From 98801506552593c9b8ac11021b0cdad12cab4f6b Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 2 Jan 2018 10:02:19 +0000 Subject: fscache: Fix the default for fscache_maybe_release_page() Fix the default for fscache_maybe_release_page() for when the cookie isn't valid or the page isn't cached. It mustn't return false as that indicates the page cannot yet be freed. The problem with the default is that if, say, there's no cache, but a network filesystem's pages are using up almost all the available memory, a system can OOM because the filesystem ->releasepage() op will not allow them to be released as fscache_maybe_release_page() incorrectly prevents it. This can be tested by writing a sequence of 512MiB files to an AFS mount. It does not affect NFS or CIFS because both of those wrap the call in a check of PG_fscache and it shouldn't bother Ceph as that only has PG_private set whilst writeback is in progress. This might be an issue for 9P, however. Note that the pages aren't entirely stuck. Removing a file or unmounting will clear things because that uses ->invalidatepage() instead. Fixes: 201a15428bd5 ("FS-Cache: Handle pages pending storage that get evicted under OOM conditions") Reported-by: Marc Dionne Signed-off-by: David Howells Reviewed-by: Jeff Layton Acked-by: Al Viro Tested-by: Marc Dionne cc: stable@vger.kernel.org # 2.6.32+ --- include/linux/fscache.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fscache.h b/include/linux/fscache.h index f4ff47d4a893..fe0c349684fa 100644 --- a/include/linux/fscache.h +++ b/include/linux/fscache.h @@ -755,7 +755,7 @@ bool fscache_maybe_release_page(struct fscache_cookie *cookie, { if (fscache_cookie_valid(cookie) && PageFsCache(page)) return __fscache_maybe_release_page(cookie, page, gfp); - return false; + return true; } /** -- cgit v1.2.3-59-g8ed1b From f24c4d478013d82bd1b943df566fff3561d52864 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Tue, 2 Jan 2018 17:21:10 +0000 Subject: efi/capsule-loader: Reinstate virtual capsule mapping Commit: 82c3768b8d68 ("efi/capsule-loader: Use a cached copy of the capsule header") ... refactored the capsule loading code that maps the capsule header, to avoid having to map it several times. However, as it turns out, the vmap() call we ended up removing did not just map the header, but the entire capsule image, and dropping this virtual mapping breaks capsules that are processed by the firmware immediately (i.e., without a reboot). Unfortunately, that change was part of a larger refactor that allowed a quirk to be implemented for Quark, which has a non-standard memory layout for capsules, and we have slightly painted ourselves into a corner by allowing quirk code to mangle the capsule header and memory layout. So we need to fix this without breaking Quark. Fortunately, Quark does not appear to care about the virtual mapping, and so we can simply do a partial revert of commit: 2a457fb31df6 ("efi/capsule-loader: Use page addresses rather than struct page pointers") ... and create a vmap() mapping of the entire capsule (including header) based on the reinstated struct page array, unless running on Quark, in which case we pass the capsule header copy as before. Reported-by: Ge Song Tested-by: Bryan O'Donoghue Tested-by: Ge Song Signed-off-by: Ard Biesheuvel Cc: Cc: Dave Young Cc: Linus Torvalds Cc: Matt Fleming Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-efi@vger.kernel.org Fixes: 82c3768b8d68 ("efi/capsule-loader: Use a cached copy of the capsule header") Link: http://lkml.kernel.org/r/20180102172110.17018-3-ard.biesheuvel@linaro.org Signed-off-by: Ingo Molnar --- arch/x86/platform/efi/quirks.c | 13 +++++++++- drivers/firmware/efi/capsule-loader.c | 45 ++++++++++++++++++++++++++++------- include/linux/efi.h | 4 +++- 3 files changed, 52 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/platform/efi/quirks.c b/arch/x86/platform/efi/quirks.c index 8a99a2e96537..5b513ccffde4 100644 --- a/arch/x86/platform/efi/quirks.c +++ b/arch/x86/platform/efi/quirks.c @@ -592,7 +592,18 @@ static int qrk_capsule_setup_info(struct capsule_info *cap_info, void **pkbuff, /* * Update the first page pointer to skip over the CSH header. */ - cap_info->pages[0] += csh->headersize; + cap_info->phys[0] += csh->headersize; + + /* + * cap_info->capsule should point at a virtual mapping of the entire + * capsule, starting at the capsule header. Our image has the Quark + * security header prepended, so we cannot rely on the default vmap() + * mapping created by the generic capsule code. + * Given that the Quark firmware does not appear to care about the + * virtual mapping, let's just point cap_info->capsule at our copy + * of the capsule header. + */ + cap_info->capsule = &cap_info->header; return 1; } diff --git a/drivers/firmware/efi/capsule-loader.c b/drivers/firmware/efi/capsule-loader.c index ec8ac5c4dd84..055e2e8f985a 100644 --- a/drivers/firmware/efi/capsule-loader.c +++ b/drivers/firmware/efi/capsule-loader.c @@ -20,10 +20,6 @@ #define NO_FURTHER_WRITE_ACTION -1 -#ifndef phys_to_page -#define phys_to_page(x) pfn_to_page((x) >> PAGE_SHIFT) -#endif - /** * efi_free_all_buff_pages - free all previous allocated buffer pages * @cap_info: pointer to current instance of capsule_info structure @@ -35,7 +31,7 @@ static void efi_free_all_buff_pages(struct capsule_info *cap_info) { while (cap_info->index > 0) - __free_page(phys_to_page(cap_info->pages[--cap_info->index])); + __free_page(cap_info->pages[--cap_info->index]); cap_info->index = NO_FURTHER_WRITE_ACTION; } @@ -71,6 +67,14 @@ int __efi_capsule_setup_info(struct capsule_info *cap_info) cap_info->pages = temp_page; + temp_page = krealloc(cap_info->phys, + pages_needed * sizeof(phys_addr_t *), + GFP_KERNEL | __GFP_ZERO); + if (!temp_page) + return -ENOMEM; + + cap_info->phys = temp_page; + return 0; } @@ -105,9 +109,24 @@ int __weak efi_capsule_setup_info(struct capsule_info *cap_info, void *kbuff, **/ static ssize_t efi_capsule_submit_update(struct capsule_info *cap_info) { + bool do_vunmap = false; int ret; - ret = efi_capsule_update(&cap_info->header, cap_info->pages); + /* + * cap_info->capsule may have been assigned already by a quirk + * handler, so only overwrite it if it is NULL + */ + if (!cap_info->capsule) { + cap_info->capsule = vmap(cap_info->pages, cap_info->index, + VM_MAP, PAGE_KERNEL); + if (!cap_info->capsule) + return -ENOMEM; + do_vunmap = true; + } + + ret = efi_capsule_update(cap_info->capsule, cap_info->phys); + if (do_vunmap) + vunmap(cap_info->capsule); if (ret) { pr_err("capsule update failed\n"); return ret; @@ -165,10 +184,12 @@ static ssize_t efi_capsule_write(struct file *file, const char __user *buff, goto failed; } - cap_info->pages[cap_info->index++] = page_to_phys(page); + cap_info->pages[cap_info->index] = page; + cap_info->phys[cap_info->index] = page_to_phys(page); cap_info->page_bytes_remain = PAGE_SIZE; + cap_info->index++; } else { - page = phys_to_page(cap_info->pages[cap_info->index - 1]); + page = cap_info->pages[cap_info->index - 1]; } kbuff = kmap(page); @@ -252,6 +273,7 @@ static int efi_capsule_release(struct inode *inode, struct file *file) struct capsule_info *cap_info = file->private_data; kfree(cap_info->pages); + kfree(cap_info->phys); kfree(file->private_data); file->private_data = NULL; return 0; @@ -281,6 +303,13 @@ static int efi_capsule_open(struct inode *inode, struct file *file) return -ENOMEM; } + cap_info->phys = kzalloc(sizeof(void *), GFP_KERNEL); + if (!cap_info->phys) { + kfree(cap_info->pages); + kfree(cap_info); + return -ENOMEM; + } + file->private_data = cap_info; return 0; diff --git a/include/linux/efi.h b/include/linux/efi.h index d813f7b04da7..29fdf8029cf6 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -140,11 +140,13 @@ struct efi_boot_memmap { struct capsule_info { efi_capsule_header_t header; + efi_capsule_header_t *capsule; int reset_type; long index; size_t count; size_t total_size; - phys_addr_t *pages; + struct page **pages; + phys_addr_t *phys; size_t page_bytes_remain; }; -- cgit v1.2.3-59-g8ed1b From 040ee69226f8a96b7943645d68f41d5d44b5ff7d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 2 Dec 2017 20:20:38 -0500 Subject: fix "netfilter: xt_bpf: Fix XT_BPF_MODE_FD_PINNED mode of 'xt_bpf_info_v1'" Descriptor table is a shared object; it's not a place where you can stick temporary references to files, especially when we don't need an opened file at all. Cc: stable@vger.kernel.org # v4.14 Fixes: 98589a0998b8 ("netfilter: xt_bpf: Fix XT_BPF_MODE_FD_PINNED mode of 'xt_bpf_info_v1'") Signed-off-by: Al Viro --- include/linux/bpf.h | 10 ++++++++++ kernel/bpf/inode.c | 40 +++++++++++++++++++++++++++++++++++++++- kernel/bpf/syscall.c | 2 +- net/netfilter/xt_bpf.c | 14 ++------------ 4 files changed, 52 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index e55e4255a210..b63a592ad29d 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -419,6 +419,8 @@ static inline int bpf_map_attr_numa_node(const union bpf_attr *attr) attr->numa_node : NUMA_NO_NODE; } +struct bpf_prog *bpf_prog_get_type_path(const char *name, enum bpf_prog_type type); + #else /* !CONFIG_BPF_SYSCALL */ static inline struct bpf_prog *bpf_prog_get(u32 ufd) { @@ -506,6 +508,12 @@ static inline int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, { return 0; } + +static inline struct bpf_prog *bpf_prog_get_type_path(const char *name, + enum bpf_prog_type type) +{ + return ERR_PTR(-EOPNOTSUPP); +} #endif /* CONFIG_BPF_SYSCALL */ static inline struct bpf_prog *bpf_prog_get_type(u32 ufd, @@ -514,6 +522,8 @@ static inline struct bpf_prog *bpf_prog_get_type(u32 ufd, return bpf_prog_get_type_dev(ufd, type, false); } +bool bpf_prog_get_ok(struct bpf_prog *, enum bpf_prog_type *, bool); + int bpf_prog_offload_compile(struct bpf_prog *prog); void bpf_prog_offload_destroy(struct bpf_prog *prog); diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c index 01aaef1a77c5..5bb5e49ef4c3 100644 --- a/kernel/bpf/inode.c +++ b/kernel/bpf/inode.c @@ -368,7 +368,45 @@ out: putname(pname); return ret; } -EXPORT_SYMBOL_GPL(bpf_obj_get_user); + +static struct bpf_prog *__get_prog_inode(struct inode *inode, enum bpf_prog_type type) +{ + struct bpf_prog *prog; + int ret = inode_permission(inode, MAY_READ | MAY_WRITE); + if (ret) + return ERR_PTR(ret); + + if (inode->i_op == &bpf_map_iops) + return ERR_PTR(-EINVAL); + if (inode->i_op != &bpf_prog_iops) + return ERR_PTR(-EACCES); + + prog = inode->i_private; + + ret = security_bpf_prog(prog); + if (ret < 0) + return ERR_PTR(ret); + + if (!bpf_prog_get_ok(prog, &type, false)) + return ERR_PTR(-EINVAL); + + return bpf_prog_inc(prog); +} + +struct bpf_prog *bpf_prog_get_type_path(const char *name, enum bpf_prog_type type) +{ + struct bpf_prog *prog; + struct path path; + int ret = kern_path(name, LOOKUP_FOLLOW, &path); + if (ret) + return ERR_PTR(ret); + prog = __get_prog_inode(d_backing_inode(path.dentry), type); + if (!IS_ERR(prog)) + touch_atime(&path); + path_put(&path); + return prog; +} +EXPORT_SYMBOL(bpf_prog_get_type_path); static void bpf_evict_inode(struct inode *inode) { diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 2c4cfeaa8d5e..5cb783fc8224 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -1057,7 +1057,7 @@ struct bpf_prog *bpf_prog_inc_not_zero(struct bpf_prog *prog) } EXPORT_SYMBOL_GPL(bpf_prog_inc_not_zero); -static bool bpf_prog_get_ok(struct bpf_prog *prog, +bool bpf_prog_get_ok(struct bpf_prog *prog, enum bpf_prog_type *attach_type, bool attach_drv) { /* not an attachment, just a refcount inc, always allow */ diff --git a/net/netfilter/xt_bpf.c b/net/netfilter/xt_bpf.c index 041da0d9c06f..fa2ca0a13619 100644 --- a/net/netfilter/xt_bpf.c +++ b/net/netfilter/xt_bpf.c @@ -52,18 +52,8 @@ static int __bpf_mt_check_fd(int fd, struct bpf_prog **ret) static int __bpf_mt_check_path(const char *path, struct bpf_prog **ret) { - mm_segment_t oldfs = get_fs(); - int retval, fd; - - set_fs(KERNEL_DS); - fd = bpf_obj_get_user(path, 0); - set_fs(oldfs); - if (fd < 0) - return fd; - - retval = __bpf_mt_check_fd(fd, ret); - sys_close(fd); - return retval; + *ret = bpf_prog_get_type_path(path, BPF_PROG_TYPE_SOCKET_FILTER); + return PTR_ERR_OR_ZERO(*ret); } static int bpf_mt_check(const struct xt_mtchk_param *par) -- cgit v1.2.3-59-g8ed1b From 5133550296d43236439494aa955bfb765a89f615 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Thu, 4 Jan 2018 21:06:49 +0300 Subject: sh_eth: fix SH7757 GEther initialization Renesas SH7757 has 2 Fast and 2 Gigabit Ether controllers, while the 'sh_eth' driver can only reset and initialize TSU of the first controller pair. Shimoda-san tried to solve that adding the 'needs_init' member to the 'struct sh_eth_plat_data', however the platform code still never sets this flag. I think that we can infer this information from the 'devno' variable (set to 'platform_device::id') and reset/init the Ether controller pair only for an even 'devno'; therefore 'sh_eth_plat_data::needs_init' can be removed... Fixes: 150647fb2c31 ("net: sh_eth: change the condition of initialization") Signed-off-by: Sergei Shtylyov Signed-off-by: David S. Miller --- drivers/net/ethernet/renesas/sh_eth.c | 4 ++-- include/linux/sh_eth.h | 1 - 2 files changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index 1bdd67a8a869..f21c1db91c3f 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -3254,8 +3254,8 @@ static int sh_eth_drv_probe(struct platform_device *pdev) ndev->features = NETIF_F_HW_VLAN_CTAG_FILTER; } - /* initialize first or needed device */ - if (!devno || pd->needs_init) { + /* Need to init only the first port of the two sharing a TSU */ + if (devno % 2 == 0) { if (mdp->cd->chip_reset) mdp->cd->chip_reset(ndev); diff --git a/include/linux/sh_eth.h b/include/linux/sh_eth.h index ff3642d267f7..94081e9a5010 100644 --- a/include/linux/sh_eth.h +++ b/include/linux/sh_eth.h @@ -17,7 +17,6 @@ struct sh_eth_plat_data { unsigned char mac_addr[ETH_ALEN]; unsigned no_ether_link:1; unsigned ether_link_active_low:1; - unsigned needs_init:1; }; #endif -- cgit v1.2.3-59-g8ed1b From 87590ce6e373d1a5401f6539f0c59ef92dd924a9 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 7 Jan 2018 22:48:00 +0100 Subject: sysfs/cpu: Add vulnerability folder As the meltdown/spectre problem affects several CPU architectures, it makes sense to have common way to express whether a system is affected by a particular vulnerability or not. If affected the way to express the mitigation should be common as well. Create /sys/devices/system/cpu/vulnerabilities folder and files for meltdown, spectre_v1 and spectre_v2. Allow architectures to override the show function. Signed-off-by: Thomas Gleixner Reviewed-by: Greg Kroah-Hartman Reviewed-by: Konrad Rzeszutek Wilk Cc: Peter Zijlstra Cc: Will Deacon Cc: Dave Hansen Cc: Linus Torvalds Cc: Borislav Petkov Cc: David Woodhouse Link: https://lkml.kernel.org/r/20180107214913.096657732@linutronix.de --- Documentation/ABI/testing/sysfs-devices-system-cpu | 16 ++++++++ drivers/base/Kconfig | 3 ++ drivers/base/cpu.c | 48 ++++++++++++++++++++++ include/linux/cpu.h | 7 ++++ 4 files changed, 74 insertions(+) (limited to 'include/linux') diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu index f3d5817c4ef0..bd3a88e16d8b 100644 --- a/Documentation/ABI/testing/sysfs-devices-system-cpu +++ b/Documentation/ABI/testing/sysfs-devices-system-cpu @@ -373,3 +373,19 @@ Contact: Linux kernel mailing list Description: information about CPUs heterogeneity. cpu_capacity: capacity of cpu#. + +What: /sys/devices/system/cpu/vulnerabilities + /sys/devices/system/cpu/vulnerabilities/meltdown + /sys/devices/system/cpu/vulnerabilities/spectre_v1 + /sys/devices/system/cpu/vulnerabilities/spectre_v2 +Date: Januar 2018 +Contact: Linux kernel mailing list +Description: Information about CPU vulnerabilities + + The files are named after the code names of CPU + vulnerabilities. The output of those files reflects the + state of the CPUs in the system. Possible output values: + + "Not affected" CPU is not affected by the vulnerability + "Vulnerable" CPU is affected and no mitigation in effect + "Mitigation: $M" CPU is affetcted and mitigation $M is in effect diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index 2f6614c9a229..37a71fd9043f 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -235,6 +235,9 @@ config GENERIC_CPU_DEVICES config GENERIC_CPU_AUTOPROBE bool +config GENERIC_CPU_VULNERABILITIES + bool + config SOC_BUS bool select GLOB diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 321cd7b4d817..825964efda1d 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -501,10 +501,58 @@ static void __init cpu_dev_register_generic(void) #endif } +#ifdef CONFIG_GENERIC_CPU_VULNERABILITIES + +ssize_t __weak cpu_show_meltdown(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return sprintf(buf, "Not affected\n"); +} + +ssize_t __weak cpu_show_spectre_v1(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return sprintf(buf, "Not affected\n"); +} + +ssize_t __weak cpu_show_spectre_v2(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return sprintf(buf, "Not affected\n"); +} + +static DEVICE_ATTR(meltdown, 0444, cpu_show_meltdown, NULL); +static DEVICE_ATTR(spectre_v1, 0444, cpu_show_spectre_v1, NULL); +static DEVICE_ATTR(spectre_v2, 0444, cpu_show_spectre_v2, NULL); + +static struct attribute *cpu_root_vulnerabilities_attrs[] = { + &dev_attr_meltdown.attr, + &dev_attr_spectre_v1.attr, + &dev_attr_spectre_v2.attr, + NULL +}; + +static const struct attribute_group cpu_root_vulnerabilities_group = { + .name = "vulnerabilities", + .attrs = cpu_root_vulnerabilities_attrs, +}; + +static void __init cpu_register_vulnerabilities(void) +{ + if (sysfs_create_group(&cpu_subsys.dev_root->kobj, + &cpu_root_vulnerabilities_group)) + pr_err("Unable to register CPU vulnerabilities\n"); +} + +#else +static inline void cpu_register_vulnerabilities(void) { } +#endif + void __init cpu_dev_init(void) { if (subsys_system_register(&cpu_subsys, cpu_root_attr_groups)) panic("Failed to register CPU subsystem"); cpu_dev_register_generic(); + cpu_register_vulnerabilities(); } diff --git a/include/linux/cpu.h b/include/linux/cpu.h index 938ea8ae0ba4..c816e6f2730c 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -47,6 +47,13 @@ extern void cpu_remove_dev_attr(struct device_attribute *attr); extern int cpu_add_dev_attr_group(struct attribute_group *attrs); extern void cpu_remove_dev_attr_group(struct attribute_group *attrs); +extern ssize_t cpu_show_meltdown(struct device *dev, + struct device_attribute *attr, char *buf); +extern ssize_t cpu_show_spectre_v1(struct device *dev, + struct device_attribute *attr, char *buf); +extern ssize_t cpu_show_spectre_v2(struct device *dev, + struct device_attribute *attr, char *buf); + extern __printf(4, 5) struct device *cpu_device_create(struct device *parent, void *drvdata, const struct attribute_group **groups, -- cgit v1.2.3-59-g8ed1b From 527187d28569e39c5d489d6306d3b79605cf85a6 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 8 Jan 2018 17:27:19 +0100 Subject: locking/lockdep: Remove cross-release leftovers There's two cross-release leftover facilities: - the crossrelease_hist_*() irq-tracing callbacks (NOPs currently) - the complete_release_commit() callback (NOP as well) Remove them. Cc: David Sterba Cc: Byungchul Park Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar --- include/linux/completion.h | 1 - include/linux/irqflags.h | 4 ---- include/linux/lockdep.h | 2 -- kernel/sched/completion.c | 5 ----- 4 files changed, 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/completion.h b/include/linux/completion.h index 94a59ba7d422..519e94915d18 100644 --- a/include/linux/completion.h +++ b/include/linux/completion.h @@ -32,7 +32,6 @@ struct completion { #define init_completion(x) __init_completion(x) static inline void complete_acquire(struct completion *x) {} static inline void complete_release(struct completion *x) {} -static inline void complete_release_commit(struct completion *x) {} #define COMPLETION_INITIALIZER(work) \ { 0, __WAIT_QUEUE_HEAD_INITIALIZER((work).wait) } diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h index 46cb57d5eb13..1b3996ff3f16 100644 --- a/include/linux/irqflags.h +++ b/include/linux/irqflags.h @@ -27,22 +27,18 @@ # define trace_hardirq_enter() \ do { \ current->hardirq_context++; \ - crossrelease_hist_start(XHLOCK_HARD); \ } while (0) # define trace_hardirq_exit() \ do { \ current->hardirq_context--; \ - crossrelease_hist_end(XHLOCK_HARD); \ } while (0) # define lockdep_softirq_enter() \ do { \ current->softirq_context++; \ - crossrelease_hist_start(XHLOCK_SOFT); \ } while (0) # define lockdep_softirq_exit() \ do { \ current->softirq_context--; \ - crossrelease_hist_end(XHLOCK_SOFT); \ } while (0) # define INIT_TRACE_IRQFLAGS .softirqs_enabled = 1, #else diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 2e75dc34bff5..3251d9c0d313 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -475,8 +475,6 @@ enum xhlock_context_t { #define STATIC_LOCKDEP_MAP_INIT(_name, _key) \ { .name = (_name), .key = (void *)(_key), } -static inline void crossrelease_hist_start(enum xhlock_context_t c) {} -static inline void crossrelease_hist_end(enum xhlock_context_t c) {} static inline void lockdep_invariant_state(bool force) {} static inline void lockdep_init_task(struct task_struct *task) {} static inline void lockdep_free_task(struct task_struct *task) {} diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c index 2ddaec40956f..0926aef10dad 100644 --- a/kernel/sched/completion.c +++ b/kernel/sched/completion.c @@ -34,11 +34,6 @@ void complete(struct completion *x) spin_lock_irqsave(&x->wait.lock, flags); - /* - * Perform commit of crossrelease here. - */ - complete_release_commit(x); - if (x->done != UINT_MAX) x->done++; __wake_up_locked(&x->wait, TASK_NORMAL, 1); -- cgit v1.2.3-59-g8ed1b From b2157399cc9898260d6031c5bfe45fe137c1fbe7 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Sun, 7 Jan 2018 17:33:02 -0800 Subject: bpf: prevent out-of-bounds speculation Under speculation, CPUs may mis-predict branches in bounds checks. Thus, memory accesses under a bounds check may be speculated even if the bounds check fails, providing a primitive for building a side channel. To avoid leaking kernel data round up array-based maps and mask the index after bounds check, so speculated load with out of bounds index will load either valid value from the array or zero from the padded area. Unconditionally mask index for all array types even when max_entries are not rounded to power of 2 for root user. When map is created by unpriv user generate a sequence of bpf insns that includes AND operation to make sure that JITed code includes the same 'index & index_mask' operation. If prog_array map is created by unpriv user replace bpf_tail_call(ctx, map, index); with if (index >= max_entries) { index &= map->index_mask; bpf_tail_call(ctx, map, index); } (along with roundup to power 2) to prevent out-of-bounds speculation. There is secondary redundant 'if (index >= max_entries)' in the interpreter and in all JITs, but they can be optimized later if necessary. Other array-like maps (cpumap, devmap, sockmap, perf_event_array, cgroup_array) cannot be used by unpriv, so no changes there. That fixes bpf side of "Variant 1: bounds check bypass (CVE-2017-5753)" on all architectures with and without JIT. v2->v3: Daniel noticed that attack potentially can be crafted via syscall commands without loading the program, so add masking to those paths as well. Signed-off-by: Alexei Starovoitov Acked-by: John Fastabend Signed-off-by: Daniel Borkmann --- include/linux/bpf.h | 2 ++ kernel/bpf/arraymap.c | 47 ++++++++++++++++++++++++++++++++++++----------- kernel/bpf/verifier.c | 36 ++++++++++++++++++++++++++++++++++++ 3 files changed, 74 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index e55e4255a210..1b985ca4ffbe 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -52,6 +52,7 @@ struct bpf_map { u32 pages; u32 id; int numa_node; + bool unpriv_array; struct user_struct *user; const struct bpf_map_ops *ops; struct work_struct work; @@ -221,6 +222,7 @@ struct bpf_prog_aux { struct bpf_array { struct bpf_map map; u32 elem_size; + u32 index_mask; /* 'ownership' of prog_array is claimed by the first program that * is going to use this map or by the first program which FD is stored * in the map to make sure that all callers and callees have the same diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 7c25426d3cf5..aaa319848e7d 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -53,9 +53,10 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) { bool percpu = attr->map_type == BPF_MAP_TYPE_PERCPU_ARRAY; int numa_node = bpf_map_attr_numa_node(attr); + u32 elem_size, index_mask, max_entries; + bool unpriv = !capable(CAP_SYS_ADMIN); struct bpf_array *array; u64 array_size; - u32 elem_size; /* check sanity of attributes */ if (attr->max_entries == 0 || attr->key_size != 4 || @@ -72,11 +73,20 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) elem_size = round_up(attr->value_size, 8); + max_entries = attr->max_entries; + index_mask = roundup_pow_of_two(max_entries) - 1; + + if (unpriv) + /* round up array size to nearest power of 2, + * since cpu will speculate within index_mask limits + */ + max_entries = index_mask + 1; + array_size = sizeof(*array); if (percpu) - array_size += (u64) attr->max_entries * sizeof(void *); + array_size += (u64) max_entries * sizeof(void *); else - array_size += (u64) attr->max_entries * elem_size; + array_size += (u64) max_entries * elem_size; /* make sure there is no u32 overflow later in round_up() */ if (array_size >= U32_MAX - PAGE_SIZE) @@ -86,6 +96,8 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) array = bpf_map_area_alloc(array_size, numa_node); if (!array) return ERR_PTR(-ENOMEM); + array->index_mask = index_mask; + array->map.unpriv_array = unpriv; /* copy mandatory map attributes */ array->map.map_type = attr->map_type; @@ -121,12 +133,13 @@ static void *array_map_lookup_elem(struct bpf_map *map, void *key) if (unlikely(index >= array->map.max_entries)) return NULL; - return array->value + array->elem_size * index; + return array->value + array->elem_size * (index & array->index_mask); } /* emit BPF instructions equivalent to C code of array_map_lookup_elem() */ static u32 array_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) { + struct bpf_array *array = container_of(map, struct bpf_array, map); struct bpf_insn *insn = insn_buf; u32 elem_size = round_up(map->value_size, 8); const int ret = BPF_REG_0; @@ -135,7 +148,12 @@ static u32 array_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) *insn++ = BPF_ALU64_IMM(BPF_ADD, map_ptr, offsetof(struct bpf_array, value)); *insn++ = BPF_LDX_MEM(BPF_W, ret, index, 0); - *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 3); + if (map->unpriv_array) { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 4); + *insn++ = BPF_ALU32_IMM(BPF_AND, ret, array->index_mask); + } else { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 3); + } if (is_power_of_2(elem_size)) { *insn++ = BPF_ALU64_IMM(BPF_LSH, ret, ilog2(elem_size)); @@ -157,7 +175,7 @@ static void *percpu_array_map_lookup_elem(struct bpf_map *map, void *key) if (unlikely(index >= array->map.max_entries)) return NULL; - return this_cpu_ptr(array->pptrs[index]); + return this_cpu_ptr(array->pptrs[index & array->index_mask]); } int bpf_percpu_array_copy(struct bpf_map *map, void *key, void *value) @@ -177,7 +195,7 @@ int bpf_percpu_array_copy(struct bpf_map *map, void *key, void *value) */ size = round_up(map->value_size, 8); rcu_read_lock(); - pptr = array->pptrs[index]; + pptr = array->pptrs[index & array->index_mask]; for_each_possible_cpu(cpu) { bpf_long_memcpy(value + off, per_cpu_ptr(pptr, cpu), size); off += size; @@ -225,10 +243,11 @@ static int array_map_update_elem(struct bpf_map *map, void *key, void *value, return -EEXIST; if (array->map.map_type == BPF_MAP_TYPE_PERCPU_ARRAY) - memcpy(this_cpu_ptr(array->pptrs[index]), + memcpy(this_cpu_ptr(array->pptrs[index & array->index_mask]), value, map->value_size); else - memcpy(array->value + array->elem_size * index, + memcpy(array->value + + array->elem_size * (index & array->index_mask), value, map->value_size); return 0; } @@ -262,7 +281,7 @@ int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value, */ size = round_up(map->value_size, 8); rcu_read_lock(); - pptr = array->pptrs[index]; + pptr = array->pptrs[index & array->index_mask]; for_each_possible_cpu(cpu) { bpf_long_memcpy(per_cpu_ptr(pptr, cpu), value + off, size); off += size; @@ -613,6 +632,7 @@ static void *array_of_map_lookup_elem(struct bpf_map *map, void *key) static u32 array_of_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) { + struct bpf_array *array = container_of(map, struct bpf_array, map); u32 elem_size = round_up(map->value_size, 8); struct bpf_insn *insn = insn_buf; const int ret = BPF_REG_0; @@ -621,7 +641,12 @@ static u32 array_of_map_gen_lookup(struct bpf_map *map, *insn++ = BPF_ALU64_IMM(BPF_ADD, map_ptr, offsetof(struct bpf_array, value)); *insn++ = BPF_LDX_MEM(BPF_W, ret, index, 0); - *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 5); + if (map->unpriv_array) { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 6); + *insn++ = BPF_ALU32_IMM(BPF_AND, ret, array->index_mask); + } else { + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 5); + } if (is_power_of_2(elem_size)) *insn++ = BPF_ALU64_IMM(BPF_LSH, ret, ilog2(elem_size)); else diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 04b24876cd23..b414d6b2d470 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1729,6 +1729,13 @@ static int check_call(struct bpf_verifier_env *env, int func_id, int insn_idx) err = check_func_arg(env, BPF_REG_2, fn->arg2_type, &meta); if (err) return err; + if (func_id == BPF_FUNC_tail_call) { + if (meta.map_ptr == NULL) { + verbose(env, "verifier bug\n"); + return -EINVAL; + } + env->insn_aux_data[insn_idx].map_ptr = meta.map_ptr; + } err = check_func_arg(env, BPF_REG_3, fn->arg3_type, &meta); if (err) return err; @@ -4456,6 +4463,35 @@ static int fixup_bpf_calls(struct bpf_verifier_env *env) */ insn->imm = 0; insn->code = BPF_JMP | BPF_TAIL_CALL; + + /* instead of changing every JIT dealing with tail_call + * emit two extra insns: + * if (index >= max_entries) goto out; + * index &= array->index_mask; + * to avoid out-of-bounds cpu speculation + */ + map_ptr = env->insn_aux_data[i + delta].map_ptr; + if (map_ptr == BPF_MAP_PTR_POISON) { + verbose(env, "tail_call obusing map_ptr\n"); + return -EINVAL; + } + if (!map_ptr->unpriv_array) + continue; + insn_buf[0] = BPF_JMP_IMM(BPF_JGE, BPF_REG_3, + map_ptr->max_entries, 2); + insn_buf[1] = BPF_ALU32_IMM(BPF_AND, BPF_REG_3, + container_of(map_ptr, + struct bpf_array, + map)->index_mask); + insn_buf[2] = *insn; + cnt = 3; + new_prog = bpf_patch_insn_data(env, i + delta, insn_buf, cnt); + if (!new_prog) + return -ENOMEM; + + delta += cnt - 1; + env->prog = prog = new_prog; + insn = new_prog->insnsi + i + delta; continue; } -- cgit v1.2.3-59-g8ed1b From be95a845cc4402272994ce290e3ad928aff06cb9 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 9 Jan 2018 13:17:44 +0100 Subject: bpf: avoid false sharing of map refcount with max_entries In addition to commit b2157399cc98 ("bpf: prevent out-of-bounds speculation") also change the layout of struct bpf_map such that false sharing of fast-path members like max_entries is avoided when the maps reference counter is altered. Therefore enforce them to be placed into separate cachelines. pahole dump after change: struct bpf_map { const struct bpf_map_ops * ops; /* 0 8 */ struct bpf_map * inner_map_meta; /* 8 8 */ void * security; /* 16 8 */ enum bpf_map_type map_type; /* 24 4 */ u32 key_size; /* 28 4 */ u32 value_size; /* 32 4 */ u32 max_entries; /* 36 4 */ u32 map_flags; /* 40 4 */ u32 pages; /* 44 4 */ u32 id; /* 48 4 */ int numa_node; /* 52 4 */ bool unpriv_array; /* 56 1 */ /* XXX 7 bytes hole, try to pack */ /* --- cacheline 1 boundary (64 bytes) --- */ struct user_struct * user; /* 64 8 */ atomic_t refcnt; /* 72 4 */ atomic_t usercnt; /* 76 4 */ struct work_struct work; /* 80 32 */ char name[16]; /* 112 16 */ /* --- cacheline 2 boundary (128 bytes) --- */ /* size: 128, cachelines: 2, members: 17 */ /* sum members: 121, holes: 1, sum holes: 7 */ }; Now all entries in the first cacheline are read only throughout the life time of the map, set up once during map creation. Overall struct size and number of cachelines doesn't change from the reordering. struct bpf_map is usually first member and embedded in map structs in specific map implementations, so also avoid those members to sit at the end where it could potentially share the cacheline with first map values e.g. in the array since remote CPUs could trigger map updates just as well for those (easily dirtying members like max_entries intentionally as well) while having subsequent values in cache. Quoting from Google's Project Zero blog [1]: Additionally, at least on the Intel machine on which this was tested, bouncing modified cache lines between cores is slow, apparently because the MESI protocol is used for cache coherence [8]. Changing the reference counter of an eBPF array on one physical CPU core causes the cache line containing the reference counter to be bounced over to that CPU core, making reads of the reference counter on all other CPU cores slow until the changed reference counter has been written back to memory. Because the length and the reference counter of an eBPF array are stored in the same cache line, this also means that changing the reference counter on one physical CPU core causes reads of the eBPF array's length to be slow on other physical CPU cores (intentional false sharing). While this doesn't 'control' the out-of-bounds speculation through masking the index as in commit b2157399cc98, triggering a manipulation of the map's reference counter is really trivial, so lets not allow to easily affect max_entries from it. Splitting to separate cachelines also generally makes sense from a performance perspective anyway in that fast-path won't have a cache miss if the map gets pinned, reused in other progs, etc out of control path, thus also avoids unintentional false sharing. [1] https://googleprojectzero.blogspot.ch/2018/01/reading-privileged-memory-with-side.html Signed-off-by: Daniel Borkmann Signed-off-by: Alexei Starovoitov --- include/linux/bpf.h | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 1b985ca4ffbe..fe2cb7c398e3 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -43,7 +43,14 @@ struct bpf_map_ops { }; struct bpf_map { - atomic_t refcnt; + /* 1st cacheline with read-mostly members of which some + * are also accessed in fast-path (e.g. ops, max_entries). + */ + const struct bpf_map_ops *ops ____cacheline_aligned; + struct bpf_map *inner_map_meta; +#ifdef CONFIG_SECURITY + void *security; +#endif enum bpf_map_type map_type; u32 key_size; u32 value_size; @@ -53,15 +60,16 @@ struct bpf_map { u32 id; int numa_node; bool unpriv_array; - struct user_struct *user; - const struct bpf_map_ops *ops; - struct work_struct work; + /* 7 bytes hole */ + + /* 2nd cacheline with misc members to avoid false sharing + * particularly with refcounting. + */ + struct user_struct *user ____cacheline_aligned; + atomic_t refcnt; atomic_t usercnt; - struct bpf_map *inner_map_meta; + struct work_struct work; char name[BPF_OBJ_NAME_LEN]; -#ifdef CONFIG_SECURITY - void *security; -#endif }; /* function argument constraints */ -- cgit v1.2.3-59-g8ed1b From 67dcf8a3e06582cb6b02952335b5612beb97889f Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 5 Jan 2018 18:09:33 +0200 Subject: ACPI: utils: Introduce acpi_dev_get_first_match_name() Sometimes the user wants to have device name of the match rather than just checking if device present or not. To make life easier for such users introduce acpi_dev_get_first_match_name() helper based on code for acpi_dev_present(). For example, GPIO driver for Intel Merrifield needs to know the device name of pin control to be able to apply GPIO mapping table to the proper device. To be more consistent with the purpose rename struct acpi_dev_present_info -> struct acpi_dev_match_info acpi_dev_present_cb() -> acpi_dev_match_cb() in the utils.c file. Tested-by: Pierre-Louis Bossart Signed-off-by: Andy Shevchenko Signed-off-by: Rafael J. Wysocki --- drivers/acpi/utils.c | 41 ++++++++++++++++++++++++++++++++++------- include/acpi/acpi_bus.h | 3 +++ include/linux/acpi.h | 6 ++++++ 3 files changed, 43 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/utils.c b/drivers/acpi/utils.c index 9d49a1acebe3..78db97687f26 100644 --- a/drivers/acpi/utils.c +++ b/drivers/acpi/utils.c @@ -737,16 +737,17 @@ bool acpi_dev_found(const char *hid) } EXPORT_SYMBOL(acpi_dev_found); -struct acpi_dev_present_info { +struct acpi_dev_match_info { + const char *dev_name; struct acpi_device_id hid[2]; const char *uid; s64 hrv; }; -static int acpi_dev_present_cb(struct device *dev, void *data) +static int acpi_dev_match_cb(struct device *dev, void *data) { struct acpi_device *adev = to_acpi_device(dev); - struct acpi_dev_present_info *match = data; + struct acpi_dev_match_info *match = data; unsigned long long hrv; acpi_status status; @@ -757,6 +758,8 @@ static int acpi_dev_present_cb(struct device *dev, void *data) strcmp(adev->pnp.unique_id, match->uid))) return 0; + match->dev_name = acpi_dev_name(adev); + if (match->hrv == -1) return 1; @@ -789,20 +792,44 @@ static int acpi_dev_present_cb(struct device *dev, void *data) */ bool acpi_dev_present(const char *hid, const char *uid, s64 hrv) { - struct acpi_dev_present_info match = {}; + struct acpi_dev_match_info match = {}; struct device *dev; strlcpy(match.hid[0].id, hid, sizeof(match.hid[0].id)); match.uid = uid; match.hrv = hrv; - dev = bus_find_device(&acpi_bus_type, NULL, &match, - acpi_dev_present_cb); - + dev = bus_find_device(&acpi_bus_type, NULL, &match, acpi_dev_match_cb); return !!dev; } EXPORT_SYMBOL(acpi_dev_present); +/** + * acpi_dev_get_first_match_name - Return name of first match of ACPI device + * @hid: Hardware ID of the device. + * @uid: Unique ID of the device, pass NULL to not check _UID + * @hrv: Hardware Revision of the device, pass -1 to not check _HRV + * + * Return device name if a matching device was present + * at the moment of invocation, or NULL otherwise. + * + * See additional information in acpi_dev_present() as well. + */ +const char * +acpi_dev_get_first_match_name(const char *hid, const char *uid, s64 hrv) +{ + struct acpi_dev_match_info match = {}; + struct device *dev; + + strlcpy(match.hid[0].id, hid, sizeof(match.hid[0].id)); + match.uid = uid; + match.hrv = hrv; + + dev = bus_find_device(&acpi_bus_type, NULL, &match, acpi_dev_match_cb); + return dev ? match.dev_name : NULL; +} +EXPORT_SYMBOL(acpi_dev_get_first_match_name); + /* * acpi_backlight= handling, this is done here rather then in video_detect.c * because __setup cannot be used in modules. diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index 79287629c888..c9608b0b80c6 100644 --- a/include/acpi/acpi_bus.h +++ b/include/acpi/acpi_bus.h @@ -91,6 +91,9 @@ acpi_evaluate_dsm_typed(acpi_handle handle, const guid_t *guid, u64 rev, bool acpi_dev_found(const char *hid); bool acpi_dev_present(const char *hid, const char *uid, s64 hrv); +const char * +acpi_dev_get_first_match_name(const char *hid, const char *uid, s64 hrv); + #ifdef CONFIG_ACPI #include diff --git a/include/linux/acpi.h b/include/linux/acpi.h index dc1ebfeeb5ec..d918f1ea84e6 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -640,6 +640,12 @@ static inline bool acpi_dev_present(const char *hid, const char *uid, s64 hrv) return false; } +static inline const char * +acpi_dev_get_first_match_name(const char *hid, const char *uid, s64 hrv) +{ + return NULL; +} + static inline bool is_acpi_node(struct fwnode_handle *fwnode) { return false; -- cgit v1.2.3-59-g8ed1b From a0b1280368d1e91ab72f849ef095b4f07a39bbf1 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Fri, 12 Jan 2018 16:53:14 -0800 Subject: kdump: write correct address of mem_section into vmcoreinfo Depending on configuration mem_section can now be an array or a pointer to an array allocated dynamically. In most cases, we can continue to refer to it as 'mem_section' regardless of what it is. But there's one exception: '&mem_section' means "address of the array" if mem_section is an array, but if mem_section is a pointer, it would mean "address of the pointer". We've stepped onto this in kdump code. VMCOREINFO_SYMBOL(mem_section) writes down address of pointer into vmcoreinfo, not array as we wanted. Let's introduce VMCOREINFO_SYMBOL_ARRAY() that would handle the situation correctly for both cases. Link: http://lkml.kernel.org/r/20180112162532.35896-1-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov Fixes: 83e3c48729d9 ("mm/sparsemem: Allocate mem_section at runtime for CONFIG_SPARSEMEM_EXTREME=y") Acked-by: Baoquan He Acked-by: Dave Young Cc: Ingo Molnar Cc: Thomas Gleixner Cc: "H. Peter Anvin" Cc: Greg Kroah-Hartman Cc: Dave Young Cc: Baoquan He Cc: Vivek Goyal Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/crash_core.h | 2 ++ kernel/crash_core.c | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index 06097ef30449..b511f6d24b42 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -42,6 +42,8 @@ phys_addr_t paddr_vmcoreinfo_note(void); vmcoreinfo_append_str("PAGESIZE=%ld\n", value) #define VMCOREINFO_SYMBOL(name) \ vmcoreinfo_append_str("SYMBOL(%s)=%lx\n", #name, (unsigned long)&name) +#define VMCOREINFO_SYMBOL_ARRAY(name) \ + vmcoreinfo_append_str("SYMBOL(%s)=%lx\n", #name, (unsigned long)name) #define VMCOREINFO_SIZE(name) \ vmcoreinfo_append_str("SIZE(%s)=%lu\n", #name, \ (unsigned long)sizeof(name)) diff --git a/kernel/crash_core.c b/kernel/crash_core.c index b3663896278e..4f63597c824d 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -410,7 +410,7 @@ static int __init crash_save_vmcoreinfo_init(void) VMCOREINFO_SYMBOL(contig_page_data); #endif #ifdef CONFIG_SPARSEMEM - VMCOREINFO_SYMBOL(mem_section); + VMCOREINFO_SYMBOL_ARRAY(mem_section); VMCOREINFO_LENGTH(mem_section, NR_SECTION_ROOTS); VMCOREINFO_STRUCT_SIZE(mem_section); VMCOREINFO_OFFSET(mem_section, section_mem_map); -- cgit v1.2.3-59-g8ed1b From 06e8ccdab15f46dfd31292e2b75d744bc5fc2a7c Mon Sep 17 00:00:00 2001 From: Dave Jiang Date: Wed, 31 Jan 2018 12:45:38 -0700 Subject: acpi: nfit: Add support for detect platform CPU cache flush on power loss In ACPI 6.2a the platform capability structure has been added to the NFIT tables. That provides software the ability to determine whether a system supports the auto flushing of CPU caches on power loss. If the capability is supported, we do not need to do dax_flush(). Plumbing the path to set the property on per region from the NFIT tables. This patch depends on the ACPI NFIT 6.2a platform capabilities support code in include/acpi/actbl1.h. Signed-off-by: Dave Jiang Reviewed-by: Ross Zwisler Signed-off-by: Ross Zwisler --- drivers/acpi/nfit/core.c | 20 ++++++++++++++++++++ drivers/acpi/nfit/nfit.h | 1 + drivers/nvdimm/pmem.c | 4 +++- include/linux/libnvdimm.h | 5 +++++ 4 files changed, 29 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c index abeb4df4f22e..dc775823aea0 100644 --- a/drivers/acpi/nfit/core.c +++ b/drivers/acpi/nfit/core.c @@ -838,6 +838,18 @@ static bool add_flush(struct acpi_nfit_desc *acpi_desc, return true; } +static bool add_platform_cap(struct acpi_nfit_desc *acpi_desc, + struct acpi_nfit_capabilities *pcap) +{ + struct device *dev = acpi_desc->dev; + u32 mask; + + mask = (1 << (pcap->highest_capability + 1)) - 1; + acpi_desc->platform_cap = pcap->capabilities & mask; + dev_dbg(dev, "%s: cap: %#x\n", __func__, acpi_desc->platform_cap); + return true; +} + static void *add_table(struct acpi_nfit_desc *acpi_desc, struct nfit_table_prev *prev, void *table, const void *end) { @@ -883,6 +895,10 @@ static void *add_table(struct acpi_nfit_desc *acpi_desc, case ACPI_NFIT_TYPE_SMBIOS: dev_dbg(dev, "%s: smbios\n", __func__); break; + case ACPI_NFIT_TYPE_CAPABILITIES: + if (!add_platform_cap(acpi_desc, table)) + return err; + break; default: dev_err(dev, "unknown table '%d' parsing nfit\n", hdr->type); break; @@ -2656,6 +2672,9 @@ static int acpi_nfit_register_region(struct acpi_nfit_desc *acpi_desc, else ndr_desc->numa_node = NUMA_NO_NODE; + if(acpi_desc->platform_cap & ACPI_NFIT_CAPABILITY_CACHE_FLUSH) + set_bit(ND_REGION_PERSIST_CACHE, &ndr_desc->flags); + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) { struct acpi_nfit_memory_map *memdev = nfit_memdev->memdev; struct nd_mapping_desc *mapping; @@ -3464,6 +3483,7 @@ static __init int nfit_init(void) BUILD_BUG_ON(sizeof(struct acpi_nfit_smbios) != 9); BUILD_BUG_ON(sizeof(struct acpi_nfit_control_region) != 80); BUILD_BUG_ON(sizeof(struct acpi_nfit_data_region) != 40); + BUILD_BUG_ON(sizeof(struct acpi_nfit_capabilities) != 16); guid_parse(UUID_VOLATILE_MEMORY, &nfit_uuid[NFIT_SPA_VOLATILE]); guid_parse(UUID_PERSISTENT_MEMORY, &nfit_uuid[NFIT_SPA_PM]); diff --git a/drivers/acpi/nfit/nfit.h b/drivers/acpi/nfit/nfit.h index f0cf18b2da8b..50d36e166d70 100644 --- a/drivers/acpi/nfit/nfit.h +++ b/drivers/acpi/nfit/nfit.h @@ -202,6 +202,7 @@ struct acpi_nfit_desc { unsigned long dimm_cmd_force_en; unsigned long bus_cmd_force_en; unsigned long bus_nfit_cmd_force_en; + unsigned int platform_cap; int (*blk_do_io)(struct nd_blk_region *ndbr, resource_size_t dpa, void *iobuf, u64 len, int rw); }; diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c index 7fbc5c5dc8e1..8aa542398db4 100644 --- a/drivers/nvdimm/pmem.c +++ b/drivers/nvdimm/pmem.c @@ -35,6 +35,7 @@ #include "pmem.h" #include "pfn.h" #include "nd.h" +#include "nd-core.h" static struct device *to_dev(struct pmem_device *pmem) { @@ -334,7 +335,8 @@ static int pmem_attach_disk(struct device *dev, dev_warn(dev, "unable to guarantee persistence of writes\n"); fua = 0; } - wbc = nvdimm_has_cache(nd_region); + wbc = nvdimm_has_cache(nd_region) && + !test_bit(ND_REGION_PERSIST_CACHE, &nd_region->flags); if (!devm_request_mem_region(dev, res->start, resource_size(res), dev_name(&ndns->dev))) { diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h index f8109ddb5ef1..f2fc0da4da04 100644 --- a/include/linux/libnvdimm.h +++ b/include/linux/libnvdimm.h @@ -47,6 +47,11 @@ enum { /* region flag indicating to direct-map persistent memory by default */ ND_REGION_PAGEMAP = 0, + /* + * Platform ensures entire CPU store data path is flushed to pmem on + * system power loss. + */ + ND_REGION_PERSIST_CACHE = 1, /* mark newly adjusted resources as requiring a label update */ DPA_RESOURCE_ADJUSTED = 1 << 0, -- cgit v1.2.3-59-g8ed1b From 30e6d7bf29daa79d80711d35211c9b60894dbc44 Mon Sep 17 00:00:00 2001 From: Dave Jiang Date: Wed, 31 Jan 2018 12:45:43 -0700 Subject: acpi: nfit: add persistent memory control flag for nd_region Propagate the ADR attribute flag from the NFIT platform capabilities sub-table to nd_region. Signed-off-by: Dave Jiang Reviewed-by: Ross Zwisler Signed-off-by: Ross Zwisler --- drivers/acpi/nfit/core.c | 3 +++ include/linux/libnvdimm.h | 6 ++++++ 2 files changed, 9 insertions(+) (limited to 'include/linux') diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c index dc775823aea0..aa9d00db763a 100644 --- a/drivers/acpi/nfit/core.c +++ b/drivers/acpi/nfit/core.c @@ -2675,6 +2675,9 @@ static int acpi_nfit_register_region(struct acpi_nfit_desc *acpi_desc, if(acpi_desc->platform_cap & ACPI_NFIT_CAPABILITY_CACHE_FLUSH) set_bit(ND_REGION_PERSIST_CACHE, &ndr_desc->flags); + if (acpi_desc->platform_cap & ACPI_NFIT_CAPABILITY_MEM_FLUSH) + set_bit(ND_REGION_PERSIST_MEMCTRL, &ndr_desc->flags); + list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) { struct acpi_nfit_memory_map *memdev = nfit_memdev->memdev; struct nd_mapping_desc *mapping; diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h index f2fc0da4da04..ff855ed965fb 100644 --- a/include/linux/libnvdimm.h +++ b/include/linux/libnvdimm.h @@ -52,6 +52,12 @@ enum { * system power loss. */ ND_REGION_PERSIST_CACHE = 1, + /* + * Platform provides mechanisms to automatically flush outstanding + * write data from memory controler to pmem on system power loss. + * (ADR) + */ + ND_REGION_PERSIST_MEMCTRL = 2, /* mark newly adjusted resources as requiring a label update */ DPA_RESOURCE_ADJUSTED = 1 << 0, -- cgit v1.2.3-59-g8ed1b