Linux kernel mirror (for testing) git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel os linux

drivers/base: fix userspace break from using bin_attributes for cpumap and cpulist

Using bin_attributes with a 0 size causes fstat and friends to return that
0 size. This breaks userspace code that retrieves the size before reading
the file. Rather than reverting 75bd50fa841 ("drivers/base/node.c: use
bin_attribute to break the size limitation of cpumap ABI") let's put in a
size value at compile time.

For cpulist the maximum size is on the order of
NR_CPUS * (ceil(log10(NR_CPUS)) + 1)/2

which for 8192 is 20480 (8192 * 5)/2. In order to get near that you'd need
a system with every other CPU on one node. For example: (0,2,4,8, ... ).
To simplify the math and support larger NR_CPUS in the future we are using
(NR_CPUS * 7)/2. We also set it to a min of PAGE_SIZE to retain the older
behavior for smaller NR_CPUS.

The cpumap file the size works out to be NR_CPUS/4 + NR_CPUS/32 - 1
(or NR_CPUS * 9/32 - 1) including the ","s.

Add a set of macros for these values to cpumask.h so they can be used in
multiple places. Apply these to the handful of such files in
drivers/base/topology.c as well as node.c.

As an example, on an 80 cpu 4-node system (NR_CPUS == 8192):

before:

-r--r--r--. 1 root root 0 Jul 12 14:08 system/node/node0/cpulist
-r--r--r--. 1 root root 0 Jul 11 17:25 system/node/node0/cpumap

after:

-r--r--r--. 1 root root 28672 Jul 13 11:32 system/node/node0/cpulist
-r--r--r--. 1 root root 4096 Jul 13 11:31 system/node/node0/cpumap

CONFIG_NR_CPUS = 16384
-r--r--r--. 1 root root 57344 Jul 13 14:03 system/node/node0/cpulist
-r--r--r--. 1 root root 4607 Jul 13 14:02 system/node/node0/cpumap

The actual number of cpus doesn't matter for the reported size since they
are based on NR_CPUS.

Fixes: 75bd50fa841d ("drivers/base/node.c: use bin_attribute to break the size limitation of cpumap ABI")
Fixes: bb9ec13d156e ("topology: use bin_attribute to break the size limitation of cpumap ABI")
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: Yury Norov <yury.norov@gmail.com>
Cc: stable@vger.kernel.org
Acked-by: Yury Norov <yury.norov@gmail.com> (for include/linux/cpumask.h)
Signed-off-by: Phil Auld <pauld@redhat.com>
Link: https://lore.kernel.org/r/20220715134924.3466194-1-pauld@redhat.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

authored by

Phil Auld and committed by
Greg Kroah-Hartman
7ee951ac 4a4e8f7f

+36 -18
+2 -2
drivers/base/node.c
··· 45 45 return n; 46 46 } 47 47 48 - static BIN_ATTR_RO(cpumap, 0); 48 + static BIN_ATTR_RO(cpumap, CPUMAP_FILE_MAX_BYTES); 49 49 50 50 static inline ssize_t cpulist_read(struct file *file, struct kobject *kobj, 51 51 struct bin_attribute *attr, char *buf, ··· 66 66 return n; 67 67 } 68 68 69 - static BIN_ATTR_RO(cpulist, 0); 69 + static BIN_ATTR_RO(cpulist, CPULIST_FILE_MAX_BYTES); 70 70 71 71 /** 72 72 * struct node_access_nodes - Access class device to hold user visible
+16 -16
drivers/base/topology.c
··· 62 62 static DEVICE_ATTR_ADMIN_RO(ppin); 63 63 64 64 define_siblings_read_func(thread_siblings, sibling_cpumask); 65 - static BIN_ATTR_RO(thread_siblings, 0); 66 - static BIN_ATTR_RO(thread_siblings_list, 0); 65 + static BIN_ATTR_RO(thread_siblings, CPUMAP_FILE_MAX_BYTES); 66 + static BIN_ATTR_RO(thread_siblings_list, CPULIST_FILE_MAX_BYTES); 67 67 68 68 define_siblings_read_func(core_cpus, sibling_cpumask); 69 - static BIN_ATTR_RO(core_cpus, 0); 70 - static BIN_ATTR_RO(core_cpus_list, 0); 69 + static BIN_ATTR_RO(core_cpus, CPUMAP_FILE_MAX_BYTES); 70 + static BIN_ATTR_RO(core_cpus_list, CPULIST_FILE_MAX_BYTES); 71 71 72 72 define_siblings_read_func(core_siblings, core_cpumask); 73 - static BIN_ATTR_RO(core_siblings, 0); 74 - static BIN_ATTR_RO(core_siblings_list, 0); 73 + static BIN_ATTR_RO(core_siblings, CPUMAP_FILE_MAX_BYTES); 74 + static BIN_ATTR_RO(core_siblings_list, CPULIST_FILE_MAX_BYTES); 75 75 76 76 #ifdef TOPOLOGY_CLUSTER_SYSFS 77 77 define_siblings_read_func(cluster_cpus, cluster_cpumask); 78 - static BIN_ATTR_RO(cluster_cpus, 0); 79 - static BIN_ATTR_RO(cluster_cpus_list, 0); 78 + static BIN_ATTR_RO(cluster_cpus, CPUMAP_FILE_MAX_BYTES); 79 + static BIN_ATTR_RO(cluster_cpus_list, CPULIST_FILE_MAX_BYTES); 80 80 #endif 81 81 82 82 #ifdef TOPOLOGY_DIE_SYSFS 83 83 define_siblings_read_func(die_cpus, die_cpumask); 84 - static BIN_ATTR_RO(die_cpus, 0); 85 - static BIN_ATTR_RO(die_cpus_list, 0); 84 + static BIN_ATTR_RO(die_cpus, CPUMAP_FILE_MAX_BYTES); 85 + static BIN_ATTR_RO(die_cpus_list, CPULIST_FILE_MAX_BYTES); 86 86 #endif 87 87 88 88 define_siblings_read_func(package_cpus, core_cpumask); 89 - static BIN_ATTR_RO(package_cpus, 0); 90 - static BIN_ATTR_RO(package_cpus_list, 0); 89 + static BIN_ATTR_RO(package_cpus, CPUMAP_FILE_MAX_BYTES); 90 + static BIN_ATTR_RO(package_cpus_list, CPULIST_FILE_MAX_BYTES); 91 91 92 92 #ifdef TOPOLOGY_BOOK_SYSFS 93 93 define_id_show_func(book_id, "%d"); 94 94 static DEVICE_ATTR_RO(book_id); 95 95 define_siblings_read_func(book_siblings, book_cpumask); 96 - static BIN_ATTR_RO(book_siblings, 0); 97 - static BIN_ATTR_RO(book_siblings_list, 0); 96 + static BIN_ATTR_RO(book_siblings, CPUMAP_FILE_MAX_BYTES); 97 + static BIN_ATTR_RO(book_siblings_list, CPULIST_FILE_MAX_BYTES); 98 98 #endif 99 99 100 100 #ifdef TOPOLOGY_DRAWER_SYSFS 101 101 define_id_show_func(drawer_id, "%d"); 102 102 static DEVICE_ATTR_RO(drawer_id); 103 103 define_siblings_read_func(drawer_siblings, drawer_cpumask); 104 - static BIN_ATTR_RO(drawer_siblings, 0); 105 - static BIN_ATTR_RO(drawer_siblings_list, 0); 104 + static BIN_ATTR_RO(drawer_siblings, CPUMAP_FILE_MAX_BYTES); 105 + static BIN_ATTR_RO(drawer_siblings_list, CPULIST_FILE_MAX_BYTES); 106 106 #endif 107 107 108 108 static struct bin_attribute *bin_attrs[] = {
+18
include/linux/cpumask.h
··· 1071 1071 [0] = 1UL \ 1072 1072 } } 1073 1073 1074 + /* 1075 + * Provide a valid theoretical max size for cpumap and cpulist sysfs files 1076 + * to avoid breaking userspace which may allocate a buffer based on the size 1077 + * reported by e.g. fstat. 1078 + * 1079 + * for cpumap NR_CPUS * 9/32 - 1 should be an exact length. 1080 + * 1081 + * For cpulist 7 is (ceil(log10(NR_CPUS)) + 1) allowing for NR_CPUS to be up 1082 + * to 2 orders of magnitude larger than 8192. And then we divide by 2 to 1083 + * cover a worst-case of every other cpu being on one of two nodes for a 1084 + * very large NR_CPUS. 1085 + * 1086 + * Use PAGE_SIZE as a minimum for smaller configurations. 1087 + */ 1088 + #define CPUMAP_FILE_MAX_BYTES ((((NR_CPUS * 9)/32 - 1) > PAGE_SIZE) \ 1089 + ? (NR_CPUS * 9)/32 - 1 : PAGE_SIZE) 1090 + #define CPULIST_FILE_MAX_BYTES (((NR_CPUS * 7)/2 > PAGE_SIZE) ? (NR_CPUS * 7)/2 : PAGE_SIZE) 1091 + 1074 1092 #endif /* __LINUX_CPUMASK_H */