user/sven/linux.git/include, branch v4.14.141

xfrm: policy: remove pcpu policy cache

2019-08-25T08:50:29Z

commit e4db5b61c572475bbbcf63e3c8a2606bfccf2c9d upstream. Kristian Evensen says: In a project I am involved in, we are running ipsec (Strongswan) on different mt7621-based routers. Each router is configured as an initiator and has around ~30 tunnels to different responders (running on misc. devices). Before the flow cache was removed (kernel 4.9), we got a combined throughput of around 70Mbit/s for all tunnels on one router. However, we recently switched to kernel 4.14 (4.14.48), and the total throughput is somewhere around 57Mbit/s (best-case). I.e., a drop of around 20%. Reverting the flow cache removal restores, as expected, performance levels to that of kernel 4.9. When pcpu xdst exists, it has to be validated first before it can be used. A negative hit thus increases cost vs. no-cache. As number of tunnels increases, hit rate decreases so this pcpu caching isn't a viable strategy. Furthermore, the xdst cache also needs to run with BH off, so when removing this the bh disable/enable pairs can be removed too. Kristian tested a 4.14.y backport of this change and reported increased performance: In our tests, the throughput reduction has been reduced from around -20% to -5%. We also see that the overall throughput is independent of the number of tunnels, while before the throughput was reduced as the number of tunnels increased. Reported-by: Kristian Evensen Signed-off-by: Florian Westphal Signed-off-by: Steffen Klassert Signed-off-by: Greg Kroah-Hartman

Revert "tcp: Clear sk_send_head after purging the write queue"

2019-08-25T08:50:23Z

This reverts commit e99e7745d03fc50ba7c5b7c91c17294fee2d5991. Ben Hutchings writes: >Sorry, this is the same issue that was already fixed by "tcp: reset >sk_send_head in tcp_write_queue_purge". You can drop my version from >the queue for 4.4 and 4.9 and revert it for 4.14. Signed-off-by: Sasha Levin

bpf: fix bpf_jit_limit knob for PAGE_SIZE >= 64K

2019-08-25T08:50:23Z

[ Upstream commit fdadd04931c2d7cd294dc5b2b342863f94be53a3 ] Michael and Sandipan report: Commit ede95a63b5 introduced a bpf_jit_limit tuneable to limit BPF JIT allocations. At compile time it defaults to PAGE_SIZE * 40000, and is adjusted again at init time if MODULES_VADDR is defined. For ppc64 kernels, MODULES_VADDR isn't defined, so we're stuck with the compile-time default at boot-time, which is 0x9c400000 when using 64K page size. This overflows the signed 32-bit bpf_jit_limit value: root@ubuntu:/tmp# cat /proc/sys/net/core/bpf_jit_limit -1673527296 and can cause various unexpected failures throughout the network stack. In one case `strace dhclient eth0` reported: setsockopt(5, SOL_SOCKET, SO_ATTACH_FILTER, {len=11, filter=0x105dd27f8}, 16) = -1 ENOTSUPP (Unknown error 524) and similar failures can be seen with tools like tcpdump. This doesn't always reproduce however, and I'm not sure why. The more consistent failure I've seen is an Ubuntu 18.04 KVM guest booted on a POWER9 host would time out on systemd/netplan configuring a virtio-net NIC with no noticeable errors in the logs. Given this and also given that in near future some architectures like arm64 will have a custom area for BPF JIT image allocations we should get rid of the BPF_JIT_LIMIT_DEFAULT fallback / default entirely. For 4.21, we have an overridable bpf_jit_alloc_exec(), bpf_jit_free_exec() so therefore add another overridable bpf_jit_alloc_exec_limit() helper function which returns the possible size of the memory area for deriving the default heuristic in bpf_jit_charge_init(). Like bpf_jit_alloc_exec() and bpf_jit_free_exec(), the new bpf_jit_alloc_exec_limit() assumes that module_alloc() is the default JIT memory provider, and therefore in case archs implement their custom module_alloc() we use MODULES_{END,_VADDR} for limits and otherwise for vmalloc_exec() cases like on ppc64 we use VMALLOC_{END,_START}. Additionally, for archs supporting large page sizes, we should change the sysctl to be handled as long to not run into sysctl restrictions in future. Fixes: ede95a63b5e8 ("bpf: add bpf_jit_limit knob to restrict unpriv allocations") Reported-by: Sandipan Das Reported-by: Michael Roth Signed-off-by: Daniel Borkmann Tested-by: Michael Roth Signed-off-by: Alexei Starovoitov Signed-off-by: Sasha Levin

KVM: arm/arm64: Sync ICH_VMCR_EL2 back when about to block

2019-08-25T08:50:18Z

commit 5eeaf10eec394b28fad2c58f1f5c3a5da0e87d1c upstream. Since commit commit 328e56647944 ("KVM: arm/arm64: vgic: Defer touching GICH_VMCR to vcpu_load/put"), we leave ICH_VMCR_EL2 (or its GICv2 equivalent) loaded as long as we can, only syncing it back when we're scheduled out. There is a small snag with that though: kvm_vgic_vcpu_pending_irq(), which is indirectly called from kvm_vcpu_check_block(), needs to evaluate the guest's view of ICC_PMR_EL1. At the point were we call kvm_vcpu_check_block(), the vcpu is still loaded, and whatever changes to PMR is not visible in memory until we do a vcpu_put(). Things go really south if the guest does the following: mov x0, #0 // or any small value masking interrupts msr ICC_PMR_EL1, x0 [vcpu preempted, then rescheduled, VMCR sampled] mov x0, #ff // allow all interrupts msr ICC_PMR_EL1, x0 wfi // traps to EL2, so samping of VMCR [interrupt arrives just after WFI] Here, the hypervisor's view of PMR is zero, while the guest has enabled its interrupts. kvm_vgic_vcpu_pending_irq() will then say that no interrupts are pending (despite an interrupt being received) and we'll block for no reason. If the guest doesn't have a periodic interrupt firing once it has blocked, it will stay there forever. To avoid this unfortuante situation, let's resync VMCR from kvm_arch_vcpu_blocking(), ensuring that a following kvm_vcpu_check_block() will observe the latest value of PMR. This has been found by booting an arm64 Linux guest with the pseudo NMI feature, and thus using interrupt priorities to mask interrupts instead of the usual PSTATE masking. Cc: stable@vger.kernel.org # 4.12 Fixes: 328e56647944 ("KVM: arm/arm64: vgic: Defer touching GICH_VMCR to vcpu_load/put") Signed-off-by: Marc Zyngier Signed-off-by: Greg Kroah-Hartman

asm-generic: fix -Wtype-limits compiler warnings

2019-08-25T08:50:18Z

[ Upstream commit cbedfe11347fe418621bd188d58a206beb676218 ] Commit d66acc39c7ce ("bitops: Optimise get_order()") introduced a compilation warning because "rx_frag_size" is an "ushort" while PAGE_SHIFT here is 16. The commit changed the get_order() to be a multi-line macro where compilers insist to check all statements in the macro even when __builtin_constant_p(rx_frag_size) will return false as "rx_frag_size" is a module parameter. In file included from ./arch/powerpc/include/asm/page_64.h:107, from ./arch/powerpc/include/asm/page.h:242, from ./arch/powerpc/include/asm/mmu.h:132, from ./arch/powerpc/include/asm/lppaca.h:47, from ./arch/powerpc/include/asm/paca.h:17, from ./arch/powerpc/include/asm/current.h:13, from ./include/linux/thread_info.h:21, from ./arch/powerpc/include/asm/processor.h:39, from ./include/linux/prefetch.h:15, from drivers/net/ethernet/emulex/benet/be_main.c:14: drivers/net/ethernet/emulex/benet/be_main.c: In function 'be_rx_cqs_create': ./include/asm-generic/getorder.h:54:9: warning: comparison is always true due to limited range of data type [-Wtype-limits] (((n) < (1UL << PAGE_SHIFT)) ? 0 : \ ^ drivers/net/ethernet/emulex/benet/be_main.c:3138:33: note: in expansion of macro 'get_order' adapter->big_page_size = (1 << get_order(rx_frag_size)) * PAGE_SIZE; ^~~~~~~~~ Fix it by moving all of this multi-line macro into a proper function, and killing __get_order() off. [akpm@linux-foundation.org: remove __get_order() altogether] [cai@lca.pw: v2] Link: http://lkml.kernel.org/r/1564000166-31428-1-git-send-email-cai@lca.pw Link: http://lkml.kernel.org/r/1563914986-26502-1-git-send-email-cai@lca.pw Fixes: d66acc39c7ce ("bitops: Optimise get_order()") Signed-off-by: Qian Cai Reviewed-by: Nathan Chancellor Cc: David S. Miller Cc: Arnd Bergmann Cc: David Howells Cc: Jakub Jelinek Cc: Nick Desaulniers Cc: Bill Wendling Cc: James Y Knight Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Sasha Levin

bpf: add bpf_jit_limit knob to restrict unpriv allocations

2019-08-25T08:50:03Z

commit ede95a63b5e84ddeea6b0c473b36ab8bfd8c6ce3 upstream. Rick reported that the BPF JIT could potentially fill the entire module space with BPF programs from unprivileged users which would prevent later attempts to load normal kernel modules or privileged BPF programs, for example. If JIT was enabled but unsuccessful to generate the image, then before commit 290af86629b2 ("bpf: introduce BPF_JIT_ALWAYS_ON config") we would always fall back to the BPF interpreter. Nowadays in the case where the CONFIG_BPF_JIT_ALWAYS_ON could be set, then the load will abort with a failure since the BPF interpreter was compiled out. Add a global limit and enforce it for unprivileged users such that in case of BPF interpreter compiled out we fail once the limit has been reached or we fall back to BPF interpreter earlier w/o using module mem if latter was compiled in. In a next step, fair share among unprivileged users can be resolved in particular for the case where we would fail hard once limit is reached. Fixes: 290af86629b2 ("bpf: introduce BPF_JIT_ALWAYS_ON config") Fixes: 0a14842f5a3c ("net: filter: Just In Time compiler for x86-64") Co-Developed-by: Rick Edgecombe Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Cc: Eric Dumazet Cc: Jann Horn Cc: Kees Cook Cc: LKML Signed-off-by: Alexei Starovoitov Signed-off-by: Ben Hutchings Signed-off-by: Greg Kroah-Hartman

KVM: Fix leak vCPU's VMCS value into other pCPU

2019-08-16T08:13:57Z

commit 17e433b54393a6269acbcb792da97791fe1592d8 upstream. After commit d73eb57b80b (KVM: Boost vCPUs that are delivering interrupts), a five years old bug is exposed. Running ebizzy benchmark in three 80 vCPUs VMs on one 80 pCPUs Skylake server, a lot of rcu_sched stall warning splatting in the VMs after stress testing: INFO: rcu_sched detected stalls on CPUs/tasks: { 4 41 57 62 77} (detected by 15, t=60004 jiffies, g=899, c=898, q=15073) Call Trace: flush_tlb_mm_range+0x68/0x140 tlb_flush_mmu.part.75+0x37/0xe0 tlb_finish_mmu+0x55/0x60 zap_page_range+0x142/0x190 SyS_madvise+0x3cd/0x9c0 system_call_fastpath+0x1c/0x21 swait_active() sustains to be true before finish_swait() is called in kvm_vcpu_block(), voluntarily preempted vCPUs are taken into account by kvm_vcpu_on_spin() loop greatly increases the probability condition kvm_arch_vcpu_runnable(vcpu) is checked and can be true, when APICv is enabled the yield-candidate vCPU's VMCS RVI field leaks(by vmx_sync_pir_to_irr()) into spinning-on-a-taken-lock vCPU's current VMCS. This patch fixes it by checking conservatively a subset of events. Cc: Paolo Bonzini Cc: Radim Krčmář Cc: Christian Borntraeger Cc: Marc Zyngier Cc: stable@vger.kernel.org Fixes: 98f4a1467 (KVM: add kvm_arch_vcpu_runnable() test to kvm_vcpu_on_spin() loop) Signed-off-by: Wanpeng Li Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman

ALSA: compress: Fix regression on compressed capture streams

2019-08-16T08:13:52Z

[ Upstream commit 4475f8c4ab7b248991a60d9c02808dbb813d6be8 ] A previous fix to the stop handling on compressed capture streams causes some knock on issues. The previous fix updated snd_compr_drain_notify to set the state back to PREPARED for capture streams. This causes some issues however as the handling for snd_compr_poll differs between the two states and some user-space applications were relying on the poll failing after the stream had been stopped. To correct this regression whilst still fixing the original problem the patch was addressing, update the capture handling to skip the PREPARED state rather than skipping the SETUP state as it has done until now. Fixes: 4f2ab5e1d13d ("ALSA: compress: Fix stop handling on compressed capture streams") Signed-off-by: Charles Keepax Acked-by: Vinod Koul Signed-off-by: Takashi Iwai Signed-off-by: Sasha Levin

tcp: Clear sk_send_head after purging the write queue

2019-08-16T08:13:47Z

Denis Andzakovic discovered a potential use-after-free in older kernel versions, using syzkaller. tcp_write_queue_purge() frees all skbs in the TCP write queue and can leave sk->sk_send_head pointing to freed memory. tcp_disconnect() clears that pointer after calling tcp_write_queue_purge(), but tcp_connect() does not. It is (surprisingly) possible to add to the write queue between disconnection and reconnection, so this needs to be done in both places. This bug was introduced by backports of commit 7f582b248d0a ("tcp: purge write queue in tcp_connect_init()") and does not exist upstream because of earlier changes in commit 75c119afe14f ("tcp: implement rb-tree based retransmit queue"). The latter is a major change that's not suitable for stable. Reported-by: Denis Andzakovic Bisected-by: Salvatore Bonaccorso Fixes: 7f582b248d0a ("tcp: purge write queue in tcp_connect_init()") Cc: # before 4.15 Cc: Eric Dumazet Signed-off-by: Ben Hutchings Signed-off-by: Greg Kroah-Hartman

crypto: ccp - Add support for valid authsize values less than 16

2019-08-16T08:13:47Z

commit 9f00baf74e4b6f79a3a3dfab44fb7bb2e797b551 upstream. AES GCM encryption allows for authsize values of 4, 8, and 12-16 bytes. Validate the requested authsize, and retain it to save in the request context. Fixes: 36cf515b9bbe2 ("crypto: ccp - Enable support for AES GCM on v5 CCPs") Cc: Signed-off-by: Gary R Hook Signed-off-by: Herbert Xu Signed-off-by: Greg Kroah-Hartman