Skip to contentSkip to conversation

Various core fixes spotted by LLMs

PATCH Applied Leon Romanovsky 4 participants 10 messages Oct 15 – Oct 18 InfiniBand

lore
Share & tools
lore
#0 lore raw

cover Various core fixes spotted by LLMs

A batch of core fixes for various issues identified by LLMs.

---
Leon Romanovsky (3):
RDMA/core: Preserve the selected source GID on failed AH modification RDMA/core: Free unpublished port statistics after initial-read failure RDMA/core: Reject LAG selection when source MAC lookup fails

 drivers/infiniband/core/lag.c   | 12 ++++++++----
 drivers/infiniband/core/sysfs.c |  9 +++++++--
 drivers/infiniband/core/verbs.c |  3 ++-
 3 files changed, 17 insertions(+), 7 deletions(-)

---
base-commit: 66196a559be5963b1ced18418c876caf9bfb76e6
change-id: 20261015-ib-core-fixes-fa78b7068672

Best regards,

Signature · 1 line
Leon Romanovsky <[email protected]>
#1 lore raw

PATCH 1/3 RDMA/core: Preserve the selected source GID on failed AH modification

From: Leon Romanovsky <[email protected]>

When an address-handle modification fails, the core replaces its source-GID reference. The old address then loses the reference which keeps its source entry alive.

Fixes 1a1f460ff151 ("RDMA: Hold the sgid_attr inside the struct ib_ah/qp")
---
 drivers/infiniband/core/verbs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/infiniband/core/verbs.c +2 -1
index edf62652f9d3..7d52d3036208 100644
@@ -1011,7 +1011,8 @@ int rdma_modify_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr)
 		ah->device->ops.modify_ah(ah, ah_attr) :
 		-EOPNOTSUPP;
 
-	ah->sgid_attr = rdma_update_sgid_attr(ah_attr, ah->sgid_attr);
+	if (!ret)
+		ah->sgid_attr = rdma_update_sgid_attr(ah_attr, ah->sgid_attr);
 	rdma_unfill_sgid_attr(ah_attr, old_sgid_attr);
 	return ret;
 }
Signature · 1 line
2.55.0
#2 lore raw

Re: [PATCH rdma-next 1/3] RDMA/core: Preserve the selected source GID on failed AH modification

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [High] rdma: memory leak and UAF due to missing sgid_attr refcount in rvt_modify_ah() --

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] rdma: memory leak and UAF due to missing sgid_attr refcount in rvt_modify_ah() --

commit cfb0c41a87200fc5df90876398650807044c23ad
Author: Leon Romanovsky <[email protected]>

RDMA/core: Preserve the selected source GID on failed AH modification

When an address-handle modification fails, the core replaces its source-GID reference. This commit ensures the selected source GID is preserved by only updating ah->sgid_attr when the modification is successful.

drivers/infiniband/core/verbs.c int rdma_modify_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr) · line 1016

Quoted 12 lines
drivers/infiniband/core/verbs.c +2 -1
index edf62652f9d3e..7d52d30362087 100644
@@ -1011,7 +1011,8 @@ int rdma_modify_ah(struct ib_ah *ah, struct rdma_ah_attr *ah_attr)
 		ah->device->ops.modify_ah(ah, ah_attr) :
 		-EOPNOTSUPP;
 
-	ah->sgid_attr = rdma_update_sgid_attr(ah_attr, ah->sgid_attr);
+	if (!ret)
+		ah->sgid_attr = rdma_update_sgid_attr(ah_attr, ah->sgid_attr);
 	rdma_unfill_sgid_attr(ah_attr, old_sgid_attr);

[Severity: High]
Does this sequence expose a memory leak and use-after-free in providers that do a shallow copy of ah_attr?

For example, in drivers/infiniband/sw/rdmavt/ah.c:rvt_modify_ah(), the structure is shallowly copied:

	ah->attr = *ah_attr;

This causes a memory leak of the old sgid_attr reference because rdma_put_gid_attr() is not called on the old pointer, and it adopts the new sgid_attr without incrementing its refcount.

When rdma_modify_ah() finishes, it calls rdma_unfill_sgid_attr() which drops the temporary refcount that rvt_modify_ah() copied but didn't pin. This leaves the driver's ah->attr exposed to a dangling pointer.

Later, rvt_destroy_ah() will unconditionally call:

	rdma_destroy_ah_attr(&ah->attr);

This calls rdma_put_gid_attr() on the unheld pointer, resulting in a refcount underflow and use-after-free.

This path is concretely reachable via the hfi1 provider (for instance, hfi1_modify_qp0_ah() uses rdma_modify_ah() with IB_AH_GRH enabled).

Should providers like rdmavt be updated to use rdma_replace_ah_attr() to properly manage these refcounts?

Quoted 2 lines
drivers/infiniband/core/verbs.c +0 -0
 	return ret;
 }
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1
#3 lore raw

PATCH 2/3 RDMA/core: Free unpublished port statistics after initial-read failure

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

An initial port counter read can fail after allocating its statistics buffer and wrapper. Free these both allocations.

Fixes 467f432a521a ("RDMA/core: Split port and device counter sysfs attributes")
---
 drivers/infiniband/core/sysfs.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/infiniband/core/sysfs.c +7 -2
index 69df8b703cbd..4cd861b5d953 100644
@@ -996,8 +996,8 @@ static int setup_hw_port_stats(struct ib_port *port,
 					    data->stats->num_counters);
 	if (ret != data->stats->num_counters) {
 		if (WARN_ON(ret >= 0))
-			return -EINVAL;
-		return ret;
+			ret = -EINVAL;
+		goto err_free_data;
 	}
 
 	data->stats->timestamp = jiffies;
@@ -1031,6 +1031,11 @@ static int setup_hw_port_stats(struct ib_port *port,
 
 	port->hw_stats_data = data;
 	return 0;
+
+err_free_data:
+	rdma_free_hw_stats_struct(data->stats);
+	kfree(data);
+	return ret;
 }
 
 struct rdma_hw_stats *ib_get_hw_stats_port(struct ib_device *ibdev,
Signature · 1 line
2.55.0
#4 lore raw

Re: [PATCH rdma-next 2/3] RDMA/core: Free unpublished port statistics after initial-read failure

Sashiko has reviewed this patch and found no issues. It looks great!
Quoted 7 lines

From: Leon Romanovsky <[email protected]>

An initial port counter read can fail after allocating its statistics buffer and wrapper. Free these both allocations.

Fixes 467f432a521a ("RDMA/core: Split port and device counter sysfs attributes")

Sashiko has reviewed this patch and found no issues. It looks great!

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=2
#5 lore raw

Re: [PATCH rdma-next 2/3] RDMA/core: Free unpublished port statistics after initial-read failure

In reply to Leon Romanovsky
Quoted 15 lines · Leon Romanovsky

From: Leon Romanovsky <[email protected]>

An initial port counter read can fail after allocating its statistics buffer and wrapper. Free these both allocations.

Fixes 467f432a521a ("RDMA/core: Split port and device counter sysfs

attributes")

LGTM,

Signature · 2 lines
Regards,
Kalesh AP
#6 lore raw

PATCH 3/3 RDMA/core: Reject LAG selection when source MAC lookup fails

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

GID deletion can clear the source device association after LAG lookup retains the bonding master. Ignoring the resulting L2 lookup failure copies uninitialized source MAC bytes into the synthetic hash packet.

Read the source MAC before allocating the packet and propagate lookup errors to the caller, so failed lookups cannot influence slave selection.

Fixes bd3920eac133 ("RDMA/core: Add LAG functionality")
---
 drivers/infiniband/core/lag.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
drivers/infiniband/core/lag.c +8 -4
index 88aac911e8ee..651c868aa04e 100644
@@ -19,6 +19,11 @@ static struct sk_buff *rdma_build_skb(struct net_device *netdev,
 	u8 smac[ETH_ALEN];
 	bool is_ipv4;
 	int hdr_len;
+	int ret;
+
+	ret = rdma_read_gid_l2_fields(ah_attr->grh.sgid_attr, NULL, smac);
+	if (ret)
+		return ERR_PTR(ret);
 
 	is_ipv4 = ipv6_addr_v4mapped((struct in6_addr *)ah_attr->grh.dgid.raw);
 	hdr_len = ETH_HLEN + sizeof(struct udphdr) + LL_RESERVED_SPACE(netdev);
@@ -26,7 +31,7 @@ static struct sk_buff *rdma_build_skb(struct net_device *netdev,
 
 	skb = alloc_skb(hdr_len, flags);
 	if (!skb)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	skb->dev = netdev;
 	skb_reserve(skb, hdr_len);
@@ -70,7 +75,6 @@ static struct sk_buff *rdma_build_skb(struct net_device *netdev,
 	skb_reset_mac_header(skb);
 	eth = eth_hdr(skb);
 	skb->protocol = eth->h_proto = htons(is_ipv4 ? ETH_P_IP : ETH_P_IPV6);
-	rdma_read_gid_l2_fields(ah_attr->grh.sgid_attr, NULL, smac);
 	memcpy(eth->h_source, smac, ETH_ALEN);
 	memcpy(eth->h_dest, ah_attr->roce.dmac, ETH_ALEN);
 
@@ -86,8 +90,8 @@ static struct net_device *rdma_get_xmit_slave_udp(struct ib_device *device,
 	struct sk_buff *skb;
 
 	skb = rdma_build_skb(master, ah_attr, flags);
-	if (!skb)
-		return ERR_PTR(-ENOMEM);
+	if (IS_ERR(skb))
+		return ERR_CAST(skb);
 
 	rcu_read_lock();
 	slave = netdev_get_xmit_slave(master, skb,
Signature · 1 line
2.55.0
#7 lore raw

Re: [PATCH rdma-next 3/3] RDMA/core: Reject LAG selection when source MAC lookup fails

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] rdma: KMSAN uninit-value in rdma_build_skb due to partial header init --

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] rdma: KMSAN uninit-value in rdma_build_skb due to partial header init --

commit fe2a587fde347fe9ebd760de282337658857e55d
Author: Leon Romanovsky <[email protected]>

RDMA/core: Reject LAG selection when source MAC lookup fails

GID deletion can clear the source device association after LAG lookup retains the bonding master. Ignoring the resulting L2 lookup failure copies uninitialized source MAC bytes into the synthetic hash packet. This commit reads the source MAC before allocating the packet and propagates lookup errors to the caller.

Quoted 4 lines
drivers/infiniband/core/lag.c +0 -0
index 88aac911e8ee1..651c868aa04e5 100644

[ ... ]

drivers/infiniband/core/lag.c static struct sk_buff *rdma_build_skb(struct net_device *netdev, · line 34

Quoted 6 lines
drivers/infiniband/core/lag.c +1 -1
@@ -26,7 +31,7 @@ static struct sk_buff *rdma_build_skb(struct net_device *netdev,
 
 	skb = alloc_skb(hdr_len, flags);
 	if (!skb)
-		return NULL;
+		return ERR_PTR(-ENOMEM);

[Severity: Medium]
Does this code expose uninitialized kernel memory to flow dissectors?

Since alloc_skb() leaves the data buffer uninitialized, bitfield assignments performed later in rdma_build_skb():

	if (is_ipv4) {
		skb_push(skb, sizeof(struct iphdr));
		skb_reset_network_header(skb);
		iph = ip_hdr(skb);
		iph->frag_off = 0;
		iph->version = 4;
		iph->protocol = IPPROTO_UDP;
		iph->ihl = 0x5;

compile down to read-modify-write instructions. These instructions will read uninitialized adjacent bits (such as the IPv6 priority field in the corresponding ipv6 branch) which immediately triggers KMSAN uninit-value warnings.

Furthermore, fields like iph->tos, iph->ttl, and the UDP checksum remain completely uninitialized. When this SKB is passed to netdev_get_xmit_slave() and into __skb_flow_dissect(), a BPF flow dissector attached in the network namespace could read these uninitialized fields, potentially leaking kernel memory.

Quoted 3 lines
drivers/infiniband/core/lag.c +0 -0
 	skb->dev = netdev;
 	skb_reserve(skb, hdr_len);
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=3
#8 lore raw
In reply to Leon Romanovsky
Quoted 8 lines · Leon Romanovsky

A batch of core fixes for various issues identified by LLMs.

---
Leon Romanovsky (3):
RDMA/core: Preserve the selected source GID on failed AH modification RDMA/core: Free unpublished port statistics after initial-read failure RDMA/core: Reject LAG selection when source MAC lookup fails

After some deliberation, I applied these patches anyway. Sashiko pointed out an issue unrelated to the patches.

Thanks

Quoted 14 lines
 drivers/infiniband/core/lag.c   | 12 ++++++++----
 drivers/infiniband/core/sysfs.c |  9 +++++++--
 drivers/infiniband/core/verbs.c |  3 ++-
 3 files changed, 17 insertions(+), 7 deletions(-)

---
base-commit: 66196a559be5963b1ced18418c876caf9bfb76e6
change-id: 20261015-ib-core-fixes-fa78b7068672

Best regards,

Signature · 1 line
Leon Romanovsky <[email protected]>
maintainer Applied
#9 lore raw
In reply to Leon Romanovsky
Quoted 1 line · Leon Romanovsky

A batch of core fixes for various issues identified by LLMs.

Applied, thanks!

[1/3] RDMA/core: Preserve the selected source GID on failed AH modification

      https://git.kernel.org/rdma/rdma/c/d01ee4c8419319
[2/3] RDMA/core: Free unpublished port statistics after initial-read failure
      https://git.kernel.org/rdma/rdma/c/d74bdcfa89fa45
[3/3] RDMA/core: Reject LAG selection when source MAC lookup fails
      https://git.kernel.org/rdma/rdma/c/fdfd099222d9de

Best regards,

Signature · 1 line
Leon Romanovsky <[email protected]>