RDMA v7.1 first rc window

- syzbot triggred crash in rxe due to concurrent plug/unplug
 
 - Possible non-zero'd memory exposed to userspace in bnxt_re
 
 - Malicous 'magic packet' with SIW causes a buffer overflow
 
 - Tighten the new uAPI validation code to not crash in debugging prints
   and have the right module dependencies in drivers
 
 - mana was missing the max_msg_sz report to userspace
 
 - UAF in rtrs on an error path
 -----BEGIN PGP SIGNATURE-----
 
 iHUEABYKAB0WIQRRRCHOFoQz/8F5bUaFwuHvBreFYQUCahDmpgAKCRCFwuHvBreF
 YRRTAP0YXV95Y5gcli55IhetKjJUzQbaREz2NueqIpf1IorMbAD+Lns4DgZCU0KW
 bC81x7cGHBSyCju9zogIdBFJhsbxeQ4=
 =MxJz
 -----END PGP SIGNATURE-----

Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma

Pull rdma fixes from Jason Gunthorpe:

 - syzbot triggred crash in rxe due to concurrent plug/unplug

 - Possible non-zero'd memory exposed to userspace in bnxt_re

 - Malicous 'magic packet' with SIW causes a buffer overflow

 - Tighten the new uAPI validation code to not crash in debugging prints
   and have the right module dependencies in drivers

 - mana was missing the max_msg_sz report to userspace

 - UAF in rtrs on an error path

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
  RDMA/rtrs: Fix use-after-free in path file creation cleanup
  RDMA/mana_ib: Report max_msg_sz in mana_ib_query_port
  RDMA/core: Do not read wild stack memory in uverbs_get_handler_fn()
  RDMA/core: Move the _ib_copy_validate_udata* functions to ib_core_uverbs
  RDMA/siw: Reject MPA FPDU length underflow before signed receive math
  RDMA/bnxt_re: zero shared page before exposing to userspace
  selftests/rdma: explicitly skip tests when required modules are missing
  RDMA/nldev: Add mutual exclusion in nldev_dellink()

drivers/infiniband/core/ib_core_uverbs.c

@@ -9,6 +9,7 @@
 #include <linux/dma-resv.h>
 #include "uverbs.h"
 #include "core_priv.h"
+#include "rdma_core.h"
 
 MODULE_IMPORT_NS("DMA_BUF");
 
@@ -416,3 +417,89 @@ struct ib_device *rdma_udata_to_dev(struct ib_udata *udata)
 }
 EXPORT_SYMBOL(rdma_udata_to_dev);
 
+#if IS_ENABLED(CONFIG_INFINIBAND_USER_ACCESS)
+uverbs_api_ioctl_handler_fn uverbs_get_handler_fn(struct ib_udata *udata)
+{
+	struct uverbs_attr_bundle *bundle =
+		rdma_udata_to_uverbs_attr_bundle(udata);
+
+	lockdep_assert_held(&bundle->ufile->device->disassociate_srcu);
+
+	return srcu_dereference(bundle->method_elm->handler,
+				&bundle->ufile->device->disassociate_srcu);
+}
+
+int _ib_copy_validate_udata_in(struct ib_udata *udata, void *req,
+			       size_t kernel_size, size_t minimum_size)
+{
+	int err;
+
+	if (udata->inlen < minimum_size) {
+		ibdev_dbg(
+			rdma_udata_to_dev(udata),
+			"System call driver input udata too small (%zu < %zu) for ioctl %ps called by %pSR\n",
+			udata->inlen, minimum_size,
+			uverbs_get_handler_fn(udata),
+			__builtin_return_address(0));
+		return -EINVAL;
+	}
+
+	err = copy_struct_from_user(req, kernel_size, udata->inbuf,
+				    udata->inlen);
+	if (err) {
+		if (err == -E2BIG) {
+			ibdev_dbg(
+				rdma_udata_to_dev(udata),
+				"System call driver input udata not zero from %zu -> %zu for ioctl %ps called by %pSR\n",
+				minimum_size, udata->inlen,
+				uverbs_get_handler_fn(udata),
+				__builtin_return_address(0));
+			return -EOPNOTSUPP;
+		}
+		ibdev_dbg(
+			rdma_udata_to_dev(udata),
+			"System call driver input udata EFAULT for ioctl %ps called by %pSR\n",
+			uverbs_get_handler_fn(udata),
+			__builtin_return_address(0));
+		return err;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(_ib_copy_validate_udata_in);
+
+int _ib_copy_validate_udata_cm_fail(struct ib_udata *udata, u64 req_cm,
+				    u64 valid_cm)
+{
+	ibdev_dbg(
+		rdma_udata_to_dev(udata),
+		"System call driver input udata has unsupported comp_mask %llx & ~%llx = %llx for ioctl %ps called by %pSR\n",
+		req_cm, valid_cm, req_cm & ~valid_cm,
+		uverbs_get_handler_fn(udata), __builtin_return_address(0));
+	return -EOPNOTSUPP;
+}
+EXPORT_SYMBOL(_ib_copy_validate_udata_cm_fail);
+
+int _ib_respond_udata(struct ib_udata *udata, const void *src, size_t len)
+{
+	size_t copy_len;
+
+	/* 0 length copy_len is a NOP for copy_to_user() and doesn't fail. */
+	copy_len = min(len, udata->outlen);
+	if (copy_to_user(udata->outbuf, src, copy_len))
+		goto err_fault;
+	if (copy_len < udata->outlen) {
+		if (clear_user(udata->outbuf + copy_len,
+			       udata->outlen - copy_len))
+			goto err_fault;
+	}
+	return 0;
+err_fault:
+	ibdev_dbg(
+		rdma_udata_to_dev(udata),
+		"System call driver out udata has EFAULT (%zu into %zu) for ioctl %ps called by %pSR\n",
+		len, udata->outlen, uverbs_get_handler_fn(udata),
+		__builtin_return_address(0));
+	return -EFAULT;
+}
+EXPORT_SYMBOL(_ib_respond_udata);
+#endif

drivers/infiniband/core/nldev.c

@@ -51,6 +51,7 @@
  * a controlled QKEY.
  */
 static bool privileged_qkey;
+static DEFINE_MUTEX(nldev_dellink_mutex);
 
 typedef int (*res_fill_func_t)(struct sk_buff*, bool,
 			       struct rdma_restrack_entry*, uint32_t);
@@ -1846,7 +1847,9 @@ static int nldev_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	 * implicitly scoped to the driver supporting dynamic link deletion like RXE.
 	 */
 	if (device->link_ops && device->link_ops->dellink) {
+		mutex_lock(&nldev_dellink_mutex);
 		err = device->link_ops->dellink(device);
+		mutex_unlock(&nldev_dellink_mutex);
 		if (err)
 			return err;
 	}

drivers/infiniband/core/uverbs.h

@@ -229,6 +229,40 @@ int uverbs_dealloc_mw(struct ib_mw *mw);
 void ib_uverbs_detach_umcast(struct ib_qp *qp,
 			     struct ib_uqp_object *uobj);
 
+struct bundle_alloc_head {
+	struct_group_tagged(bundle_alloc_head_hdr, hdr,
+		struct bundle_alloc_head *next;
+	);
+	u8 data[];
+};
+
+struct bundle_priv {
+	/* Must be first */
+	struct bundle_alloc_head_hdr alloc_head;
+	struct bundle_alloc_head *allocated_mem;
+	size_t internal_avail;
+	size_t internal_used;
+
+	struct radix_tree_root *radix;
+	void __rcu **radix_slots;
+	unsigned long radix_slots_len;
+	u32 method_key;
+
+	struct ib_uverbs_attr __user *user_attrs;
+	struct ib_uverbs_attr *uattrs;
+
+	DECLARE_BITMAP(uobj_finalize, UVERBS_API_ATTR_BKEY_LEN);
+	DECLARE_BITMAP(spec_finalize, UVERBS_API_ATTR_BKEY_LEN);
+	DECLARE_BITMAP(uobj_hw_obj_valid, UVERBS_API_ATTR_BKEY_LEN);
+
+	/*
+	 * Must be last. bundle ends in a flex array which overlaps
+	 * internal_buffer.
+	 */
+	struct uverbs_attr_bundle_hdr bundle;
+	u64 internal_buffer[32];
+};
+
 long ib_uverbs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg);
 
 struct ib_uverbs_flow_spec {

drivers/infiniband/core/uverbs_ioctl.c

@@ -35,54 +35,6 @@
 #include "rdma_core.h"
 #include "uverbs.h"
 
-struct bundle_alloc_head {
-	struct_group_tagged(bundle_alloc_head_hdr, hdr,
-		struct bundle_alloc_head *next;
-	);
-	u8 data[];
-};
-
-struct bundle_priv {
-	/* Must be first */
-	struct bundle_alloc_head_hdr alloc_head;
-	struct bundle_alloc_head *allocated_mem;
-	size_t internal_avail;
-	size_t internal_used;
-
-	struct radix_tree_root *radix;
-	const struct uverbs_api_ioctl_method *method_elm;
-	void __rcu **radix_slots;
-	unsigned long radix_slots_len;
-	u32 method_key;
-
-	struct ib_uverbs_attr __user *user_attrs;
-	struct ib_uverbs_attr *uattrs;
-
-	DECLARE_BITMAP(uobj_finalize, UVERBS_API_ATTR_BKEY_LEN);
-	DECLARE_BITMAP(spec_finalize, UVERBS_API_ATTR_BKEY_LEN);
-	DECLARE_BITMAP(uobj_hw_obj_valid, UVERBS_API_ATTR_BKEY_LEN);
-
-	/*
-	 * Must be last. bundle ends in a flex array which overlaps
-	 * internal_buffer.
-	 */
-	struct uverbs_attr_bundle_hdr bundle;
-	u64 internal_buffer[32];
-};
-
-uverbs_api_ioctl_handler_fn uverbs_get_handler_fn(struct ib_udata *udata)
-{
-	struct uverbs_attr_bundle *bundle =
-		rdma_udata_to_uverbs_attr_bundle(udata);
-	struct bundle_priv *pbundle =
-		container_of(&bundle->hdr, struct bundle_priv, bundle);
-
-	lockdep_assert_held(&bundle->ufile->device->disassociate_srcu);
-
-	return srcu_dereference(pbundle->method_elm->handler,
-				&bundle->ufile->device->disassociate_srcu);
-}
-
 /*
  * Each method has an absolute minimum amount of memory it needs to allocate,
  * precompute that amount and determine if the onstack memory can be used or
@@ -445,13 +397,13 @@ static int ib_uverbs_run_method(struct bundle_priv *pbundle,
 	struct uverbs_attr_bundle *bundle =
 		container_of(&pbundle->bundle, struct uverbs_attr_bundle, hdr);
 	size_t uattrs_size = array_size(sizeof(*pbundle->uattrs), num_attrs);
-	unsigned int destroy_bkey = pbundle->method_elm->destroy_bkey;
+	unsigned int destroy_bkey = bundle->method_elm->destroy_bkey;
 	unsigned int i;
 	int ret;
 
 	/* See uverbs_disassociate_api() */
 	handler = srcu_dereference(
-		pbundle->method_elm->handler,
+		bundle->method_elm->handler,
 		&pbundle->bundle.ufile->device->disassociate_srcu);
 	if (!handler)
 		return -EIO;
@@ -469,12 +421,12 @@ static int ib_uverbs_run_method(struct bundle_priv *pbundle,
 	}
 
 	/* User space did not provide all the mandatory attributes */
-	if (unlikely(!bitmap_subset(pbundle->method_elm->attr_mandatory,
+	if (unlikely(!bitmap_subset(bundle->method_elm->attr_mandatory,
 				    pbundle->bundle.attr_present,
-				    pbundle->method_elm->key_bitmap_len)))
+				    bundle->method_elm->key_bitmap_len)))
 		return -EINVAL;
 
-	if (pbundle->method_elm->has_udata)
+	if (bundle->method_elm->has_udata)
 		uverbs_fill_udata(bundle, &pbundle->bundle.driver_udata,
 				  UVERBS_ATTR_UHW_IN, UVERBS_ATTR_UHW_OUT);
 	else
@@ -499,7 +451,7 @@ static int ib_uverbs_run_method(struct bundle_priv *pbundle,
 	 * assume that the driver wrote to its UHW_OUT and flag userspace
 	 * appropriately.
 	 */
-	if (!ret && pbundle->method_elm->has_udata) {
+	if (!ret && bundle->method_elm->has_udata) {
 		const struct uverbs_attr *attr =
 			uverbs_attr_get(bundle, UVERBS_ATTR_UHW_OUT);
 
@@ -520,7 +472,7 @@ static int ib_uverbs_run_method(struct bundle_priv *pbundle,
 
 static void bundle_destroy(struct bundle_priv *pbundle, bool commit)
 {
-	unsigned int key_bitmap_len = pbundle->method_elm->key_bitmap_len;
+	unsigned int key_bitmap_len = pbundle->bundle.method_elm->key_bitmap_len;
 	struct uverbs_attr_bundle *bundle =
 		container_of(&pbundle->bundle, struct uverbs_attr_bundle, hdr);
 	struct bundle_alloc_head *memblock;
@@ -608,7 +560,7 @@ static int ib_uverbs_cmd_verbs(struct ib_uverbs_file *ufile,
 	}
 
 	/* Space for the pbundle->bundle.attrs flex array */
-	pbundle->method_elm = method_elm;
+	pbundle->bundle.method_elm = method_elm;
 	pbundle->method_key = attrs_iter.index;
 	pbundle->bundle.ufile = ufile;
 	pbundle->bundle.context = NULL; /* only valid if bundle has uobject */
@@ -617,10 +569,12 @@ static int ib_uverbs_cmd_verbs(struct ib_uverbs_file *ufile,
 	pbundle->radix_slots_len = radix_tree_chunk_size(&attrs_iter);
 	pbundle->user_attrs = user_attrs;
 
-	pbundle->internal_used = ALIGN(pbundle->method_elm->key_bitmap_len *
-					       sizeof(*container_of(&pbundle->bundle,
-							struct uverbs_attr_bundle, hdr)->attrs),
-					       sizeof(*pbundle->internal_buffer));
+	pbundle->internal_used = ALIGN(
+		pbundle->bundle.method_elm->key_bitmap_len *
+			sizeof(*container_of(&pbundle->bundle,
+					     struct uverbs_attr_bundle, hdr)
+					->attrs),
+		sizeof(*pbundle->internal_buffer));
 	memset(pbundle->bundle.attr_present, 0,
 	       sizeof(pbundle->bundle.attr_present));
 	memset(pbundle->uobj_finalize, 0, sizeof(pbundle->uobj_finalize));
@@ -860,77 +814,3 @@ void uverbs_finalize_uobj_create(const struct uverbs_attr_bundle *bundle,
 		  pbundle->uobj_hw_obj_valid);
 }
 EXPORT_SYMBOL(uverbs_finalize_uobj_create);
-
-int _ib_copy_validate_udata_in(struct ib_udata *udata, void *req,
-			       size_t kernel_size, size_t minimum_size)
-{
-	int err;
-
-	if (udata->inlen < minimum_size) {
-		ibdev_dbg(
-			rdma_udata_to_dev(udata),
-			"System call driver input udata too small (%zu < %zu) for ioctl %ps called by %pSR\n",
-			udata->inlen, minimum_size,
-			uverbs_get_handler_fn(udata),
-			__builtin_return_address(0));
-		return -EINVAL;
-	}
-
-	err = copy_struct_from_user(req, kernel_size, udata->inbuf,
-				    udata->inlen);
-	if (err) {
-		if (err == -E2BIG) {
-			ibdev_dbg(
-				rdma_udata_to_dev(udata),
-				"System call driver input udata not zero from %zu -> %zu for ioctl %ps called by %pSR\n",
-				minimum_size, udata->inlen,
-				uverbs_get_handler_fn(udata),
-				__builtin_return_address(0));
-			return -EOPNOTSUPP;
-		}
-		ibdev_dbg(
-			rdma_udata_to_dev(udata),
-			"System call driver input udata EFAULT for ioctl %ps called by %pSR\n",
-			uverbs_get_handler_fn(udata),
-			__builtin_return_address(0));
-		return err;
-	}
-	return 0;
-}
-EXPORT_SYMBOL(_ib_copy_validate_udata_in);
-
-int _ib_copy_validate_udata_cm_fail(struct ib_udata *udata, u64 req_cm,
-				    u64 valid_cm)
-{
-	ibdev_dbg(
-		rdma_udata_to_dev(udata),
-		"System call driver input udata has unsupported comp_mask %llx & ~%llx = %llx for ioctl %ps called by %pSR\n",
-		req_cm, valid_cm, req_cm & ~valid_cm,
-		uverbs_get_handler_fn(udata), __builtin_return_address(0));
-	return -EOPNOTSUPP;
-}
-EXPORT_SYMBOL(_ib_copy_validate_udata_cm_fail);
-
-int _ib_respond_udata(struct ib_udata *udata, const void *src, size_t len)
-{
-	size_t copy_len;
-
-	/* 0 length copy_len is a NOP for copy_to_user() and doesn't fail. */
-	copy_len = min(len, udata->outlen);
-	if (copy_to_user(udata->outbuf, src, copy_len))
-		goto err_fault;
-	if (copy_len < udata->outlen) {
-		if (clear_user(udata->outbuf + copy_len,
-			       udata->outlen - copy_len))
-			goto err_fault;
-	}
-	return 0;
-err_fault:
-	ibdev_dbg(
-		rdma_udata_to_dev(udata),
-		"System call driver out udata has EFAULT (%zu into %zu) for ioctl %ps called by %pSR\n",
-		len, udata->outlen, uverbs_get_handler_fn(udata),
-		__builtin_return_address(0));
-	return -EFAULT;
-}
-EXPORT_SYMBOL(_ib_respond_udata);

drivers/infiniband/hw/bnxt_re/ib_verbs.c

@@ -4638,7 +4638,7 @@ int bnxt_re_alloc_ucontext(struct ib_ucontext *ctx, struct ib_udata *udata)
 
 	uctx->rdev = rdev;
 
-	uctx->shpg = (void *)__get_free_page(GFP_KERNEL);
+	uctx->shpg = (void *)get_zeroed_page(GFP_KERNEL);
 	if (!uctx->shpg) {
 		rc = -ENOMEM;
 		goto fail;

drivers/infiniband/hw/mana/main.c

@@ -606,6 +606,7 @@ int mana_ib_query_port(struct ib_device *ibdev, u32 port,
 	if (mana_ib_is_rnic(dev)) {
 		props->gid_tbl_len = 16;
 		props->ip_gids = true;
+		props->max_msg_sz = SZ_16M;
 		if (port == 1)
 			props->port_cap_flags = IB_PORT_CM_SUP;
 	}

drivers/infiniband/sw/siw/siw_qp_rx.c

@@ -1081,6 +1081,21 @@ static int siw_get_hdr(struct siw_rx_stream *srx)
 			return -EAGAIN;
 	}
 
+	/*
+	 * Peer-controlled mpa_len must not underflow srx->fpdu_part_rem
+	 * in siw_tcp_rx_data(); a negative value flows as a signed copy
+	 * length into siw_check_mem() and skb_copy_bits().
+	 */
+	if (unlikely(be16_to_cpu(c_hdr->mpa_len) + MPA_HDR_SIZE <
+		     iwarp_pktinfo[opcode].hdr_len)) {
+		pr_warn_ratelimited("siw: short mpa_len %u for opcode %u (hdr_len %u)\n",
+				    be16_to_cpu(c_hdr->mpa_len), opcode,
+				    iwarp_pktinfo[opcode].hdr_len);
+		siw_init_terminate(rx_qp(srx), TERM_ERROR_LAYER_LLP,
+				   LLP_ETYPE_MPA, LLP_ECODE_FPDU_START, 0);
+		return -EINVAL;
+	}
+
 	/*
 	 * DDP/RDMAP header receive completed. Check if the current
 	 * DDP segment starts a new RDMAP message or continues a previously

drivers/infiniband/ulp/rtrs/rtrs-srv-sysfs.c

@@ -295,8 +295,8 @@ int rtrs_srv_create_path_files(struct rtrs_srv_path *srv_path)
 put_kobj:
 	kobject_del(&srv_path->kobj);
 destroy_root:
-	kobject_put(&srv_path->kobj);
 	rtrs_srv_destroy_once_sysfs_root_folders(srv_path);
+	kobject_put(&srv_path->kobj);
 
 	return err;
 }

include/rdma/uverbs_ioctl.h

@@ -635,6 +635,7 @@ struct uverbs_attr_bundle {
 		struct ib_uverbs_file *ufile;
 		struct ib_ucontext *context;
 		struct ib_uobject *uobject;
+		const struct uverbs_api_ioctl_method *method_elm;
 		DECLARE_BITMAP(attr_present, UVERBS_API_ATTR_BKEY_LEN);
 	);
 	struct uverbs_attr attrs[];

tools/testing/selftests/rdma/rxe_ipv6.sh

@@ -8,6 +8,8 @@ RXE_NAME="rxe6"
 PORT=4791
 IP6_ADDR="2001:db8::1/64"
 
+source "$(dirname "$0")/../kselftest/ktap_helpers.sh"
+
 exec > /dev/null
 
 # Cleanup function to run on exit (even on failure)
@@ -21,8 +23,8 @@ trap cleanup EXIT
 # 1. Prerequisites check
 for mod in tun veth rdma_rxe; do
     if ! modinfo "$mod" >/dev/null 2>&1; then
-        echo "Error: Kernel module '$mod' not found."
-        exit 1
+        echo "SKIP: Kernel module '$mod' not found." >&2
+        exit $KSFT_SKIP
     fi
 done
 

tools/testing/selftests/rdma/rxe_rping_between_netns.sh

@@ -8,6 +8,8 @@ IP_A="1.1.1.1"
 IP_B="1.1.1.2"
 PORT=4791
 
+source "$(dirname "$0")/../kselftest/ktap_helpers.sh"
+
 exec > /dev/null
 
 # --- Cleanup Routine ---
@@ -27,6 +29,11 @@ if [[ $EUID -ne 0 ]]; then
    exit 1
 fi
 
+if ! modinfo rdma_rxe >/dev/null 2>&1; then
+    echo "SKIP: Kernel module 'rdma_rxe' not found." >&2
+    exit $KSFT_SKIP
+fi
+
 modprobe rdma_rxe || { echo "Failed to load rdma_rxe"; exit 1; }
 
 # --- Setup Network Topology ---

tools/testing/selftests/rdma/rxe_socket_with_netns.sh

@@ -4,6 +4,8 @@
 PORT=4791
 MODS=("tun" "rdma_rxe")
 
+source "$(dirname "$0")/../kselftest/ktap_helpers.sh"
+
 exec > /dev/null
 
 # --- Helper: Cleanup Routine ---
@@ -26,6 +28,10 @@ if [[ $EUID -ne 0 ]]; then
 fi
 
 for m in "${MODS[@]}"; do
+    if ! modinfo "$m" >/dev/null 2>&1; then
+        echo "SKIP: Kernel module '$m' not found." >&2
+        exit $KSFT_SKIP
+    fi
     modprobe "$m" || { echo "Error: Failed to load $m"; exit 1; }
 done
 

tools/testing/selftests/rdma/rxe_test_NETDEV_UNREGISTER.sh

@@ -5,6 +5,8 @@ DEV_NAME="tun0"
 RXE_NAME="rxe0"
 RDMA_PORT=4791
 
+source "$(dirname "$0")/../kselftest/ktap_helpers.sh"
+
 exec > /dev/null
 
 # --- Cleanup Routine ---
@@ -19,8 +21,8 @@ trap cleanup EXIT
 
 # 1. Dependency Check
 if ! modinfo rdma_rxe >/dev/null 2>&1; then
-    echo "Error: rdma_rxe module not found."
-    exit 1
+    echo "SKIP: rdma_rxe module not found." >&2
+    exit $KSFT_SKIP
 fi
 
 modprobe rdma_rxe