Commit a7210354 authored by NeilBrown's avatar NeilBrown Committed by Trond Myklebust

SUNRPC/xprt: async tasks mustn't block waiting for memory

When memory is short, new worker threads cannot be created and we depend
on the minimum one rpciod thread to be able to handle everything.  So it
must not block waiting for memory.

xprt_dynamic_alloc_slot can block indefinitely.  This can tie up all
workqueue threads and NFS can deadlock.  So when called from a
workqueue, set __GFP_NORETRY.

The rdma alloc_slot already does not block.  However it sets the error
to -EAGAIN suggesting this will trigger a sleep.  It does not.  As we
can see in call_reserveresult(), only -ENOMEM causes a sleep.  -EAGAIN
causes immediate retry.
Signed-off-by: default avatarNeilBrown <neilb@suse.de>
Signed-off-by: default avatarTrond Myklebust <trond.myklebust@hammerspace.com>
parent a41b05ed
...@@ -1687,12 +1687,15 @@ static bool xprt_throttle_congested(struct rpc_xprt *xprt, struct rpc_task *task ...@@ -1687,12 +1687,15 @@ static bool xprt_throttle_congested(struct rpc_xprt *xprt, struct rpc_task *task
static struct rpc_rqst *xprt_dynamic_alloc_slot(struct rpc_xprt *xprt) static struct rpc_rqst *xprt_dynamic_alloc_slot(struct rpc_xprt *xprt)
{ {
struct rpc_rqst *req = ERR_PTR(-EAGAIN); struct rpc_rqst *req = ERR_PTR(-EAGAIN);
gfp_t gfp_mask = GFP_KERNEL;
if (xprt->num_reqs >= xprt->max_reqs) if (xprt->num_reqs >= xprt->max_reqs)
goto out; goto out;
++xprt->num_reqs; ++xprt->num_reqs;
spin_unlock(&xprt->reserve_lock); spin_unlock(&xprt->reserve_lock);
req = kzalloc(sizeof(struct rpc_rqst), GFP_KERNEL); if (current->flags & PF_WQ_WORKER)
gfp_mask |= __GFP_NORETRY | __GFP_NOWARN;
req = kzalloc(sizeof(*req), gfp_mask);
spin_lock(&xprt->reserve_lock); spin_lock(&xprt->reserve_lock);
if (req != NULL) if (req != NULL)
goto out; goto out;
......
...@@ -517,7 +517,7 @@ xprt_rdma_alloc_slot(struct rpc_xprt *xprt, struct rpc_task *task) ...@@ -517,7 +517,7 @@ xprt_rdma_alloc_slot(struct rpc_xprt *xprt, struct rpc_task *task)
return; return;
out_sleep: out_sleep:
task->tk_status = -EAGAIN; task->tk_status = -ENOMEM;
xprt_add_backlog(xprt, task); xprt_add_backlog(xprt, task);
} }
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment