diff --git a/NEWS b/NEWS
index 3c3ab0d0d8c130b80acc10a3a1cd7160ae31d324..934ce76b54e7c34a725a36039c2dda47ec6a0db4 100644
--- a/NEWS
+++ b/NEWS
@@ -78,8 +78,6 @@ documents those changes that are of interest to users and administrators.
  -- Disable a configuration of SelectTypeParameters=CR_ONE_TASK_PER_CORE with
     SelectType=select/cons_tres. This will be addressed later.
  -- job_submit/lua - expose more fields off the partition record.
- -- task/cgroup - prevent setting a memory.soft_limit_in_bytes higher than the
-    memory.limit_in_bytes since the hard limit will take precedence anyway.
 
 * Changes in Slurm 19.05.0pre1
 ==============================
diff --git a/doc/man/man5/cgroup.conf.5 b/doc/man/man5/cgroup.conf.5
index 8d00ea91e0fede19f3e8fa294b58d3ce1ff470c5..477ef181191c73a82271008edc82baa9475e088a 100644
--- a/doc/man/man5/cgroup.conf.5
+++ b/doc/man/man5/cgroup.conf.5
@@ -57,19 +57,11 @@ it will be reset to that upper or lower limit, whichever has been exceeded.
 
 .TP
 \fBAllowedRAMSpace\fR=<number>
-Constrain the job/step cgroup RAM to this percentage of the allocated memory.
-The percentage supplied may be expressed as floating point number, e.g. 101.5.
-Sets the cgroup soft memory limit at the allocated memory size and then sets the
-job/step hard memory limit at the (AllowedRAMSpace/100) * allocated memory. If
-the job/step exceeds the hard limit, then it might trigger Out Of Memory (OOM)
-events (including oom-kill) which will be logged to kernel log ringbuffer
-(dmesg in Linux). Setting AllowedRAMSpace above 100 may cause system Out of
-Memory (OOM) events as it allows job/step to allocate more memory than
-configured to the nodes.  Reducing configured node available memory to avoid
-system OOM events is suggested.  Setting AllowedRAMSpace below 100 will result
-in jobs recieving less memory than allocated and soft memory limit will set to
-the same value as the hard limit.
-Also see \fBConstrainRAMSpace\fR.
+Constrain the job cgroup RAM to this percentage of the allocated memory.
+The percentage supplied may be expressed as floating point
+number, e.g. 98.5. If the \fBAllowedRAMSpace\fR limit is exceeded, the
+job steps will be killed and a warning message will be written to standard
+error.  Also see \fBConstrainRAMSpace\fR.
 The default value is 100.
 
 .TP
diff --git a/src/plugins/task/cgroup/task_cgroup_memory.c b/src/plugins/task/cgroup/task_cgroup_memory.c
index ba96d6dc09eb45b8240bf89fe292e2c9816bebe1..7c03b6d0e3bd85982d21f8edc91e04f5e0372dd4 100644
--- a/src/plugins/task/cgroup/task_cgroup_memory.c
+++ b/src/plugins/task/cgroup/task_cgroup_memory.c
@@ -367,16 +367,6 @@ static int memcg_initialize (xcgroup_ns_t *ns, xcgroup_t *cg,
 	uint64_t mlb_soft = mem_limit_in_bytes(mem_limit, false);
 	uint64_t mls = swap_limit_in_bytes  (mem_limit);
 
-	if (mlb_soft > mlb) {
-		/*
-		 * NOTE: It is recommended to set the soft limit always below
-		 * the hard limit, otherwise the hard one will take precedence.
-		 */
-		debug2("%s: Setting memory.soft_limit_in_bytes (%"PRIu64" bytes) to the same value as memory.limit_in_bytes (%"PRIu64" bytes) for cgroup: %s",
-		       __func__, mlb_soft, mlb, path);
-		mlb_soft = mlb;
-	}
-
 	if (xcgroup_create (ns, cg, path, uid, gid) != XCGROUP_SUCCESS)
 		return -1;