[PATCH net-next v2] net: napi: Prevent overflow of napi_defer_hard_irqs

[PATCH net-next v2] net: napi: Prevent overflow of napi_defer_hard_irqs

[Joe Damato]

In commit 6f8b12d661d0 ("net: napi: add hard irqs deferral feature")
napi_defer_irqs was added to net_device and napi_defer_irqs_count was
added to napi_struct, both as type int.

This value never goes below zero, so there is not reason for it to be a
signed int. Change the type for both from int to u32, and add an
overflow check to sysfs to limit the value to S32_MAX.

The limit of S32_MAX was chosen because the practical limit before this
patch was S32_MAX (anything larger was an overflow) and thus there are
no behavioral changes introduced. If the extra bit is needed in the
future, the limit can be raised.

Before this patch:

$ sudo bash -c 'echo 2147483649 > /sys/class/net/eth4/napi_defer_hard_irqs'
$ cat /sys/class/net/eth4/napi_defer_hard_irqs
-2147483647

After this patch:

$ sudo bash -c 'echo 2147483649 > /sys/class/net/eth4/napi_defer_hard_irqs'
bash: line 0: echo: write error: Numerical result out of range

Similarly, /sys/class/net/XXXXX/tx_queue_len is defined as unsigned:

include/linux/netdevice.h:      unsigned int            tx_queue_len;

And has an overflow check:

dev_change_tx_queue_len(..., unsigned long new_len):

  if (new_len != (unsigned int)new_len)
          return -ERANGE;

Cc: Eric Dumazet <edumazet@google.com>
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Joe Damato <jdamato@fastly.com>
---
 Documentation/networking/net_cachelines/net_device.rst | 2 +-
 include/linux/netdevice.h                              | 4 ++--
 net/core/net-sysfs.c                                   | 6 +++++-
 3 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/Documentation/networking/net_cachelines/net_device.rst b/Documentation/networking/net_cachelines/net_device.rst
index a0e0fab8161a..615baddb398c 100644
--- a/Documentation/networking/net_cachelines/net_device.rst
+++ b/Documentation/networking/net_cachelines/net_device.rst
@@ -98,7 +98,7 @@ unsigned_int                        num_rx_queues
 unsigned_int                        real_num_rx_queues      -                   read_mostly         get_rps_cpu
 struct_bpf_prog*                    xdp_prog                -                   read_mostly         netif_elide_gro()
 unsigned_long                       gro_flush_timeout       -                   read_mostly         napi_complete_done
-int                                 napi_defer_hard_irqs    -                   read_mostly         napi_complete_done
+u32                                 napi_defer_hard_irqs    -                   read_mostly         napi_complete_done
 unsigned_int                        gro_max_size            -                   read_mostly         skb_gro_receive
 unsigned_int                        gro_ipv4_max_size       -                   read_mostly         skb_gro_receive
 rx_handler_func_t*                  rx_handler              read_mostly         -                   __netif_receive_skb_core
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index fce70990b209..971a24a2d117 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -356,7 +356,7 @@ struct napi_struct {
 
 	unsigned long		state;
 	int			weight;
-	int			defer_hard_irqs_count;
+	u32			defer_hard_irqs_count;
 	unsigned long		gro_bitmask;
 	int			(*poll)(struct napi_struct *, int);
 #ifdef CONFIG_NETPOLL
@@ -2065,7 +2065,7 @@ struct net_device {
 	unsigned int		real_num_rx_queues;
 	struct netdev_rx_queue	*_rx;
 	unsigned long		gro_flush_timeout;
-	int			napi_defer_hard_irqs;
+	u32			napi_defer_hard_irqs;
 	unsigned int		gro_max_size;
 	unsigned int		gro_ipv4_max_size;
 	rx_handler_func_t __rcu	*rx_handler;
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 444f23e74f8e..b34d731524d5 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -32,6 +32,7 @@
 #ifdef CONFIG_SYSFS
 static const char fmt_hex[] = "%#x\n";
 static const char fmt_dec[] = "%d\n";
+static const char fmt_uint[] = "%u\n";
 static const char fmt_ulong[] = "%lu\n";
 static const char fmt_u64[] = "%llu\n";
 
@@ -425,6 +426,9 @@ NETDEVICE_SHOW_RW(gro_flush_timeout, fmt_ulong);
 
 static int change_napi_defer_hard_irqs(struct net_device *dev, unsigned long val)
 {
+	if (val > S32_MAX)
+		return -ERANGE;
+
 	WRITE_ONCE(dev->napi_defer_hard_irqs, val);
 	return 0;
 }
@@ -438,7 +442,7 @@ static ssize_t napi_defer_hard_irqs_store(struct device *dev,
 
 	return netdev_store(dev, attr, buf, len, change_napi_defer_hard_irqs);
 }
-NETDEVICE_SHOW_RW(napi_defer_hard_irqs, fmt_dec);
+NETDEVICE_SHOW_RW(napi_defer_hard_irqs, fmt_uint);
 
 static ssize_t ifalias_store(struct device *dev, struct device_attribute *attr,
 			     const char *buf, size_t len)
-- 
2.25.1

Re: [PATCH net-next v2] net: napi: Prevent overflow of napi_defer_hard_irqs

[Eric Dumazet]

On Wed, Sep 4, 2024 at 5:34 PM Joe Damato <jdamato@fastly.com> wrote:
>
> In commit 6f8b12d661d0 ("net: napi: add hard irqs deferral feature")
> napi_defer_irqs was added to net_device and napi_defer_irqs_count was
> added to napi_struct, both as type int.
>
> This value never goes below zero, so there is not reason for it to be a
> signed int. Change the type for both from int to u32, and add an
> overflow check to sysfs to limit the value to S32_MAX.
>
> The limit of S32_MAX was chosen because the practical limit before this
> patch was S32_MAX (anything larger was an overflow) and thus there are
> no behavioral changes introduced. If the extra bit is needed in the
> future, the limit can be raised.

Reviewed-by: Eric Dumazet <edumazet@google.com>

Re: [PATCH net-next v2] net: napi: Prevent overflow of napi_defer_hard_irqs

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Wed,  4 Sep 2024 15:34:30 +0000 you wrote:
> In commit 6f8b12d661d0 ("net: napi: add hard irqs deferral feature")
> napi_defer_irqs was added to net_device and napi_defer_irqs_count was
> added to napi_struct, both as type int.
> 
> This value never goes below zero, so there is not reason for it to be a
> signed int. Change the type for both from int to u32, and add an
> overflow check to sysfs to limit the value to S32_MAX.
> 
> [...]

Here is the summary with links:
  - [net-next,v2] net: napi: Prevent overflow of napi_defer_hard_irqs
    https://git.kernel.org/netdev/net-next/c/08062af0a521

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html