From: NeilBrown <neilb@suse.de>
Date: Tue, 16 Nov 2021 09:23:04 +1100
Subject: [PATCH] SUNRPC/xprt: async tasks mustn't block waiting for memory
Patch-mainline: Not yet - undergoing review
References: bsc#1191876 bsc#1192866

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: NeilBrown <neilb@suse.de>
Acked-by: NeilBrown <neilb@suse.com>

---
 net/sunrpc/xprt.c               |    5 ++++-
 net/sunrpc/xprtrdma/transport.c |    2 +-
 2 files changed, 5 insertions(+), 2 deletions(-)

--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -1140,12 +1140,15 @@ out:
 static struct rpc_rqst *xprt_dynamic_alloc_slot(struct rpc_xprt *xprt)
 {
 	struct rpc_rqst *req = ERR_PTR(-EAGAIN);
+	gfp_t gfp_mask = GFP_NOFS;
 
 	if (xprt->num_reqs >= xprt->max_reqs)
 		goto out;
 	++xprt->num_reqs;
 	spin_unlock(&xprt->reserve_lock);
-	req = kzalloc(sizeof(struct rpc_rqst), GFP_NOFS);
+	if (current->flags & PF_WQ_WORKER)
+		gfp_mask |= __GFP_NORETRY | __GFP_NOWARN;
+	req = kzalloc(sizeof(struct rpc_rqst), gfp_mask);
 	spin_lock(&xprt->reserve_lock);
 	if (req != NULL)
 		goto out;
--- a/net/sunrpc/xprtrdma/transport.c
+++ b/net/sunrpc/xprtrdma/transport.c
@@ -571,7 +571,7 @@ xprt_rdma_alloc_slot(struct rpc_xprt *xp
 	return;
 
 out_sleep:
-	task->tk_status = -EAGAIN;
+	task->tk_status = -ENOMEM;
 	xprt_add_backlog(xprt, task);
 }