wireguard-linux, branch jd/new-archs

wireguard: selftests: support OpenRISC

2022-07-30T00:48:15Z

Stafford and I have been using this to shake out OpenRISC bugs, and it's been a great help, so it's time OpenRISC support for the WireGuard test suite is made into a proper commit. The QEMU changes necessary for this to work should also be around the corner now, and they seem some what stationary in their interface too. Cc: Stafford Horne Signed-off-by: Jason A. Donenfeld

wireguard: selftests: support UML

2022-07-30T00:48:15Z

This shoud open up various possibilities like time travel execution, and is also just another platform to help shake out bugs. Cc: Johannes Berg Signed-off-by: Jason A. Donenfeld

wireguard: allowedips: don't corrupt stack when detecting overflow

2022-07-30T00:47:02Z

In case push_rcu() and related functions are buggy, there's a WARN_ON(len >= 128), which the selftest tries to hit by being tricky. In case it is hit, we shouldn't corrupt the kernel's stack, though; otherwise it may be hard to even receive the report that it's buggy. So conditionalize the stack write based on that WARN_ON()'s return value. Note that this never *actually* happens anyway. The WARN_ON() in the first place is bounded by IS_ENABLED(DEBUG), and isn't expected to ever actually hit. This is just a debugging sanity check. Additionally, hoist the constant 128 into a named enum, MAX_ALLOWEDIPS_BITS, so that it's clear why this value is chosen. Suggested-by: Linus Torvalds Link: https://lore.kernel.org/all/CAHk-=wjJZGA6w_DxA+k7Ejbqsq+uGK==koPai3sqdsfJqemvag@mail.gmail.com/ Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld

wireguard: selftests: update config fragments

2022-07-29T23:45:14Z

The kernel.config and debug.config fragments in wireguard selftests mention some config symbols that have been reworked: Commit c5665868183f ("mm: kmemleak: use the memory pool for early allocations") removes the config DEBUG_KMEMLEAK_EARLY_LOG_SIZE and since then, the config's feature is available without further configuration. Commit 4675ff05de2d ("kmemcheck: rip it out") removes kmemcheck and the corresponding arch config HAVE_ARCH_KMEMCHECK. There is no need for this config. Commit 3bf195ae6037 ("netfilter: nat: merge nf_nat_ipv4,6 into nat core") removes the config NF_NAT_IPV4 and since then, the config's feature is available without further configuration. Commit 41a2901e7d22 ("rcu: Remove SPARSE_RCU_POINTER Kconfig option") removes the config SPARSE_RCU_POINTER and since then, the config's feature is enabled by default. Commit dfb4357da6dd ("time: Remove CONFIG_TIMER_STATS") removes the feature and config CONFIG_TIMER_STATS without any replacement. Commit 3ca17b1f3628 ("lib/ubsan: remove null-pointer checks") removes the check and config UBSAN_NULL without any replacement. Adjust the config fragments to those changes in configs. Signed-off-by: Lukas Bulwahn Signed-off-by: Jason A. Donenfeld

wireguard: ratelimiter: use hrtimer in selftest

2022-07-29T23:45:13Z

Using msleep() is problematic because it's compared against ratelimiter.c's ktime_get_coarse_boottime_ns(), which means on systems with slow jiffies (such as UML's forced HZ=100), the result is inaccurate. So switch to using schedule_hrtimeout(). However, hrtimer gives us access only to the traditional posix timers, and none of the _COARSE variants. So now, rather than being too imprecise like jiffies, it's too precise. One solution would be to give it a large "range" value, but this will still fire early on a loaded system. A better solution is to align the timeout to the actual coarse timer, and then round up to the nearest tick, plus change. So add the timeout to the current coarse time, and then schedule_hrtimer() until the absolute computed time. This should hopefully reduce flakes in CI as well. Note that we keep the retry loop in case the entire function is running behind, because the test could still be scheduled out, by either the kernel or by the hypervisor's kernel, in which case restarting the test and hoping to not be scheduled out still helps. Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Suggested-by: Thomas Gleixner Signed-off-by: Jason A. Donenfeld

Merge branch 'netdevsim-fib-route-delete-leak'

2022-07-29T11:21:02Z

Ido Schimmel says: ==================== netdevsim: fib: Fix reference count leak on route deletion failure Fix a recently reported netdevsim bug found using syzkaller. Patch #1 fixes the bug. Patch #2 adds a debugfs knob to allow us to test the fix. Patch #3 adds test cases. ==================== Signed-off-by: David S. Miller

selftests: netdevsim: Add test cases for route deletion failure

2022-07-29T11:21:02Z

Add IPv4 and IPv6 test cases that ensure that we are not leaking a reference on the nexthop device when we are unable to delete its associated route. Without the fix in a previous patch ("netdevsim: fib: Fix reference count leak on route deletion failure") both test cases get stuck, waiting for the reference to be released from the dummy device [1][2]. [1] unregister_netdevice: waiting for dummy1 to become free. Usage count = 5 leaked reference. fib_check_nh+0x275/0x620 fib_create_info+0x237c/0x4d30 fib_table_insert+0x1dd/0x1d20 inet_rtm_newroute+0x11b/0x200 rtnetlink_rcv_msg+0x43b/0xd20 netlink_rcv_skb+0x15e/0x430 netlink_unicast+0x53b/0x800 netlink_sendmsg+0x945/0xe40 ____sys_sendmsg+0x747/0x960 ___sys_sendmsg+0x11d/0x190 __sys_sendmsg+0x118/0x1e0 do_syscall_64+0x34/0x80 entry_SYSCALL_64_after_hwframe+0x63/0xcd [2] unregister_netdevice: waiting for dummy1 to become free. Usage count = 5 leaked reference. fib6_nh_init+0xc46/0x1ca0 ip6_route_info_create+0x1167/0x19a0 ip6_route_add+0x27/0x150 inet6_rtm_newroute+0x161/0x170 rtnetlink_rcv_msg+0x43b/0xd20 netlink_rcv_skb+0x15e/0x430 netlink_unicast+0x53b/0x800 netlink_sendmsg+0x945/0xe40 ____sys_sendmsg+0x747/0x960 ___sys_sendmsg+0x11d/0x190 __sys_sendmsg+0x118/0x1e0 do_syscall_64+0x34/0x80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Signed-off-by: Ido Schimmel Reviewed-by: Amit Cohen Reviewed-by: David Ahern Signed-off-by: David S. Miller

netdevsim: fib: Add debugfs knob to simulate route deletion failure

2022-07-29T11:21:02Z

The previous patch ("netdevsim: fib: Fix reference count leak on route deletion failure") fixed a reference count leak that happens on route deletion failure. Such failures can only be simulated by injecting slab allocation failures, which cannot be surgically injected. In order to be able to specifically test this scenario, add a debugfs knob that allows user space to fail route deletion requests when enabled. Signed-off-by: Ido Schimmel Reviewed-by: Amit Cohen Reviewed-by: David Ahern Signed-off-by: David S. Miller

netdevsim: fib: Fix reference count leak on route deletion failure

2022-07-29T11:21:02Z

As part of FIB offload simulation, netdevsim stores IPv4 and IPv6 routes and holds a reference on FIB info structures that in turn hold a reference on the associated nexthop device(s). In the unlikely case where we are unable to allocate memory to process a route deletion request, netdevsim will not release the reference from the associated FIB info structure, thereby preventing the associated nexthop device(s) from ever being removed [1]. Fix this by scheduling a work item that will flush netdevsim's FIB table upon route deletion failure. This will cause netdevsim to release its reference from all the FIB info structures in its table. Reported by Lucas Leong of Trend Micro Zero Day Initiative. Fixes: 0ae3eb7b4611 ("netdevsim: fib: Perform the route programming in a non-atomic context") Signed-off-by: Ido Schimmel Reviewed-by: Amit Cohen Reviewed-by: David Ahern Signed-off-by: David S. Miller

net: allow unbound socket for packets in VRF when tcp_l3mdev_accept set

2022-07-29T10:58:54Z

The commit 3c82a21f4320 ("net: allow binding socket in a VRF when there's an unbound socket") changed the inet socket lookup to avoid packets in a VRF from matching an unbound socket. This is to ensure the necessary isolation between the default and other VRFs for routing and forwarding. VRF-unaware processes running in the default VRF cannot access another VRF and have to be run with 'ip vrf exec '. This is to be expected with tcp_l3mdev_accept disabled, but could be reallowed when this sysctl option is enabled. So instead of directly checking dif and sdif in inet[6]_match, here call inet_sk_bound_dev_eq(). This allows a match on unbound socket for non-zero sdif i.e. for packets in a VRF, if tcp_l3mdev_accept is enabled. Fixes: 3c82a21f4320 ("net: allow binding socket in a VRF when there's an unbound socket") Signed-off-by: Mike Manning Link: https://lore.kernel.org/netdev/a54c149aed38fded2d3b5fdb1a6c89e36a083b74.camel@lasnet.de/ Reviewed-by: David Ahern Signed-off-by: David S. Miller