All memory in GVE is currently allocated without regard for the NUMA
node of the device. Because access to NUMA-local memory access is
significantly cheaper than access to a remote node, this change attempts
to ensure that page frags used in the RX path, including page pool
frags, are allocated on the NUMA node local to the gVNIC device. Note
that this attempt is best-effort. If necessary, the driver will still
allocate non-local memory, as __GFP_THISNODE is not passed. Descriptor
ring allocations are not updated, as dma_alloc_coherent handles that.
This change also modifies the IRQ affinity setting to only select CPUs
from the node local to the device, preserving the behavior that TX and
RX queues of the same index share CPU affinity.
Signed-off-by: Bailey Forrest <bcf@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Signed-off-by: Jeroen de Borst <jeroendb@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250707210107.2742029-1-jeroendb@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch adds support for XDP_TX and XDP_REDIRECT for the DQ RDA
queue format. To appropriately support transmission of XDP frames, a
new pending packet type GVE_TX_PENDING_PACKET_DQO_XDP_FRAME is
introduced for completion handling, as there was a previous assumption
that completed packets would be SKBs.
XDP_TX handling completes the basic XDP actions, so the feature is
recorded accordingly. This patch also enables the ndo_xdp_xmit callback
allowing DQ to handle XDP_REDIRECT packets originating from another
interface.
The XDP spinlock is moved to common TX ring fields so that it can be
used in both GQ and DQ. Originally, it was in a section which was
mutually exclusive for GQ and DQ.
In summary, 3 XDP features are exposed for the DQ RDA queue format:
1) NETDEV_XDP_ACT_BASIC
2) NETDEV_XDP_ACT_NDO_XMIT
3) NETDEV_XDP_ACT_REDIRECT
Note that XDP and header-data split are mutually exclusive for the time
being due to lack of multi-buffer XDP support.
This patch does not add support for the DQ QPL format. That is to come
in a future patch series.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch performs various minor DQO TX datapath refactors in
preparation for adding XDP_TX and XDP_REDIRECT support. The following
refactors are performed:
1) gve_tx_fill_pkt_desc_dqo() relies on a SKB pointer to
get whether checksum offloading should be enabled. This won't work
for the XDP case, which does not have a SKB. This patch updates the
method to use a boolean representing whether checksum offloading
should be enabled directly.
2) gve_maybe_stop_dqo() contains some synchronization between the true
TX head and the cached value, a synchronization which is common for
XDP queues and normal netdev queues. However, that method is reserved
for netdev TX queues. To avoid duplicate code, this logic is factored
out into a new method, gve_has_tx_slots_available().
3) gve_tx_update_tail() is added to update the TX tail, a functionality
that will be common between normal TX and XDP TX codepaths.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In preparation for XDP DQ support, the gve_xdp_xmit callback needs to
be generalized for all queue formats. This patch renames the GQ-specific
function to gve_xdp_xmit_gqi, and introduces a new gve_xdp_xmit callback
which branches on queue format.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In gve_adminq_issue_cmd(), return -EINVAL instead of 0 when an unknown
admin queue command opcode is encountered.
This prevents the function from silently succeeding on invalid input
and prevents undefined behavior by ensuring the function fails gracefully
when an unrecognized opcode is provided.
These changes improve error handling.
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com>
Link: https://patch.msgid.link/20250616054504.1644770-2-alok.a.tiwari@oracle.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
- Correct spelling and improves the clarity of comments
"confiugration" -> "configuration"
"spilt" -> "split"
"It if is 0" -> "If it is 0"
"DQ" -> "DQO" (correct abbreviation)
- Clarify BIT(0) flag usage in gve_get_priv_flags()
- Replaced hardcoded array size with GVE_NUM_PTYPES
for clarity and maintainability.
These changes are purely cosmetic and do not affect functionality.
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com>
Reviewed-by: Joe Damato <joe@dama.to>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Link: https://patch.msgid.link/20250616054504.1644770-1-alok.a.tiwari@oracle.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Implement ndo_hwtstamp_get/set to enable hardware RX timestamping,
providing support for SIOC[SG]HWTSTAMP IOCTLs. Included with this support
is the small change necessary to read the rx timestamp out of the rx
descriptor, now that timestamps start being enabled. The gve clock is
only used for hardware timestamps, so started when timestamps are
requested and stopped when not needed.
This version only supports RX hardware timestamping with the rx filter
HWTSTAMP_FILTER_ALL. If the user attempts to configure a more
restrictive filter, the filter will be set to HWTSTAMP_FILTER_ALL in the
returned structure.
Signed-off-by: John Fraker <jfraker@google.com>
Signed-off-by: Ziwei Xiao <ziweixiao@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Link: https://patch.msgid.link/20250614000754.164827-8-hramamurthy@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Allow the rx path to recover the high 32 bits of the full 64 bit rx
timestamp.
Use the low 32 bits of the last synced nic time and the 32 bits of the
timestamp provided in the rx descriptor to generate a difference, which
is then applied to the last synced nic time to reconstruct the complete
64-bit timestamp.
This scheme remains accurate as long as no more than ~2 seconds have
passed between the last read of the nic clock and the timestamping
application of the received packet.
Signed-off-by: John Fraker <jfraker@google.com>
Signed-off-by: Ziwei Xiao <ziweixiao@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Link: https://patch.msgid.link/20250614000754.164827-7-hramamurthy@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Adminq commands for queues creation and destruction were not
consistently protected by the driver's adminq_lock. This was previously
benign as these operations were always initiated from contexts holding
kernel-level locks (e.g., rtnl_lock, netdev_lock), which provided
serialization.
Upcoming PTP aux_work will issue adminq commands directly from the
driver to read the NIC clock, without such kernel lock protection.
To prevent race conditions with this new PTP work, this patch ensures
the adminq_lock is held during queues creation and destruction.
Signed-off-by: Ziwei Xiao <ziweixiao@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Link: https://patch.msgid.link/20250614000754.164827-5-hramamurthy@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
gve_alloc_pending_packet() can return NULL, but gve_tx_add_skb_dqo()
did not check for this case before dereferencing the returned pointer.
Add a missing NULL check to prevent a potential NULL pointer
dereference when allocation fails.
This improves robustness in low-memory scenarios.
Fixes: a57e5de476 ("gve: DQO: Add TX path")
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 03df156dd3 ("xdp: double protect netdev->xdp_flags with
netdev->lock") introduces the netdev lock to xdp_set_features_flag().
The change includes a _locked version of the method, as it is possible
for a driver to have already acquired the netdev lock before calling
this helper. However, the same applies to
xdp_features_(set|clear)_redirect_flags(), which ends up calling the
unlocked version of xdp_set_features_flags() leading to deadlocks in
GVE, which grabs the netdev lock as part of its suspend, reset, and
shutdown processes:
[ 833.265543] WARNING: possible recursive locking detected
[ 833.270949] 6.15.0-rc1 #6 Tainted: G E
[ 833.276271] --------------------------------------------
[ 833.281681] systemd-shutdow/1 is trying to acquire lock:
[ 833.287090] ffff949d2b148c68 (&dev->lock){+.+.}-{4:4}, at: xdp_set_features_flag+0x29/0x90
[ 833.295470]
[ 833.295470] but task is already holding lock:
[ 833.301400] ffff949d2b148c68 (&dev->lock){+.+.}-{4:4}, at: gve_shutdown+0x44/0x90 [gve]
[ 833.309508]
[ 833.309508] other info that might help us debug this:
[ 833.316130] Possible unsafe locking scenario:
[ 833.316130]
[ 833.322142] CPU0
[ 833.324681] ----
[ 833.327220] lock(&dev->lock);
[ 833.330455] lock(&dev->lock);
[ 833.333689]
[ 833.333689] *** DEADLOCK ***
[ 833.333689]
[ 833.339701] May be due to missing lock nesting notation
[ 833.339701]
[ 833.346582] 5 locks held by systemd-shutdow/1:
[ 833.351205] #0: ffffffffa9c89130 (system_transition_mutex){+.+.}-{4:4}, at: __se_sys_reboot+0xe6/0x210
[ 833.360695] #1: ffff93b399e5c1b8 (&dev->mutex){....}-{4:4}, at: device_shutdown+0xb4/0x1f0
[ 833.369144] #2: ffff949d19a471b8 (&dev->mutex){....}-{4:4}, at: device_shutdown+0xc2/0x1f0
[ 833.377603] #3: ffffffffa9eca050 (rtnl_mutex){+.+.}-{4:4}, at: gve_shutdown+0x33/0x90 [gve]
[ 833.386138] #4: ffff949d2b148c68 (&dev->lock){+.+.}-{4:4}, at: gve_shutdown+0x44/0x90 [gve]
Introduce xdp_features_(set|clear)_redirect_target_locked() versions
which assume that the netdev lock has already been acquired before
setting the XDP feature flag and update GVE to use the locked version.
Fixes: 03df156dd3 ("xdp: double protect netdev->xdp_flags with netdev->lock")
Tested-by: Mina Almasry <almasrymina@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Acked-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20250422011643.3509287-1-joshwash@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Many drivers populate the stats buffer using C-String based APIs (e.g.
ethtool_sprintf() and ethtool_puts()), usually when building up the
list of stats individually (i.e. with a for() loop). This, however,
requires that the source strings be populated in such a way as to have
a terminating NUL byte in the source.
Other drivers populate the stats buffer directly using one big memcpy()
of an entire array of strings. No NUL termination is needed here, as the
bytes are being directly passed through. Yet others will build up the
stats buffer individually, but also use memcpy(). This, too, does not
need NUL termination of the source strings.
However, there are cases where the strings that populate the
source stats strings are exactly ETH_GSTRING_LEN long, and GCC
15's -Wunterminated-string-initialization option complains that the
trailing NUL byte has been truncated. This situation is fine only if the
driver is using the memcpy() approach. If the C-String APIs are used,
the destination string name will have its final byte truncated by the
required trailing NUL byte applied by the C-string API.
For drivers that are already using memcpy() but have initializers that
truncate the NUL terminator, mark their source strings as __nonstring to
silence the GCC warnings.
For drivers that have initializers that truncate the NUL terminator and
are using the C-String APIs, switch to memcpy() to avoid destination
string truncation and mark their source strings as __nonstring to silence
the GCC warnings. (Also introduce ethtool_cpy() as a helper to make this
an easy replacement).
Specifically the following warnings were investigated and addressed:
../drivers/net/ethernet/chelsio/cxgb/cxgb2.c:364:9: warning: initializer-string for array of 'char' truncates NUL terminator but destination lacks 'nonstring' attribute (33 chars into 32 available) [-Wunterminated-string-initialization]
364 | "TxFramesAbortedDueToXSCollisions",
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../drivers/net/ethernet/freescale/enetc/enetc_ethtool.c:165:33: warning: initializer-string for array of 'char' truncates NUL terminator but destination lacks 'nonstring' attribute (33 chars into 32 available) [-Wunterminated-string-initialization]
165 | { ENETC_PM_R1523X(0), "MAC rx 1523 to max-octet packets" },
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../drivers/net/ethernet/freescale/enetc/enetc_ethtool.c:190:33: warning: initializer-string for array of 'char' truncates NUL terminator but destination lacks 'nonstring' attribute (33 chars into 32 available) [-Wunterminated-string-initialization]
190 | { ENETC_PM_T1523X(0), "MAC tx 1523 to max-octet packets" },
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../drivers/net/ethernet/google/gve/gve_ethtool.c:76:9: warning: initializer-string for array of 'char' truncates NUL terminator but destination lacks 'nonstring' attribute (33 chars into 32 available) [-Wunterminated-string-initialization]
76 | "adminq_dcfg_device_resources_cnt", "adminq_set_driver_parameter_cnt",
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c:117:53: warning: initializer-string for array of 'char' truncates NUL terminator but destination lacks 'nonstring' attribute (33 chars into 32 available) [-Wunterminated-string-initialization]
117 | STMMAC_STAT(ptp_rx_msg_type_pdelay_follow_up),
| ^
../drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c:46:12: note: in definition of macro 'STMMAC_STAT'
46 | { #m, sizeof_field(struct stmmac_extra_stats, m), \
| ^
../drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c:328:24: warning: initializer-string for array of 'char' truncates NUL terminator but destination lacks 'nonstring' attribute (33 chars into 32 available) [-Wunterminated-string-initialization]
328 | .str = "a_mac_control_frames_transmitted",
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c:340:24: warning: initializer-string for array of 'char' truncates NUL terminator but destination lacks 'nonstring' attribute (33 chars into 32 available) [-Wunterminated-string-initialization]
340 | .str = "a_pause_mac_ctrl_frames_received",
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Kees Cook <kees@kernel.org>
Reviewed-by: Petr Machata <petrm@nvidia.com> # for mlxsw
Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com>
Link: https://patch.msgid.link/20250416010210.work.904-kees@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Protect xdp_features with netdev->lock. This way pure readers
no longer have to take rtnl_lock to access the field.
This includes calling NETDEV_XDP_FEAT_CHANGE under the lock.
Looks like that's fine for bonding, the only "real" listener,
it's the same as ethtool feature change.
In terms of normal drivers - only GVE need special consideration
(other drivers don't use instance lock or don't support XDP).
It calls xdp_set_features_flag() helper from gve_init_priv() which
in turn is called from gve_reset_recovery() (locked), or prior
to netdev registration. So switch to _locked.
Reviewed-by: Joe Damato <jdamato@fastly.com>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Acked-by: Harshitha Ramamurthy <hramamurthy@google.com>
Acked-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20250408195956.412733-6-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
timer_delete[_sync]() replaces del_timer[_sync](). Convert the whole tree
over and remove the historical wrapper inlines.
Conversion was done with coccinelle plus manual fixups where necessary.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
All the misc entry points end up calling into either gve_open()
or gve_close(), they take rtnl_lock today but since the recent
instance locking changes should also take the instance lock.
Found by code inspection and untested.
Fixes: cae03e5bdd ("net: hold netdev instance lock during queue operations")
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com>
Link: https://patch.msgid.link/20250328164742.1268069-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Merge in late fixes to prepare for the 6.15 net-next PR.
No conflicts, adjacent changes:
drivers/net/ethernet/broadcom/bnxt/bnxt.c
919f9f497d ("eth: bnxt: fix out-of-range access of vnic_info array")
fe96d717d3 ("bnxt_en: Extend queue stop/start for TX rings")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
In order to support installing an XDP program on DQ, RX buffers need to
be reposted using 4K buffers, which is larger than the default packet
buffer size of 2K. This is needed to accommodate the extra head and tail
that accompanies the data portion of an XDP buffer. Continuing to use 2K
buffers would mean that the packet buffer size for the NIC would have to
be restricted to 2048 - 320 - 256 = 1472B. However, this is problematic
for two reasons: first, 1472 is not a packet buffer size accepted by
GVE; second, at least 1474B of buffer space is needed to accommodate an
MTU of 1460, which is the default on GCP. As such, we allocate 4K
buffers, and post a 2K section of those 4K buffers (offset relative to
the XDP headroom) to the NIC for DMA to avoid a potential extra copy.
Because the GQ-QPL datapath requires copies regardless, this change was
not needed to support XDP in that case.
To capture this subtlety, a new field, packet_buffer_truesize, has been
added to the rx ring struct to represent size of the allocated buffer,
while packet_buffer_size has been left to represent the portion of the
buffer posted to the NIC.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Link: https://patch.msgid.link/20250321002910.1343422-6-hramamurthy@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The data_buffer_size_dqo field in gve_priv and the packet_buffer_size
field in gve_rx_ring theoretically have the same meaning, but they are
defined in two different places and used in two separate contexts. There
is no good reason for this, so this change merges those fields into the
packet_buffer_size field in the RX ring.
This change also introduces a packet_buffer_size field to struct
gve_rx_queue_config to account for cases where queues are not allocated,
such as when the interface is down.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Link: https://patch.msgid.link/20250321002910.1343422-5-hramamurthy@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit ebdfae0d37 ("gve: adopt page pool for DQ RDA mode") introduced
a buf_size field to the gve_rx_slot_page_info struct, which can be used
in the datapath to take the place of the packet_buffer_size field, as it
will already be hot in the cache due to its extensive use. Using the
buf_size field in the datapath frees up the packet_buffer_size field in
the GQ-specific RX cacheline to be generalized for GQ and DQ (in the
next patch), as there is currently no common packet buffer size field
between the two queue formats.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Link: https://patch.msgid.link/20250321002910.1343422-4-hramamurthy@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
An earlier patch series[1] introduced RX/TX ring allocation configuration
structs which contained metadata used to allocate and configure new RX
and TX rings. This led to a much cleaner and safer allocation pattern
wherein queue resources were not deallocated until new queue resources
were successfully allocated.
Migrate the XDP allocation path to use the same pattern to allow for the
existence of a single allocation path instead of relying on XDP-specific
allocation methods. These extra allocation methods result in the
duplication of many existing behaviors while being prone to error when
configuration changes unrelated to XDP occur.
Link: https://lore.kernel.org/netdev/20240122182632.1102721-1-shailend@google.com/ [1]
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Link: https://patch.msgid.link/20250321002910.1343422-3-hramamurthy@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
To add netmem support to the gve driver, add a union
to the struct gve_rx_slot_page_info. netmem_ref is used for
DQO queue format's raw DMA addressing(RDA) mode. The struct
page is retained for other usecases.
Then, switch to using relevant netmem helper functions for
page pool and skb frag management.
Reviewed-by: Mina Almasry <almasrymina@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com>
Link: https://patch.msgid.link/20250307003905.601175-1-hramamurthy@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
For the drivers that use queue management API, switch to the mode where
core stack holds the netdev instance lock. This affects the following
drivers:
- bnxt
- gve
- netdevsim
Originally I locked only start/stop, but switched to holding the
lock over all iterations to make them look atomic to the device
(feels like it should be easier to reason about).
Reviewed-by: Eric Dumazet <edumazet@google.com>
Cc: Saeed Mahameed <saeed@kernel.org>
Signed-off-by: Stanislav Fomichev <sdf@fomichev.me>
Link: https://patch.msgid.link/20250305163732.2766420-6-sdf@fomichev.me
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Not all the devices have the capability for the driver to query for the
registered RSS configuration. The driver can discover this by checking
the relevant device option during setup. If it cannot, the driver needs
to store the RSS config cache and directly return such cache when
queried by the ethtool. RSS config is inited when driver probes. Also the
default RSS config will be adjusted when there is RX queue count change.
At this point, only keys of GVE_RSS_KEY_SIZE and indirection tables of
GVE_RSS_INDIR_SIZE are supported.
Signed-off-by: Ziwei Xiao <ziweixiao@google.com>
Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Signed-off-by: Jeroen de Borst <jeroendb@google.com>
Link: https://patch.msgid.link/20250219200451.3348166-1-jeroendb@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Before this patch the NETDEV_XDP_ACT_NDO_XMIT XDP feature flag is set by
default as part of driver initialization, and is never cleared. However,
this flag differs from others in that it is used as an indicator for
whether the driver is ready to perform the ndo_xdp_xmit operation as
part of an XDP_REDIRECT. Kernel helpers
xdp_features_(set|clear)_redirect_target exist to convey this meaning.
This patch ensures that the netdev is only reported as a redirect target
when XDP queues exist to forward traffic.
Fixes: 39a7f4aa3e ("gve: Add XDP REDIRECT support for GQI-QPL format")
Cc: stable@vger.kernel.org
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Jeroen de Borst <jeroendb@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Link: https://patch.msgid.link/20250214224417.1237818-1-joshwash@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit f85949f982 ("xdp: add xdp_set_features_flag utility routine")
added routines to inform the core about XDP flag changes.
GVE support was added around the same time and missed using them.
GVE only changes the flags on error recover or resume.
Presumably the flags may change during resume if VM migrated.
User would not get the notification and upper devices would
not get a chance to recalculate their flags.
Fixes: 75eaae158b ("gve: Add XDP DROP and TX support for GQI-QPL format")
Reviewed-By: Jeroen de Borst <jeroendb@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20250106180210.1861784-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit ba0925c34e ("gve: process XSK TX descriptors as part of RX NAPI")
moved XSK TX processing to be part of the RX NAPI. However, that commit
did not include triggering the RX NAPI in gve_xsk_wakeup. This is
necessary because the TX NAPI only processes TX completions, meaning
that a TX wakeup would not actually trigger XSK descriptor processing.
Also, the branch on XDP_WAKEUP_TX was supposed to have been removed, as
the NAPI should be scheduled whether the wakeup is for RX or TX.
Fixes: ba0925c34e ("gve: process XSK TX descriptors as part of RX NAPI")
Cc: stable@vger.kernel.org
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Link: https://patch.msgid.link/20241221032807.302244-1-pkaligineedi@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch fixes a number of consistency issues in the queue allocation
path related to XDP.
As it stands, the number of allocated XDP queues changes in three
different scenarios.
1) Adding an XDP program while the interface is up via
gve_add_xdp_queues
2) Removing an XDP program while the interface is up via
gve_remove_xdp_queues
3) After queues have been allocated and the old queue memory has been
removed in gve_queues_start.
However, the requirement for the interface to be up for
gve_(add|remove)_xdp_queues to be called, in conjunction with the fact
that the number of queues stored in priv isn't updated until _after_ XDP
queues have been allocated in the normal queue allocation path means
that if an XDP program is added while the interface is down, XDP queues
won't be added until the _second_ if_up, not the first.
Given the expectation that the number of XDP queues is equal to the
number of RX queues, scenario (3) has another problematic implication.
When changing the number of queues while an XDP program is loaded, the
number of XDP queues must be updated as well, as there is logic in the
driver (gve_xdp_tx_queue_id()) which relies on every RX queue having a
corresponding XDP TX queue. However, the number of XDP queues stored in
priv would not be updated until _after_ a close/open leading to a
mismatch in the number of XDP queues reported vs the number of XDP
queues which actually exist after the queue count update completes.
This patch remedies these issues by doing the following:
1) The allocation config getter function is set up to retrieve the
_expected_ number of XDP queues to allocate instead of relying
on the value stored in `priv` which is only updated once the queues
have been allocated.
2) When adjusting queues, XDP queues are adjusted to match the number of
RX queues when XDP is enabled. This only works in the case when
queues are live, so part (1) of the fix must still be available in
the case that queues are adjusted when there is an XDP program and
the interface is down.
Fixes: 5f08cd3d64 ("gve: Alloc before freeing when adjusting queues")
Cc: stable@vger.kernel.org
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Shailend Chand <shailend@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When busy polling is enabled, xsk_sendmsg for AF_XDP zero copy marks
the NAPI ID corresponding to the memory pool allocated for the socket.
In GVE, this NAPI ID will never correspond to a NAPI ID of one of the
dedicated XDP TX queues registered with the umem because XDP TX is not
set up to share a NAPI with a corresponding RX queue.
This patch moves XSK TX descriptor processing from the TX NAPI to the RX
NAPI, and the gve_xsk_wakeup callback is updated to use the RX NAPI
instead of the TX NAPI, accordingly. The branch on if the wakeup is for
TX is removed, as the NAPI poll should be invoked whether the wakeup is
for TX or for RX.
Fixes: fd8e40321a ("gve: Add AF_XDP zero-copy support for GQI-QPL format")
Cc: stable@vger.kernel.org
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Signed-off-by: Joshua Washington <joshwash@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch predicates the enabling and disabling of XSK pools on the
existence of queues. As it stands, if the interface is down, disabling
or enabling XSK pools would result in a crash, as the RX queue pointer
would be NULL. XSK pool registration will occur as part of the next
interface up.
Similarly, xsk_wakeup needs be guarded against queues disappearing
while the function is executing, so a check against the
GVE_PRIV_FLAGS_NAPI_ENABLED flag is added to synchronize with the
disabling of the bit and the synchronize_net() in gve_turndown.
Fixes: fd8e40321a ("gve: Add AF_XDP zero-copy support for GQI-QPL format")
Cc: stable@vger.kernel.org
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Shailend Chand <shailend@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Larysa Zaremba <larysa.zaremba@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In GVE, dedicated XDP queues only exist when an XDP program is installed
and the interface is up. As such, the NDO XDP XMIT callback should
return early if either of these conditions are false.
In the case of no loaded XDP program, priv->num_xdp_queues=0 which can
cause a divide-by-zero error, and in the case of interface down,
num_xdp_queues remains untouched to persist XDP queue count for the next
interface up, but the TX pointer itself would be NULL.
The XDP xmit callback also needs to synchronize with a device
transitioning from open to close. This synchronization will happen via
the GVE_PRIV_FLAGS_NAPI_ENABLED bit along with a synchronize_net() call,
which waits for any RCU critical sections at call-time to complete.
Fixes: 39a7f4aa3e ("gve: Add XDP REDIRECT support for GQI-QPL format")
Cc: stable@vger.kernel.org
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Shailend Chand <shailend@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When stopping XDP TX rings, the XDP clean function needs to be called to
clean out the entire queue, similar to what happens in the normal TX
queue case. Otherwise, the FIFO won't be cleared correctly, and
xsk_tx_completed won't be reported.
Fixes: 75eaae158b ("gve: Add XDP DROP and TX support for GQI-QPL format")
Cc: stable@vger.kernel.org
Signed-off-by: Joshua Washington <joshwash@google.com>
Signed-off-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
