summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* nfp: flower: add missing clean up call to avoid memory leaksJakub Kicinski2017-07-071-0/+1
| | | | | | | | | | nfp_flower_metadata_cleanup() is defined but never invoked, not calling it will cause us to leak mask and statistics queue memory on the host. Fixes: 43f84b72c50d ("nfp: add metadata to each flow offload") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* vrf: fix bug_on triggered by rx when destroying a vrfNikolay Aleksandrov2017-07-061-5/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When destroying a VRF device we cleanup the slaves in its ndo_uninit() function, but that causes packets to be switched (skb->dev == vrf being destroyed) even though we're pass the point where the VRF should be receiving any packets while it is being dismantled. This causes a BUG_ON to trigger if we have raw sockets (trace below). The reason is that the inetdev of the VRF has been destroyed but we're still sending packets up the stack with it, so let's free the slaves in the dellink callback as David Ahern suggested. Note that this fix doesn't prevent packets from going up when the VRF device is admin down. [ 35.631371] ------------[ cut here ]------------ [ 35.631603] kernel BUG at net/ipv4/fib_frontend.c:285! [ 35.631854] invalid opcode: 0000 [#1] SMP [ 35.631977] Modules linked in: [ 35.632081] CPU: 2 PID: 22 Comm: ksoftirqd/2 Not tainted 4.12.0-rc7+ #45 [ 35.632247] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [ 35.632477] task: ffff88005ad68000 task.stack: ffff88005ad64000 [ 35.632632] RIP: 0010:fib_compute_spec_dst+0xfc/0x1ee [ 35.632769] RSP: 0018:ffff88005ad67978 EFLAGS: 00010202 [ 35.632910] RAX: 0000000000000001 RBX: ffff880059a7f200 RCX: 0000000000000000 [ 35.633084] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffffffff82274af0 [ 35.633256] RBP: ffff88005ad679f8 R08: 000000000001ef70 R09: 0000000000000046 [ 35.633430] R10: ffff88005ad679f8 R11: ffff880037731cb0 R12: 0000000000000001 [ 35.633603] R13: ffff8800599e3000 R14: 0000000000000000 R15: ffff8800599cb852 [ 35.634114] FS: 0000000000000000(0000) GS:ffff88005d900000(0000) knlGS:0000000000000000 [ 35.634306] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 35.634456] CR2: 00007f3563227095 CR3: 000000000201d000 CR4: 00000000000406e0 [ 35.634632] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 35.634865] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 35.635055] Call Trace: [ 35.635271] ? __lock_acquire+0xf0d/0x1117 [ 35.635522] ipv4_pktinfo_prepare+0x82/0x151 [ 35.635831] raw_rcv_skb+0x17/0x3c [ 35.636062] raw_rcv+0xe5/0xf7 [ 35.636287] raw_local_deliver+0x169/0x1d9 [ 35.636534] ip_local_deliver_finish+0x87/0x1c4 [ 35.636820] ip_local_deliver+0x63/0x7f [ 35.637058] ip_rcv_finish+0x340/0x3a1 [ 35.637295] ip_rcv+0x314/0x34a [ 35.637525] __netif_receive_skb_core+0x49f/0x7c5 [ 35.637780] ? lock_acquire+0x13f/0x1d7 [ 35.638018] ? lock_acquire+0x15e/0x1d7 [ 35.638259] __netif_receive_skb+0x1e/0x94 [ 35.638502] ? __netif_receive_skb+0x1e/0x94 [ 35.638748] netif_receive_skb_internal+0x74/0x300 [ 35.639002] ? dev_gro_receive+0x2ed/0x411 [ 35.639246] ? lock_is_held_type+0xc4/0xd2 [ 35.639491] napi_gro_receive+0x105/0x1a0 [ 35.639736] receive_buf+0xc32/0xc74 [ 35.639965] ? detach_buf+0x67/0x153 [ 35.640201] ? virtqueue_get_buf_ctx+0x120/0x176 [ 35.640453] virtnet_poll+0x128/0x1c5 [ 35.640690] net_rx_action+0x103/0x343 [ 35.640932] __do_softirq+0x1c7/0x4b7 [ 35.641171] run_ksoftirqd+0x23/0x5c [ 35.641403] smpboot_thread_fn+0x24f/0x26d [ 35.641646] ? sort_range+0x22/0x22 [ 35.641878] kthread+0x129/0x131 [ 35.642104] ? __list_add+0x31/0x31 [ 35.642335] ? __list_add+0x31/0x31 [ 35.642568] ret_from_fork+0x2a/0x40 [ 35.642804] Code: 05 bd 87 a3 00 01 e8 1f ef 98 ff 4d 85 f6 48 c7 c7 f0 4a 27 82 41 0f 94 c4 31 c9 31 d2 41 0f b6 f4 e8 04 71 a1 ff 45 84 e4 74 02 <0f> 0b 0f b7 93 c4 00 00 00 4d 8b a5 80 05 00 00 48 03 93 d0 00 [ 35.644342] RIP: fib_compute_spec_dst+0xfc/0x1ee RSP: ffff88005ad67978 Fixes: 193125dbd8eb ("net: Introduce VRF device driver") Reported-by: Chris Cormier <chriscormier@cumulusnetworks.com> Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nfDavid S. Miller2017-07-063-0/+15
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pablo Neira Ayuso says: ==================== Netfilter fixes for net The following patchset contains two Netfilter fixes for your net tree, they are: 1) Fix memleak from netns release path of conntrack protocol trackers, patch from Liping Zhang. 2) Uninitialized flags field in ebt_log, that results in unpredictable logging format in ebtables, also from Liping. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
| * netfilter: ebt_nflog: fix unexpected truncated packetLiping Zhang2017-06-291-0/+1
| | | | | | | | | | | | | | | | | | | | | | "struct nf_loginfo li;" is a local variable, so we should set the flags to 0 explicitly, else, packets maybe truncated unexpectedly when copied to the userspace. Fixes: 7643507fe8b5 ("netfilter: xt_NFLOG: nflog-range does not truncate packets") Cc: Vishwanath Pai <vpai@akamai.com> Signed-off-by: Liping Zhang <zlpnobody@gmail.com> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
| * netfilter: nf_ct_dccp/sctp: fix memory leak after netns cleanupLiping Zhang2017-06-292-0/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After running the following commands for a while, kmemleak reported that "1879 new suspected memory leaks" happened: # while : ; do ip netns add test ip netns delete test done unreferenced object 0xffff88006342fa38 (size 1024): comm "ip", pid 15477, jiffies 4295982857 (age 957.836s) hex dump (first 32 bytes): b8 b0 4d a0 ff ff ff ff c0 34 c3 59 00 88 ff ff ..M......4.Y.... 04 00 00 00 a4 01 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [<ffffffff8190510a>] kmemleak_alloc+0x4a/0xa0 [<ffffffff81284130>] __kmalloc_track_caller+0x150/0x300 [<ffffffff812302d0>] kmemdup+0x20/0x50 [<ffffffffa04d598a>] dccp_init_net+0x8a/0x160 [nf_conntrack] [<ffffffffa04cf9f5>] nf_ct_l4proto_pernet_register_one+0x25/0x90 ... unreferenced object 0xffff88006342da58 (size 1024): comm "ip", pid 15477, jiffies 4295982857 (age 957.836s) hex dump (first 32 bytes): 10 b3 4d a0 ff ff ff ff 04 35 c3 59 00 88 ff ff ..M......5.Y.... 04 00 00 00 a4 01 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [<ffffffff8190510a>] kmemleak_alloc+0x4a/0xa0 [<ffffffff81284130>] __kmalloc_track_caller+0x150/0x300 [<ffffffff812302d0>] kmemdup+0x20/0x50 [<ffffffffa04d6a9d>] sctp_init_net+0x5d/0x130 [nf_conntrack] [<ffffffffa04cf9f5>] nf_ct_l4proto_pernet_register_one+0x25/0x90 ... This is because we forgot to implement the get_net_proto for sctp and dccp, so we won't invoke the nf_ct_unregister_sysctl to free the ctl_table when do netns cleanup. Also note, we will fail to register the sysctl for dccp/sctp either due to the lack of get_net_proto. Fixes: c51d39010a1b ("netfilter: conntrack: built-in support for DCCP") Fixes: a85406afeb3e ("netfilter: conntrack: built-in support for SCTP") Cc: Davide Caratti <dcaratti@redhat.com> Signed-off-by: Liping Zhang <zlpnobody@gmail.com> Acked-by: Davide Caratti <dcaratti@redhat.com> Acked-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
* | ptp: dte: Use LL suffix for 64-bit constantsGeert Uytterhoeven2017-07-061-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With gcc 4.1.2: drivers/ptp/ptp_dte.c: In function ‘dte_write_nco_delta’: drivers/ptp/ptp_dte.c:105: warning: integer constant is too large for ‘long’ type drivers/ptp/ptp_dte.c:112: warning: integer constant is too large for ‘long’ type drivers/ptp/ptp_dte.c:114: warning: integer constant is too large for ‘long’ type Add the missing "LL" suffix to fix this. Fixes: 8a56aa107f1e8123 ("ptp: Add a ptp clock driver for Broadcom DTE") Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* | sctp: set the value of flowi6_oif to sk_bound_dev_if to make sctp_v6_get_dst ↵Zheng Li2017-07-061-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | to find the correct route entry. if there are several same route entries with different outgoing net device, application's socket specifies the oif through setsockopt with SO_BINDTODEVICE, sctpv6 should choose the route entry whose outgoing net device is the oif which was specified by socket, set the value of flowi6_oif to sk->sk_bound_dev_if to make sctp_v6_get_dst to find the correct route entry. Signed-off-by: Zheng Li <james.z.li@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | TLS: Fix length check in do_tls_getsockopt_tx()Matthias Rosenfelder2017-07-061-1/+1
| | | | | | | | | | | | | | | | | | | | copy_to_user() copies the struct the pointer is pointing to, but the length check compares against sizeof(pointer) and not sizeof(struct). On 32-bit the size is probably the same, so it might have worked accidentally. Signed-off-by: Matthias Rosenfelder <mrosenfelder.lkml@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | tcp: md5: tcp_md5_do_lookup_exact() can be staticWu Fengguang2017-07-061-3/+3
| | | | | | | | | | | | Fixes: 6797318e623d ("tcp: md5: add an address prefix for key lookup") Signed-off-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | net: ipv6: Compare lwstate in detecting duplicate nexthopsDavid Ahern2017-07-063-11/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Lennert reported a failure to add different mpls encaps in a multipath route: $ ip -6 route add 1234::/16 \ nexthop encap mpls 10 via fe80::1 dev ens3 \ nexthop encap mpls 20 via fe80::1 dev ens3 RTNETLINK answers: File exists The problem is that the duplicate nexthop detection does not compare lwtunnel configuration. Add it. Fixes: 19e42e451506 ("ipv6: support for fib route lwtunnel encap attributes") Signed-off-by: David Ahern <dsahern@gmail.com> Reported-by: João Taveira Araújo <joao.taveira@gmail.com> Reported-by: Lennert Buytenhek <buytenh@wantstofly.org> Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com> Tested-by: Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* | liquidio: fix bug in soft reset failure detectionDerek Chickles2017-07-062-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The code that detects a failed soft reset of Octeon is comparing the wrong value against the reset value of the Octeon SLI_SCRATCH_1 register, resulting in an inability to detect a soft reset failure. Fix it by using the correct value in the comparison, which is any non-zero value. Fixes: f21fb3ed364b ("Add support of Cavium Liquidio ethernet adapters") Fixes: c0eab5b3580a ("liquidio: CN23XX firmware download") Signed-off-by: Derek Chickles <derek.chickles@cavium.com> Signed-off-by: Satanand Burla <satananda.burla@cavium.com> Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@cavium.com> Signed-off-by: Felix Manlunas <felix.manlunas@cavium.com> Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | Merge branch 'parisc-4.13-2' of ↵Linus Torvalds2017-07-063-4/+18
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux Pull another parisc update from Helge Deller: "Christoph Hellwig provided one patch for the parisc architecture to drop the DMA_ERROR_CODE define from the parisc architecture" * 'parisc-4.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux: parisc: ->mapping_error
| * | parisc: ->mapping_errorChristoph Hellwig2017-07-053-4/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | DMA_ERROR_CODE already went away in linux-next, but parisc unfortunately added a new instance of it without any review as far as I can tell. Move the two iommu drivers to report errors through ->mapping_error. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Helge Deller <deller@gmx.de>
* | | Merge tag 'arm64-upstream' of ↵Linus Torvalds2017-07-0662-341/+2190
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux Pull arm64 updates from Will Deacon: - RAS reporting via GHES/APEI (ACPI) - Indirect ftrace trampolines for modules - Improvements to kernel fault reporting - Page poisoning - Sigframe cleanups and preparation for SVE context - Core dump fixes - Sparse fixes (mainly relating to endianness) - xgene SoC PMU v3 driver - Misc cleanups and non-critical fixes * tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (75 commits) arm64: fix endianness annotation for 'struct jit_ctx' and friends arm64: cpuinfo: constify attribute_group structures. arm64: ptrace: Fix incorrect get_user() use in compat_vfp_set() arm64: ptrace: Remove redundant overrun check from compat_vfp_set() arm64: ptrace: Avoid setting compat FP[SC]R to garbage if get_user fails arm64: fix endianness annotation for __apply_alternatives()/get_alt_insn() arm64: fix endianness annotation in get_kaslr_seed() arm64: add missing conversion to __wsum in ip_fast_csum() arm64: fix endianness annotation in acpi_parking_protocol.c arm64: use readq() instead of readl() to read 64bit entry_point arm64: fix endianness annotation for reloc_insn_movw() & reloc_insn_imm() arm64: fix endianness annotation for aarch64_insn_write() arm64: fix endianness annotation in aarch64_insn_read() arm64: fix endianness annotation in call_undef_hook() arm64: fix endianness annotation for debug-monitors.c ras: mark stub functions as 'inline' arm64: pass endianness info to sparse arm64: ftrace: fix !CONFIG_ARM64_MODULE_PLTS kernels arm64: signal: Allow expansion of the signal frame acpi: apei: check for pending errors when probing GHES entries ...
| * | | arm64: fix endianness annotation for 'struct jit_ctx' and friendsLuc Van Oostenryck2017-06-301-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | struct jit_ctx::image is used the store a pointer to the jitted intructions, which are always little-endian. These instructions are thus correctly converted from native order to little-endian before being stored but the pointer 'image' is declared as for native order values. Fix this by declaring the field as __le32* instead of u32*. Same for the pointer used in jit_fill_hole() to initialize the image. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: cpuinfo: constify attribute_group structures.Arvind Yadav2017-06-301-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | attribute_groups are not supposed to change at runtime. All functions working with attribute_groups provided by <linux/sysfs.h> work with const attribute_group. So mark the non-const structs as const. Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: ptrace: Fix incorrect get_user() use in compat_vfp_set()Dave Martin2017-06-291-3/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Now that compat_vfp_get() uses the regset API to copy the FPSCR value out to userspace, compat_vfp_set() looks inconsistent. In particular, compat_vfp_set() will fail if called with kbuf != NULL && ubuf == NULL (which is valid usage according to the regset API). This patch fixes compat_vfp_set() to use user_regset_copyin(), similarly to compat_vfp_get(). This also squashes a sparse warning triggered by the cast that drops __user when calling get_user(). Signed-off-by: Dave Martin <Dave.Martin@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: ptrace: Remove redundant overrun check from compat_vfp_set()Dave Martin2017-06-291-3/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | compat_vfp_set() checks for userspace trying to write an excessive amount of data to the regset. However this check is conspicuous for its absence from every other _set() in the arm64 ptrace implementation. In fact, the core ptrace_regset() already clamps userspace's iov_len to the regset size before the individual regset .{get,set}() methods get called. This patch removes the redundant check. Signed-off-by: Dave Martin <Dave.Martin@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: ptrace: Avoid setting compat FP[SC]R to garbage if get_user failsDave Martin2017-06-291-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If get_user() fails when reading the new FPSCR value from userspace in compat_vfp_get(), then garbage* will be written to the task's FPSR and FPCR registers. This patch prevents this by checking the return from get_user() first. [*] Actually, zero, due to the behaviour of get_user() on error, but that's still not what userspace expects. Fixes: 478fcb2cdb23 ("arm64: Debugging support") Signed-off-by: Dave Martin <Dave.Martin@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: fix endianness annotation for __apply_alternatives()/get_alt_insn()Luc Van Oostenryck2017-06-291-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | get_alt_insn() is used to read and create ARM instructions, which are always stored in memory in little-endian order. These values are thus correctly converted to/from native order when processed but the pointers used to hold the address of these instructions are declared as for native order values. Fix this by declaring the pointers as __le32* instead of u32* and make the few appropriate needed changes like removing the unneeded cast '(u32*)' in front of __ALT_PTR()'s definition. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: fix endianness annotation in get_kaslr_seed()Luc Van Oostenryck2017-06-291-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In the flattened device tree format, all integer properties are in big-endian order. Here the property "kaslr-seed" is read from the fdt and then correctly converted to native order (via fdt64_to_cpu()) but the pointer used for this is not annotated as being for big-endian. Fix this by declaring the pointer as fdt64_t instead of u64 (fdt64_t being itself typedefed to __be64). Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: add missing conversion to __wsum in ip_fast_csum()Luc Van Oostenryck2017-06-291-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ARM64 implementation of ip_fast_csum() do most of the work in 128 or 64 bit and call csum_fold() to finalize. csum_fold() itself take a __wsum argument, to insure that this value is always a 32bit native-order value. Fix this by adding the sadly needed '__force' to cast the native 'sum' to the type '__wsum'. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: fix endianness annotation in acpi_parking_protocol.cLuc Van Oostenryck2017-06-291-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Here both variables 'cpu_id' and 'entry_point' are read via read[lq]_relaxed(), from a little-endian annotated pointer and then used as a native endian value. This is correct since the read[lq]() family of function internally do a little-to-native endian conversion. But in this case, it is wrong to declare these variable as little-endian since there are native ones. Fix this by changing the declaration of these variables as 'u32' or 'u64' instead of '__le32' / '__le64'. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: use readq() instead of readl() to read 64bit entry_pointLuc Van Oostenryck2017-06-291-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Here the entrypoint, declared as a 64 bit integer, is read from a pointer to 64bit integer but the read is done via readl_relaxed() which is for 32bit quantities. All the high bits will thus be lost which change the meaning of the test against zero done later. Fix this by using readq_relaxed() instead as it should be for 64bit quantities. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: fix endianness annotation for reloc_insn_movw() & reloc_insn_imm()Luc Van Oostenryck2017-06-291-7/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Here the functions reloc_insn_movw() & reloc_insn_imm() are used to read, modify and write back ARM instructions, which are always stored in memory in little-endian order. These values are thus correctly converted to/from native order but the pointers used to hold their addresses are declared as for native order values. Fix this by declaring the pointers as __le32* and remove the casts that are now unneeded. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: fix endianness annotation for aarch64_insn_write()Luc Van Oostenryck2017-06-291-3/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | aarch64_insn_write() is used to write an instruction. As on ARM64 in-memory instructions are always stored in little-endian order, this function, taking the instruction opcode in native order, correctly convert it to little-endian before sending it to an helper function __aarch64_insn_write() which will do the effective write. This is all good, but the variable and argument holding the converted value are not annotated for a little-endian value but left for native values. Fix this by adjusting the prototype of the helper and directly using the result of cpu_to_le32() without passing by an intermediate variable (which was not a distinct one but the same as the one holding the native value). Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: fix endianness annotation in aarch64_insn_read()Luc Van Oostenryck2017-06-291-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The function arch64_insn_read() is used to read an instruction. On AM64 instructions are always stored in little-endian order and thus the function correctly do a little-to-native endian conversion to the value just read. However, the variable used to hold the value before the conversion is not declared for a little-endian value but for a native one. Fix this by using the correct type for the declaration: __le32 Note: This only works because the function reading the value, probe_kernel_read((), takes a void pointer and void pointers are endian-agnostic. Otherwise probe_kernel_read() should also be properly annotated (or worse, need to be specialized). Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: fix endianness annotation in call_undef_hook()Luc Van Oostenryck2017-06-291-6/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Here we're reading thumb or ARM instructions, which are always stored in memory in little-endian order. These values are thus correctly converted to native order but the intermediate value should be annotated as for little-endian values. Fix this by declaring the intermediate var as __le32 or __le16. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | arm64: fix endianness annotation for debug-monitors.cLuc Van Oostenryck2017-06-291-6/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Here we're reading thumb or ARM instructions, which are always stored in memory in little-endian order. These values are thus correctly converted to native order but the intermediate value should be annotated as for little-endian values. Fix this by declaring the intermediate var as __le32 or __le16. Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | ras: mark stub functions as 'inline'Arnd Bergmann2017-06-291-5/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With CONFIG_RAS disabled, we get two harmless warnings about unused functions: include/linux/ras.h:37:13: error: 'log_arm_hw_error' defined but not used [-Werror=unused-function] static void log_arm_hw_error(struct cper_sec_proc_arm *err) { return; } include/linux/ras.h:33:13: error: 'log_non_standard_event' defined but not used [-Werror=unused-function] static void log_non_standard_event(const guid_t *sec_type, Clearly these are meant to be 'inline', like the other stubs in the same header. Fixes: 297b64c74385 ("ras: acpi / apei: generate trace event for unrecognized CPER section") Fixes: e9279e83ad1f ("trace, ras: add ARM processor error trace event") Acked-by: Borislav Petkov <bp@suse.de> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Will Deacon <will.deacon@arm.com>
| * | | Merge branch 'aarch64/for-next/ras-apei' into aarch64/for-next/coreWill Deacon2017-06-26314-1762/+3162
| |\ \ \ | | | | | | | | | | | | | | | Merge in arm64 ACPI RAS support (APEI/GHES) from Tyler Baicar.
| | * | | acpi: apei: check for pending errors when probing GHES entriesTyler Baicar2017-06-221-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Check for pending errors when probing GHES entries. It is possible that a fatal error is already pending at this point, so we should handle it as soon as the driver is probed. This also avoids a potential issue if there was an interrupt that was already cleared for an error since the GHES driver wasn't present. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | arm/arm64: KVM: add guest SEA supportTyler Baicar2017-06-228-12/+92
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently external aborts are unsupported by the guest abort handling. Add handling for SEAs so that the host kernel reports SEAs which occur in the guest kernel. When an SEA occurs in the guest kernel, the guest exits and is routed to kvm_handle_guest_abort(). Prior to this patch, a print message of an unsupported FSC would be printed and nothing else would happen. With this patch, the code gets routed to the APEI handling of SEAs in the host kernel to report the SEA information. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> Acked-by: Catalin Marinas <catalin.marinas@arm.com> Acked-by: Marc Zyngier <marc.zyngier@arm.com> Acked-by: Christoffer Dall <cdall@linaro.org> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | trace, ras: add ARM processor error trace eventTyler Baicar2017-06-225-1/+60
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently there are trace events for the various RAS errors with the exception of ARM processor type errors. Add a new trace event for such errors so that the user will know when they occur. These trace events are consistent with the ARM processor error section type defined in UEFI 2.6 spec section N.2.4.4. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> Acked-by: Steven Rostedt <rostedt@goodmis.org> Reviewed-by: Xie XiuQi <xiexiuqi@huawei.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | ras: acpi / apei: generate trace event for unrecognized CPER sectionTyler Baicar2017-06-225-2/+87
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The UEFI spec includes non-standard section type support in the Common Platform Error Record. This is defined in section N.2.3 of UEFI version 2.5. Currently if the CPER section's type (UUID) does not match any section type that the kernel knows how to parse, a trace event is not generated. Generate a trace event which contains the raw error data for non-standard section type error records. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> CC: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org> Tested-by: Shiju Jose <shiju.jose@huawei.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | efi: print unrecognized CPER sectionTyler Baicar2017-06-221-2/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | UEFI spec allows for non-standard section in Common Platform Error Record. This is defined in section N.2.3 of UEFI version 2.5. Currently if the CPER section's type (UUID) does not match with one of the section types that the kernel knows how to parse, the section is skipped. Therefore, user is not able to see such CPER data, for instance, error record of non-standard section. This change prints out the raw data in hex in the dmesg buffer so that non-standard sections are reported to the user. Non-standard section type errors should be reported to the user because these can include errors which are vendor specific. The data length is taken from Error Data length field of Generic Error Data Entry. The following is a sample output from dmesg: Hardware error from APEI Generic Hardware Error Source: 2 It has been corrected by h/w and requires no further action event severity: corrected time: precise 2017-03-15 20:37:35 Error 0, type: corrected section type: unknown, d2e2621c-f936-468d-0d84-15a4ed015c8b section length: 0x238 00000000: 4d415201 4d492031 453a4d45 435f4343 .RAM1 IMEM:ECC_C 00000010: 53515f45 44525f42 00000000 00000000 E_QSB_RD........ 00000020: 00000000 00000000 00000000 00000000 ................ 00000030: 00000000 00000000 01010000 01010000 ................ 00000040: 00000000 00000000 00000005 00000000 ................ 00000050: 01010000 00000000 00000001 00dddd00 ................ ... The raw data from the error can then be decoded using vendor specific tools. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> CC: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org> Reviewed-by: James Morse <james.morse@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | acpi: apei: panic OS with fatal error status blockJonathan (Zhixiong) Zhang2017-06-221-15/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Even if an error status block's severity is fatal, the kernel does not honor the severity level and panic. With the firmware first model, the platform could inform the OS about a fatal hardware error through the non-NMI GHES notification type. The OS should panic when a hardware error record is received with this severity. Call panic() after CPER data in error status block is printed if severity is fatal, before each error section is handled. Signed-off-by: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org> Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> Reviewed-by: James Morse <james.morse@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | acpi: apei: handle SEA notification type for ARMv8Tyler Baicar2017-06-225-6/+105
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ARM APEI extension proposal added SEA (Synchronous External Abort) notification type for ARMv8. Add a new GHES error source handling function for SEA. If an error source's notification type is SEA, then this function can be registered into the SEA exception handler. That way GHES will parse and report SEA exceptions when they occur. An SEA can interrupt code that had interrupts masked and is treated as an NMI. To aid this the page of address space for mapping APEI buffers while in_nmi() is always reserved, and ghes_ioremap_pfn_nmi() is changed to use the helper methods to find the prot_t to map with in the same way as ghes_ioremap_pfn_irq(). Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> CC: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org> Reviewed-by: James Morse <james.morse@arm.com> Acked-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | arm64: exception: handle Synchronous External AbortTyler Baicar2017-06-222-10/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | SEA exceptions are often caused by an uncorrected hardware error, and are handled when data abort and instruction abort exception classes have specific values for their Fault Status Code. When SEA occurs, before killing the process, report the error in the kernel logs. Update fault_info[] with specific SEA faults so that the new SEA handler is used. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> CC: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org> Reviewed-by: James Morse <james.morse@arm.com> Acked-by: Catalin Marinas <catalin.marinas@arm.com> [will: use NULL instead of 0 when assigning si_addr] Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | efi: parse ARM processor errorTyler Baicar2017-06-222-0/+183
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add support for ARM Common Platform Error Record (CPER). UEFI 2.6 specification adds support for ARM specific processor error information to be reported as part of the CPER records. This provides more detail on for processor error logs. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> CC: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org> Reviewed-by: James Morse <james.morse@arm.com> Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | cper: add timestamp print to CPER status printingTyler Baicar2017-06-221-0/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The ACPI 6.1 spec added a timestamp to the generic error data entry structure. Print the timestamp out when printing out the error information. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> CC: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | ras: acpi/apei: cper: add support for generic data v3 structureTyler Baicar2017-06-223-21/+63
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The ACPI 6.1 spec adds a new revision of the generic error data entry structure. Add support to handle the new structure as well as properly verify and iterate through the generic data entries. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> CC: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | acpi: apei: read ack upon ghes record consumptionTyler Baicar2017-06-223-6/+65
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A RAS (Reliability, Availability, Serviceability) controller may be a separate processor running in parallel with OS execution, and may generate error records for consumption by the OS. If the RAS controller produces multiple error records, then they may be overwritten before the OS has consumed them. The Generic Hardware Error Source (GHES) v2 structure introduces the capability for the OS to acknowledge the consumption of the error record generated by the RAS controller. A RAS controller supporting GHESv2 shall wait for the acknowledgment before writing a new error record, thus eliminating the race condition. Add support for parsing of GHESv2 sub-tables as well. Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org> CC: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org> Reviewed-by: James Morse <james.morse@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | Merge branch 'uuid-types' of git://git.infradead.org/users/hch/uuid into ↵Will Deacon2017-06-2076-500/+373
| | |\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | aarch64/for-next/ras-apei Pull in uuid-types branch from Christoph, since this conflicts heavily with the ACPI/APEI RAS work from Tyler Baicer and was created as an immutable branch to avoid conflicts with ACPI development.
| * | \ \ \ Merge branch 'perf/updates' into aarch64/for-next/coreWill Deacon2017-06-264-62/+635
| |\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Merge in arm64 perf updates: * xgene system PMUv3 support * 16-bit events for ARMv8.1
| | * | | | | perf: xgene: Add support for SoC PMU version 3Hoan Tran2017-06-221-29/+534
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds support for SoC-wide (AKA uncore) Performance Monitoring Unit version 3. It can support up to - 2 IOB PMU instances - 8 L3C PMU instances - 2 MCB PMU instances - 8 MCU PMU instances and these PMUs support 64 bit counter Signed-off-by: Hoan Tran <hotran@apm.com> [Mark: stop counters in _xgene_pmu_isr()] Signed-off-by: Mark Rutland <mark.rutland@arm.com> [will: make xgene_pmu_v3_ops static] Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | | | perf: xgene: Move PMU leaf functions into function pointer structureHoan Tran2017-06-221-18/+65
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch moves PMU leaf functions into a function pointer structure. It helps code maintain and expasion easier. Signed-off-by: Hoan Tran <hotran@apm.com> [Mark: remove redundant cast] Signed-off-by: Mark Rutland <mark.rutland@arm.com> [will: make xgene_pmu_ops static] Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | | | perf: xgene: Parse PMU subnode from the match tableHoan Tran2017-06-221-10/+30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch parses PMU Subnode from a match table. Signed-off-by: Hoan Tran <hotran@apm.com> Signed-off-by: Mark Rutland <mark.rutland@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | | | drivers/perf: commonise PERF_EVENTS dependencyMark Rutland2017-06-151-4/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | All PMU drivers are going to depend on PERF_EVENTS, so let's make this dependency common and simplify the individual Kconfig entries. Signed-off-by: Mark Rutland <mark.rutland@arm.com> Cc: Will Deacon <will.deacon@arm.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
| | * | | | | arm: perf: make of_device_ids constArvind Yadav2017-06-151-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | of_device_ids are not supposed to change at runtime. All functions working with of_device_ids provided by <linux/of.h> work with const of_device_ids. So mark the non-const structs as const. Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com> Signed-off-by: Will Deacon <will.deacon@arm.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-01-14 23:19:18 +0100