From: Joshua Washington joshwash@google.com
When the tx tail is less than the head (in cases of wraparound), the TX consumed descriptor statistic in DQ will be reported as UINT32_MAX - head + tail, which is incorrect. Mask the difference of head and tail according to the ring size when reporting the statistic.
Cc: stable@vger.kernel.org Fixes: 2c9198356d56 ("gve: Add consumed counts to ethtool stats") Signed-off-by: Joshua Washington joshwash@google.com Signed-off-by: Harshitha Ramamurthy hramamurthy@google.com --- drivers/net/ethernet/google/gve/gve_ethtool.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c index 31a21ccf4863..4dea1fdce748 100644 --- a/drivers/net/ethernet/google/gve/gve_ethtool.c +++ b/drivers/net/ethernet/google/gve/gve_ethtool.c @@ -392,7 +392,9 @@ gve_get_ethtool_stats(struct net_device *netdev, */ data[i++] = 0; data[i++] = 0; - data[i++] = tx->dqo_tx.tail - tx->dqo_tx.head; + data[i++] = + (tx->dqo_tx.tail - tx->dqo_tx.head) & + tx->mask; } do { start =
On Wed, Apr 02, 2025 at 12:10:37AM +0000, Harshitha Ramamurthy wrote:
From: Joshua Washington joshwash@google.com
When the tx tail is less than the head (in cases of wraparound), the TX consumed descriptor statistic in DQ will be reported as UINT32_MAX - head + tail, which is incorrect. Mask the difference of head and tail according to the ring size when reporting the statistic.
Cc: stable@vger.kernel.org Fixes: 2c9198356d56 ("gve: Add consumed counts to ethtool stats") Signed-off-by: Joshua Washington joshwash@google.com Signed-off-by: Harshitha Ramamurthy hramamurthy@google.com
drivers/net/ethernet/google/gve/gve_ethtool.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c index 31a21ccf4863..4dea1fdce748 100644 --- a/drivers/net/ethernet/google/gve/gve_ethtool.c +++ b/drivers/net/ethernet/google/gve/gve_ethtool.c @@ -392,7 +392,9 @@ gve_get_ethtool_stats(struct net_device *netdev, */ data[i++] = 0; data[i++] = 0;
data[i++] = tx->dqo_tx.tail - tx->dqo_tx.head;
data[i++] =(tx->dqo_tx.tail - tx->dqo_tx.head) &tx->mask;
Reviewed-by: Michal Swiatkowski michal.swiatkowski@linux.intel.com
I will add it in gve_tx_dqo.c as num_used_tx_slots() and simplify num_avail_tx_slots() { return tx->mask - num_used_tx_slots(); } but it isn't needed, even maybe unwanted as this is just fix.
Thanks
} do { start =-- 2.49.0.472.ge94155a9ec-goog
On Wed, Apr 02, 2025 at 07:10:18AM +0200, Michal Swiatkowski wrote:
On Wed, Apr 02, 2025 at 12:10:37AM +0000, Harshitha Ramamurthy wrote:
From: Joshua Washington joshwash@google.com
When the tx tail is less than the head (in cases of wraparound), the TX consumed descriptor statistic in DQ will be reported as UINT32_MAX - head + tail, which is incorrect. Mask the difference of head and tail according to the ring size when reporting the statistic.
Cc: stable@vger.kernel.org Fixes: 2c9198356d56 ("gve: Add consumed counts to ethtool stats") Signed-off-by: Joshua Washington joshwash@google.com Signed-off-by: Harshitha Ramamurthy hramamurthy@google.com
drivers/net/ethernet/google/gve/gve_ethtool.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c index 31a21ccf4863..4dea1fdce748 100644 --- a/drivers/net/ethernet/google/gve/gve_ethtool.c +++ b/drivers/net/ethernet/google/gve/gve_ethtool.c @@ -392,7 +392,9 @@ gve_get_ethtool_stats(struct net_device *netdev, */ data[i++] = 0; data[i++] = 0;
data[i++] = tx->dqo_tx.tail - tx->dqo_tx.head;
data[i++] =(tx->dqo_tx.tail - tx->dqo_tx.head) &tx->mask;Reviewed-by: Michal Swiatkowski michal.swiatkowski@linux.intel.com
I will add it in gve_tx_dqo.c as num_used_tx_slots() and simplify num_avail_tx_slots() { return tx->mask - num_used_tx_slots(); } but it isn't needed, even maybe unwanted as this is just fix.
I think that would be a nice cleanup, but should be as a follow-up to this minimal fix.
Reviewed-by: Simon Horman horms@kernel.org
Hello:
This patch was applied to netdev/net.git (main) by Jakub Kicinski kuba@kernel.org:
On Wed, 2 Apr 2025 00:10:37 +0000 you wrote:
From: Joshua Washington joshwash@google.com
When the tx tail is less than the head (in cases of wraparound), the TX consumed descriptor statistic in DQ will be reported as UINT32_MAX - head + tail, which is incorrect. Mask the difference of head and tail according to the ring size when reporting the statistic.
[...]
Here is the summary with links: - [net] gve: handle overflow when reporting TX consumed descriptors https://git.kernel.org/netdev/net/c/15970e1b23f5
You are awesome, thank you!
linux-stable-mirror@lists.linaro.org
-
Harshitha Ramamurthy -
Michal Swiatkowski -
patchwork-bot+netdevbpf＠kernel.org -
Simon Horman