In the Linux kernel, the following vulnerability has been resolved:
RDMA/iwcm: Fix WARNING:atkernel/workqueue.c:#checkflush_dependency
In the commit aee2424246f9 ("RDMA/iwcm: Fix a use-after-free related to destroying CM IDs"), the function flushworkqueue is invoked to flush the work queue iwcmwq.
But at that time, the work queue iwcmwq was created via the function allocorderedworkqueue without the flag WQMEM_RECLAIM.
Because the current process is trying to flush the whole iwcmwq, if iwcmwq doesn't have the flag WQMEMRECLAIM, verify that the current process is not reclaiming memory or running on a workqueue which doesn't have the flag WQMEMRECLAIM as that can break forward-progress guarantee leading to a deadlock.
The call trace is as below:
[ 125.350876][ T1430] Call Trace: [ 125.356281][ T1430] <TASK> [ 125.361285][ T1430] ? warn (kernel/panic.c:693) [ 125.367640][ T1430] ? checkflushdependency (kernel/workqueue.c:3706 (discriminator 9)) [ 125.375689][ T1430] ? reportbug (lib/bug.c:180 lib/bug.c:219) [ 125.382505][ T1430] ? handlebug (arch/x86/kernel/traps.c:239) [ 125.388987][ T1430] ? excinvalidop (arch/x86/kernel/traps.c:260 (discriminator 1)) [ 125.395831][ T1430] ? asmexcinvalidop (arch/x86/include/asm/idtentry.h:621) [ 125.403125][ T1430] ? checkflushdependency (kernel/workqueue.c:3706 (discriminator 9)) [ 125.410984][ T1430] ? checkflushdependency (kernel/workqueue.c:3706 (discriminator 9)) [ 125.418764][ T1430] _flushworkqueue (kernel/workqueue.c:3970) [ 125.426021][ T1430] ? _pfxmightresched (kernel/sched/core.c:10151) [ 125.433431][ T1430] ? destroycmid (drivers/infiniband/core/iwcm.c:375) iwcm [ 125.441209][ T1430] ? pfxflushworkqueue (kernel/workqueue.c:3910) [ 125.473900][ T1430] ? _rawspinlockirqsave (arch/x86/include/asm/atomic.h:107 include/linux/atomic/atomic-arch-fallback.h:2170 include/linux/atomic/atomic-instrumented.h:1302 include/asm-generic/qspinlock.h:111 include/linux/spinlock.h:187 include/linux/spinlockapismp.h:111 kernel/locking/spinlock.c:162) [ 125.473909][ T1430] ? pfxrawspinlockirqsave (kernel/locking/spinlock.c:161) [ 125.482537][ T1430] _destroyid (drivers/infiniband/core/cma.c:2044) rdmacm [ 125.495072][ T1430] nvmerdmafreequeue (drivers/nvme/host/rdma.c:656 drivers/nvme/host/rdma.c:650) nvmerdma [ 125.505827][ T1430] nvmerdmaresetctrlwork (drivers/nvme/host/rdma.c:2180) nvmerdma [ 125.505831][ T1430] processonework (kernel/workqueue.c:3231) [ 125.515122][ T1430] workerthread (kernel/workqueue.c:3306 kernel/workqueue.c:3393) [ 125.515127][ T1430] ? _pfxworkerthread (kernel/workqueue.c:3339) [ 125.531837][ T1430] kthread (kernel/kthread.c:389) [ 125.539864][ T1430] ? _pfxkthread (kernel/kthread.c:342) [ 125.550628][ T1430] retfromfork (arch/x86/kernel/process.c:147) [ 125.558840][ T1430] ? _pfxkthread (kernel/kthread.c:342) [ 125.558844][ T1430] retfromforkasm (arch/x86/entry/entry64.S:257) [ 125.566487][ T1430] </TASK> [ 125.566488][ T1430] ---[ end trace 0000000000000000 ]---