From c1aca3080e382886e2e58e809787441984a2f89b Mon Sep 17 00:00:00 2001 From: Yan Yan Date: Tue, 18 Jan 2022 16:00:13 -0800 Subject: xfrm: Check if_id in xfrm_migrate This patch enables distinguishing SAs and SPs based on if_id during the xfrm_migrate flow. This ensures support for xfrm interfaces throughout the SA/SP lifecycle. When there are multiple existing SPs with the same direction, the same xfrm_selector and different endpoint addresses, xfrm_migrate might fail with ENODATA. Specifically, the code path for performing xfrm_migrate is: Stage 1: find policy to migrate with xfrm_migrate_policy_find(sel, dir, type, net) Stage 2: find and update state(s) with xfrm_migrate_state_find(mp, net) Stage 3: update endpoint address(es) of template(s) with xfrm_policy_migrate(pol, m, num_migrate) Currently "Stage 1" always returns the first xfrm_policy that matches, and "Stage 3" looks for the xfrm_tmpl that matches the old endpoint address. Thus if there are multiple xfrm_policy with same selector, direction, type and net, "Stage 1" might rertun a wrong xfrm_policy and "Stage 3" will fail with ENODATA because it cannot find a xfrm_tmpl with the matching endpoint address. The fix is to allow userspace to pass an if_id and add if_id to the matching rule in Stage 1 and Stage 2 since if_id is a unique ID for xfrm_policy and xfrm_state. For compatibility, if_id will only be checked if the attribute is set. Tested with additions to Android's kernel unit test suite: https://android-review.googlesource.com/c/kernel/tests/+/1668886 Signed-off-by: Yan Yan Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index fdb41e8bb626..743dd1da506e 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1681,14 +1681,15 @@ int km_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_bundles, const struct xfrm_kmaddress *k, const struct xfrm_encap_tmpl *encap); -struct xfrm_state *xfrm_migrate_state_find(struct xfrm_migrate *m, struct net *net); +struct xfrm_state *xfrm_migrate_state_find(struct xfrm_migrate *m, struct net *net, + u32 if_id); struct xfrm_state *xfrm_state_migrate(struct xfrm_state *x, struct xfrm_migrate *m, struct xfrm_encap_tmpl *encap); int xfrm_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, struct xfrm_migrate *m, int num_bundles, struct xfrm_kmaddress *k, struct net *net, - struct xfrm_encap_tmpl *encap); + struct xfrm_encap_tmpl *encap, u32 if_id); #endif int km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, __be16 sport); -- cgit v1.2.3 From a6d95c5a628a09be129f25d5663a7e9db8261f51 Mon Sep 17 00:00:00 2001 From: Jiri Bohac Date: Wed, 26 Jan 2022 16:00:18 +0100 Subject: Revert "xfrm: xfrm_state_mtu should return at least 1280 for ipv6" This reverts commit b515d2637276a3810d6595e10ab02c13bfd0b63a. Commit b515d2637276a3810d6595e10ab02c13bfd0b63a ("xfrm: xfrm_state_mtu should return at least 1280 for ipv6") in v5.14 breaks the TCP MSS calculation in ipsec transport mode, resulting complete stalls of TCP connections. This happens when the (P)MTU is 1280 or slighly larger. The desired formula for the MSS is: MSS = (MTU - ESP_overhead) - IP header - TCP header However, the above commit clamps the (MTU - ESP_overhead) to a minimum of 1280, turning the formula into MSS = max(MTU - ESP overhead, 1280) - IP header - TCP header With the (P)MTU near 1280, the calculated MSS is too large and the resulting TCP packets never make it to the destination because they are over the actual PMTU. The above commit also causes suboptimal double fragmentation in xfrm tunnel mode, as described in https://lore.kernel.org/netdev/20210429202529.codhwpc7w6kbudug@dwarf.suse.cz/ The original problem the above commit was trying to fix is now fixed by commit 6596a0229541270fb8d38d989f91b78838e5e9da ("xfrm: fix MTU regression"). Signed-off-by: Jiri Bohac Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 1 - net/ipv4/esp4.c | 2 +- net/ipv6/esp6.c | 2 +- net/xfrm/xfrm_state.c | 14 ++------------ 4 files changed, 4 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 743dd1da506e..76aa6f11a540 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1568,7 +1568,6 @@ void xfrm_sad_getinfo(struct net *net, struct xfrmk_sadinfo *si); void xfrm_spd_getinfo(struct net *net, struct xfrmk_spdinfo *si); u32 xfrm_replay_seqhi(struct xfrm_state *x, __be32 net_seq); int xfrm_init_replay(struct xfrm_state *x); -u32 __xfrm_state_mtu(struct xfrm_state *x, int mtu); u32 xfrm_state_mtu(struct xfrm_state *x, int mtu); int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload); int xfrm_init_state(struct xfrm_state *x); diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 851f542928a3..e1b1d080e908 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -671,7 +671,7 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb) struct xfrm_dst *dst = (struct xfrm_dst *)skb_dst(skb); u32 padto; - padto = min(x->tfcpad, __xfrm_state_mtu(x, dst->child_mtu_cached)); + padto = min(x->tfcpad, xfrm_state_mtu(x, dst->child_mtu_cached)); if (skb->len < padto) esp.tfclen = padto - skb->len; } diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 8bb2c407b46b..7591160edce1 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -707,7 +707,7 @@ static int esp6_output(struct xfrm_state *x, struct sk_buff *skb) struct xfrm_dst *dst = (struct xfrm_dst *)skb_dst(skb); u32 padto; - padto = min(x->tfcpad, __xfrm_state_mtu(x, dst->child_mtu_cached)); + padto = min(x->tfcpad, xfrm_state_mtu(x, dst->child_mtu_cached)); if (skb->len < padto) esp.tfclen = padto - skb->len; } diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 1ba6fbfe8cdb..b749935152ba 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -2579,7 +2579,7 @@ void xfrm_state_delete_tunnel(struct xfrm_state *x) } EXPORT_SYMBOL(xfrm_state_delete_tunnel); -u32 __xfrm_state_mtu(struct xfrm_state *x, int mtu) +u32 xfrm_state_mtu(struct xfrm_state *x, int mtu) { const struct xfrm_type *type = READ_ONCE(x->type); struct crypto_aead *aead; @@ -2610,17 +2610,7 @@ u32 __xfrm_state_mtu(struct xfrm_state *x, int mtu) return ((mtu - x->props.header_len - crypto_aead_authsize(aead) - net_adj) & ~(blksize - 1)) + net_adj - 2; } -EXPORT_SYMBOL_GPL(__xfrm_state_mtu); - -u32 xfrm_state_mtu(struct xfrm_state *x, int mtu) -{ - mtu = __xfrm_state_mtu(x, mtu); - - if (x->props.family == AF_INET6 && mtu < IPV6_MIN_MTU) - return IPV6_MIN_MTU; - - return mtu; -} +EXPORT_SYMBOL_GPL(xfrm_state_mtu); int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload) { -- cgit v1.2.3 From 7c76ecd9c99b6e9a771d813ab1aa7fa428b3ade1 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Tue, 8 Feb 2022 16:14:32 +0200 Subject: xfrm: enforce validity of offload input flags struct xfrm_user_offload has flags variable that received user input, but kernel didn't check if valid bits were provided. It caused a situation where not sanitized input was forwarded directly to the drivers. For example, XFRM_OFFLOAD_IPV6 define that was exposed, was used by strongswan, but not implemented in the kernel at all. As a solution, check and sanitize input flags to forward XFRM_OFFLOAD_INBOUND to the drivers. Fixes: d77e38e612a0 ("xfrm: Add an IPsec hardware offloading API") Signed-off-by: Leon Romanovsky Signed-off-by: Steffen Klassert --- include/uapi/linux/xfrm.h | 6 ++++++ net/xfrm/xfrm_device.c | 6 +++++- 2 files changed, 11 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/uapi/linux/xfrm.h b/include/uapi/linux/xfrm.h index 4e29d7851890..65e13a099b1a 100644 --- a/include/uapi/linux/xfrm.h +++ b/include/uapi/linux/xfrm.h @@ -511,6 +511,12 @@ struct xfrm_user_offload { int ifindex; __u8 flags; }; +/* This flag was exposed without any kernel code that supporting it. + * Unfortunately, strongswan has the code that uses sets this flag, + * which makes impossible to reuse this bit. + * + * So leave it here to make sure that it won't be reused by mistake. + */ #define XFRM_OFFLOAD_IPV6 1 #define XFRM_OFFLOAD_INBOUND 2 diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c index 3fa066419d37..39bce5d764de 100644 --- a/net/xfrm/xfrm_device.c +++ b/net/xfrm/xfrm_device.c @@ -223,6 +223,9 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x, if (x->encap || x->tfcpad) return -EINVAL; + if (xuo->flags & ~(XFRM_OFFLOAD_IPV6 | XFRM_OFFLOAD_INBOUND)) + return -EINVAL; + dev = dev_get_by_index(net, xuo->ifindex); if (!dev) { if (!(xuo->flags & XFRM_OFFLOAD_INBOUND)) { @@ -262,7 +265,8 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x, netdev_tracker_alloc(dev, &xso->dev_tracker, GFP_ATOMIC); xso->real_dev = dev; xso->num_exthdrs = 1; - xso->flags = xuo->flags; + /* Don't forward bit that is not implemented */ + xso->flags = xuo->flags & ~XFRM_OFFLOAD_IPV6; err = dev->xfrmdev_ops->xdo_dev_state_add(x); if (err) { -- cgit v1.2.3 From ba2689234be92024e5635d30fe744f4853ad97db Mon Sep 17 00:00:00 2001 From: James Morse Date: Thu, 18 Nov 2021 13:59:46 +0000 Subject: arm64: entry: Add vectors that have the bhb mitigation sequences Some CPUs affected by Spectre-BHB need a sequence of branches, or a firmware call to be run before any indirect branch. This needs to go in the vectors. No CPU needs both. While this can be patched in, it would run on all CPUs as there is a single set of vectors. If only one part of a big/little combination is affected, the unaffected CPUs have to run the mitigation too. Create extra vectors that include the sequence. Subsequent patches will allow affected CPUs to select this set of vectors. Later patches will modify the loop count to match what the CPU requires. Reviewed-by: Catalin Marinas Signed-off-by: James Morse --- arch/arm64/include/asm/assembler.h | 24 +++++++++++++++++ arch/arm64/include/asm/vectors.h | 34 ++++++++++++++++++++++++ arch/arm64/kernel/entry.S | 53 +++++++++++++++++++++++++++++++------- arch/arm64/kernel/proton-pack.c | 16 ++++++++++++ include/linux/arm-smccc.h | 5 ++++ 5 files changed, 123 insertions(+), 9 deletions(-) create mode 100644 arch/arm64/include/asm/vectors.h (limited to 'include') diff --git a/arch/arm64/include/asm/assembler.h b/arch/arm64/include/asm/assembler.h index e8bd0af0141c..046c38ee2841 100644 --- a/arch/arm64/include/asm/assembler.h +++ b/arch/arm64/include/asm/assembler.h @@ -850,4 +850,28 @@ alternative_endif #endif /* GNU_PROPERTY_AARCH64_FEATURE_1_DEFAULT */ + .macro __mitigate_spectre_bhb_loop tmp +#ifdef CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY + mov \tmp, #32 +.Lspectre_bhb_loop\@: + b . + 4 + subs \tmp, \tmp, #1 + b.ne .Lspectre_bhb_loop\@ + sb +#endif /* CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY */ + .endm + + /* Save/restores x0-x3 to the stack */ + .macro __mitigate_spectre_bhb_fw +#ifdef CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY + stp x0, x1, [sp, #-16]! + stp x2, x3, [sp, #-16]! + mov w0, #ARM_SMCCC_ARCH_WORKAROUND_3 +alternative_cb smccc_patch_fw_mitigation_conduit + nop // Patched to SMC/HVC #0 +alternative_cb_end + ldp x2, x3, [sp], #16 + ldp x0, x1, [sp], #16 +#endif /* CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY */ + .endm #endif /* __ASM_ASSEMBLER_H */ diff --git a/arch/arm64/include/asm/vectors.h b/arch/arm64/include/asm/vectors.h new file mode 100644 index 000000000000..bac53fad037d --- /dev/null +++ b/arch/arm64/include/asm/vectors.h @@ -0,0 +1,34 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2022 ARM Ltd. + */ +#ifndef __ASM_VECTORS_H +#define __ASM_VECTORS_H + +/* + * Note: the order of this enum corresponds to two arrays in entry.S: + * tramp_vecs and __bp_harden_el1_vectors. By default the canonical + * 'full fat' vectors are used directly. + */ +enum arm64_bp_harden_el1_vectors { +#ifdef CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY + /* + * Perform the BHB loop mitigation, before branching to the canonical + * vectors. + */ + EL1_VECTOR_BHB_LOOP, + + /* + * Make the SMC call for firmware mitigation, before branching to the + * canonical vectors. + */ + EL1_VECTOR_BHB_FW, +#endif /* CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY */ + + /* + * Remap the kernel before branching to the canonical vectors. + */ + EL1_VECTOR_KPTI, ++}; + +#endif /* __ASM_VECTORS_H */ diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S index 9c4ff75f983e..2ceb0c3647b4 100644 --- a/arch/arm64/kernel/entry.S +++ b/arch/arm64/kernel/entry.S @@ -646,13 +646,26 @@ alternative_else_nop_endif sub \dst, \dst, PAGE_SIZE .endm - .macro tramp_ventry, vector_start, regsize, kpti + +#define BHB_MITIGATION_NONE 0 +#define BHB_MITIGATION_LOOP 1 +#define BHB_MITIGATION_FW 2 + + .macro tramp_ventry, vector_start, regsize, kpti, bhb .align 7 1: .if \regsize == 64 msr tpidrro_el0, x30 // Restored in kernel_ventry .endif + .if \bhb == BHB_MITIGATION_LOOP + /* + * This sequence must appear before the first indirect branch. i.e. the + * ret out of tramp_ventry. It appears here because x30 is free. + */ + __mitigate_spectre_bhb_loop x30 + .endif // \bhb == BHB_MITIGATION_LOOP + .if \kpti == 1 /* * Defend against branch aliasing attacks by pushing a dummy @@ -680,6 +693,15 @@ alternative_else_nop_endif ldr x30, =vectors .endif // \kpti == 1 + .if \bhb == BHB_MITIGATION_FW + /* + * The firmware sequence must appear before the first indirect branch. + * i.e. the ret out of tramp_ventry. But it also needs the stack to be + * mapped to save/restore the registers the SMC clobbers. + */ + __mitigate_spectre_bhb_fw + .endif // \bhb == BHB_MITIGATION_FW + add x30, x30, #(1b - \vector_start + 4) ret .org 1b + 128 // Did we overflow the ventry slot? @@ -687,6 +709,9 @@ alternative_else_nop_endif .macro tramp_exit, regsize = 64 adr x30, tramp_vectors +#ifdef CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY + add x30, x30, SZ_4K +#endif msr vbar_el1, x30 ldr lr, [sp, #S_LR] tramp_unmap_kernel x29 @@ -698,26 +723,32 @@ alternative_else_nop_endif sb .endm - .macro generate_tramp_vector, kpti + .macro generate_tramp_vector, kpti, bhb .Lvector_start\@: .space 0x400 .rept 4 - tramp_ventry .Lvector_start\@, 64, \kpti + tramp_ventry .Lvector_start\@, 64, \kpti, \bhb .endr .rept 4 - tramp_ventry .Lvector_start\@, 32, \kpti + tramp_ventry .Lvector_start\@, 32, \kpti, \bhb .endr .endm #ifdef CONFIG_UNMAP_KERNEL_AT_EL0 /* * Exception vectors trampoline. + * The order must match __bp_harden_el1_vectors and the + * arm64_bp_harden_el1_vectors enum. */ .pushsection ".entry.tramp.text", "ax" .align 11 SYM_CODE_START_NOALIGN(tramp_vectors) - generate_tramp_vector kpti=1 +#ifdef CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY + generate_tramp_vector kpti=1, bhb=BHB_MITIGATION_LOOP + generate_tramp_vector kpti=1, bhb=BHB_MITIGATION_FW +#endif /* CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY */ + generate_tramp_vector kpti=1, bhb=BHB_MITIGATION_NONE SYM_CODE_END(tramp_vectors) SYM_CODE_START(tramp_exit_native) @@ -744,7 +775,7 @@ SYM_DATA_END(__entry_tramp_data_start) * Exception vectors for spectre mitigations on entry from EL1 when * kpti is not in use. */ - .macro generate_el1_vector + .macro generate_el1_vector, bhb .Lvector_start\@: kernel_ventry 1, t, 64, sync // Synchronous EL1t kernel_ventry 1, t, 64, irq // IRQ EL1t @@ -757,17 +788,21 @@ SYM_DATA_END(__entry_tramp_data_start) kernel_ventry 1, h, 64, error // Error EL1h .rept 4 - tramp_ventry .Lvector_start\@, 64, kpti=0 + tramp_ventry .Lvector_start\@, 64, 0, \bhb .endr .rept 4 - tramp_ventry .Lvector_start\@, 32, kpti=0 + tramp_ventry .Lvector_start\@, 32, 0, \bhb .endr .endm +/* The order must match tramp_vecs and the arm64_bp_harden_el1_vectors enum. */ .pushsection ".entry.text", "ax" .align 11 SYM_CODE_START(__bp_harden_el1_vectors) - generate_el1_vector +#ifdef CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY + generate_el1_vector bhb=BHB_MITIGATION_LOOP + generate_el1_vector bhb=BHB_MITIGATION_FW +#endif /* CONFIG_MITIGATE_SPECTRE_BRANCH_HISTORY */ SYM_CODE_END(__bp_harden_el1_vectors) .popsection diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c index 9394f21d7566..6a5eeb8beea3 100644 --- a/arch/arm64/kernel/proton-pack.c +++ b/arch/arm64/kernel/proton-pack.c @@ -770,3 +770,19 @@ int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which) return -ENODEV; } } + +/* Patched to NOP when enabled */ +void noinstr spectre_bhb_patch_loop_mitigation_enable(struct alt_instr *alt, + __le32 *origptr, + __le32 *updptr, int nr_inst) +{ + BUG_ON(nr_inst != 1); +} + +/* Patched to NOP when enabled */ +void noinstr spectre_bhb_patch_fw_mitigation_enabled(struct alt_instr *alt, + __le32 *origptr, + __le32 *updptr, int nr_inst) +{ + BUG_ON(nr_inst != 1); +} diff --git a/include/linux/arm-smccc.h b/include/linux/arm-smccc.h index 63ccb5252190..220c8c60e021 100644 --- a/include/linux/arm-smccc.h +++ b/include/linux/arm-smccc.h @@ -92,6 +92,11 @@ ARM_SMCCC_SMC_32, \ 0, 0x7fff) +#define ARM_SMCCC_ARCH_WORKAROUND_3 \ + ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL, \ + ARM_SMCCC_SMC_32, \ + 0, 0x3fff) + #define ARM_SMCCC_VENDOR_HYP_CALL_UID_FUNC_ID \ ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL, \ ARM_SMCCC_SMC_32, \ -- cgit v1.2.3 From 988f0a9045b0058a43ccee764a671dfab81e6d15 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Wed, 10 Nov 2021 12:59:52 +0200 Subject: soc: fsl: Replace kernel.h with the necessary inclusions When kernel.h is used in the headers it adds a lot into dependency hell, especially when there are circular dependencies are involved. Replace kernel.h inclusion with the list of what is really being used. Signed-off-by: Andy Shevchenko Signed-off-by: Li Yang --- include/soc/fsl/dpaa2-fd.h | 3 ++- include/soc/fsl/qe/immap_qe.h | 3 ++- include/soc/fsl/qe/qe_tdm.h | 4 +++- include/soc/fsl/qe/ucc_fast.h | 2 +- include/soc/fsl/qe/ucc_slow.h | 2 +- 5 files changed, 9 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/soc/fsl/dpaa2-fd.h b/include/soc/fsl/dpaa2-fd.h index 90ae8d191f1a..bae490cac0aa 100644 --- a/include/soc/fsl/dpaa2-fd.h +++ b/include/soc/fsl/dpaa2-fd.h @@ -7,7 +7,8 @@ #ifndef __FSL_DPAA2_FD_H #define __FSL_DPAA2_FD_H -#include +#include +#include /** * DOC: DPAA2 FD - Frame Descriptor APIs for DPAA2 diff --git a/include/soc/fsl/qe/immap_qe.h b/include/soc/fsl/qe/immap_qe.h index 7614fee532f1..edd601f53f5d 100644 --- a/include/soc/fsl/qe/immap_qe.h +++ b/include/soc/fsl/qe/immap_qe.h @@ -13,7 +13,8 @@ #define _ASM_POWERPC_IMMAP_QE_H #ifdef __KERNEL__ -#include +#include + #include #define QE_IMMAP_SIZE (1024 * 1024) /* 1MB from 1MB+IMMR */ diff --git a/include/soc/fsl/qe/qe_tdm.h b/include/soc/fsl/qe/qe_tdm.h index b6febe225071..43ea830cfe1f 100644 --- a/include/soc/fsl/qe/qe_tdm.h +++ b/include/soc/fsl/qe/qe_tdm.h @@ -10,8 +10,8 @@ #ifndef _QE_TDM_H_ #define _QE_TDM_H_ -#include #include +#include #include #include @@ -19,6 +19,8 @@ #include #include +struct device_node; + /* SI RAM entries */ #define SIR_LAST 0x0001 #define SIR_BYTE 0x0002 diff --git a/include/soc/fsl/qe/ucc_fast.h b/include/soc/fsl/qe/ucc_fast.h index 9696a5b9b5d1..ad60b87a3c69 100644 --- a/include/soc/fsl/qe/ucc_fast.h +++ b/include/soc/fsl/qe/ucc_fast.h @@ -10,7 +10,7 @@ #ifndef __UCC_FAST_H__ #define __UCC_FAST_H__ -#include +#include #include #include diff --git a/include/soc/fsl/qe/ucc_slow.h b/include/soc/fsl/qe/ucc_slow.h index 11a216e4e919..7548ce8a202d 100644 --- a/include/soc/fsl/qe/ucc_slow.h +++ b/include/soc/fsl/qe/ucc_slow.h @@ -11,7 +11,7 @@ #ifndef __UCC_SLOW_H__ #define __UCC_SLOW_H__ -#include +#include #include #include -- cgit v1.2.3 From 44a3918c8245ab10c6c9719dd12e7a8d291980d8 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Fri, 18 Feb 2022 11:49:08 -0800 Subject: x86/speculation: Include unprivileged eBPF status in Spectre v2 mitigation reporting With unprivileged eBPF enabled, eIBRS (without retpoline) is vulnerable to Spectre v2 BHB-based attacks. When both are enabled, print a warning message and report it in the 'spectre_v2' sysfs vulnerabilities file. Signed-off-by: Josh Poimboeuf Signed-off-by: Borislav Petkov Reviewed-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 35 +++++++++++++++++++++++++++++------ include/linux/bpf.h | 11 +++++++++++ kernel/sysctl.c | 7 +++++++ 3 files changed, 47 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 79c52dd6c597..0a4267c63d3b 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include @@ -650,6 +651,16 @@ static inline const char *spectre_v2_module_string(void) static inline const char *spectre_v2_module_string(void) { return ""; } #endif +#define SPECTRE_V2_EIBRS_EBPF_MSG "WARNING: Unprivileged eBPF is enabled with eIBRS on, data leaks possible via Spectre v2 BHB attacks!\n" + +#ifdef CONFIG_BPF_SYSCALL +void unpriv_ebpf_notify(int new_state) +{ + if (spectre_v2_enabled == SPECTRE_V2_EIBRS && !new_state) + pr_err(SPECTRE_V2_EIBRS_EBPF_MSG); +} +#endif + static inline bool match_option(const char *arg, int arglen, const char *opt) { int len = strlen(opt); @@ -994,6 +1005,9 @@ static void __init spectre_v2_select_mitigation(void) break; } + if (mode == SPECTRE_V2_EIBRS && unprivileged_ebpf_enabled()) + pr_err(SPECTRE_V2_EIBRS_EBPF_MSG); + if (spectre_v2_in_eibrs_mode(mode)) { /* Force it so VMEXIT will restore correctly */ x86_spec_ctrl_base |= SPEC_CTRL_IBRS; @@ -1780,6 +1794,20 @@ static char *ibpb_state(void) return ""; } +static ssize_t spectre_v2_show_state(char *buf) +{ + if (spectre_v2_enabled == SPECTRE_V2_EIBRS && unprivileged_ebpf_enabled()) + return sprintf(buf, "Vulnerable: Unprivileged eBPF enabled\n"); + + return sprintf(buf, "%s%s%s%s%s%s\n", + spectre_v2_strings[spectre_v2_enabled], + ibpb_state(), + boot_cpu_has(X86_FEATURE_USE_IBRS_FW) ? ", IBRS_FW" : "", + stibp_state(), + boot_cpu_has(X86_FEATURE_RSB_CTXSW) ? ", RSB filling" : "", + spectre_v2_module_string()); +} + static ssize_t srbds_show_state(char *buf) { return sprintf(buf, "%s\n", srbds_strings[srbds_mitigation]); @@ -1805,12 +1833,7 @@ static ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr return sprintf(buf, "%s\n", spectre_v1_strings[spectre_v1_mitigation]); case X86_BUG_SPECTRE_V2: - return sprintf(buf, "%s%s%s%s%s%s\n", spectre_v2_strings[spectre_v2_enabled], - ibpb_state(), - boot_cpu_has(X86_FEATURE_USE_IBRS_FW) ? ", IBRS_FW" : "", - stibp_state(), - boot_cpu_has(X86_FEATURE_RSB_CTXSW) ? ", RSB filling" : "", - spectre_v2_module_string()); + return spectre_v2_show_state(buf); case X86_BUG_SPEC_STORE_BYPASS: return sprintf(buf, "%s\n", ssb_strings[ssb_mode]); diff --git a/include/linux/bpf.h b/include/linux/bpf.h index fa517ae604ad..1f56806d8eb9 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -1793,6 +1793,11 @@ struct bpf_core_ctx { int bpf_core_apply(struct bpf_core_ctx *ctx, const struct bpf_core_relo *relo, int relo_idx, void *insn); +static inline bool unprivileged_ebpf_enabled(void) +{ + return !sysctl_unprivileged_bpf_disabled; +} + #else /* !CONFIG_BPF_SYSCALL */ static inline struct bpf_prog *bpf_prog_get(u32 ufd) { @@ -2012,6 +2017,12 @@ bpf_jit_find_kfunc_model(const struct bpf_prog *prog, { return NULL; } + +static inline bool unprivileged_ebpf_enabled(void) +{ + return false; +} + #endif /* CONFIG_BPF_SYSCALL */ void __bpf_free_used_btfs(struct bpf_prog_aux *aux, diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 5ae443b2882e..730ab56d9e92 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -180,6 +180,10 @@ static int bpf_stats_handler(struct ctl_table *table, int write, return ret; } +void __weak unpriv_ebpf_notify(int new_state) +{ +} + static int bpf_unpriv_handler(struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { @@ -197,6 +201,9 @@ static int bpf_unpriv_handler(struct ctl_table *table, int write, return -EPERM; *(int *)table->data = unpriv_enable; } + + unpriv_ebpf_notify(unpriv_enable); + return ret; } #endif /* CONFIG_BPF_SYSCALL && CONFIG_SYSCTL */ -- cgit v1.2.3 From c086df4902573e2f06c6a2a83452c13a8bc603f5 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Mon, 10 Jan 2022 18:52:52 -0500 Subject: fuse: move FUSE_SUPER_MAGIC definition to magic.h ...to help userland apps that need to identify FUSE mounts. Signed-off-by: Jeff Layton Signed-off-by: Miklos Szeredi --- fs/fuse/inode.c | 3 +-- include/uapi/linux/magic.h | 1 + 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index ee846ce371d8..9ee36aa73251 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -23,6 +23,7 @@ #include #include #include +#include MODULE_AUTHOR("Miklos Szeredi "); MODULE_DESCRIPTION("Filesystem in Userspace"); @@ -50,8 +51,6 @@ MODULE_PARM_DESC(max_user_congthresh, "Global limit for the maximum congestion threshold an " "unprivileged user can set"); -#define FUSE_SUPER_MAGIC 0x65735546 - #define FUSE_DEFAULT_BLKSIZE 512 /** Maximum number of outstanding background requests */ diff --git a/include/uapi/linux/magic.h b/include/uapi/linux/magic.h index 0425cd79af9a..f724129c0425 100644 --- a/include/uapi/linux/magic.h +++ b/include/uapi/linux/magic.h @@ -36,6 +36,7 @@ #define EFIVARFS_MAGIC 0xde5e81e4 #define HOSTFS_SUPER_MAGIC 0x00c0ffee #define OVERLAYFS_SUPER_MAGIC 0x794c7630 +#define FUSE_SUPER_MAGIC 0x65735546 #define MINIX_SUPER_MAGIC 0x137F /* minix v1 fs, 14 char names */ #define MINIX_SUPER_MAGIC2 0x138F /* minix v1 fs, 30 char names */ -- cgit v1.2.3 From 29fb608396d6a62c1b85acc421ad7a4399085b9f Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 14 Feb 2022 17:59:38 -0800 Subject: Bluetooth: Fix bt_skb_sendmmsg not allocating partial chunks Since bt_skb_sendmmsg can be used with the likes of SOCK_STREAM it shall return the partial chunks it could allocate instead of freeing everything as otherwise it can cause problems like bellow. Fixes: 81be03e026dc ("Bluetooth: RFCOMM: Replace use of memcpy_from_msg with bt_skb_sendmmsg") Reported-by: Paul Menzel Link: https://lore.kernel.org/r/d7206e12-1b99-c3be-84f4-df22af427ef5@molgen.mpg.de BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=215594 Signed-off-by: Luiz Augusto von Dentz Tested-by: Paul Menzel (Nokia N9 (MeeGo/Harmattan) Signed-off-by: Marcel Holtmann --- include/net/bluetooth/bluetooth.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h index 4b3d0b16c185..a647e5fabdbd 100644 --- a/include/net/bluetooth/bluetooth.h +++ b/include/net/bluetooth/bluetooth.h @@ -506,8 +506,7 @@ static inline struct sk_buff *bt_skb_sendmmsg(struct sock *sk, tmp = bt_skb_sendmsg(sk, msg, len, mtu, headroom, tailroom); if (IS_ERR(tmp)) { - kfree_skb(skb); - return tmp; + return skb; } len -= tmp->len; -- cgit v1.2.3 From a56a1138cbd85e4d565356199d60e1cb94e5a77a Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Thu, 17 Feb 2022 13:10:38 -0800 Subject: Bluetooth: hci_sync: Fix not using conn_timeout When using hci_le_create_conn_sync it shall wait for the conn_timeout since the connection complete may take longer than just 2 seconds. Also fix the masking of HCI_EV_LE_ENHANCED_CONN_COMPLETE and HCI_EV_LE_CONN_COMPLETE so they are never both set so we can predict which one the controller will use in case of HCI_OP_LE_CREATE_CONN. Fixes: 6cd29ec6ae5e3 ("Bluetooth: hci_sync: Wait for proper events when connecting LE") Signed-off-by: Luiz Augusto von Dentz Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci_core.h | 8 ++++++++ net/bluetooth/hci_sync.c | 21 +++++++++++++++------ 2 files changed, 23 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 586f69d084a2..e336e9c1dda4 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1489,6 +1489,14 @@ void hci_conn_del_sysfs(struct hci_conn *conn); /* Extended advertising support */ #define ext_adv_capable(dev) (((dev)->le_features[1] & HCI_LE_EXT_ADV)) +/* BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E page 1789: + * + * C24: Mandatory if the LE Controller supports Connection State and either + * LE Feature (LL Privacy) or LE Feature (Extended Advertising) is supported + */ +#define use_enhanced_conn_complete(dev) (ll_privacy_capable(dev) || \ + ext_adv_capable(dev)) + /* ----- HCI protocols ----- */ #define HCI_PROTO_DEFER 0x01 diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 9d8490570b42..9ba2a1a7d481 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -3265,10 +3265,10 @@ static int hci_le_set_event_mask_sync(struct hci_dev *hdev) if (hdev->le_features[0] & HCI_LE_DATA_LEN_EXT) events[0] |= 0x40; /* LE Data Length Change */ - /* If the controller supports LL Privacy feature, enable - * the corresponding event. + /* If the controller supports LL Privacy feature or LE Extended Adv, + * enable the corresponding event. */ - if (hdev->le_features[0] & HCI_LE_LL_PRIVACY) + if (use_enhanced_conn_complete(hdev)) events[1] |= 0x02; /* LE Enhanced Connection Complete */ /* If the controller supports Extended Scanner Filter @@ -5188,7 +5188,7 @@ int hci_le_ext_create_conn_sync(struct hci_dev *hdev, struct hci_conn *conn, return __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_EXT_CREATE_CONN, plen, data, HCI_EV_LE_ENHANCED_CONN_COMPLETE, - HCI_CMD_TIMEOUT, NULL); + conn->conn_timeout, NULL); } int hci_le_create_conn_sync(struct hci_dev *hdev, struct hci_conn *conn) @@ -5273,9 +5273,18 @@ int hci_le_create_conn_sync(struct hci_dev *hdev, struct hci_conn *conn) cp.min_ce_len = cpu_to_le16(0x0000); cp.max_ce_len = cpu_to_le16(0x0000); + /* BLUETOOTH CORE SPECIFICATION Version 5.3 | Vol 4, Part E page 2261: + * + * If this event is unmasked and the HCI_LE_Connection_Complete event + * is unmasked, only the HCI_LE_Enhanced_Connection_Complete event is + * sent when a new connection has been created. + */ err = __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_CREATE_CONN, - sizeof(cp), &cp, HCI_EV_LE_CONN_COMPLETE, - HCI_CMD_TIMEOUT, NULL); + sizeof(cp), &cp, + use_enhanced_conn_complete(hdev) ? + HCI_EV_LE_ENHANCED_CONN_COMPLETE : + HCI_EV_LE_CONN_COMPLETE, + conn->conn_timeout, NULL); done: /* Re-enable advertising after the connection attempt is finished. */ -- cgit v1.2.3 From 17a8f31bba7bac8cce4bd12bab50697da96e7710 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 28 Feb 2022 04:18:05 +0100 Subject: netfilter: egress: silence egress hook lockdep splats Netfilter assumes its called with rcu_read_lock held, but in egress hook case it may be called with BH readlock. This triggers lockdep splat. In order to avoid to change all rcu_dereference() to rcu_dereference_check(..., rcu_read_lock_bh_held()), wrap nf_hook_slow with read lock/unlock pair. Reported-by: Eric Dumazet Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter_netdev.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/linux/netfilter_netdev.h b/include/linux/netfilter_netdev.h index b4dd96e4dc8d..e6487a691136 100644 --- a/include/linux/netfilter_netdev.h +++ b/include/linux/netfilter_netdev.h @@ -101,7 +101,11 @@ static inline struct sk_buff *nf_hook_egress(struct sk_buff *skb, int *rc, nf_hook_state_init(&state, NF_NETDEV_EGRESS, NFPROTO_NETDEV, dev, NULL, NULL, dev_net(dev), NULL); + + /* nf assumes rcu_read_lock, not just read_lock_bh */ + rcu_read_lock(); ret = nf_hook_slow(skb, &state, e, 0); + rcu_read_unlock(); if (ret == 1) { return skb; -- cgit v1.2.3 From 50bb467c9e76743fbc8441d29113cdad62dbc4fe Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 18 Feb 2022 09:38:58 +0000 Subject: rfkill: define rfill_soft_blocked() if !RFKILL If CONFIG_RFKILL is not set, the Intel WiFi driver will not build the iw_mvm driver part due to the missing rfill_soft_blocked() call. Adding a inline declaration of rfill_soft_blocked() if CONFIG_RFKILL=n fixes the following error: drivers/net/wireless/intel/iwlwifi/mvm/mvm.h: In function 'iwl_mvm_mei_set_sw_rfkill_state': drivers/net/wireless/intel/iwlwifi/mvm/mvm.h:2215:38: error: implicit declaration of function 'rfkill_soft_blocked'; did you mean 'rfkill_blocked'? [-Werror=implicit-function-declaration] 2215 | mvm->hw_registered ? rfkill_soft_blocked(mvm->hw->wiphy->rfkill) : false; | ^~~~~~~~~~~~~~~~~~~ | rfkill_blocked Signed-off-by: Ben Dooks Reported-by: Neill Whillans Fixes: 5bc9a9dd7535 ("rfkill: allow to get the software rfkill state") Link: https://lore.kernel.org/r/20220218093858.1245677-1-ben.dooks@codethink.co.uk Signed-off-by: Johannes Berg --- include/linux/rfkill.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h index c35f3962dc4f..373003ace639 100644 --- a/include/linux/rfkill.h +++ b/include/linux/rfkill.h @@ -308,6 +308,11 @@ static inline bool rfkill_blocked(struct rfkill *rfkill) return false; } +static inline bool rfkill_soft_blocked(struct rfkill *rfkill) +{ + return false; +} + static inline enum rfkill_type rfkill_find_type(const char *name) { return RFKILL_TYPE_ALL; -- cgit v1.2.3 From c3873070247d9e3c7a6b0cf9bf9b45e8018427b1 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 28 Feb 2022 06:22:22 +0100 Subject: netfilter: nf_queue: fix possible use-after-free Eric Dumazet says: The sock_hold() side seems suspect, because there is no guarantee that sk_refcnt is not already 0. On failure, we cannot queue the packet and need to indicate an error. The packet will be dropped by the caller. v2: split skb prefetch hunk into separate change Fixes: 271b72c7fa82c ("udp: RCU handling for Unicast packets.") Reported-by: Eric Dumazet Reviewed-by: Eric Dumazet Signed-off-by: Florian Westphal --- include/net/netfilter/nf_queue.h | 2 +- net/netfilter/nf_queue.c | 13 +++++++++---- net/netfilter/nfnetlink_queue.c | 12 +++++++++--- 3 files changed, 19 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_queue.h b/include/net/netfilter/nf_queue.h index 9eed51e920e8..980daa6e1e3a 100644 --- a/include/net/netfilter/nf_queue.h +++ b/include/net/netfilter/nf_queue.h @@ -37,7 +37,7 @@ void nf_register_queue_handler(const struct nf_queue_handler *qh); void nf_unregister_queue_handler(void); void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict); -void nf_queue_entry_get_refs(struct nf_queue_entry *entry); +bool nf_queue_entry_get_refs(struct nf_queue_entry *entry); void nf_queue_entry_free(struct nf_queue_entry *entry); static inline void init_hashrandom(u32 *jhash_initval) diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index 5ab0680db445..e39549c55945 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -96,19 +96,21 @@ static void __nf_queue_entry_init_physdevs(struct nf_queue_entry *entry) } /* Bump dev refs so they don't vanish while packet is out */ -void nf_queue_entry_get_refs(struct nf_queue_entry *entry) +bool nf_queue_entry_get_refs(struct nf_queue_entry *entry) { struct nf_hook_state *state = &entry->state; + if (state->sk && !refcount_inc_not_zero(&state->sk->sk_refcnt)) + return false; + dev_hold(state->in); dev_hold(state->out); - if (state->sk) - sock_hold(state->sk); #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) dev_hold(entry->physin); dev_hold(entry->physout); #endif + return true; } EXPORT_SYMBOL_GPL(nf_queue_entry_get_refs); @@ -196,7 +198,10 @@ static int __nf_queue(struct sk_buff *skb, const struct nf_hook_state *state, __nf_queue_entry_init_physdevs(entry); - nf_queue_entry_get_refs(entry); + if (!nf_queue_entry_get_refs(entry)) { + kfree(entry); + return -ENOTCONN; + } switch (entry->state.pf) { case AF_INET: diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index ea2d9c2a44cf..64a6acb6aeae 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -710,9 +710,15 @@ static struct nf_queue_entry * nf_queue_entry_dup(struct nf_queue_entry *e) { struct nf_queue_entry *entry = kmemdup(e, e->size, GFP_ATOMIC); - if (entry) - nf_queue_entry_get_refs(entry); - return entry; + + if (!entry) + return NULL; + + if (nf_queue_entry_get_refs(entry)) + return entry; + + kfree(entry); + return NULL; } #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) -- cgit v1.2.3 From db6140e5e35a48405e669353bd54042c1d4c3841 Mon Sep 17 00:00:00 2001 From: Paul Blakey Date: Mon, 28 Feb 2022 11:23:49 +0200 Subject: net/sched: act_ct: Fix flow table lookup failure with no originating ifindex After cited commit optimizted hw insertion, flow table entries are populated with ifindex information which was intended to only be used for HW offload. This tuple ifindex is hashed in the flow table key, so it must be filled for lookup to be successful. But tuple ifindex is only relevant for the netfilter flowtables (nft), so it's not filled in act_ct flow table lookup, resulting in lookup failure, and no SW offload and no offload teardown for TCP connection FIN/RST packets. To fix this, add new tc ifindex field to tuple, which will only be used for offloading, not for lookup, as it will not be part of the tuple hash. Fixes: 9795ded7f924 ("net/sched: act_ct: Fill offloading tuple iifidx") Signed-off-by: Paul Blakey Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_flow_table.h | 6 +++++- net/netfilter/nf_flow_table_offload.c | 6 +++++- net/sched/act_ct.c | 13 +++++++++---- 3 files changed, 19 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h index a3647fadf1cc..bd59e950f4d6 100644 --- a/include/net/netfilter/nf_flow_table.h +++ b/include/net/netfilter/nf_flow_table.h @@ -96,6 +96,7 @@ enum flow_offload_xmit_type { FLOW_OFFLOAD_XMIT_NEIGH, FLOW_OFFLOAD_XMIT_XFRM, FLOW_OFFLOAD_XMIT_DIRECT, + FLOW_OFFLOAD_XMIT_TC, }; #define NF_FLOW_TABLE_ENCAP_MAX 2 @@ -127,7 +128,7 @@ struct flow_offload_tuple { struct { } __hash; u8 dir:2, - xmit_type:2, + xmit_type:3, encap_num:2, in_vlan_ingress:2; u16 mtu; @@ -142,6 +143,9 @@ struct flow_offload_tuple { u8 h_source[ETH_ALEN]; u8 h_dest[ETH_ALEN]; } out; + struct { + u32 iifidx; + } tc; }; }; diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c index b561e0a44a45..fc4265acd9c4 100644 --- a/net/netfilter/nf_flow_table_offload.c +++ b/net/netfilter/nf_flow_table_offload.c @@ -110,7 +110,11 @@ static int nf_flow_rule_match(struct nf_flow_match *match, nf_flow_rule_lwt_match(match, tun_info); } - key->meta.ingress_ifindex = tuple->iifidx; + if (tuple->xmit_type == FLOW_OFFLOAD_XMIT_TC) + key->meta.ingress_ifindex = tuple->tc.iifidx; + else + key->meta.ingress_ifindex = tuple->iifidx; + mask->meta.ingress_ifindex = 0xffffffff; if (tuple->encap_num > 0 && !(tuple->in_vlan_ingress & BIT(0)) && diff --git a/net/sched/act_ct.c b/net/sched/act_ct.c index 33e70d60f0bf..ec19f625863a 100644 --- a/net/sched/act_ct.c +++ b/net/sched/act_ct.c @@ -361,6 +361,13 @@ static void tcf_ct_flow_table_put(struct tcf_ct_params *params) } } +static void tcf_ct_flow_tc_ifidx(struct flow_offload *entry, + struct nf_conn_act_ct_ext *act_ct_ext, u8 dir) +{ + entry->tuplehash[dir].tuple.xmit_type = FLOW_OFFLOAD_XMIT_TC; + entry->tuplehash[dir].tuple.tc.iifidx = act_ct_ext->ifindex[dir]; +} + static void tcf_ct_flow_table_add(struct tcf_ct_flow_table *ct_ft, struct nf_conn *ct, bool tcp) @@ -385,10 +392,8 @@ static void tcf_ct_flow_table_add(struct tcf_ct_flow_table *ct_ft, act_ct_ext = nf_conn_act_ct_ext_find(ct); if (act_ct_ext) { - entry->tuplehash[FLOW_OFFLOAD_DIR_ORIGINAL].tuple.iifidx = - act_ct_ext->ifindex[IP_CT_DIR_ORIGINAL]; - entry->tuplehash[FLOW_OFFLOAD_DIR_REPLY].tuple.iifidx = - act_ct_ext->ifindex[IP_CT_DIR_REPLY]; + tcf_ct_flow_tc_ifidx(entry, act_ct_ext, FLOW_OFFLOAD_DIR_ORIGINAL); + tcf_ct_flow_tc_ifidx(entry, act_ct_ext, FLOW_OFFLOAD_DIR_REPLY); } err = flow_offload_add(&ct_ft->nf_ft, entry); -- cgit v1.2.3 From 2d3916f3189172d5c69d33065c3c21119fe539fc Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 3 Mar 2022 09:37:28 -0800 Subject: ipv6: fix skb drops in igmp6_event_query() and igmp6_event_report() While investigating on why a synchronize_net() has been added recently in ipv6_mc_down(), I found that igmp6_event_query() and igmp6_event_report() might drop skbs in some cases. Discussion about removing synchronize_net() from ipv6_mc_down() will happen in a different thread. Fixes: f185de28d9ae ("mld: add new workqueues for process mld events") Signed-off-by: Eric Dumazet Cc: Taehee Yoo Cc: Cong Wang Cc: David Ahern Link: https://lore.kernel.org/r/20220303173728.937869-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski --- include/net/ndisc.h | 4 ++-- net/ipv6/mcast.c | 32 ++++++++++++-------------------- 2 files changed, 14 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/include/net/ndisc.h b/include/net/ndisc.h index 53cb8de0e589..47ffb360ddfa 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -475,9 +475,9 @@ int igmp6_late_init(void); void igmp6_cleanup(void); void igmp6_late_cleanup(void); -int igmp6_event_query(struct sk_buff *skb); +void igmp6_event_query(struct sk_buff *skb); -int igmp6_event_report(struct sk_buff *skb); +void igmp6_event_report(struct sk_buff *skb); #ifdef CONFIG_SYSCTL diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index a8861db52c18..909f937befd7 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1371,27 +1371,23 @@ static void mld_process_v2(struct inet6_dev *idev, struct mld2_query *mld, } /* called with rcu_read_lock() */ -int igmp6_event_query(struct sk_buff *skb) +void igmp6_event_query(struct sk_buff *skb) { struct inet6_dev *idev = __in6_dev_get(skb->dev); - if (!idev) - return -EINVAL; - - if (idev->dead) { - kfree_skb(skb); - return -ENODEV; - } + if (!idev || idev->dead) + goto out; spin_lock_bh(&idev->mc_query_lock); if (skb_queue_len(&idev->mc_query_queue) < MLD_MAX_SKBS) { __skb_queue_tail(&idev->mc_query_queue, skb); if (!mod_delayed_work(mld_wq, &idev->mc_query_work, 0)) in6_dev_hold(idev); + skb = NULL; } spin_unlock_bh(&idev->mc_query_lock); - - return 0; +out: + kfree_skb(skb); } static void __mld_query_work(struct sk_buff *skb) @@ -1542,27 +1538,23 @@ static void mld_query_work(struct work_struct *work) } /* called with rcu_read_lock() */ -int igmp6_event_report(struct sk_buff *skb) +void igmp6_event_report(struct sk_buff *skb) { struct inet6_dev *idev = __in6_dev_get(skb->dev); - if (!idev) - return -EINVAL; - - if (idev->dead) { - kfree_skb(skb); - return -ENODEV; - } + if (!idev || idev->dead) + goto out; spin_lock_bh(&idev->mc_report_lock); if (skb_queue_len(&idev->mc_report_queue) < MLD_MAX_SKBS) { __skb_queue_tail(&idev->mc_report_queue, skb); if (!mod_delayed_work(mld_wq, &idev->mc_report_work, 0)) in6_dev_hold(idev); + skb = NULL; } spin_unlock_bh(&idev->mc_report_lock); - - return 0; +out: + kfree_skb(skb); } static void __mld_report_work(struct sk_buff *skb) -- cgit v1.2.3 From bfa26ba343c727e055223be04e08f2ebdd43c293 Mon Sep 17 00:00:00 2001 From: William Mahon Date: Thu, 3 Mar 2022 18:23:42 -0800 Subject: HID: add mapping for KEY_DICTATE Numerous keyboards are adding dictate keys which allows for text messages to be dictated by a microphone. This patch adds a new key definition KEY_DICTATE and maps 0x0c/0x0d8 usage code to this new keycode. Additionally hid-debug is adjusted to recognize this new usage code as well. Signed-off-by: William Mahon Acked-by: Benjamin Tissoires Link: https://lore.kernel.org/r/20220303021501.1.I5dbf50eb1a7a6734ee727bda4a8573358c6d3ec0@changeid Signed-off-by: Dmitry Torokhov --- drivers/hid/hid-debug.c | 1 + drivers/hid/hid-input.c | 1 + include/uapi/linux/input-event-codes.h | 1 + 3 files changed, 3 insertions(+) (limited to 'include') diff --git a/drivers/hid/hid-debug.c b/drivers/hid/hid-debug.c index 26c31d759914..8aa68416b1d7 100644 --- a/drivers/hid/hid-debug.c +++ b/drivers/hid/hid-debug.c @@ -969,6 +969,7 @@ static const char *keys[KEY_MAX + 1] = { [KEY_ASSISTANT] = "Assistant", [KEY_KBD_LAYOUT_NEXT] = "KbdLayoutNext", [KEY_EMOJI_PICKER] = "EmojiPicker", + [KEY_DICTATE] = "Dictate", [KEY_BRIGHTNESS_MIN] = "BrightnessMin", [KEY_BRIGHTNESS_MAX] = "BrightnessMax", [KEY_BRIGHTNESS_AUTO] = "BrightnessAuto", diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c index 112901d2d8d2..ce2b75a67cb8 100644 --- a/drivers/hid/hid-input.c +++ b/drivers/hid/hid-input.c @@ -992,6 +992,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel case 0x0cd: map_key_clear(KEY_PLAYPAUSE); break; case 0x0cf: map_key_clear(KEY_VOICECOMMAND); break; + case 0x0d8: map_key_clear(KEY_DICTATE); break; case 0x0d9: map_key_clear(KEY_EMOJI_PICKER); break; case 0x0e0: map_abs_clear(ABS_VOLUME); break; diff --git a/include/uapi/linux/input-event-codes.h b/include/uapi/linux/input-event-codes.h index 225ec87d4f22..4db5d41848e4 100644 --- a/include/uapi/linux/input-event-codes.h +++ b/include/uapi/linux/input-event-codes.h @@ -612,6 +612,7 @@ #define KEY_ASSISTANT 0x247 /* AL Context-aware desktop assistant */ #define KEY_KBD_LAYOUT_NEXT 0x248 /* AC Next Keyboard Layout Select */ #define KEY_EMOJI_PICKER 0x249 /* Show/hide emoji picker (HUTRR101) */ +#define KEY_DICTATE 0x24a /* Start or Stop Voice Dictation Session (HUTRR99) */ #define KEY_BRIGHTNESS_MIN 0x250 /* Set Brightness to Minimum */ #define KEY_BRIGHTNESS_MAX 0x251 /* Set Brightness to Maximum */ -- cgit v1.2.3 From 327b89f0acc4c20a06ed59e4d9af7f6d804dc2e2 Mon Sep 17 00:00:00 2001 From: William Mahon Date: Thu, 3 Mar 2022 18:26:22 -0800 Subject: HID: add mapping for KEY_ALL_APPLICATIONS This patch adds a new key definition for KEY_ALL_APPLICATIONS and aliases KEY_DASHBOARD to it. It also maps the 0x0c/0x2a2 usage code to KEY_ALL_APPLICATIONS. Signed-off-by: William Mahon Acked-by: Benjamin Tissoires Link: https://lore.kernel.org/r/20220303035618.1.I3a7746ad05d270161a18334ae06e3b6db1a1d339@changeid Signed-off-by: Dmitry Torokhov --- drivers/hid/hid-debug.c | 4 +++- drivers/hid/hid-input.c | 2 ++ include/uapi/linux/input-event-codes.h | 3 ++- 3 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/hid/hid-debug.c b/drivers/hid/hid-debug.c index 8aa68416b1d7..81e7e404a5fc 100644 --- a/drivers/hid/hid-debug.c +++ b/drivers/hid/hid-debug.c @@ -860,7 +860,9 @@ static const char *keys[KEY_MAX + 1] = { [KEY_F22] = "F22", [KEY_F23] = "F23", [KEY_F24] = "F24", [KEY_PLAYCD] = "PlayCD", [KEY_PAUSECD] = "PauseCD", [KEY_PROG3] = "Prog3", - [KEY_PROG4] = "Prog4", [KEY_SUSPEND] = "Suspend", + [KEY_PROG4] = "Prog4", + [KEY_ALL_APPLICATIONS] = "AllApplications", + [KEY_SUSPEND] = "Suspend", [KEY_CLOSE] = "Close", [KEY_PLAY] = "Play", [KEY_FASTFORWARD] = "FastForward", [KEY_BASSBOOST] = "BassBoost", [KEY_PRINT] = "Print", [KEY_HP] = "HP", diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c index ce2b75a67cb8..56ec27398a00 100644 --- a/drivers/hid/hid-input.c +++ b/drivers/hid/hid-input.c @@ -1084,6 +1084,8 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel case 0x29d: map_key_clear(KEY_KBD_LAYOUT_NEXT); break; + case 0x2a2: map_key_clear(KEY_ALL_APPLICATIONS); break; + case 0x2c7: map_key_clear(KEY_KBDINPUTASSIST_PREV); break; case 0x2c8: map_key_clear(KEY_KBDINPUTASSIST_NEXT); break; case 0x2c9: map_key_clear(KEY_KBDINPUTASSIST_PREVGROUP); break; diff --git a/include/uapi/linux/input-event-codes.h b/include/uapi/linux/input-event-codes.h index 4db5d41848e4..7989d9483ea7 100644 --- a/include/uapi/linux/input-event-codes.h +++ b/include/uapi/linux/input-event-codes.h @@ -278,7 +278,8 @@ #define KEY_PAUSECD 201 #define KEY_PROG3 202 #define KEY_PROG4 203 -#define KEY_DASHBOARD 204 /* AL Dashboard */ +#define KEY_ALL_APPLICATIONS 204 /* AC Desktop Show All Applications */ +#define KEY_DASHBOARD KEY_ALL_APPLICATIONS #define KEY_SUSPEND 205 #define KEY_CLOSE 206 /* AC Close */ #define KEY_PLAY 207 -- cgit v1.2.3 From 838d6d3461db0fdbf33fc5f8a69c27b50b4a46da Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Fri, 14 Jan 2022 14:56:15 -0500 Subject: virtio: unexport virtio_finalize_features virtio_finalize_features is only used internally within virtio. No reason to export it. Signed-off-by: Michael S. Tsirkin Reviewed-by: Cornelia Huck Acked-by: Jason Wang --- drivers/virtio/virtio.c | 3 +-- include/linux/virtio.h | 1 - 2 files changed, 1 insertion(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c index 00ac9db792a4..d891b0a354b0 100644 --- a/drivers/virtio/virtio.c +++ b/drivers/virtio/virtio.c @@ -166,7 +166,7 @@ void virtio_add_status(struct virtio_device *dev, unsigned int status) } EXPORT_SYMBOL_GPL(virtio_add_status); -int virtio_finalize_features(struct virtio_device *dev) +static int virtio_finalize_features(struct virtio_device *dev) { int ret = dev->config->finalize_features(dev); unsigned status; @@ -202,7 +202,6 @@ int virtio_finalize_features(struct virtio_device *dev) } return 0; } -EXPORT_SYMBOL_GPL(virtio_finalize_features); void virtio_reset_device(struct virtio_device *dev) { diff --git a/include/linux/virtio.h b/include/linux/virtio.h index 72292a62cd90..5464f398912a 100644 --- a/include/linux/virtio.h +++ b/include/linux/virtio.h @@ -133,7 +133,6 @@ bool is_virtio_device(struct device *dev); void virtio_break_device(struct virtio_device *dev); void virtio_config_changed(struct virtio_device *dev); -int virtio_finalize_features(struct virtio_device *dev); #ifdef CONFIG_PM_SLEEP int virtio_device_freeze(struct virtio_device *dev); int virtio_device_restore(struct virtio_device *dev); -- cgit v1.2.3 From 4fa59ede95195f267101a1b8916992cf3f245cdb Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Fri, 14 Jan 2022 14:58:41 -0500 Subject: virtio: acknowledge all features before access The feature negotiation was designed in a way that makes it possible for devices to know which config fields will be accessed by drivers. This is broken since commit 404123c2db79 ("virtio: allow drivers to validate features") with fallout in at least block and net. We have a partial work-around in commit 2f9a174f918e ("virtio: write back F_VERSION_1 before validate") which at least lets devices find out which format should config space have, but this is a partial fix: guests should not access config space without acknowledging features since otherwise we'll never be able to change the config space format. To fix, split finalize_features from virtio_finalize_features and call finalize_features with all feature bits before validation, and then - if validation changed any bits - once again after. Since virtio_finalize_features no longer writes out features rename it to virtio_features_ok - since that is what it does: checks that features are ok with the device. As a side effect, this also reduces the amount of hypervisor accesses - we now only acknowledge features once unless we are clearing any features when validating (which is uncommon). IRC I think that this was more or less always the intent in the spec but unfortunately the way the spec is worded does not say this explicitly, I plan to address this at the spec level, too. Acked-by: Jason Wang Cc: stable@vger.kernel.org Fixes: 404123c2db79 ("virtio: allow drivers to validate features") Fixes: 2f9a174f918e ("virtio: write back F_VERSION_1 before validate") Cc: "Halil Pasic" Signed-off-by: Michael S. Tsirkin --- drivers/virtio/virtio.c | 39 ++++++++++++++++++++++----------------- include/linux/virtio_config.h | 3 ++- 2 files changed, 24 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c index d891b0a354b0..d6396be0ea83 100644 --- a/drivers/virtio/virtio.c +++ b/drivers/virtio/virtio.c @@ -166,14 +166,13 @@ void virtio_add_status(struct virtio_device *dev, unsigned int status) } EXPORT_SYMBOL_GPL(virtio_add_status); -static int virtio_finalize_features(struct virtio_device *dev) +/* Do some validation, then set FEATURES_OK */ +static int virtio_features_ok(struct virtio_device *dev) { - int ret = dev->config->finalize_features(dev); unsigned status; + int ret; might_sleep(); - if (ret) - return ret; ret = arch_has_restricted_virtio_memory_access(); if (ret) { @@ -244,17 +243,6 @@ static int virtio_dev_probe(struct device *_d) driver_features_legacy = driver_features; } - /* - * Some devices detect legacy solely via F_VERSION_1. Write - * F_VERSION_1 to force LE config space accesses before FEATURES_OK for - * these when needed. - */ - if (drv->validate && !virtio_legacy_is_little_endian() - && device_features & BIT_ULL(VIRTIO_F_VERSION_1)) { - dev->features = BIT_ULL(VIRTIO_F_VERSION_1); - dev->config->finalize_features(dev); - } - if (device_features & (1ULL << VIRTIO_F_VERSION_1)) dev->features = driver_features & device_features; else @@ -265,13 +253,26 @@ static int virtio_dev_probe(struct device *_d) if (device_features & (1ULL << i)) __virtio_set_bit(dev, i); + err = dev->config->finalize_features(dev); + if (err) + goto err; + if (drv->validate) { + u64 features = dev->features; + err = drv->validate(dev); if (err) goto err; + + /* Did validation change any features? Then write them again. */ + if (features != dev->features) { + err = dev->config->finalize_features(dev); + if (err) + goto err; + } } - err = virtio_finalize_features(dev); + err = virtio_features_ok(dev); if (err) goto err; @@ -495,7 +496,11 @@ int virtio_device_restore(struct virtio_device *dev) /* We have a driver! */ virtio_add_status(dev, VIRTIO_CONFIG_S_DRIVER); - ret = virtio_finalize_features(dev); + ret = dev->config->finalize_features(dev); + if (ret) + goto err; + + ret = virtio_features_ok(dev); if (ret) goto err; diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h index 4d107ad31149..dafdc7f48c01 100644 --- a/include/linux/virtio_config.h +++ b/include/linux/virtio_config.h @@ -64,8 +64,9 @@ struct virtio_shm_region { * Returns the first 64 feature bits (all we currently need). * @finalize_features: confirm what device features we'll be using. * vdev: the virtio_device - * This gives the final feature bits for the device: it can change + * This sends the driver feature bits to the device: it can change * the dev->feature bits if it wants. + * Note: despite the name this can be called any number of times. * Returns 0 on success or error status * @bus_name: return the bus name associated with the device (optional) * vdev: the virtio_device -- cgit v1.2.3 From e0077cc13b831f8fad5557442f73bf7728683713 Mon Sep 17 00:00:00 2001 From: Si-Wei Liu Date: Fri, 14 Jan 2022 19:27:59 -0500 Subject: vdpa: factor out vdpa_set_features_unlocked for vdpa internal use No functional change introduced. vdpa bus driver such as virtio_vdpa or vhost_vdpa is not supposed to take care of the locking for core by its own. The locked API vdpa_set_features should suffice the bus driver's need. Signed-off-by: Si-Wei Liu Reviewed-by: Eli Cohen Link: https://lore.kernel.org/r/1642206481-30721-2-git-send-email-si-wei.liu@oracle.com Signed-off-by: Michael S. Tsirkin Acked-by: Jason Wang --- drivers/vdpa/vdpa.c | 2 +- drivers/vhost/vdpa.c | 2 +- drivers/virtio/virtio_vdpa.c | 2 +- include/linux/vdpa.h | 18 ++++++++++++------ 4 files changed, 15 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/drivers/vdpa/vdpa.c b/drivers/vdpa/vdpa.c index 9846c9de4bfa..1ea525433a5c 100644 --- a/drivers/vdpa/vdpa.c +++ b/drivers/vdpa/vdpa.c @@ -393,7 +393,7 @@ static void vdpa_get_config_unlocked(struct vdpa_device *vdev, * If it does happen we assume a legacy guest. */ if (!vdev->features_valid) - vdpa_set_features(vdev, 0, true); + vdpa_set_features_unlocked(vdev, 0); ops->get_config(vdev, offset, buf, len); } diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index 851539807bc9..ec5249e8c32d 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -286,7 +286,7 @@ static long vhost_vdpa_set_features(struct vhost_vdpa *v, u64 __user *featurep) if (copy_from_user(&features, featurep, sizeof(features))) return -EFAULT; - if (vdpa_set_features(vdpa, features, false)) + if (vdpa_set_features(vdpa, features)) return -EINVAL; return 0; diff --git a/drivers/virtio/virtio_vdpa.c b/drivers/virtio/virtio_vdpa.c index 7767a7f0119b..76504559bc25 100644 --- a/drivers/virtio/virtio_vdpa.c +++ b/drivers/virtio/virtio_vdpa.c @@ -317,7 +317,7 @@ static int virtio_vdpa_finalize_features(struct virtio_device *vdev) /* Give virtio_ring a chance to accept features. */ vring_transport_features(vdev); - return vdpa_set_features(vdpa, vdev->features, false); + return vdpa_set_features(vdpa, vdev->features); } static const char *virtio_vdpa_bus_name(struct virtio_device *vdev) diff --git a/include/linux/vdpa.h b/include/linux/vdpa.h index 2de442ececae..721089bb4c84 100644 --- a/include/linux/vdpa.h +++ b/include/linux/vdpa.h @@ -401,18 +401,24 @@ static inline int vdpa_reset(struct vdpa_device *vdev) return ret; } -static inline int vdpa_set_features(struct vdpa_device *vdev, u64 features, bool locked) +static inline int vdpa_set_features_unlocked(struct vdpa_device *vdev, u64 features) { const struct vdpa_config_ops *ops = vdev->config; int ret; - if (!locked) - mutex_lock(&vdev->cf_mutex); - vdev->features_valid = true; ret = ops->set_driver_features(vdev, features); - if (!locked) - mutex_unlock(&vdev->cf_mutex); + + return ret; +} + +static inline int vdpa_set_features(struct vdpa_device *vdev, u64 features) +{ + int ret; + + mutex_lock(&vdev->cf_mutex); + ret = vdpa_set_features_unlocked(vdev, features); + mutex_unlock(&vdev->cf_mutex); return ret; } -- cgit v1.2.3 From 5c26f6ac9416b63d093e29c30e79b3297e425472 Mon Sep 17 00:00:00 2001 From: Suren Baghdasaryan Date: Fri, 4 Mar 2022 20:28:51 -0800 Subject: mm: refactor vm_area_struct::anon_vma_name usage code Avoid mixing strings and their anon_vma_name referenced pointers by using struct anon_vma_name whenever possible. This simplifies the code and allows easier sharing of anon_vma_name structures when they represent the same name. [surenb@google.com: fix comment] Link: https://lkml.kernel.org/r/20220223153613.835563-1-surenb@google.com Link: https://lkml.kernel.org/r/20220224231834.1481408-1-surenb@google.com Signed-off-by: Suren Baghdasaryan Suggested-by: Matthew Wilcox Suggested-by: Michal Hocko Acked-by: Michal Hocko Cc: Colin Cross Cc: Sumit Semwal Cc: Dave Hansen Cc: Kees Cook Cc: "Kirill A. Shutemov" Cc: Vlastimil Babka Cc: Johannes Weiner Cc: "Eric W. Biederman" Cc: Christian Brauner Cc: Alexey Gladkov Cc: Sasha Levin Cc: Chris Hyser Cc: Davidlohr Bueso Cc: Peter Collingbourne Cc: Xiaofeng Cao Cc: David Hildenbrand Cc: Cyrill Gorcunov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/task_mmu.c | 6 ++-- fs/userfaultfd.c | 6 ++-- include/linux/mm.h | 7 ++-- include/linux/mm_inline.h | 87 ++++++++++++++++++++++++++++++++--------------- include/linux/mm_types.h | 5 ++- kernel/fork.c | 4 +-- kernel/sys.c | 19 +++++++---- mm/madvise.c | 87 ++++++++++++++++------------------------------- mm/mempolicy.c | 2 +- mm/mlock.c | 2 +- mm/mmap.c | 12 +++---- mm/mprotect.c | 2 +- 12 files changed, 125 insertions(+), 114 deletions(-) (limited to 'include') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 6e97ed775074..2c48b1eaaa9c 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -309,7 +309,7 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma) name = arch_vma_name(vma); if (!name) { - const char *anon_name; + struct anon_vma_name *anon_name; if (!mm) { name = "[vdso]"; @@ -327,10 +327,10 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma) goto done; } - anon_name = vma_anon_name(vma); + anon_name = anon_vma_name(vma); if (anon_name) { seq_pad(m, ' '); - seq_printf(m, "[anon:%s]", anon_name); + seq_printf(m, "[anon:%s]", anon_name->name); } } diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index e26b10132d47..8e03b3d3f5fa 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -878,7 +878,7 @@ static int userfaultfd_release(struct inode *inode, struct file *file) new_flags, vma->anon_vma, vma->vm_file, vma->vm_pgoff, vma_policy(vma), - NULL_VM_UFFD_CTX, vma_anon_name(vma)); + NULL_VM_UFFD_CTX, anon_vma_name(vma)); if (prev) vma = prev; else @@ -1438,7 +1438,7 @@ static int userfaultfd_register(struct userfaultfd_ctx *ctx, vma->anon_vma, vma->vm_file, vma->vm_pgoff, vma_policy(vma), ((struct vm_userfaultfd_ctx){ ctx }), - vma_anon_name(vma)); + anon_vma_name(vma)); if (prev) { vma = prev; goto next; @@ -1615,7 +1615,7 @@ static int userfaultfd_unregister(struct userfaultfd_ctx *ctx, prev = vma_merge(mm, prev, start, vma_end, new_flags, vma->anon_vma, vma->vm_file, vma->vm_pgoff, vma_policy(vma), - NULL_VM_UFFD_CTX, vma_anon_name(vma)); + NULL_VM_UFFD_CTX, anon_vma_name(vma)); if (prev) { vma = prev; goto next; diff --git a/include/linux/mm.h b/include/linux/mm.h index 213cc569b192..5744a3fc4716 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -2626,7 +2626,7 @@ static inline int vma_adjust(struct vm_area_struct *vma, unsigned long start, extern struct vm_area_struct *vma_merge(struct mm_struct *, struct vm_area_struct *prev, unsigned long addr, unsigned long end, unsigned long vm_flags, struct anon_vma *, struct file *, pgoff_t, - struct mempolicy *, struct vm_userfaultfd_ctx, const char *); + struct mempolicy *, struct vm_userfaultfd_ctx, struct anon_vma_name *); extern struct anon_vma *find_mergeable_anon_vma(struct vm_area_struct *); extern int __split_vma(struct mm_struct *, struct vm_area_struct *, unsigned long addr, int new_below); @@ -3372,11 +3372,12 @@ static inline int seal_check_future_write(int seals, struct vm_area_struct *vma) #ifdef CONFIG_ANON_VMA_NAME int madvise_set_anon_name(struct mm_struct *mm, unsigned long start, - unsigned long len_in, const char *name); + unsigned long len_in, + struct anon_vma_name *anon_name); #else static inline int madvise_set_anon_name(struct mm_struct *mm, unsigned long start, - unsigned long len_in, const char *name) { + unsigned long len_in, struct anon_vma_name *anon_name) { return 0; } #endif diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h index b725839dfe71..dd3accaa4e6d 100644 --- a/include/linux/mm_inline.h +++ b/include/linux/mm_inline.h @@ -140,50 +140,81 @@ static __always_inline void del_page_from_lru_list(struct page *page, #ifdef CONFIG_ANON_VMA_NAME /* - * mmap_lock should be read-locked when calling vma_anon_name() and while using - * the returned pointer. + * mmap_lock should be read-locked when calling anon_vma_name(). Caller should + * either keep holding the lock while using the returned pointer or it should + * raise anon_vma_name refcount before releasing the lock. */ -extern const char *vma_anon_name(struct vm_area_struct *vma); +extern struct anon_vma_name *anon_vma_name(struct vm_area_struct *vma); +extern struct anon_vma_name *anon_vma_name_alloc(const char *name); +extern void anon_vma_name_free(struct kref *kref); -/* - * mmap_lock should be read-locked for orig_vma->vm_mm. - * mmap_lock should be write-locked for new_vma->vm_mm or new_vma should be - * isolated. - */ -extern void dup_vma_anon_name(struct vm_area_struct *orig_vma, - struct vm_area_struct *new_vma); +/* mmap_lock should be read-locked */ +static inline void anon_vma_name_get(struct anon_vma_name *anon_name) +{ + if (anon_name) + kref_get(&anon_name->kref); +} -/* - * mmap_lock should be write-locked or vma should have been isolated under - * write-locked mmap_lock protection. - */ -extern void free_vma_anon_name(struct vm_area_struct *vma); +static inline void anon_vma_name_put(struct anon_vma_name *anon_name) +{ + if (anon_name) + kref_put(&anon_name->kref, anon_vma_name_free); +} -/* mmap_lock should be read-locked */ -static inline bool is_same_vma_anon_name(struct vm_area_struct *vma, - const char *name) +static inline void dup_anon_vma_name(struct vm_area_struct *orig_vma, + struct vm_area_struct *new_vma) +{ + struct anon_vma_name *anon_name = anon_vma_name(orig_vma); + + if (anon_name) { + anon_vma_name_get(anon_name); + new_vma->anon_name = anon_name; + } +} + +static inline void free_anon_vma_name(struct vm_area_struct *vma) { - const char *vma_name = vma_anon_name(vma); + /* + * Not using anon_vma_name because it generates a warning if mmap_lock + * is not held, which might be the case here. + */ + if (!vma->vm_file) + anon_vma_name_put(vma->anon_name); +} - /* either both NULL, or pointers to same string */ - if (vma_name == name) +static inline bool anon_vma_name_eq(struct anon_vma_name *anon_name1, + struct anon_vma_name *anon_name2) +{ + if (anon_name1 == anon_name2) return true; - return name && vma_name && !strcmp(name, vma_name); + return anon_name1 && anon_name2 && + !strcmp(anon_name1->name, anon_name2->name); } + #else /* CONFIG_ANON_VMA_NAME */ -static inline const char *vma_anon_name(struct vm_area_struct *vma) +static inline struct anon_vma_name *anon_vma_name(struct vm_area_struct *vma) { return NULL; } -static inline void dup_vma_anon_name(struct vm_area_struct *orig_vma, - struct vm_area_struct *new_vma) {} -static inline void free_vma_anon_name(struct vm_area_struct *vma) {} -static inline bool is_same_vma_anon_name(struct vm_area_struct *vma, - const char *name) + +static inline struct anon_vma_name *anon_vma_name_alloc(const char *name) +{ + return NULL; +} + +static inline void anon_vma_name_get(struct anon_vma_name *anon_name) {} +static inline void anon_vma_name_put(struct anon_vma_name *anon_name) {} +static inline void dup_anon_vma_name(struct vm_area_struct *orig_vma, + struct vm_area_struct *new_vma) {} +static inline void free_anon_vma_name(struct vm_area_struct *vma) {} + +static inline bool anon_vma_name_eq(struct anon_vma_name *anon_name1, + struct anon_vma_name *anon_name2) { return true; } + #endif /* CONFIG_ANON_VMA_NAME */ static inline void init_tlb_flush_pending(struct mm_struct *mm) diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 5140e5feb486..0f549870da6a 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -416,7 +416,10 @@ struct vm_area_struct { struct rb_node rb; unsigned long rb_subtree_last; } shared; - /* Serialized by mmap_sem. */ + /* + * Serialized by mmap_sem. Never use directly because it is + * valid only when vm_file is NULL. Use anon_vma_name instead. + */ struct anon_vma_name *anon_name; }; diff --git a/kernel/fork.c b/kernel/fork.c index a024bf6254df..f1e89007f228 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -366,14 +366,14 @@ struct vm_area_struct *vm_area_dup(struct vm_area_struct *orig) *new = data_race(*orig); INIT_LIST_HEAD(&new->anon_vma_chain); new->vm_next = new->vm_prev = NULL; - dup_vma_anon_name(orig, new); + dup_anon_vma_name(orig, new); } return new; } void vm_area_free(struct vm_area_struct *vma) { - free_vma_anon_name(vma); + free_anon_vma_name(vma); kmem_cache_free(vm_area_cachep, vma); } diff --git a/kernel/sys.c b/kernel/sys.c index 97dc9e5d6bf9..5b0e172c4d47 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -7,6 +7,7 @@ #include #include +#include #include #include #include @@ -2286,15 +2287,16 @@ static int prctl_set_vma(unsigned long opt, unsigned long addr, { struct mm_struct *mm = current->mm; const char __user *uname; - char *name, *pch; + struct anon_vma_name *anon_name = NULL; int error; switch (opt) { case PR_SET_VMA_ANON_NAME: uname = (const char __user *)arg; if (uname) { - name = strndup_user(uname, ANON_VMA_NAME_MAX_LEN); + char *name, *pch; + name = strndup_user(uname, ANON_VMA_NAME_MAX_LEN); if (IS_ERR(name)) return PTR_ERR(name); @@ -2304,15 +2306,18 @@ static int prctl_set_vma(unsigned long opt, unsigned long addr, return -EINVAL; } } - } else { - /* Reset the name */ - name = NULL; + /* anon_vma has its own copy */ + anon_name = anon_vma_name_alloc(name); + kfree(name); + if (!anon_name) + return -ENOMEM; + } mmap_write_lock(mm); - error = madvise_set_anon_name(mm, addr, size, name); + error = madvise_set_anon_name(mm, addr, size, anon_name); mmap_write_unlock(mm); - kfree(name); + anon_vma_name_put(anon_name); break; default: error = -EINVAL; diff --git a/mm/madvise.c b/mm/madvise.c index 5604064df464..081b1cded21e 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -65,7 +65,7 @@ static int madvise_need_mmap_write(int behavior) } #ifdef CONFIG_ANON_VMA_NAME -static struct anon_vma_name *anon_vma_name_alloc(const char *name) +struct anon_vma_name *anon_vma_name_alloc(const char *name) { struct anon_vma_name *anon_name; size_t count; @@ -81,78 +81,49 @@ static struct anon_vma_name *anon_vma_name_alloc(const char *name) return anon_name; } -static void vma_anon_name_free(struct kref *kref) +void anon_vma_name_free(struct kref *kref) { struct anon_vma_name *anon_name = container_of(kref, struct anon_vma_name, kref); kfree(anon_name); } -static inline bool has_vma_anon_name(struct vm_area_struct *vma) +struct anon_vma_name *anon_vma_name(struct vm_area_struct *vma) { - return !vma->vm_file && vma->anon_name; -} - -const char *vma_anon_name(struct vm_area_struct *vma) -{ - if (!has_vma_anon_name(vma)) - return NULL; - mmap_assert_locked(vma->vm_mm); - return vma->anon_name->name; -} - -void dup_vma_anon_name(struct vm_area_struct *orig_vma, - struct vm_area_struct *new_vma) -{ - if (!has_vma_anon_name(orig_vma)) - return; - - kref_get(&orig_vma->anon_name->kref); - new_vma->anon_name = orig_vma->anon_name; -} - -void free_vma_anon_name(struct vm_area_struct *vma) -{ - struct anon_vma_name *anon_name; - - if (!has_vma_anon_name(vma)) - return; + if (vma->vm_file) + return NULL; - anon_name = vma->anon_name; - vma->anon_name = NULL; - kref_put(&anon_name->kref, vma_anon_name_free); + return vma->anon_name; } /* mmap_lock should be write-locked */ -static int replace_vma_anon_name(struct vm_area_struct *vma, const char *name) +static int replace_anon_vma_name(struct vm_area_struct *vma, + struct anon_vma_name *anon_name) { - const char *anon_name; + struct anon_vma_name *orig_name = anon_vma_name(vma); - if (!name) { - free_vma_anon_name(vma); + if (!anon_name) { + vma->anon_name = NULL; + anon_vma_name_put(orig_name); return 0; } - anon_name = vma_anon_name(vma); - if (anon_name) { - /* Same name, nothing to do here */ - if (!strcmp(name, anon_name)) - return 0; + if (anon_vma_name_eq(orig_name, anon_name)) + return 0; - free_vma_anon_name(vma); - } - vma->anon_name = anon_vma_name_alloc(name); - if (!vma->anon_name) - return -ENOMEM; + anon_vma_name_get(anon_name); + vma->anon_name = anon_name; + anon_vma_name_put(orig_name); return 0; } #else /* CONFIG_ANON_VMA_NAME */ -static int replace_vma_anon_name(struct vm_area_struct *vma, const char *name) +static int replace_anon_vma_name(struct vm_area_struct *vma, + struct anon_vma_name *anon_name) { - if (name) + if (anon_name) return -EINVAL; return 0; @@ -165,13 +136,13 @@ static int replace_vma_anon_name(struct vm_area_struct *vma, const char *name) static int madvise_update_vma(struct vm_area_struct *vma, struct vm_area_struct **prev, unsigned long start, unsigned long end, unsigned long new_flags, - const char *name) + struct anon_vma_name *anon_name) { struct mm_struct *mm = vma->vm_mm; int error; pgoff_t pgoff; - if (new_flags == vma->vm_flags && is_same_vma_anon_name(vma, name)) { + if (new_flags == vma->vm_flags && anon_vma_name_eq(anon_vma_name(vma), anon_name)) { *prev = vma; return 0; } @@ -179,7 +150,7 @@ static int madvise_update_vma(struct vm_area_struct *vma, pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT); *prev = vma_merge(mm, *prev, start, end, new_flags, vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma), - vma->vm_userfaultfd_ctx, name); + vma->vm_userfaultfd_ctx, anon_name); if (*prev) { vma = *prev; goto success; @@ -209,7 +180,7 @@ success: */ vma->vm_flags = new_flags; if (!vma->vm_file) { - error = replace_vma_anon_name(vma, name); + error = replace_anon_vma_name(vma, anon_name); if (error) return error; } @@ -1041,7 +1012,7 @@ static int madvise_vma_behavior(struct vm_area_struct *vma, } error = madvise_update_vma(vma, prev, start, end, new_flags, - vma_anon_name(vma)); + anon_vma_name(vma)); out: /* @@ -1225,7 +1196,7 @@ int madvise_walk_vmas(struct mm_struct *mm, unsigned long start, static int madvise_vma_anon_name(struct vm_area_struct *vma, struct vm_area_struct **prev, unsigned long start, unsigned long end, - unsigned long name) + unsigned long anon_name) { int error; @@ -1234,7 +1205,7 @@ static int madvise_vma_anon_name(struct vm_area_struct *vma, return -EBADF; error = madvise_update_vma(vma, prev, start, end, vma->vm_flags, - (const char *)name); + (struct anon_vma_name *)anon_name); /* * madvise() returns EAGAIN if kernel resources, such as @@ -1246,7 +1217,7 @@ static int madvise_vma_anon_name(struct vm_area_struct *vma, } int madvise_set_anon_name(struct mm_struct *mm, unsigned long start, - unsigned long len_in, const char *name) + unsigned long len_in, struct anon_vma_name *anon_name) { unsigned long end; unsigned long len; @@ -1266,7 +1237,7 @@ int madvise_set_anon_name(struct mm_struct *mm, unsigned long start, if (end == start) return 0; - return madvise_walk_vmas(mm, start, end, (unsigned long)name, + return madvise_walk_vmas(mm, start, end, (unsigned long)anon_name, madvise_vma_anon_name); } #endif /* CONFIG_ANON_VMA_NAME */ diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 028e8dd82b44..69284d3b5e53 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -814,7 +814,7 @@ static int mbind_range(struct mm_struct *mm, unsigned long start, prev = vma_merge(mm, prev, vmstart, vmend, vma->vm_flags, vma->anon_vma, vma->vm_file, pgoff, new_pol, vma->vm_userfaultfd_ctx, - vma_anon_name(vma)); + anon_vma_name(vma)); if (prev) { vma = prev; next = vma->vm_next; diff --git a/mm/mlock.c b/mm/mlock.c index 8f584eddd305..25934e7db3e1 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -512,7 +512,7 @@ static int mlock_fixup(struct vm_area_struct *vma, struct vm_area_struct **prev, pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT); *prev = vma_merge(mm, *prev, start, end, newflags, vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma), - vma->vm_userfaultfd_ctx, vma_anon_name(vma)); + vma->vm_userfaultfd_ctx, anon_vma_name(vma)); if (*prev) { vma = *prev; goto success; diff --git a/mm/mmap.c b/mm/mmap.c index d445c1b9d606..f61a15474dd6 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1031,7 +1031,7 @@ again: static inline int is_mergeable_vma(struct vm_area_struct *vma, struct file *file, unsigned long vm_flags, struct vm_userfaultfd_ctx vm_userfaultfd_ctx, - const char *anon_name) + struct anon_vma_name *anon_name) { /* * VM_SOFTDIRTY should not prevent from VMA merging, if we @@ -1049,7 +1049,7 @@ static inline int is_mergeable_vma(struct vm_area_struct *vma, return 0; if (!is_mergeable_vm_userfaultfd_ctx(vma, vm_userfaultfd_ctx)) return 0; - if (!is_same_vma_anon_name(vma, anon_name)) + if (!anon_vma_name_eq(anon_vma_name(vma), anon_name)) return 0; return 1; } @@ -1084,7 +1084,7 @@ can_vma_merge_before(struct vm_area_struct *vma, unsigned long vm_flags, struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff, struct vm_userfaultfd_ctx vm_userfaultfd_ctx, - const char *anon_name) + struct anon_vma_name *anon_name) { if (is_mergeable_vma(vma, file, vm_flags, vm_userfaultfd_ctx, anon_name) && is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { @@ -1106,7 +1106,7 @@ can_vma_merge_after(struct vm_area_struct *vma, unsigned long vm_flags, struct anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff, struct vm_userfaultfd_ctx vm_userfaultfd_ctx, - const char *anon_name) + struct anon_vma_name *anon_name) { if (is_mergeable_vma(vma, file, vm_flags, vm_userfaultfd_ctx, anon_name) && is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { @@ -1167,7 +1167,7 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm, struct anon_vma *anon_vma, struct file *file, pgoff_t pgoff, struct mempolicy *policy, struct vm_userfaultfd_ctx vm_userfaultfd_ctx, - const char *anon_name) + struct anon_vma_name *anon_name) { pgoff_t pglen = (end - addr) >> PAGE_SHIFT; struct vm_area_struct *area, *next; @@ -3256,7 +3256,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap, return NULL; /* should never get here */ new_vma = vma_merge(mm, prev, addr, addr + len, vma->vm_flags, vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma), - vma->vm_userfaultfd_ctx, vma_anon_name(vma)); + vma->vm_userfaultfd_ctx, anon_vma_name(vma)); if (new_vma) { /* * Source vma may have been merged into new_vma diff --git a/mm/mprotect.c b/mm/mprotect.c index 5ca3fbcb1495..2887644fd150 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -464,7 +464,7 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev, pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT); *pprev = vma_merge(mm, *pprev, start, end, newflags, vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma), - vma->vm_userfaultfd_ctx, vma_anon_name(vma)); + vma->vm_userfaultfd_ctx, anon_vma_name(vma)); if (*pprev) { vma = *pprev; VM_WARN_ON((vma->vm_flags ^ newflags) & ~VM_SOFTDIRTY); -- cgit v1.2.3 From 96403e11283def1d1c465c8279514c9a504d8630 Mon Sep 17 00:00:00 2001 From: Suren Baghdasaryan Date: Fri, 4 Mar 2022 20:28:55 -0800 Subject: mm: prevent vm_area_struct::anon_name refcount saturation A deep process chain with many vmas could grow really high. With default sysctl_max_map_count (64k) and default pid_max (32k) the max number of vmas in the system is 2147450880 and the refcounter has headroom of 1073774592 before it reaches REFCOUNT_SATURATED (3221225472). Therefore it's unlikely that an anonymous name refcounter will overflow with these defaults. Currently the max for pid_max is PID_MAX_LIMIT (4194304) and for sysctl_max_map_count it's INT_MAX (2147483647). In this configuration anon_vma_name refcount overflow becomes theoretically possible (that still require heavy sharing of that anon_vma_name between processes). kref refcounting interface used in anon_vma_name structure will detect a counter overflow when it reaches REFCOUNT_SATURATED value but will only generate a warning and freeze the ref counter. This would lead to the refcounted object never being freed. A determined attacker could leak memory like that but it would be rather expensive and inefficient way to do so. To ensure anon_vma_name refcount does not overflow, stop anon_vma_name sharing when the refcount reaches REFCOUNT_MAX (2147483647), which still leaves INT_MAX/2 (1073741823) values before the counter reaches REFCOUNT_SATURATED. This should provide enough headroom for raising the refcounts temporarily. Link: https://lkml.kernel.org/r/20220223153613.835563-2-surenb@google.com Signed-off-by: Suren Baghdasaryan Suggested-by: Michal Hocko Acked-by: Michal Hocko Cc: Alexey Gladkov Cc: Chris Hyser Cc: Christian Brauner Cc: Colin Cross Cc: Cyrill Gorcunov Cc: Dave Hansen Cc: David Hildenbrand Cc: Davidlohr Bueso Cc: "Eric W. Biederman" Cc: Johannes Weiner Cc: Kees Cook Cc: "Kirill A. Shutemov" Cc: Matthew Wilcox Cc: Peter Collingbourne Cc: Sasha Levin Cc: Sumit Semwal Cc: Vlastimil Babka Cc: Xiaofeng Cao Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm_inline.h | 18 ++++++++++++++---- mm/madvise.c | 3 +-- 2 files changed, 15 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h index dd3accaa4e6d..cf90b1fa2c60 100644 --- a/include/linux/mm_inline.h +++ b/include/linux/mm_inline.h @@ -161,15 +161,25 @@ static inline void anon_vma_name_put(struct anon_vma_name *anon_name) kref_put(&anon_name->kref, anon_vma_name_free); } +static inline +struct anon_vma_name *anon_vma_name_reuse(struct anon_vma_name *anon_name) +{ + /* Prevent anon_name refcount saturation early on */ + if (kref_read(&anon_name->kref) < REFCOUNT_MAX) { + anon_vma_name_get(anon_name); + return anon_name; + + } + return anon_vma_name_alloc(anon_name->name); +} + static inline void dup_anon_vma_name(struct vm_area_struct *orig_vma, struct vm_area_struct *new_vma) { struct anon_vma_name *anon_name = anon_vma_name(orig_vma); - if (anon_name) { - anon_vma_name_get(anon_name); - new_vma->anon_name = anon_name; - } + if (anon_name) + new_vma->anon_name = anon_vma_name_reuse(anon_name); } static inline void free_anon_vma_name(struct vm_area_struct *vma) diff --git a/mm/madvise.c b/mm/madvise.c index 081b1cded21e..1f2693dccf7b 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -113,8 +113,7 @@ static int replace_anon_vma_name(struct vm_area_struct *vma, if (anon_vma_name_eq(orig_name, anon_name)) return 0; - anon_vma_name_get(anon_name); - vma->anon_name = anon_name; + vma->anon_name = anon_vma_name_reuse(anon_name); anon_vma_name_put(orig_name); return 0; -- cgit v1.2.3 From 6b1775f26a2da2b05a6dc8ec2b5d14e9a4701a1a Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Mon, 7 Mar 2022 09:48:54 +0100 Subject: xen/grant-table: add gnttab_try_end_foreign_access() Add a new grant table function gnttab_try_end_foreign_access(), which will remove and free a grant if it is not in use. Its main use case is to either free a grant if it is no longer in use, or to take some other action if it is still in use. This other action can be an error exit, or (e.g. in the case of blkfront persistent grant feature) some special handling. This is CVE-2022-23036, CVE-2022-23038 / part of XSA-396. Reported-by: Demi Marie Obenour Signed-off-by: Juergen Gross Reviewed-by: Jan Beulich --- V2: - new patch V4: - add comments to header (Jan Beulich) --- drivers/xen/grant-table.c | 14 ++++++++++++-- include/xen/grant_table.h | 12 ++++++++++++ 2 files changed, 24 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c index 3729bea0c989..1b82e7a3722a 100644 --- a/drivers/xen/grant-table.c +++ b/drivers/xen/grant-table.c @@ -435,11 +435,21 @@ static void gnttab_add_deferred(grant_ref_t ref, bool readonly, what, ref, page ? page_to_pfn(page) : -1); } +int gnttab_try_end_foreign_access(grant_ref_t ref) +{ + int ret = _gnttab_end_foreign_access_ref(ref, 0); + + if (ret) + put_free_entry(ref); + + return ret; +} +EXPORT_SYMBOL_GPL(gnttab_try_end_foreign_access); + void gnttab_end_foreign_access(grant_ref_t ref, int readonly, unsigned long page) { - if (gnttab_end_foreign_access_ref(ref, readonly)) { - put_free_entry(ref); + if (gnttab_try_end_foreign_access(ref)) { if (page != 0) put_page(virt_to_page(page)); } else diff --git a/include/xen/grant_table.h b/include/xen/grant_table.h index cb854df031ce..358d2817741b 100644 --- a/include/xen/grant_table.h +++ b/include/xen/grant_table.h @@ -104,10 +104,22 @@ int gnttab_end_foreign_access_ref(grant_ref_t ref, int readonly); * access has been ended, free the given page too. Access will be ended * immediately iff the grant entry is not in use, otherwise it will happen * some time later. page may be 0, in which case no freeing will occur. + * Note that the granted page might still be accessed (read or write) by the + * other side after gnttab_end_foreign_access() returns, so even if page was + * specified as 0 it is not allowed to just reuse the page for other + * purposes immediately. */ void gnttab_end_foreign_access(grant_ref_t ref, int readonly, unsigned long page); +/* + * End access through the given grant reference, iff the grant entry is + * no longer in use. In case of success ending foreign access, the + * grant reference is deallocated. + * Return 1 if the grant entry was freed, 0 if it is still in use. + */ +int gnttab_try_end_foreign_access(grant_ref_t ref); + int gnttab_grant_foreign_transfer(domid_t domid, unsigned long pfn); unsigned long gnttab_end_foreign_transfer_ref(grant_ref_t ref); -- cgit v1.2.3 From 1dbd11ca75fe664d3e54607547771d021f531f59 Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Mon, 7 Mar 2022 09:48:54 +0100 Subject: xen: remove gnttab_query_foreign_access() Remove gnttab_query_foreign_access(), as it is unused and unsafe to use. All previous use cases assumed a grant would not be in use after gnttab_query_foreign_access() returned 0. This information is useless in best case, as it only refers to a situation in the past, which could have changed already. Signed-off-by: Juergen Gross Reviewed-by: Jan Beulich --- drivers/xen/grant-table.c | 25 ------------------------- include/xen/grant_table.h | 2 -- 2 files changed, 27 deletions(-) (limited to 'include') diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c index 1b82e7a3722a..e6548910e79f 100644 --- a/drivers/xen/grant-table.c +++ b/drivers/xen/grant-table.c @@ -133,13 +133,6 @@ struct gnttab_ops { * return the frame. */ unsigned long (*end_foreign_transfer_ref)(grant_ref_t ref); - /* - * Query the status of a grant entry. Ref parameter is reference of - * queried grant entry, return value is the status of queried entry. - * Detailed status(writing/reading) can be gotten from the return value - * by bit operations. - */ - int (*query_foreign_access)(grant_ref_t ref); }; struct unmap_refs_callback_data { @@ -284,22 +277,6 @@ int gnttab_grant_foreign_access(domid_t domid, unsigned long frame, } EXPORT_SYMBOL_GPL(gnttab_grant_foreign_access); -static int gnttab_query_foreign_access_v1(grant_ref_t ref) -{ - return gnttab_shared.v1[ref].flags & (GTF_reading|GTF_writing); -} - -static int gnttab_query_foreign_access_v2(grant_ref_t ref) -{ - return grstatus[ref] & (GTF_reading|GTF_writing); -} - -int gnttab_query_foreign_access(grant_ref_t ref) -{ - return gnttab_interface->query_foreign_access(ref); -} -EXPORT_SYMBOL_GPL(gnttab_query_foreign_access); - static int gnttab_end_foreign_access_ref_v1(grant_ref_t ref, int readonly) { u16 flags, nflags; @@ -1427,7 +1404,6 @@ static const struct gnttab_ops gnttab_v1_ops = { .update_entry = gnttab_update_entry_v1, .end_foreign_access_ref = gnttab_end_foreign_access_ref_v1, .end_foreign_transfer_ref = gnttab_end_foreign_transfer_ref_v1, - .query_foreign_access = gnttab_query_foreign_access_v1, }; static const struct gnttab_ops gnttab_v2_ops = { @@ -1439,7 +1415,6 @@ static const struct gnttab_ops gnttab_v2_ops = { .update_entry = gnttab_update_entry_v2, .end_foreign_access_ref = gnttab_end_foreign_access_ref_v2, .end_foreign_transfer_ref = gnttab_end_foreign_transfer_ref_v2, - .query_foreign_access = gnttab_query_foreign_access_v2, }; static bool gnttab_need_v2(void) diff --git a/include/xen/grant_table.h b/include/xen/grant_table.h index 358d2817741b..ab9e692a0ef4 100644 --- a/include/xen/grant_table.h +++ b/include/xen/grant_table.h @@ -125,8 +125,6 @@ int gnttab_grant_foreign_transfer(domid_t domid, unsigned long pfn); unsigned long gnttab_end_foreign_transfer_ref(grant_ref_t ref); unsigned long gnttab_end_foreign_transfer(grant_ref_t ref); -int gnttab_query_foreign_access(grant_ref_t ref); - /* * operations on reserved batches of grant references */ -- cgit v1.2.3 From 42baefac638f06314298087394b982ead9ec444b Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Mon, 7 Mar 2022 09:48:55 +0100 Subject: xen/gnttab: fix gnttab_end_foreign_access() without page specified gnttab_end_foreign_access() is used to free a grant reference and optionally to free the associated page. In case the grant is still in use by the other side processing is being deferred. This leads to a problem in case no page to be freed is specified by the caller: the caller doesn't know that the page is still mapped by the other side and thus should not be used for other purposes. The correct way to handle this situation is to take an additional reference to the granted page in case handling is being deferred and to drop that reference when the grant reference could be freed finally. This requires that there are no users of gnttab_end_foreign_access() left directly repurposing the granted page after the call, as this might result in clobbered data or information leaks via the not yet freed grant reference. This is part of CVE-2022-23041 / XSA-396. Reported-by: Simon Gaiser Signed-off-by: Juergen Gross Reviewed-by: Jan Beulich --- V4: - expand comment in header V5: - get page ref in case of kmalloc() failure, too --- drivers/xen/grant-table.c | 36 +++++++++++++++++++++++++++++------- include/xen/grant_table.h | 7 ++++++- 2 files changed, 35 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c index e6548910e79f..5c83d41766c8 100644 --- a/drivers/xen/grant-table.c +++ b/drivers/xen/grant-table.c @@ -133,6 +133,10 @@ struct gnttab_ops { * return the frame. */ unsigned long (*end_foreign_transfer_ref)(grant_ref_t ref); + /* + * Read the frame number related to a given grant reference. + */ + unsigned long (*read_frame)(grant_ref_t ref); }; struct unmap_refs_callback_data { @@ -330,6 +334,16 @@ int gnttab_end_foreign_access_ref(grant_ref_t ref, int readonly) } EXPORT_SYMBOL_GPL(gnttab_end_foreign_access_ref); +static unsigned long gnttab_read_frame_v1(grant_ref_t ref) +{ + return gnttab_shared.v1[ref].frame; +} + +static unsigned long gnttab_read_frame_v2(grant_ref_t ref) +{ + return gnttab_shared.v2[ref].full_page.frame; +} + struct deferred_entry { struct list_head list; grant_ref_t ref; @@ -359,12 +373,9 @@ static void gnttab_handle_deferred(struct timer_list *unused) spin_unlock_irqrestore(&gnttab_list_lock, flags); if (_gnttab_end_foreign_access_ref(entry->ref, entry->ro)) { put_free_entry(entry->ref); - if (entry->page) { - pr_debug("freeing g.e. %#x (pfn %#lx)\n", - entry->ref, page_to_pfn(entry->page)); - put_page(entry->page); - } else - pr_info("freeing g.e. %#x\n", entry->ref); + pr_debug("freeing g.e. %#x (pfn %#lx)\n", + entry->ref, page_to_pfn(entry->page)); + put_page(entry->page); kfree(entry); entry = NULL; } else { @@ -389,9 +400,18 @@ static void gnttab_handle_deferred(struct timer_list *unused) static void gnttab_add_deferred(grant_ref_t ref, bool readonly, struct page *page) { - struct deferred_entry *entry = kmalloc(sizeof(*entry), GFP_ATOMIC); + struct deferred_entry *entry; + gfp_t gfp = (in_atomic() || irqs_disabled()) ? GFP_ATOMIC : GFP_KERNEL; const char *what = KERN_WARNING "leaking"; + entry = kmalloc(sizeof(*entry), gfp); + if (!page) { + unsigned long gfn = gnttab_interface->read_frame(ref); + + page = pfn_to_page(gfn_to_pfn(gfn)); + get_page(page); + } + if (entry) { unsigned long flags; @@ -1404,6 +1424,7 @@ static const struct gnttab_ops gnttab_v1_ops = { .update_entry = gnttab_update_entry_v1, .end_foreign_access_ref = gnttab_end_foreign_access_ref_v1, .end_foreign_transfer_ref = gnttab_end_foreign_transfer_ref_v1, + .read_frame = gnttab_read_frame_v1, }; static const struct gnttab_ops gnttab_v2_ops = { @@ -1415,6 +1436,7 @@ static const struct gnttab_ops gnttab_v2_ops = { .update_entry = gnttab_update_entry_v2, .end_foreign_access_ref = gnttab_end_foreign_access_ref_v2, .end_foreign_transfer_ref = gnttab_end_foreign_transfer_ref_v2, + .read_frame = gnttab_read_frame_v2, }; static bool gnttab_need_v2(void) diff --git a/include/xen/grant_table.h b/include/xen/grant_table.h index ab9e692a0ef4..c9fea9389ebe 100644 --- a/include/xen/grant_table.h +++ b/include/xen/grant_table.h @@ -107,7 +107,12 @@ int gnttab_end_foreign_access_ref(grant_ref_t ref, int readonly); * Note that the granted page might still be accessed (read or write) by the * other side after gnttab_end_foreign_access() returns, so even if page was * specified as 0 it is not allowed to just reuse the page for other - * purposes immediately. + * purposes immediately. gnttab_end_foreign_access() will take an additional + * reference to the granted page in this case, which is dropped only after + * the grant is no longer in use. + * This requires that multi page allocations for areas subject to + * gnttab_end_foreign_access() are done via alloc_pages_exact() (and freeing + * via free_pages_exact()) in order to avoid high order pages. */ void gnttab_end_foreign_access(grant_ref_t ref, int readonly, unsigned long page); -- cgit v1.2.3 From ebe48d368e97d007bfeb76fcb065d6cfc4c96645 Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Mon, 7 Mar 2022 13:11:39 +0100 Subject: esp: Fix possible buffer overflow in ESP transformation The maximum message size that can be send is bigger than the maximum site that skb_page_frag_refill can allocate. So it is possible to write beyond the allocated buffer. Fix this by doing a fallback to COW in that case. v2: Avoid get get_order() costs as suggested by Linus Torvalds. Fixes: cac2661c53f3 ("esp4: Avoid skb_cow_data whenever possible") Fixes: 03e2a30f6a27 ("esp6: Avoid skb_cow_data whenever possible") Reported-by: valis Signed-off-by: Steffen Klassert --- include/net/esp.h | 2 ++ net/ipv4/esp4.c | 5 +++++ net/ipv6/esp6.c | 5 +++++ 3 files changed, 12 insertions(+) (limited to 'include') diff --git a/include/net/esp.h b/include/net/esp.h index 9c5637d41d95..90cd02ff77ef 100644 --- a/include/net/esp.h +++ b/include/net/esp.h @@ -4,6 +4,8 @@ #include +#define ESP_SKB_FRAG_MAXSIZE (PAGE_SIZE << SKB_FRAG_PAGE_ORDER) + struct ip_esp_hdr; static inline struct ip_esp_hdr *ip_esp_hdr(const struct sk_buff *skb) diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index e1b1d080e908..70e6c87fbe3d 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -446,6 +446,7 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * struct page *page; struct sk_buff *trailer; int tailen = esp->tailen; + unsigned int allocsz; /* this is non-NULL only with TCP/UDP Encapsulation */ if (x->encap) { @@ -455,6 +456,10 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * return err; } + allocsz = ALIGN(skb->data_len + tailen, L1_CACHE_BYTES); + if (allocsz > ESP_SKB_FRAG_MAXSIZE) + goto cow; + if (!skb_cloned(skb)) { if (tailen <= skb_tailroom(skb)) { nfrags = 1; diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 7591160edce1..b0ffbcd5432d 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -482,6 +482,7 @@ int esp6_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info struct page *page; struct sk_buff *trailer; int tailen = esp->tailen; + unsigned int allocsz; if (x->encap) { int err = esp6_output_encap(x, skb, esp); @@ -490,6 +491,10 @@ int esp6_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info return err; } + allocsz = ALIGN(skb->data_len + tailen, L1_CACHE_BYTES); + if (allocsz > ESP_SKB_FRAG_MAXSIZE) + goto cow; + if (!skb_cloned(skb)) { if (tailen <= skb_tailroom(skb)) { nfrags = 1; -- cgit v1.2.3 From 23c7f8d7989e1646aac82f75761b7648c355cb8a Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Mon, 7 Mar 2022 13:11:41 +0100 Subject: net: Fix esp GSO on inter address family tunnels. The esp tunnel GSO handlers use skb_mac_gso_segment to push the inner packet to the segmentation handlers. However, skb_mac_gso_segment takes the Ethernet Protocol ID from 'skb->protocol' which is wrong for inter address family tunnels. We fix this by introducing a new skb_eth_gso_segment function. This function can be used if it is necessary to pass the Ethernet Protocol ID directly to the segmentation handler. First users of this function will be the esp4 and esp6 tunnel segmentation handlers. Fixes: c35fe4106b92 ("xfrm: Add mode handlers for IPsec on layer 2") Signed-off-by: Steffen Klassert --- include/linux/netdevice.h | 2 ++ net/core/gro.c | 25 +++++++++++++++++++++++++ net/ipv4/esp4_offload.c | 3 +-- net/ipv6/esp6_offload.c | 3 +-- 4 files changed, 29 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 8b5a314db167..f53ea7038441 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -4602,6 +4602,8 @@ int skb_csum_hwoffload_help(struct sk_buff *skb, struct sk_buff *__skb_gso_segment(struct sk_buff *skb, netdev_features_t features, bool tx_path); +struct sk_buff *skb_eth_gso_segment(struct sk_buff *skb, + netdev_features_t features, __be16 type); struct sk_buff *skb_mac_gso_segment(struct sk_buff *skb, netdev_features_t features); diff --git a/net/core/gro.c b/net/core/gro.c index a11b286d1495..b7d2b0dc59a2 100644 --- a/net/core/gro.c +++ b/net/core/gro.c @@ -92,6 +92,31 @@ void dev_remove_offload(struct packet_offload *po) } EXPORT_SYMBOL(dev_remove_offload); +/** + * skb_eth_gso_segment - segmentation handler for ethernet protocols. + * @skb: buffer to segment + * @features: features for the output path (see dev->features) + * @type: Ethernet Protocol ID + */ +struct sk_buff *skb_eth_gso_segment(struct sk_buff *skb, + netdev_features_t features, __be16 type) +{ + struct sk_buff *segs = ERR_PTR(-EPROTONOSUPPORT); + struct packet_offload *ptype; + + rcu_read_lock(); + list_for_each_entry_rcu(ptype, &offload_base, list) { + if (ptype->type == type && ptype->callbacks.gso_segment) { + segs = ptype->callbacks.gso_segment(skb, features); + break; + } + } + rcu_read_unlock(); + + return segs; +} +EXPORT_SYMBOL(skb_eth_gso_segment); + /** * skb_mac_gso_segment - mac layer segmentation handler. * @skb: buffer to segment diff --git a/net/ipv4/esp4_offload.c b/net/ipv4/esp4_offload.c index 146d4d54830c..935026f4c807 100644 --- a/net/ipv4/esp4_offload.c +++ b/net/ipv4/esp4_offload.c @@ -110,8 +110,7 @@ static struct sk_buff *xfrm4_tunnel_gso_segment(struct xfrm_state *x, struct sk_buff *skb, netdev_features_t features) { - __skb_push(skb, skb->mac_len); - return skb_mac_gso_segment(skb, features); + return skb_eth_gso_segment(skb, features, htons(ETH_P_IP)); } static struct sk_buff *xfrm4_transport_gso_segment(struct xfrm_state *x, diff --git a/net/ipv6/esp6_offload.c b/net/ipv6/esp6_offload.c index e61172d50817..3a293838a91d 100644 --- a/net/ipv6/esp6_offload.c +++ b/net/ipv6/esp6_offload.c @@ -145,8 +145,7 @@ static struct sk_buff *xfrm6_tunnel_gso_segment(struct xfrm_state *x, struct sk_buff *skb, netdev_features_t features) { - __skb_push(skb, skb->mac_len); - return skb_mac_gso_segment(skb, features); + return skb_eth_gso_segment(skb, features, htons(ETH_P_IPV6)); } static struct sk_buff *xfrm6_transport_gso_segment(struct xfrm_state *x, -- cgit v1.2.3 From aa6f8dcbab473f3a3c7454b74caa46d36cdc5d13 Mon Sep 17 00:00:00 2001 From: Halil Pasic Date: Sat, 5 Mar 2022 18:07:14 +0100 Subject: swiotlb: rework "fix info leak with DMA_FROM_DEVICE" Unfortunately, we ended up merging an old version of the patch "fix info leak with DMA_FROM_DEVICE" instead of merging the latest one. Christoph (the swiotlb maintainer), he asked me to create an incremental fix (after I have pointed this out the mix up, and asked him for guidance). So here we go. The main differences between what we got and what was agreed are: * swiotlb_sync_single_for_device is also required to do an extra bounce * We decided not to introduce DMA_ATTR_OVERWRITE until we have exploiters * The implantation of DMA_ATTR_OVERWRITE is flawed: DMA_ATTR_OVERWRITE must take precedence over DMA_ATTR_SKIP_CPU_SYNC Thus this patch removes DMA_ATTR_OVERWRITE, and makes swiotlb_sync_single_for_device() bounce unconditionally (that is, also when dir == DMA_TO_DEVICE) in order do avoid synchronising back stale data from the swiotlb buffer. Let me note, that if the size used with dma_sync_* API is less than the size used with dma_[un]map_*, under certain circumstances we may still end up with swiotlb not being transparent. In that sense, this is no perfect fix either. To get this bullet proof, we would have to bounce the entire mapping/bounce buffer. For that we would have to figure out the starting address, and the size of the mapping in swiotlb_sync_single_for_device(). While this does seem possible, there seems to be no firm consensus on how things are supposed to work. Signed-off-by: Halil Pasic Fixes: ddbd89deb7d3 ("swiotlb: fix info leak with DMA_FROM_DEVICE") Cc: stable@vger.kernel.org Reviewed-by: Christoph Hellwig Signed-off-by: Linus Torvalds --- Documentation/core-api/dma-attributes.rst | 8 -------- include/linux/dma-mapping.h | 8 -------- kernel/dma/swiotlb.c | 23 +++++++++++++++-------- 3 files changed, 15 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/Documentation/core-api/dma-attributes.rst b/Documentation/core-api/dma-attributes.rst index 17706dc91ec9..1887d92e8e92 100644 --- a/Documentation/core-api/dma-attributes.rst +++ b/Documentation/core-api/dma-attributes.rst @@ -130,11 +130,3 @@ accesses to DMA buffers in both privileged "supervisor" and unprivileged subsystem that the buffer is fully accessible at the elevated privilege level (and ideally inaccessible or at least read-only at the lesser-privileged levels). - -DMA_ATTR_OVERWRITE ------------------- - -This is a hint to the DMA-mapping subsystem that the device is expected to -overwrite the entire mapped size, thus the caller does not require any of the -previous buffer contents to be preserved. This allows bounce-buffering -implementations to optimise DMA_FROM_DEVICE transfers. diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index 6150d11a607e..dca2b1355bb1 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -61,14 +61,6 @@ */ #define DMA_ATTR_PRIVILEGED (1UL << 9) -/* - * This is a hint to the DMA-mapping subsystem that the device is expected - * to overwrite the entire mapped size, thus the caller does not require any - * of the previous buffer contents to be preserved. This allows - * bounce-buffering implementations to optimise DMA_FROM_DEVICE transfers. - */ -#define DMA_ATTR_OVERWRITE (1UL << 10) - /* * A dma_addr_t can hold any valid DMA or bus address for the platform. It can * be given to a device to use as a DMA source or target. It is specific to a diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c index bfc56cb21705..6db1c475ec82 100644 --- a/kernel/dma/swiotlb.c +++ b/kernel/dma/swiotlb.c @@ -627,10 +627,14 @@ phys_addr_t swiotlb_tbl_map_single(struct device *dev, phys_addr_t orig_addr, for (i = 0; i < nr_slots(alloc_size + offset); i++) mem->slots[index + i].orig_addr = slot_addr(orig_addr, i); tlb_addr = slot_addr(mem->start, index) + offset; - if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC) && - (!(attrs & DMA_ATTR_OVERWRITE) || dir == DMA_TO_DEVICE || - dir == DMA_BIDIRECTIONAL)) - swiotlb_bounce(dev, tlb_addr, mapping_size, DMA_TO_DEVICE); + /* + * When dir == DMA_FROM_DEVICE we could omit the copy from the orig + * to the tlb buffer, if we knew for sure the device will + * overwirte the entire current content. But we don't. Thus + * unconditional bounce may prevent leaking swiotlb content (i.e. + * kernel memory) to user-space. + */ + swiotlb_bounce(dev, tlb_addr, mapping_size, DMA_TO_DEVICE); return tlb_addr; } @@ -697,10 +701,13 @@ void swiotlb_tbl_unmap_single(struct device *dev, phys_addr_t tlb_addr, void swiotlb_sync_single_for_device(struct device *dev, phys_addr_t tlb_addr, size_t size, enum dma_data_direction dir) { - if (dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL) - swiotlb_bounce(dev, tlb_addr, size, DMA_TO_DEVICE); - else - BUG_ON(dir != DMA_FROM_DEVICE); + /* + * Unconditional bounce is necessary to avoid corruption on + * sync_*_for_cpu or dma_ummap_* when the device didn't overwrite + * the whole lengt of the bounce buffer. + */ + swiotlb_bounce(dev, tlb_addr, size, DMA_TO_DEVICE); + BUG_ON(!valid_dma_direction(dir)); } void swiotlb_sync_single_for_cpu(struct device *dev, phys_addr_t tlb_addr, -- cgit v1.2.3 From ac77998b7ac3044f0509b097da9637184598980d Mon Sep 17 00:00:00 2001 From: Mohammad Kabat Date: Thu, 25 Mar 2021 14:38:55 +0200 Subject: net/mlx5: Fix size field in bufferx_reg struct According to HW spec the field "size" should be 16 bits in bufferx register. Fixes: e281682bf294 ("net/mlx5_core: HW data structs/types definitions cleanup") Signed-off-by: Mohammad Kabat Reviewed-by: Moshe Shemesh Signed-off-by: Saeed Mahameed --- include/linux/mlx5/mlx5_ifc.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 598ac3bcc901..5743f5b3414b 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -9900,8 +9900,8 @@ struct mlx5_ifc_bufferx_reg_bits { u8 reserved_at_0[0x6]; u8 lossy[0x1]; u8 epsb[0x1]; - u8 reserved_at_8[0xc]; - u8 size[0xc]; + u8 reserved_at_8[0x8]; + u8 size[0x10]; u8 xoff_threshold[0x10]; u8 xon_threshold[0x10]; -- cgit v1.2.3 From 99a2b9be077ae3a5d97fbf5f7782e0f2e9812978 Mon Sep 17 00:00:00 2001 From: Ben Ben-Ishay Date: Wed, 2 Mar 2022 17:07:08 +0200 Subject: net/mlx5e: SHAMPO, reduce TIR indication SHAMPO is an RQ / WQ feature, an indication was added to the TIR in the first place to enforce suitability between connected TIR and RQ, this enforcement does not exist in current the Firmware implementation and was redundant in the first place. Fixes: 83439f3c37aa ("net/mlx5e: Add HW-GRO offload") Signed-off-by: Ben Ben-Ishay Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en/tir.c | 3 --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 3 +-- include/linux/mlx5/mlx5_ifc.h | 1 - 3 files changed, 1 insertion(+), 6 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tir.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tir.c index da169b816665..d4239e3b3c88 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tir.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tir.c @@ -88,9 +88,6 @@ void mlx5e_tir_builder_build_packet_merge(struct mlx5e_tir_builder *builder, (MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ - rough_max_l2_l3_hdr_sz) >> 8); MLX5_SET(tirc, tirc, lro_timeout_period_usecs, pkt_merge_param->timeout); break; - case MLX5E_PACKET_MERGE_SHAMPO: - MLX5_SET(tirc, tirc, packet_merge_mask, MLX5_TIRC_PACKET_MERGE_MASK_SHAMPO); - break; default: break; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index bf80fb612449..3667f5ef5990 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -3616,8 +3616,7 @@ static int set_feature_hw_gro(struct net_device *netdev, bool enable) goto out; } - err = mlx5e_safe_switch_params(priv, &new_params, - mlx5e_modify_tirs_packet_merge_ctx, NULL, reset); + err = mlx5e_safe_switch_params(priv, &new_params, NULL, NULL, reset); out: mutex_unlock(&priv->state_lock); return err; diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 5743f5b3414b..49a48d7709ac 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -3434,7 +3434,6 @@ enum { enum { MLX5_TIRC_PACKET_MERGE_MASK_IPV4_LRO = BIT(0), MLX5_TIRC_PACKET_MERGE_MASK_IPV6_LRO = BIT(1), - MLX5_TIRC_PACKET_MERGE_MASK_SHAMPO = BIT(2), }; enum { -- cgit v1.2.3 From 26183cfe478c1d1d5cd1e3920a4b2c5b1980849d Mon Sep 17 00:00:00 2001 From: Colin Foster Date: Tue, 8 Mar 2022 22:25:44 -0800 Subject: net: phy: correct spelling error of media in documentation The header file incorrectly referenced "median-independant interface" instead of media. Correct this typo. Signed-off-by: Colin Foster Fixes: 4069a572d423 ("net: phy: Document core PHY structures") Reviewed-by: Russell King (Oracle) Link: https://lore.kernel.org/r/20220309062544.3073-1-colin.foster@in-advantage.com Signed-off-by: Jakub Kicinski --- include/linux/phy.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/phy.h b/include/linux/phy.h index 6de8d7a90d78..8fa70ba063a5 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -87,8 +87,8 @@ extern const int phy_10gbit_features_array[1]; * * @PHY_INTERFACE_MODE_NA: Not Applicable - don't touch * @PHY_INTERFACE_MODE_INTERNAL: No interface, MAC and PHY combined - * @PHY_INTERFACE_MODE_MII: Median-independent interface - * @PHY_INTERFACE_MODE_GMII: Gigabit median-independent interface + * @PHY_INTERFACE_MODE_MII: Media-independent interface + * @PHY_INTERFACE_MODE_GMII: Gigabit media-independent interface * @PHY_INTERFACE_MODE_SGMII: Serial gigabit media-independent interface * @PHY_INTERFACE_MODE_TBI: Ten Bit Interface * @PHY_INTERFACE_MODE_REVMII: Reverse Media Independent Interface -- cgit v1.2.3 From c993ee0f9f81caf5767a50d1faeba39a0dc82af2 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 11 Mar 2022 13:23:31 +0000 Subject: watch_queue: Fix filter limit check In watch_queue_set_filter(), there are a couple of places where we check that the filter type value does not exceed what the type_filter bitmap can hold. One place calculates the number of bits by: if (tf[i].type >= sizeof(wfilter->type_filter) * 8) which is fine, but the second does: if (tf[i].type >= sizeof(wfilter->type_filter) * BITS_PER_LONG) which is not. This can lead to a couple of out-of-bounds writes due to a too-large type: (1) __set_bit() on wfilter->type_filter (2) Writing more elements in wfilter->filters[] than we allocated. Fix this by just using the proper WATCH_TYPE__NR instead, which is the number of types we actually know about. The bug may cause an oops looking something like: BUG: KASAN: slab-out-of-bounds in watch_queue_set_filter+0x659/0x740 Write of size 4 at addr ffff88800d2c66bc by task watch_queue_oob/611 ... Call Trace: dump_stack_lvl+0x45/0x59 print_address_description.constprop.0+0x1f/0x150 ... kasan_report.cold+0x7f/0x11b ... watch_queue_set_filter+0x659/0x740 ... __x64_sys_ioctl+0x127/0x190 do_syscall_64+0x43/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae Allocated by task 611: kasan_save_stack+0x1e/0x40 __kasan_kmalloc+0x81/0xa0 watch_queue_set_filter+0x23a/0x740 __x64_sys_ioctl+0x127/0x190 do_syscall_64+0x43/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae The buggy address belongs to the object at ffff88800d2c66a0 which belongs to the cache kmalloc-32 of size 32 The buggy address is located 28 bytes inside of 32-byte region [ffff88800d2c66a0, ffff88800d2c66c0) Fixes: c73be61cede5 ("pipe: Add general notification queue support") Reported-by: Jann Horn Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- include/linux/watch_queue.h | 3 ++- kernel/watch_queue.c | 4 ++-- 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/watch_queue.h b/include/linux/watch_queue.h index c994d1b2cdba..3b9a40ae8bdb 100644 --- a/include/linux/watch_queue.h +++ b/include/linux/watch_queue.h @@ -28,7 +28,8 @@ struct watch_type_filter { struct watch_filter { union { struct rcu_head rcu; - unsigned long type_filter[2]; /* Bitmask of accepted types */ + /* Bitmask of accepted types */ + DECLARE_BITMAP(type_filter, WATCH_TYPE__NR); }; u32 nr_filters; /* Number of filters */ struct watch_type_filter filters[]; diff --git a/kernel/watch_queue.c b/kernel/watch_queue.c index 9c9eb20dd2c5..427b0318e303 100644 --- a/kernel/watch_queue.c +++ b/kernel/watch_queue.c @@ -320,7 +320,7 @@ long watch_queue_set_filter(struct pipe_inode_info *pipe, tf[i].info_mask & WATCH_INFO_LENGTH) goto err_filter; /* Ignore any unknown types */ - if (tf[i].type >= sizeof(wfilter->type_filter) * 8) + if (tf[i].type >= WATCH_TYPE__NR) continue; nr_filter++; } @@ -336,7 +336,7 @@ long watch_queue_set_filter(struct pipe_inode_info *pipe, q = wfilter->filters; for (i = 0; i < filter.nr_filters; i++) { - if (tf[i].type >= sizeof(wfilter->type_filter) * BITS_PER_LONG) + if (tf[i].type >= WATCH_TYPE__NR) continue; q->type = tf[i].type; -- cgit v1.2.3 From 413a4a6b0b5553f2423d210f65e98c211b99c3f8 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 11 Mar 2022 16:02:18 +0000 Subject: cachefiles: Fix volume coherency attribute A network filesystem may set coherency data on a volume cookie, and if given, cachefiles will store this in an xattr on the directory in the cache corresponding to the volume. The function that sets the xattr just stores the contents of the volume coherency buffer directly into the xattr, with nothing added; the checking function, on the other hand, has a cut'n'paste error whereby it tries to interpret the xattr contents as would be the xattr on an ordinary file (using the cachefiles_xattr struct). This results in a failure to match the coherency data because the buffer ends up being shifted by 18 bytes. Fix this by defining a structure specifically for the volume xattr and making both the setting and checking functions use it. Since the volume coherency doesn't work if used, take the opportunity to insert a reserved field for future use, set it to 0 and check that it is 0. Log mismatch through the appropriate tracepoint. Note that this only affects cifs; 9p, afs, ceph and nfs don't use the volume coherency data at the moment. Fixes: 32e150037dce ("fscache, cachefiles: Store the volume coherency data") Reported-by: Rohith Surabattula Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Steve French cc: linux-cifs@vger.kernel.org cc: linux-cachefs@redhat.com Signed-off-by: Linus Torvalds --- fs/cachefiles/xattr.c | 23 ++++++++++++++++++++--- include/trace/events/cachefiles.h | 2 ++ 2 files changed, 22 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/fs/cachefiles/xattr.c b/fs/cachefiles/xattr.c index 83f41bd0c3a9..35465109d9c4 100644 --- a/fs/cachefiles/xattr.c +++ b/fs/cachefiles/xattr.c @@ -28,6 +28,11 @@ struct cachefiles_xattr { static const char cachefiles_xattr_cache[] = XATTR_USER_PREFIX "CacheFiles.cache"; +struct cachefiles_vol_xattr { + __be32 reserved; /* Reserved, should be 0 */ + __u8 data[]; /* netfs volume coherency data */ +} __packed; + /* * set the state xattr on a cache file */ @@ -185,6 +190,7 @@ void cachefiles_prepare_to_write(struct fscache_cookie *cookie) */ bool cachefiles_set_volume_xattr(struct cachefiles_volume *volume) { + struct cachefiles_vol_xattr *buf; unsigned int len = volume->vcookie->coherency_len; const void *p = volume->vcookie->coherency; struct dentry *dentry = volume->dentry; @@ -192,10 +198,17 @@ bool cachefiles_set_volume_xattr(struct cachefiles_volume *volume) _enter("%x,#%d", volume->vcookie->debug_id, len); + len += sizeof(*buf); + buf = kmalloc(len, GFP_KERNEL); + if (!buf) + return false; + buf->reserved = cpu_to_be32(0); + memcpy(buf->data, p, len); + ret = cachefiles_inject_write_error(); if (ret == 0) ret = vfs_setxattr(&init_user_ns, dentry, cachefiles_xattr_cache, - p, len, 0); + buf, len, 0); if (ret < 0) { trace_cachefiles_vfs_error(NULL, d_inode(dentry), ret, cachefiles_trace_setxattr_error); @@ -209,6 +222,7 @@ bool cachefiles_set_volume_xattr(struct cachefiles_volume *volume) cachefiles_coherency_vol_set_ok); } + kfree(buf); _leave(" = %d", ret); return ret == 0; } @@ -218,7 +232,7 @@ bool cachefiles_set_volume_xattr(struct cachefiles_volume *volume) */ int cachefiles_check_volume_xattr(struct cachefiles_volume *volume) { - struct cachefiles_xattr *buf; + struct cachefiles_vol_xattr *buf; struct dentry *dentry = volume->dentry; unsigned int len = volume->vcookie->coherency_len; const void *p = volume->vcookie->coherency; @@ -228,6 +242,7 @@ int cachefiles_check_volume_xattr(struct cachefiles_volume *volume) _enter(""); + len += sizeof(*buf); buf = kmalloc(len, GFP_KERNEL); if (!buf) return -ENOMEM; @@ -245,7 +260,9 @@ int cachefiles_check_volume_xattr(struct cachefiles_volume *volume) "Failed to read xattr with error %zd", xlen); } why = cachefiles_coherency_vol_check_xattr; - } else if (memcmp(buf->data, p, len) != 0) { + } else if (buf->reserved != cpu_to_be32(0)) { + why = cachefiles_coherency_vol_check_resv; + } else if (memcmp(buf->data, p, len - sizeof(*buf)) != 0) { why = cachefiles_coherency_vol_check_cmp; } else { why = cachefiles_coherency_vol_check_ok; diff --git a/include/trace/events/cachefiles.h b/include/trace/events/cachefiles.h index c6f5aa74db89..2c530637e10a 100644 --- a/include/trace/events/cachefiles.h +++ b/include/trace/events/cachefiles.h @@ -56,6 +56,7 @@ enum cachefiles_coherency_trace { cachefiles_coherency_set_ok, cachefiles_coherency_vol_check_cmp, cachefiles_coherency_vol_check_ok, + cachefiles_coherency_vol_check_resv, cachefiles_coherency_vol_check_xattr, cachefiles_coherency_vol_set_fail, cachefiles_coherency_vol_set_ok, @@ -139,6 +140,7 @@ enum cachefiles_error_trace { EM(cachefiles_coherency_set_ok, "SET ok ") \ EM(cachefiles_coherency_vol_check_cmp, "VOL BAD cmp ") \ EM(cachefiles_coherency_vol_check_ok, "VOL OK ") \ + EM(cachefiles_coherency_vol_check_resv, "VOL BAD resv") \ EM(cachefiles_coherency_vol_check_xattr,"VOL BAD xatt") \ EM(cachefiles_coherency_vol_set_fail, "VOL SET fail") \ E_(cachefiles_coherency_vol_set_ok, "VOL SET ok ") -- cgit v1.2.3