* [PATCH net] net: bcmgenet: reject undersized RX classification rule buffers
@ 2026-11-21 0:15 Artemii Mashanov
2026-11-21 0:17 ` netdev-bot+sinfo
2026-11-21 20:48 ` Nicolai Buchwitz
0 siblings, 2 replies; 4+ messages in thread
From: Artemii Mashanov @ 2026-11-21 0:15 UTC (permalink / raw)
To: netdev
Cc: Doug Berger, Florian Fainelli, Nicolai Buchwitz,
Broadcom internal kernel review list, Andrew Lunn,
David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni
bcmgenet_get_rxnfc() returns success with an incomplete list of RX
classification rules when ETHTOOL_GRXCLSRLALL receives fewer slots
than the number of installed rules.
Return -EMSGSIZE when the next rule cannot fit in the caller's
buffer. On success, report the actual number of copied rules.
The source-extracted ASan/UBSan regression tests passed.
Fixes: 3e370952287c ("net: bcmgenet: add support for ethtool rxnfc flows")
Signed-off-by: Artemii Mashanov <ralerrdirsardx@gmail.com>
---
drivers/net/ethernet/broadcom/genet/bcmgenet.c | 10 ++++++----
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 4d2be2a..6a4a7fd 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -1664,7 +1664,7 @@ static int bcmgenet_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
struct bcmgenet_priv *priv = netdev_priv(dev);
struct bcmgenet_rxnfc_rule *rule;
int err = 0;
- int i = 0;
+ u32 i = 0;
switch (cmd->cmd) {
case ETHTOOL_GRXCLSRLCNT:
@@ -1675,9 +1675,11 @@ static int bcmgenet_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
err = bcmgenet_get_flow(dev, cmd, cmd->fs.location);
break;
case ETHTOOL_GRXCLSRLALL:
- list_for_each_entry(rule, &priv->rxnfc_list, list)
- if (i < cmd->rule_cnt)
- rule_locs[i++] = rule->fs.location;
+ list_for_each_entry(rule, &priv->rxnfc_list, list) {
+ if (i == cmd->rule_cnt)
+ return -EMSGSIZE;
+ rule_locs[i++] = rule->fs.location;
+ }
cmd->rule_cnt = i;
cmd->data = MAX_NUM_OF_FS_RULES;
break;
--
2.56.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH net] net: bcmgenet: reject undersized RX classification rule buffers
2026-11-21 0:15 [PATCH net] net: bcmgenet: reject undersized RX classification rule buffers Artemii Mashanov
@ 2026-11-21 0:17 ` netdev-bot+sinfo
2026-11-21 18:58 ` Artemii Mashanov
2026-11-21 20:48 ` Nicolai Buchwitz
1 sibling, 1 reply; 4+ messages in thread
From: netdev-bot+sinfo @ 2026-11-21 0:17 UTC (permalink / raw)
To: Artemii Mashanov
Cc: netdev, Doug Berger, Florian Fainelli, Nicolai Buchwitz,
Broadcom internal kernel review list, Andrew Lunn,
David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni
Hi!
This is an automated message. This series looks like a fix, but its
commit messages seem to be missing some information:
- How the issue was discovered, e.g. hit in production, hit during
development, syzbot report, manual code inspection, LLM or static
analysis tool scan.
- Whether the issue was actually triggered, or is only theoretical
(e.g. found by LLM and not proven at runtime). If it was triggered
please include the symptoms, like the stack trace or kernel messages.
- What hardware the change was tested on. For driver fixes please
mention the device (and if relevant firmware version) used for
testing, or say that the change was not tested on real hardware.
Please do not repost the series just to address the above. Instead,
reply to this email with the missing information, so that reviewers
can take it into account.
*IMPORTANT*: If you do repost (for other reasons) - please include
the relevant info in the cover letter or commit messages.
--
The evaluation is done by an LLM so it may be wrong, if you think
that is the case please reply and explain.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH net] net: bcmgenet: reject undersized RX classification rule buffers
2026-11-21 0:17 ` netdev-bot+sinfo
@ 2026-11-21 18:58 ` Artemii Mashanov
0 siblings, 0 replies; 4+ messages in thread
From: Artemii Mashanov @ 2026-11-21 18:58 UTC (permalink / raw)
To: netdev-bot+sinfo
Cc: netdev, Doug Berger, Florian Fainelli, Nicolai Buchwitz,
Broadcom internal kernel review list, Andrew Lunn,
David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni
Hi,
A scripted source audit of `ETHTOOL_GRXCLSRLALL` handlers flagged this
path. Reviewing `bcmgenet_get_rxnfc()` showed that it stops copying
rule locations when the caller's buffer fills, then returns success
with a partial list.
The original behavior was identified by source inspection. The
ASan/UBSan tests exercised the patched handler in a standalone
userspace harness. They verified that an undersized buffer returns
`-EMSGSIZE` and that a successful call reports the actual number of
locations copied.
An earlier revision built as `genet.ko` against Linux 7.4-rc2. I did
not reproduce the issue through the ioctl on a running kernel or test
the patch on GENET hardware.
Thanks,
Artemii
сб, 21 нояб. 2026 г. в 03:17, <netdev-bot+sinfo@kernel.org>:
>
> Hi!
>
> This is an automated message. This series looks like a fix, but its
> commit messages seem to be missing some information:
>
> - How the issue was discovered, e.g. hit in production, hit during
> development, syzbot report, manual code inspection, LLM or static
> analysis tool scan.
>
> - Whether the issue was actually triggered, or is only theoretical
> (e.g. found by LLM and not proven at runtime). If it was triggered
> please include the symptoms, like the stack trace or kernel messages.
>
> - What hardware the change was tested on. For driver fixes please
> mention the device (and if relevant firmware version) used for
> testing, or say that the change was not tested on real hardware.
>
> Please do not repost the series just to address the above. Instead,
> reply to this email with the missing information, so that reviewers
> can take it into account.
>
> *IMPORTANT*: If you do repost (for other reasons) - please include
> the relevant info in the cover letter or commit messages.
> --
> The evaluation is done by an LLM so it may be wrong, if you think
> that is the case please reply and explain.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH net] net: bcmgenet: reject undersized RX classification rule buffers
2026-11-21 0:15 [PATCH net] net: bcmgenet: reject undersized RX classification rule buffers Artemii Mashanov
2026-11-21 0:17 ` netdev-bot+sinfo
@ 2026-11-21 20:48 ` Nicolai Buchwitz
1 sibling, 0 replies; 4+ messages in thread
From: Nicolai Buchwitz @ 2026-11-21 20:48 UTC (permalink / raw)
To: Artemii Mashanov
Cc: netdev, Doug Berger, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni
Hi Artemii
Thanks for your patch.
On 21.11.2026 01:15, Artemii Mashanov wrote:
> bcmgenet_get_rxnfc() returns success with an incomplete list of RX
> classification rules when ETHTOOL_GRXCLSRLALL receives fewer slots
> than the number of installed rules.
>
> Return -EMSGSIZE when the next rule cannot fit in the caller's
> buffer. On success, report the actual number of copied rules.
The loop already stops at rule_cnt, so nothing overflows. The short
list only can happen if a rule is added between the count and dump
ioctls - and then ethtool -n just shows one rule fewer.
Fine as a cleanup, but please resend for net-next without the Fixes tag.
> The source-extracted ASan/UBSan regression tests passed.
Please drop this line, it doesn't apply to the kernel change.
> [...]
Thanks,
Nicolai
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-11-21 20:48 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-21 0:15 [PATCH net] net: bcmgenet: reject undersized RX classification rule buffers Artemii Mashanov
2026-11-21 0:17 ` netdev-bot+sinfo
2026-11-21 18:58 ` Artemii Mashanov
2026-11-21 20:48 ` Nicolai Buchwitz
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox