Wind River Support Network

Description

In the Linux kernel, the following vulnerability has been resolved:[EOL][EOL]blk-mq: fix possible memleak when register 'hctx' failed[EOL][EOL]There's issue as follows when do fault injection test:[EOL]unreferenced object 0xffff888132a9f400 (size 512):[EOL]  comm "insmod", pid 308021, jiffies 4324277909 (age 509.733s)[EOL]  hex dump (first 32 bytes):[EOL]    00 00 00 00 00 00 00 00 08 f4 a9 32 81 88 ff ff  ...........2....[EOL]    08 f4 a9 32 81 88 ff ff 00 00 00 00 00 00 00 00  ...2............[EOL]  backtrace:[EOL]    [<00000000e8952bb4>] kmalloc_node_trace+0x22/0xa0[EOL]    [<00000000f9980e0f>] blk_mq_alloc_and_init_hctx+0x3f1/0x7e0[EOL]    [<000000002e719efa>] blk_mq_realloc_hw_ctxs+0x1e6/0x230[EOL]    [<000000004f1fda40>] blk_mq_init_allocated_queue+0x27e/0x910[EOL]    [<00000000287123ec>] __blk_mq_alloc_disk+0x67/0xf0[EOL]    [<00000000a2a34657>] 0xffffffffa2ad310f[EOL]    [<00000000b173f718>] 0xffffffffa2af824a[EOL]    [<0000000095a1dabb>] do_one_initcall+0x87/0x2a0[EOL]    [<00000000f32fdf93>] do_init_module+0xdf/0x320[EOL]    [<00000000cbe8541e>] load_module+0x3006/0x3390[EOL]    [<0000000069ed1bdb>] __do_sys_finit_module+0x113/0x1b0[EOL]    [<00000000a1a29ae8>] do_syscall_64+0x35/0x80[EOL]    [<000000009cd878b0>] entry_SYSCALL_64_after_hwframe+0x46/0xb0[EOL][EOL]Fault injection context as follows:[EOL] kobject_add[EOL] blk_mq_register_hctx[EOL] blk_mq_sysfs_register[EOL] blk_register_queue[EOL] device_add_disk[EOL] null_add_dev.part.0 [null_blk][EOL][EOL]As 'blk_mq_register_hctx' may already add some objects when failed halfway,[EOL]but there isn't do fallback, caller don't know which objects add failed.[EOL]To solve above issue just do fallback when add objects failed halfway in[EOL]'blk_mq_register_hctx'.