kernel/linux.git/drivers/net/ethernet/google, branch v5.10.257

gve: Correct ethtool rx_dropped calculation

2026-02-11T12:34:25+00:00

[ Upstream commit c7db85d579a1dccb624235534508c75fbf2dfe46 ] The gve driver's "rx_dropped" statistic, exposed via `ethtool -S`, incorrectly includes `rx_buf_alloc_fail` counts. These failures represent an inability to allocate receive buffers, not true packet drops where a received packet is discarded. This misrepresentation can lead to inaccurate diagnostics. This patch rectifies the ethtool "rx_dropped" calculation. It removes `rx_buf_alloc_fail` from the total and adds `xdp_tx_errors` and `xdp_redirect_errors`, which represent legitimate packet drops within the XDP path. Cc: stable@vger.kernel.org Fixes: 433e274b8f7b ("gve: Add stats for gve.") Signed-off-by: Max Yuan Reviewed-by: Jordan Rhee Reviewed-by: Joshua Washington Reviewed-by: Matt Olson Signed-off-by: Harshitha Ramamurthy Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20260202193925.3106272-3-hramamurthy@google.com Signed-off-by: Jakub Kicinski [ removed rx_buf_alloc_fail from rx_dropped calculation ] Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

gve: Fix stats report corruption on queue count change

2026-02-11T12:34:25+00:00

[ Upstream commit 7b9ebcce0296e104a0d82a6b09d68564806158ff ] The driver and the NIC share a region in memory for stats reporting. The NIC calculates its offset into this region based on the total size of the stats region and the size of the NIC's stats. When the number of queues is changed, the driver's stats region is resized. If the queue count is increased, the NIC can write past the end of the allocated stats region, causing memory corruption. If the queue count is decreased, there is a gap between the driver and NIC stats, leading to incorrect stats reporting. This change fixes the issue by allocating stats region with maximum size, and the offset calculation for NIC stats is changed to match with the calculation of the NIC. Cc: stable@vger.kernel.org Fixes: 24aeb56f2d38 ("gve: Add Gvnic stats AQ command and ethtool show/set-priv-flags.") Signed-off-by: Debarghya Kundu Reviewed-by: Joshua Washington Signed-off-by: Harshitha Ramamurthy Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20260202193925.3106272-2-hramamurthy@google.com Signed-off-by: Jakub Kicinski [ Same changes as 6.1 + context ] Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

gve: Return error for unknown admin queue command

2025-08-28T14:22:43+00:00

[ Upstream commit b11344f63fdd9e8c5121148a6965b41079071dd2 ] 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 Link: https://patch.msgid.link/20250616054504.1644770-2-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

gve: Fix stuck TX queue for DQ queue format

2025-08-28T14:22:30+00:00

commit b03f15c0192b184078206760c839054ae6eb4eaa upstream. gve_tx_timeout was calculating missed completions in a way that is only relevant in the GQ queue format. Additionally, it was attempting to disable device interrupts, which is not needed in either GQ or DQ queue formats. As a result, TX timeouts with the DQ queue format likely would have triggered early resets without kicking the queue at all. This patch drops the check for pending work altogether and always kicks the queue after validating the queue has not seen a TX timeout too recently. Cc: stable@vger.kernel.org Fixes: 87a7f321bb6a ("gve: Recover from queue stall due to missed IRQ") Co-developed-by: Tim Hostetler Signed-off-by: Tim Hostetler Signed-off-by: Praveen Kaligineedi Signed-off-by: Harshitha Ramamurthy Link: https://patch.msgid.link/20250717192024.1820931-1-hramamurthy@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Greg Kroah-Hartman

gve: Fix RX_BUFFERS_POSTED stat to report per-queue fill_cnt

2025-06-27T10:04:07+00:00

[ Upstream commit f41a94aade120dc60322865f363cee7865f2df01 ] Previously, the RX_BUFFERS_POSTED stat incorrectly reported the fill_cnt from RX queue 0 for all queues, resulting in inaccurate per-queue statistics. Fix this by correctly indexing priv->rx[idx].fill_cnt for each RX queue. Fixes: 24aeb56f2d38 ("gve: Add Gvnic stats AQ command and ethtool show/set-priv-flags.") Signed-off-by: Alok Tiwari Link: https://patch.msgid.link/20250527130830.1812903-1-alok.a.tiwari@oracle.com Signed-off-by: Paolo Abeni Signed-off-by: Sasha Levin

gve: Use size_add() in call to struct_size()

2023-11-20T10:06:44+00:00

[ Upstream commit d692873cbe861a870cdc9cbfb120eefd113c3dfd ] If, for any reason, `tx_stats_num + rx_stats_num` wraps around, the protection that struct_size() adds against potential integer overflows is defeated. Fix this by hardening call to struct_size() with size_add(). Fixes: 691f4077d560 ("gve: Replace zero-length array with flexible-array member") Signed-off-by: Gustavo A. R. Silva Reviewed-by: Kees Cook Signed-off-by: David S. Miller Signed-off-by: Sasha Levin

gve: Set default duplex configuration to full

2023-07-27T06:44:20+00:00

[ Upstream commit 0503efeadbf6bb8bf24397613a73b67e665eac5f ] Current duplex mode was unset in the driver, resulting in the default parameter being set to 0, which corresponds to half duplex. It might mislead users to have incorrect expectation about the driver's transmission capabilities. Set the default duplex configuration to full, as the driver runs in full duplex mode at this point. Fixes: 7e074d5a76ca ("gve: Enable Link Speed Reporting in the driver.") Signed-off-by: Junfeng Guo Reviewed-by: Leon Romanovsky Message-ID: <20230706044128.2726747-1-junfeng.guo@intel.com> Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

gve: Cache link_speed value from device

2023-04-05T09:23:36+00:00

[ Upstream commit 68c3e4fc8628b1487c965aabb29207249657eb5f ] The link speed is never changed for the uptime of a VM, and the current implementation sends an admin queue command for each call. Admin queue command invocations have nontrivial overhead (e.g., VM exits), which can be disruptive to users if triggered frequently. Our telemetry data shows that there are VMs that make frequent calls to this admin queue command. Caching the result of the original admin queue command would eliminate the need to send multiple admin queue commands on subsequent calls to retrieve link speed. Fixes: 7e074d5a76ca ("gve: Enable Link Speed Reporting in the driver.") Signed-off-by: Joshua Washington Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20230321172332.91678-1-joshwash@google.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

net: Use u64_stats_fetch_begin_irq() for stats fetch.

2022-09-08T09:11:40+00:00

commit 278d3ba61563ceed3cb248383ced19e14ec7bc1f upstream. On 32bit-UP u64_stats_fetch_begin() disables only preemption. If the reader is in preemptible context and the writer side (u64_stats_update_begin*()) runs in an interrupt context (IRQ or softirq) then the writer can update the stats during the read operation. This update remains undetected. Use u64_stats_fetch_begin_irq() to ensure the stats fetch on 32bit-UP are not interrupted by a writer. 32bit-SMP remains unaffected by this change. Cc: "David S. Miller" Cc: Catherine Sullivan Cc: David Awogbemila Cc: Dimitris Michailidis Cc: Eric Dumazet Cc: Hans Ulli Kroll Cc: Jakub Kicinski Cc: Jeroen de Borst Cc: Johannes Berg Cc: Linus Walleij Cc: Paolo Abeni Cc: Simon Horman Cc: linux-arm-kernel@lists.infradead.org Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Cc: oss-drivers@corigine.com Cc: stable@vger.kernel.org Signed-off-by: Sebastian Andrzej Siewior Reviewed-by: Simon Horman Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

gve: fix the wrong AdminQ buffer queue index check

2022-02-08T17:30:39+00:00

commit 1f84a9450d75e08af70d9e2f2d5e1c0ac0c881d2 upstream. The 'tail' and 'head' are 'unsigned int' type free-running count, when 'head' is overflow, the 'int i (= tail) < u32 head' will be false: Only '- loop 0: idx = 63' result is shown, so it needs to use 'int' type to compare, it can handle the overflow correctly. typedef uint32_t u32; int main() { u32 tail, head; int stail, shead; int i, loop; tail = 0xffffffff; head = 0x00000000; for (i = tail, loop = 0; i < head; i++) { unsigned int idx = i & 63; printf("+ loop %d: idx = %u\n", loop++, idx); } stail = tail; shead = head; for (i = stail, loop = 0; i < shead; i++) { unsigned int idx = i & 63; printf("- loop %d: idx = %u\n", loop++, idx); } return 0; } Fixes: 5cdad90de62c ("gve: Batch AQ commands for creating and destroying queues.") Signed-off-by: Haiyue Wang Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman