IB/addr: Fix setting source address in addr6_resolve()
PATCH 1/1 IB/addr: Fix setting source address in addr6_resolve()
From: Roland Dreier <[email protected]>
Commit eea40b8f624f ("infiniband: call ipv6 route lookup via the stub interface") introduced a regression in address resolution when connecting to IPv6 destination addresses. The old code called ip6_route_output(), while the new code calls ipv6_stub->ipv6_dst_lookup(). The two are almost the same, except that ipv6_dst_lookup() also calls ip6_route_get_saddr() if the source address is in6addr_any.
This means that the test of ipv6_addr_any(&fl6.saddr) now never succeeds, and so we never copy the source address out. This ends up causing rdma_resolve_addr() to fail, because without a resolved source address, cma_acquire_dev() will fail to find an RDMA device to use. For me, this causes connecting to an NVMe over Fabrics target via RoCE / IPv6 to fail.
Fix this by copying out fl6.saddr if ipv6_addr_any() is true for the original source address passed into addr6_resolve(). We can drop our call to ipv6_dev_get_saddr() because ipv6_dst_lookup() already does that work.
--- drivers/infiniband/core/addr.c | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-)
drivers/infiniband/core/addr.c +1 -9
@@ -449,12 +449,7 @@ static int addr6_resolve(struct sockaddr_in6 *src_in,return ret; rt = (struct rt6_info *)dst; -if (ipv6_addr_any (&fl6.saddr)) { - ret =ipv6_dev_get_saddr (addr->net,ip6_dst_idev (dst)->dev, - &fl6.daddr,0 , &fl6.saddr); -if (ret) - goto put; - +if (ipv6_addr_any (&src_in->sin6_addr)) { src_in->sin6_family =AF_INET6 ; src_in->sin6_addr = fl6.saddr; }
@@ -471,9 +466,6 @@ static int addr6_resolve(struct sockaddr_in6 *src_in, *pdst = dst;return 0 ; -put: -dst_release (dst); -return ret; }#else static int addr6_resolve(struct sockaddr_in6 *src_in,
Signature · 6 lines
2.11.0 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
A message that is not in the archive (sent off-list or not yet received): 20170606162200.24629-1-roland-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org
Paolo Abeni
Acked-byQuoted 45 lines · Roland Dreier
From: Roland Dreier <[email protected]>
Commit
eea40b8f624f("infiniband: call ipv6 route lookup via the stub interface") introduced a regression in address resolution when connecting to IPv6 destination addresses. The old code calledip6_route_output(), while the new code callsipv6_stub->ipv6_dst_lookup(). The two are almost the same, except thatipv6_dst_lookup()also callsip6_route_get_saddr()if the source address isin6addr_any.This means that the test of
ipv6_addr_any(&fl6.saddr)now never succeeds, and so we never copy the source address out. This ends up causingrdma_resolve_addr()to fail, because without a resolved source address,cma_acquire_dev()will fail to find an RDMA device to use. For me, this causes connecting to an NVMe over Fabrics target via RoCE / IPv6 to fail.Fix this by copying out fl6.saddr if
ipv6_addr_any()is true for the original source address passed intoaddr6_resolve(). We can drop our call toipv6_dev_get_saddr()becauseipv6_dst_lookup()already does that work.Fixes eea40b8f624 ("infiniband: call ipv6 route lookup via the stub interface")Cc <[email protected]> # 3.12+Signed-off-by Roland Dreier <[email protected]>--- drivers/infiniband/core/addr.c | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-)drivers/infiniband/core/addr.c +1
-6@@ -449,12 +449,7 @@ static int addr6_resolve(struct sockaddr_in6 *src_in,return ret; rt = (struct rt6_info *)dst; -if (ipv6_addr_any (&fl6.saddr)) { - ret =ipv6_dev_get_saddr (addr->net,ip6_dst_idev (dst)->dev, - &fl6.daddr,0 , &fl6.saddr); -if (ret) - goto put; - +if (ipv6_addr_any (&src_in->sin6_addr)) { src_in->sin6_family =AF_INET6 ; src_in->sin6_addr = fl6.saddr; }
drivers/infiniband/core/addr.c static int addr6_resolve(struct sockaddr_in6 *src_in, · line 471
Quoted 9 lines
drivers/infiniband/core/addr.c +0
-3@@ -471,9 +466,6 @@ static int addr6_resolve(struct sockaddr_in6 *src_in, *pdst = dst;return 0 ; -put: -dst_release (dst); -return ret; }#else static int addr6_resolve(struct sockaddr_in6 *src_in,
LGTM, thank you for fixing this!
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 29 lines · Roland Dreier
From: Roland Dreier <[email protected]>
Commit
eea40b8f624f("infiniband: call ipv6 route lookup via the stub interface") introduced a regression in address resolution when connecting
to IPv6 destination addresses. The old code calledip6_route_output(),
while the new code calls ipv6_stub->ipv6_dst_lookup(). The two are almost
the same, except thatipv6_dst_lookup()also callsip6_route_get_saddr()
if the source address isin6addr_any.This means that the test of
ipv6_addr_any(&fl6.saddr)now never succeeds,
and so we never copy the source address out. This ends up causingrdma_resolve_addr()to fail, because without a resolved source address,cma_acquire_dev()will fail to find an RDMA device to use. For me, this
causes connecting to an NVMe over Fabrics target via RoCE / IPv6 to fail.Fix this by copying out fl6.saddr if
ipv6_addr_any()is true for the original
source address passed into addr6_resolve(). We can drop our call toipv6_dev_get_saddr()becauseipv6_dst_lookup()already does that work.
Quoted 4 lines
Fixes eea40b8f624 ("infiniband: call ipv6 route lookup via the stubinterface")
Cc <[email protected]> # 3.12+Signed-off-by Roland Dreier <[email protected]>
Thanks Roland, applied.
Signature · 9 lines
Doug Ledford <[email protected]> GPG KeyID: B826A3330E572FDD Key fingerprint = AE6B 1BDA 122B 23B4 265B 1274 B826 A333 0E57 2FDD -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html