/801f5faa4cec87c65b2c6e27fb220c944bce593a.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Each context stores the RXFH settings (indir, key, and hfunc) as well
as optionally some driver private data.
Delete any still-existing contexts at netdev unregister time.
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/cbd1c402cec38f2e03124f2ab65b4ae4e08bd90d.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net_dev->ethtool is a pointer to new struct ethtool_netdev_state, which
currently contains only the wol_enabled field.
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/293a562278371de7534ed1eb17531838ca090633.1719502239.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jakub Kicinski says:
====================
selftests: drv-net: add ability to schedule cleanup with defer()
Introduce a defer / cleanup mechanism for driver selftests.
More detailed info in the second patch.
====================
Link: https://patch.msgid.link/20240627185502.3069139-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This implements what I was describing in [1]. When writing a test
author can schedule cleanup / undo actions right after the creation
completes, eg:
cmd("touch /tmp/file")
defer(cmd, "rm /tmp/file")
defer() takes the function name as first argument, and the rest are
arguments for that function. defer()red functions are called in
inverse order after test exits. It's also possible to capture them
and execute earlier (in which case they get automatically de-queued).
undo = defer(cmd, "rm /tmp/file")
# ... some unsafe code ...
undo.exec()
As a nice safety all exceptions from defer()ed calls are captured,
printed, and ignored (they do make the test fail, however).
This addresses the common problem of exceptions in cleanup paths
often being unhandled, leading to potential leaks.
There is a global action queue, flushed by ksft_run(). We could support
function level defers too, I guess, but there's no immediate need..
Link: https://lore.kernel.org/all/877cedb2ki.fsf@nvidia.com/ # [1]
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/20240627185502.3069139-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Exception handlers print the result and use continue
to skip the non-exception result printing. This makes
inserting common post-test code hard. Refactor to
avoid the continues and have only one ktap_result() call.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/20240627185502.3069139-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add .get_channel to enic_ethtool_ops to enable basic ethtool -l
support to get the current channel configuration.
Note that the driver does not support dynamically changing queue
configuration, so .set_channel is intentionally unused. Instead, users
should use Cisco's hardware management tools (UCSM/IMC) to modify
virtual interface card configuration out of band.
Signed-off-by: Jon Kohler <jon@nutanix.com>
Link: https://patch.msgid.link/20240627202013.2398217-1-jon@nutanix.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Extend the existing test to exercise UDP GSO egress through devices with
various offload capabilities, including lack of checksum offload, which is
the default case for TUN/TAP devices.
Test against a dummy device because it is simpler to set up then TUN/TAP.
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626-linux-udpgso-v2-2-422dfcbd6b48@cloudflare.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Today sending a UDP GSO packet from a TUN device results in an EIO error:
import fcntl, os, struct
from socket import *
TUNSETIFF = 0x400454CA
IFF_TUN = 0x0001
IFF_NO_PI = 0x1000
UDP_SEGMENT = 103
tun_fd = os.open("/dev/net/tun", os.O_RDWR)
ifr = struct.pack("16sH", b"tun0", IFF_TUN | IFF_NO_PI)
fcntl.ioctl(tun_fd, TUNSETIFF, ifr)
os.system("ip addr add 192.0.2.1/24 dev tun0")
os.system("ip link set dev tun0 up")
s = socket(AF_INET, SOCK_DGRAM)
s.setsockopt(SOL_UDP, UDP_SEGMENT, 1200)
s.sendto(b"x" * 3000, ("192.0.2.2", 9)) # EIO
This is due to a check in the udp stack if the egress device offers
checksum offload. While TUN/TAP devices, by default, don't advertise this
capability because it requires support from the TUN/TAP reader.
However, the GSO stack has a software fallback for checksum calculation,
which we can use. This way we don't force UDP_SEGMENT users to handle the
EIO error and implement a segmentation fallback.
Lift the restriction so that UDP_SEGMENT can be used with any egress
device. We also need to adjust the UDP GSO code to match the GSO stack
expectation about ip_summed field, as set in commit 8d63bee643 ("net:
avoid skb_warn_bad_offload false positives on UFO"). Otherwise we will hit
the bad offload check.
Users should, however, expect a potential performance impact when
batch-sending packets with UDP_SEGMENT without checksum offload on the
egress device. In such case the packet payload is read twice: first during
the sendmsg syscall when copying data from user memory, and then in the GSO
stack for checksum computation. This double memory read can be less
efficient than a regular sendmsg where the checksum is calculated during
the initial data copy from user memory.
Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626-linux-udpgso-v2-1-422dfcbd6b48@cloudflare.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Marc Kleine-Budde <mkl@pengutronix.de> says:
This patch series tries to work around erratum DS80000789E 6 of the
mcp2518fd, found by Stefan Althöfer, the other variants of the chip
family (mcp2517fd and mcp251863) are probably also affected.
Erratum DS80000789E 6 says "reading of the FIFOCI bits in the FIFOSTA
register for an RX FIFO may be corrupted". However observation shows
that this problem is not limited to RX FIFOs but also effects the TEF
FIFO.
In the bad case, the driver reads a too large head index. In the
original code, the driver always trusted the read value.
For the RX FIDO this caused old, already processed CAN frames or new,
incompletely written CAN frames to be (re-)processed.
To work around this issue, keep a per FIFO timestamp of the last valid
received CAN frame and compare against the timestamp of every received
CAN frame.
Further tests showed that this workaround can recognize old CAN
frames, but a small time window remains in which partially written CAN
frames are not recognized but then processed. These CAN frames have
the correct data and time stamps, but the DLC has not yet been
updated.
For the TEF FIFO the original driver already detects the error, update
the error handling with the knowledge that it is causes by this erratum.
Link: https://lore.kernel.org/all/20240628-mcp251xfd-workaround-erratum-6-v4-0-53586f168524@pengutronix.de
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
This patch updates the workaround for a problem similar to erratum
DS80000789E 6 of the mcp2518fd, the other variants of the chip
family (mcp2517fd and mcp251863) are probably also affected.
Erratum DS80000789E 6 says "reading of the FIFOCI bits in the FIFOSTA
register for an RX FIFO may be corrupted". However observation shows
that this problem is not limited to RX FIFOs but also effects the TEF
FIFO.
In the bad case, the driver reads a too large head index. As the FIFO
is implemented as a ring buffer, this results in re-handling old CAN
transmit complete events.
Every transmit complete event contains with a sequence number that
equals to the sequence number of the corresponding TX request. This
way old TX complete events can be detected.
If the original driver detects a non matching sequence number, it
prints an info message and tries again later. As wrong sequence
numbers can be explained by the erratum DS80000789E 6, demote the info
message to debug level, streamline the code and update the comments.
Keep the behavior: If an old CAN TX complete event is detected, abort
the iteration and mark the number of valid CAN TX complete events as
processed in the chip by incrementing the FIFO's tail index.
Cc: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Cc: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
This is a preparatory patch to work around a problem similar to
erratum DS80000789E 6 of the mcp2518fd, the other variants of the chip
family (mcp2517fd and mcp251863) are probably also affected.
Erratum DS80000789E 6 says "reading of the FIFOCI bits in the FIFOSTA
register for an RX FIFO may be corrupted". However observation shows
that this problem is not limited to RX FIFOs but also effects the TEF
FIFO.
When handling the TEF interrupt, the driver reads the FIFO header
index from the TEF FIFO STA register of the chip.
In the bad case, the driver reads a too large head index. In the
original code, the driver always trusted the read value, which caused
old CAN transmit complete events that were already processed to be
re-processed.
Instead of reading and trusting the head index, read the head index
and calculate the number of CAN frames that were supposedly received -
replace mcp251xfd_tef_ring_update() with mcp251xfd_get_tef_len().
The mcp251xfd_handle_tefif() function reads the CAN transmit complete
events from the chip, iterates over them and pushes them into the
network stack. The original driver already contains code to detect old
CAN transmit complete events, that will be updated in the next patch.
Cc: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Cc: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
This patch tries to works around erratum DS80000789E 6 of the
mcp2518fd, the other variants of the chip family (mcp2517fd and
mcp251863) are probably also affected.
In the bad case, the driver reads a too large head index. In the
original code, the driver always trusted the read value, which caused
old, already processed CAN frames or new, incompletely written CAN
frames to be (re-)processed.
To work around this issue, keep a per FIFO timestamp [1] of the last
valid received CAN frame and compare against the timestamp of every
received CAN frame. If an old CAN frame is detected, abort the
iteration and mark the number of valid CAN frames as processed in the
chip by incrementing the FIFO's tail index.
Further tests showed that this workaround can recognize old CAN
frames, but a small time window remains in which partially written CAN
frames [2] are not recognized but then processed. These CAN frames
have the correct data and time stamps, but the DLC has not yet been
updated.
[1] As the raw timestamp overflows every 107 seconds (at the usual
clock rate of 40 MHz) convert it to nanoseconds with the
timecounter framework and use this to detect stale CAN frames.
Link: https://lore.kernel.org/all/BL3PR11MB64844C1C95CA3BDADAE4D8CCFBC99@BL3PR11MB6484.namprd11.prod.outlook.com [2]
Reported-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Closes: https://lore.kernel.org/all/FR0P281MB1966273C216630B120ABB6E197E89@FR0P281MB1966.DEUP281.PROD.OUTLOOK.COM
Tested-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Tested-by: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
This is a preparatory patch to work around erratum DS80000789E 6 of
the mcp2518fd, the other variants of the chip family (mcp2517fd and
mcp251863) are probably also affected.
When handling the RX interrupt, the driver iterates over all pending
FIFOs (which are implemented as ring buffers in hardware) and reads
the FIFO header index from the RX FIFO STA register of the chip.
In the bad case, the driver reads a too large head index. In the
original code, the driver always trusted the read value, which caused
old CAN frames that were already processed, or new, incompletely
written CAN frames to be (re-)processed.
Instead of reading and trusting the head index, read the head index
and calculate the number of CAN frames that were supposedly received -
replace mcp251xfd_rx_ring_update() with mcp251xfd_get_rx_len().
The mcp251xfd_handle_rxif_ring() function reads the received CAN
frames from the chip, iterates over them and pushes them into the
network stack. Prepare that the iteration can be stopped if an old CAN
frame is detected. The actual code to detect old or incomplete frames
and abort will be added in the next patch.
Link: https://lore.kernel.org/all/BL3PR11MB64844C1C95CA3BDADAE4D8CCFBC99@BL3PR11MB6484.namprd11.prod.outlook.com
Reported-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Closes: https://lore.kernel.org/all/FR0P281MB1966273C216630B120ABB6E197E89@FR0P281MB1966.DEUP281.PROD.OUTLOOK.COM
Tested-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Tested-by: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
This is a preparation patch.
Sending the UINC messages followed by incrementing the tail pointer
will be called in more than one place in upcoming patches, so factor
this out into a separate function.
Also make mcp251xfd_handle_rxif_ring_uinc() safe to be called with a
"len" of 0.
Tested-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Tested-by: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
The mcp251xfd chip is configured to provide a timestamp with each
received and transmitted CAN frame. The timestamp is derived from the
internal free-running timer, which can also be read from the TBC
register via SPI. The timer is 32 bits wide and is clocked by the
external oscillator (typically 20 or 40 MHz).
To avoid confusion, we call this timestamp "timestamp_raw" or "ts_raw"
for short.
Using the timecounter framework, the "ts_raw" is converted to 64 bit
nanoseconds since the epoch. This is what we call "timestamp".
This is a preparation for the next patches which use the "timestamp"
to work around a bug where so far only the "ts_raw" is used.
Tested-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Tested-by: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
The mcp251xfd wakes up from Low Power or Sleep Mode when SPI activity
is detected. To avoid this, make sure that the timestamp worker is
stopped before shutting down the chip.
Split the starting of the timestamp worker out of
mcp251xfd_timestamp_init() into the separate function
mcp251xfd_timestamp_start().
Call mcp251xfd_timestamp_init() before mcp251xfd_chip_start(), move
mcp251xfd_timestamp_start() to mcp251xfd_chip_start(). In this way,
mcp251xfd_timestamp_stop() can be called unconditionally by
mcp251xfd_chip_stop().
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
