diff --git a/NEWS b/NEWS
index fc3ec021c5f080af4b36f07514f1db715fd19b1e..c617b980b832a2836095de79156096302758cc89 100644
--- a/NEWS
+++ b/NEWS
@@ -215,6 +215,10 @@ documents those changes that are of interest to users and administrators.
had --network= specified.
-- Defer job step initiation of required GRES are in use by other steps rather
than immediately returning an error.
+ -- Deprecate --cpu_bind from sbatch and salloc. These never worked correctly
+ and only caused confusion since the cpu_bind options mostly refer to a
+ step we opted to only allow srun to set them in future versions.
+ -- Modify sgather to work if Nodename and NodeHostname differ.
* Changes in Slurm 14.03.7
==========================
diff --git a/contribs/sgather/sgather b/contribs/sgather/sgather
index 27a1132f23d702c7c0a5528844864cf7a3b543c0..af2739488d451766e53e9a03562b297d0b3ef2db 100755
--- a/contribs/sgather/sgather
+++ b/contribs/sgather/sgather
@@ -241,8 +241,7 @@ if [ -z "$_SGATHER_SPAWNED" ]; then
# ... or in multiple steps with regard to $_SGATHER_FANOUT
#
else
- nodelist=$($SRUN --ntasks=$SLURM_NNODES --ntasks-per-node=1 -l hostname) || exit $?
- nodelist=$(echo "$nodelist" | cut -d ' ' -f 2 | sort)
+ nodelist=$($SCONTROL show hostnames $SLURM_NODELIST | sort)
set $nodelist
nodesublist=""
nodesubcnt=0
diff --git a/doc/man/man1/salloc.1 b/doc/man/man1/salloc.1
index 3ff5dc021908019f660dc767cda8c67ba9a8c0a3..349aa6897ead335b77ee8e16122ec46d2c69957c 100644
--- a/doc/man/man1/salloc.1
+++ b/doc/man/man1/salloc.1
@@ -114,10 +114,6 @@ options if desired:
\fB\-\-cores\-per\-socket\fR=<\fIcores\fR>
\fB\-\-threads\-per\-core\fR=<\fIthreads\fR>
.fi
-If task/affinity plugin is enabled, then specifying an allocation in this
-manner also sets a default \fB\-\-cpu_bind\fR option of \fIthreads\fR
-if the \fB\-B\fR option specifies a thread count, otherwise an option of
-\fIcores\fR if a core count is specified, otherwise an option of \fIsockets\fR.
If SelectType is configured to select/cons_res, it must have a parameter of
CR_Core, CR_Core_Memory, CR_Socket, or CR_Socket_Memory for this option
to be honored.
@@ -243,125 +239,6 @@ Restrict node selection to nodes with at least the specified number of
cores per socket. See additional information under \fB\-B\fR option
above when task/affinity plugin is enabled.
-.TP
-\fB\-\-cpu_bind\fR=[{\fIquiet,verbose\fR},]\fItype\fR
-Bind tasks to CPUs.
-Used only when the task/affinity or task/cgroup plugin is enabled.
-The configuration parameter \fBTaskPluginParam\fR may override these options.
-For example, if \fBTaskPluginParam\fR is configured to bind to cores,
-your job will not be able to bind tasks to sockets.
-NOTE: To have SLURM always report on the selected CPU binding for all
-commands executed in a shell, you can enable verbose mode by setting
-the SLURM_CPU_BIND environment variable value to "verbose".
-
-The following informational environment variables are set when \fB\-\-cpu_bind\fR
-is in use:
-.nf
- SLURM_CPU_BIND_VERBOSE
- SLURM_CPU_BIND_TYPE
- SLURM_CPU_BIND_LIST
-.fi
-
-See the \fBENVIRONMENT VARIABLE\fR section for a more detailed description
-of the individual SLURM_CPU_BIND* variables.
-
-When using \fB\-\-cpus\-per\-task\fR to run multithreaded tasks, be aware that
-CPU binding is inherited from the parent of the process. This means that
-the multithreaded task should either specify or clear the CPU binding
-itself to avoid having all threads of the multithreaded task use the same
-mask/CPU as the parent. Alternatively, fat masks (masks which specify more
-than one allowed CPU) could be used for the tasks in order to provide
-multiple CPUs for the multithreaded tasks.
-
-By default, a job step has access to every CPU allocated to the job.
-To ensure that distinct CPUs are allocated to each job step, use the
-\fB\-\-exclusive\fR option.
-
-If the job step allocation includes an allocation with a number of
-sockets, cores, or threads equal to the number of tasks to be started
-then the tasks will by default be bound to the appropriate resources.
-Disable this mode of operation by explicitly setting "-\-cpu_bind=none".
-
-Note that a job step can be allocated different numbers of CPUs on each node
-or be allocated CPUs not starting at location zero. Therefore one of the
-options which automatically generate the task binding is recommended.
-Explicitly specified masks or bindings are only honored when the job step
-has been allocated every available CPU on the node.
-
-Binding a task to a NUMA locality domain means to bind the task to the set of
-CPUs that belong to the NUMA locality domain or "NUMA node".
-If NUMA locality domain options are used on systems with no NUMA support, then
-each socket is considered a locality domain.
-
-Supported options include:
-.PD 1
-.RS
-.TP
-.B q[uiet]
-Quietly bind before task runs (default)
-.TP
-.B v[erbose]
-Verbosely report binding before task runs
-.TP
-.B no[ne]
-Do not bind tasks to CPUs (default)
-.TP
-.B rank
-Automatically bind by task rank.
-The lowest numbered task on each node is bound to socket (or core or thread) zero, etc.
-Not supported unless the entire node is allocated to the job.
-.TP
-.B map_cpu:<list>
-Bind by mapping CPU IDs to tasks as specified
-where <list> is <cpuid1>,<cpuid2>,...<cpuidN>.
-The mapping is specified for a node and identical mapping is applied to the
-tasks on every node (i.e. the lowest task ID on each node is mapped to the
-first CPU ID specified in the list, etc.).
-CPU IDs are interpreted as decimal values unless they are preceded
-with '0x' in which case they are interpreted as hexadecimal values.
-Not supported unless the entire node is allocated to the job.
-.TP
-.B mask_cpu:<list>
-Bind by setting CPU masks on tasks as specified
-where <list> is <mask1>,<mask2>,...<maskN>.
-The mapping is specified for a node and identical mapping is applied to the
-tasks on every node (i.e. the lowest task ID on each node is mapped to the
-first mask specified in the list, etc.).
-CPU masks are \fBalways\fR interpreted as hexadecimal values but can be
-preceded with an optional '0x'. Not supported unless the entire node is
-allocated to the job.
-.TP
-.B sockets
-Automatically generate masks binding tasks to sockets.
-Only the CPUs on the socket which have been allocated to the job will be used.
-If the number of tasks differs from the number of allocated sockets
-this can result in sub\-optimal binding.
-.TP
-.B cores
-Automatically generate masks binding tasks to cores.
-If the number of tasks differs from the number of allocated cores
-this can result in sub\-optimal binding.
-.TP
-.B threads
-Automatically generate masks binding tasks to threads.
-If the number of tasks differs from the number of allocated threads
-this can result in sub\-optimal binding.
-.TP
-.B ldoms
-Automatically generate masks binding tasks to NUMA locality domains.
-If the number of tasks differs from the number of allocated locality domains
-this can result in sub\-optimal binding.
-.TP
-.B boards
-Automatically generate masks binding tasks to boards.
-If the number of tasks differs from the number of allocated boards
-this can result in sub\-optimal binding. This option is supported by the
-task/cgroup plugin only.
-.TP
-.B help
-Show this help message
-.RE
-
.TP
\fB\-c\fR, \fB\-\-cpus\-per\-task\fR=<\fIncpus\fR>
Advise the SLURM controller that ensuing job steps will require \fIncpus\fR
@@ -719,7 +596,7 @@ be performed at the level of nodes, where the definition of "nodes"
may differ from system to system. \fBThe use of any type other than
"none" or "local" is not recommended.\fR
If you want greater control, try running a simple test code with the
-options "\-\-cpu_bind=verbose,none \-\-mem_bind=verbose,none" to determine
+options "\-\-mem_bind=verbose,none" to determine
the specific configuration.
NOTE: To have SLURM always report on the selected memory binding for
@@ -969,9 +846,7 @@ ignored if \fISchedulerType=sched/wiki\fR or
Request the maximum \fIntasks\fR be invoked on each core.
Meant to be used with the \fB\-\-ntasks\fR option.
Related to \fB\-\-ntasks\-per\-node\fR except at the core level
-instead of the node level. Masks will automatically be generated
-to bind the tasks to specific core unless \fB\-\-cpu_bind=none\fR
-is specified.
+instead of the node level.
NOTE: This option is not supported unless
\fISelectTypeParameters=CR_Core\fR or
\fISelectTypeParameters=CR_Core_Memory\fR is configured.
@@ -981,9 +856,7 @@ NOTE: This option is not supported unless
Request the maximum \fIntasks\fR be invoked on each socket.
Meant to be used with the \fB\-\-ntasks\fR option.
Related to \fB\-\-ntasks\-per\-node\fR except at the socket level
-instead of the node level. Masks will automatically be generated
-to bind the tasks to specific sockets unless \fB\-\-cpu_bind=none\fR
-is specified.
+instead of the node level.
NOTE: This option is not supported unless
\fISelectTypeParameters=CR_Socket\fR or
\fISelectTypeParameters=CR_Socket_Memory\fR is configured.
@@ -1361,9 +1234,6 @@ Same as \fB\-\-conn\-type\fR
\fBSALLOC_CORE_SPEC\fR
Same as \fB\-\-core\-spec\fR
.TP
-\fBSALLOC_CPU_BIND\fR
-Same as \fB\-\-cpu_bind\fR
-.TP
\fBSALLOC_DEBUG\fR
Same as \fB\-v, \-\-verbose\fR
.TP
@@ -1454,15 +1324,6 @@ the executed program:
\fBBASIL_RESERVATION_ID\fR
The reservation ID on Cray systems running ALPS/BASIL only.
.TP
-\fBSLURM_CPU_BIND\fR
-Set to value of the \-\-cpu_bind\fR option.
-.TP
-\fBSLURM_CPU_BIND_LIST\fR
-\-\-cpu_bind map or mask list (list of SLURM CPU IDs or masks for this node,
-CPU_ID = Board_ID x threads_per_board +
-Socket_ID x threads_per_socket +
-Core_ID x threads_per_core + Thread_ID).
-.TP
\fBSLURM_DISTRIBUTION\fR
Same as \fB\-m, \-\-distribution\fR
.TP
diff --git a/doc/man/man1/sbatch.1 b/doc/man/man1/sbatch.1
index cb6fc34ab42799c8cf7392eba7c9965c68a45170..ac0abcb8b5d1bc465d6ca2aef900fec76cae9206 100644
--- a/doc/man/man1/sbatch.1
+++ b/doc/man/man1/sbatch.1
@@ -133,10 +133,6 @@ options if desired:
\fB\-\-cores\-per\-socket\fR=<\fIcores\fR>
\fB\-\-threads\-per\-core\fR=<\fIthreads\fR>
.fi
-If task/affinity plugin is enabled, then specifying an allocation in this
-manner also sets a default \fB\-\-cpu_bind\fR option of \fIthreads\fR
-if the \fB\-B\fR option specifies a thread count, otherwise an option of
-\fIcores\fR if a core count is specified, otherwise an option of \fIsockets\fR.
If SelectType is configured to select/cons_res, it must have a parameter of
CR_Core, CR_Core_Memory, CR_Socket, or CR_Socket_Memory for this option
to be honored.
@@ -272,125 +268,6 @@ Restrict node selection to nodes with at least the specified number of
cores per socket. See additional information under \fB\-B\fR option
above when task/affinity plugin is enabled.
-.TP
-\fB\-\-cpu_bind\fR=[{\fIquiet,verbose\fR},]\fItype\fR
-Bind tasks to CPUs.
-Used only when the task/affinity or task/cgroup plugin is enabled.
-The configuration parameter \fBTaskPluginParam\fR may override these options.
-For example, if \fBTaskPluginParam\fR is configured to bind to cores,
-your job will not be able to bind tasks to sockets.
-NOTE: To have SLURM always report on the selected CPU binding for all
-commands executed in a shell, you can enable verbose mode by setting
-the SLURM_CPU_BIND environment variable value to "verbose".
-
-The following informational environment variables are set when \fB\-\-cpu_bind\fR
-is in use:
-.nf
- SLURM_CPU_BIND_VERBOSE
- SLURM_CPU_BIND_TYPE
- SLURM_CPU_BIND_LIST
-.fi
-
-See the \fBENVIRONMENT VARIABLE\fR section for a more detailed description
-of the individual SLURM_CPU_BIND* variables.
-
-When using \fB\-\-cpus\-per\-task\fR to run multithreaded tasks, be aware that
-CPU binding is inherited from the parent of the process. This means that
-the multithreaded task should either specify or clear the CPU binding
-itself to avoid having all threads of the multithreaded task use the same
-mask/CPU as the parent. Alternatively, fat masks (masks which specify more
-than one allowed CPU) could be used for the tasks in order to provide
-multiple CPUs for the multithreaded tasks.
-
-By default, a job step has access to every CPU allocated to the job.
-To ensure that distinct CPUs are allocated to each job step, use the
-\fB\-\-exclusive\fR option.
-
-If the job step allocation includes an allocation with a number of
-sockets, cores, or threads equal to the number of tasks to be started
-then the tasks will by default be bound to the appropriate resources.
-Disable this mode of operation by explicitly setting "-\-cpu_bind=none".
-
-Note that a job step can be allocated different numbers of CPUs on each node
-or be allocated CPUs not starting at location zero. Therefore one of the
-options which automatically generate the task binding is recommended.
-Explicitly specified masks or bindings are only honored when the job step
-has been allocated every available CPU on the node.
-
-Binding a task to a NUMA locality domain means to bind the task to the set of
-CPUs that belong to the NUMA locality domain or "NUMA node".
-If NUMA locality domain options are used on systems with no NUMA support, then
-each socket is considered a locality domain.
-
-Supported options include:
-.PD 1
-.RS
-.TP
-.B q[uiet]
-Quietly bind before task runs (default)
-.TP
-.B v[erbose]
-Verbosely report binding before task runs
-.TP
-.B no[ne]
-Do not bind tasks to CPUs (default)
-.TP
-.B rank
-Automatically bind by task rank.
-The lowest numbered task on each node is bound to socket (or core or thread) zero, etc.
-Not supported unless the entire node is allocated to the job.
-.TP
-.B map_cpu:<list>
-Bind by mapping CPU IDs to tasks as specified
-where <list> is <cpuid1>,<cpuid2>,...<cpuidN>.
-The mapping is specified for a node and identical mapping is applied to the
-tasks on every node (i.e. the lowest task ID on each node is mapped to the
-first CPU ID specified in the list, etc.).
-CPU IDs are interpreted as decimal values unless they are preceded
-with '0x' in which case they are interpreted as hexadecimal values.
-Not supported unless the entire node is allocated to the job.
-.TP
-.B mask_cpu:<list>
-Bind by setting CPU masks on tasks as specified
-where <list> is <mask1>,<mask2>,...<maskN>.
-The mapping is specified for a node and identical mapping is applied to the
-tasks on every node (i.e. the lowest task ID on each node is mapped to the
-first mask specified in the list, etc.).
-CPU masks are \fBalways\fR interpreted as hexadecimal values but can be
-preceded with an optional '0x'. Not supported unless the entire node is
-allocated to the job.
-.TP
-.B sockets
-Automatically generate masks binding tasks to sockets.
-Only the CPUs on the socket which have been allocated to the job will be used.
-If the number of tasks differs from the number of allocated sockets
-this can result in sub\-optimal binding.
-.TP
-.B cores
-Automatically generate masks binding tasks to cores.
-If the number of tasks differs from the number of allocated cores
-this can result in sub\-optimal binding.
-.TP
-.B threads
-Automatically generate masks binding tasks to threads.
-If the number of tasks differs from the number of allocated threads
-this can result in sub\-optimal binding.
-.TP
-.B ldoms
-Automatically generate masks binding tasks to NUMA locality domains.
-If the number of tasks differs from the number of allocated locality domains
-this can result in sub\-optimal binding.
-.TP
-.B boards
-Automatically generate masks binding tasks to boards.
-If the number of tasks differs from the number of allocated boards
-this can result in sub\-optimal binding. This option is supported by the
-task/cgroup plugin only.
-.TP
-.B help
-Show this help message
-.RE
-
.TP
\fB\-c\fR, \fB\-\-cpus\-per\-task\fR=<\fIncpus\fR>
Advise the SLURM controller that ensuing job steps will require \fIncpus\fR
@@ -810,7 +687,7 @@ be performed at the level of nodes, where the definition of "nodes"
may differ from system to system. \fBThe use of any type other than
"none" or "local" is not recommended.\fR
If you want greater control, try running a simple test code with the
-options "\-\-cpu_bind=verbose,none \-\-mem_bind=verbose,none" to determine
+options "\-\-mem_bind=verbose,none" to determine
the specific configuration.
NOTE: To have SLURM always report on the selected memory binding for
@@ -1070,9 +947,7 @@ behavior on the cluster.
Request the maximum \fIntasks\fR be invoked on each core.
Meant to be used with the \fB\-\-ntasks\fR option.
Related to \fB\-\-ntasks\-per\-node\fR except at the core level
-instead of the node level. Masks will automatically be generated
-to bind the tasks to specific core unless \fB\-\-cpu_bind=none\fR
-is specified.
+instead of the node level.
NOTE: This option is not supported unless
\fISelectTypeParameters=CR_Core\fR or
\fISelectTypeParameters=CR_Core_Memory\fR is configured.
@@ -1082,9 +957,7 @@ NOTE: This option is not supported unless
Request the maximum \fIntasks\fR be invoked on each socket.
Meant to be used with the \fB\-\-ntasks\fR option.
Related to \fB\-\-ntasks\-per\-node\fR except at the socket level
-instead of the node level. Masks will automatically be generated
-to bind the tasks to specific sockets unless \fB\-\-cpu_bind=none\fR
-is specified.
+instead of the node level.
NOTE: This option is not supported unless
\fISelectTypeParameters=CR_Socket\fR or
\fISelectTypeParameters=CR_Socket_Memory\fR is configured.
@@ -1552,8 +1425,6 @@ Same as \fB\-\-conn\-type\fR
\fBSBATCH_CORE_SPEC\fR
Same as \fB\-\-core\-spec\fR
.TP
-\fBSBATCH_CPU_BIND\fR
-Same as \fB\-\-cpu_bind\fR
.TP
\fBSBATCH_DEBUG\fR
Same as \fB\-v, \-\-verbose\fR
@@ -1690,15 +1561,6 @@ Job array's master job ID number.
Directory into which checkpoint images should be written if specified
on the execute line.
.TP
-\fBSLURM_CPU_BIND\fR
-Set to value of the \-\-cpu_bind\fR option.
-.TP
-\fBSLURM_CPU_BIND_LIST\fR
-\-\-cpu_bind map or mask list (list of SLURM CPU IDs or masks for this node,
-CPU_ID = Board_ID x threads_per_board +
-Socket_ID x threads_per_socket +
-Core_ID x threads_per_core + Thread_ID).
-.TP
\fBSLURM_CPUS_ON_NODE\fR
Number of CPUS on the allocated node.
.TP
diff --git a/src/salloc/opt.c b/src/salloc/opt.c
index 9217b53a02756ccca6eeea0776b020af2c38b8b6..24f433720d100a82da5a6908e17cc1f5db9c2bc3 100644
--- a/src/salloc/opt.c
+++ b/src/salloc/opt.c
@@ -544,6 +544,9 @@ _process_env_var(env_vars_t *e, const char *val)
opt.overcommit = true;
break;
case OPT_CPU_BIND:
+ verbose("The --cpu_bind option has been deprecated in "
+ "salloc, --cpu_bind is for srun only going "
+ "forward.");
if (slurm_verify_cpu_bind(val, &opt.cpu_bind,
&opt.cpu_bind_type))
exit(error_exit);
@@ -1151,6 +1154,9 @@ void set_options(const int argc, char **argv)
opt.network = xstrdup(optarg);
break;
case LONG_OPT_CPU_BIND:
+ verbose("The --cpu_bind option has been deprecated in "
+ "salloc, --cpu_bind is for srun only going "
+ "forward.");
if (slurm_verify_cpu_bind(optarg, &opt.cpu_bind,
&opt.cpu_bind_type))
exit(error_exit);
@@ -1857,8 +1863,6 @@ static void _opt_list(void)
info("ntasks-per-socket : %d", opt.ntasks_per_socket);
info("ntasks-per-core : %d", opt.ntasks_per_core);
info("plane_size : %u", opt.plane_size);
- info("cpu_bind : %s",
- opt.cpu_bind == NULL ? "default" : opt.cpu_bind);
info("mem_bind : %s",
opt.mem_bind == NULL ? "default" : opt.mem_bind);
str = print_commandline(command_argc, command_argv);
@@ -1899,7 +1903,7 @@ static void _usage(void)
" [--bell] [--no-bell] [--kill-command[=signal]]\n"
" [--nodefile=file] [--nodelist=hosts] [--exclude=hosts]\n"
" [--network=type] [--mem-per-cpu=MB] [--qos=qos]\n"
-" [--cpu_bind=...] [--mem_bind=...] [--reservation=name]\n"
+" [--mem_bind=...] [--reservation=name]\n"
" [--time-min=minutes] [--gres=list] [--profile=...]\n"
" [--switches=max-switches[@max-time-to-wait]]\n"
" [--core-spec=cores] [--reboot]\n"
@@ -1993,8 +1997,6 @@ static void _help(void)
if (conf->task_plugin != NULL
&& strcasecmp(conf->task_plugin, "task/affinity") == 0) {
printf(
-" --cpu_bind= Bind tasks to CPUs\n"
-" (see \"--cpu_bind=help\" for options)\n"
" --hint= Bind tasks according to application hints\n"
" (see \"--hint=help\" for options)\n"
" --mem_bind= Bind memory to locality domains (ldom)\n"
diff --git a/src/sbatch/opt.c b/src/sbatch/opt.c
index bc36c8e99ff64129a4a5667ff6843676355a3867..4aa2e1583ed1ad688a05cc6d0e522197066386ea 100644
--- a/src/sbatch/opt.c
+++ b/src/sbatch/opt.c
@@ -1350,6 +1350,9 @@ static void _set_options(int argc, char **argv)
opt.shared = 0;
break;
case LONG_OPT_CPU_BIND:
+ verbose("The --cpu_bind option has been deprecated in "
+ "sbatch, --cpu_bind is for srun only going "
+ "forward.");
if (slurm_verify_cpu_bind(optarg, &opt.cpu_bind,
&opt.cpu_bind_type))
exit(error_exit);
@@ -2916,8 +2919,6 @@ static void _opt_list(void)
info("ntasks-per-node : %d", opt.ntasks_per_node);
info("ntasks-per-socket : %d", opt.ntasks_per_socket);
info("ntasks-per-core : %d", opt.ntasks_per_core);
- info("cpu_bind : %s",
- opt.cpu_bind == NULL ? "default" : opt.cpu_bind);
info("mem_bind : %s",
opt.mem_bind == NULL ? "default" : opt.mem_bind);
info("plane_size : %u", opt.plane_size);
@@ -2963,7 +2964,7 @@ static void _usage(void)
" [--requeue] [--no-requeue] [--ntasks-per-node=n] [--propagate]\n"
" [--nodefile=file] [--nodelist=hosts] [--exclude=hosts]\n"
" [--network=type] [--mem-per-cpu=MB] [--qos=qos] [--gres=list]\n"
-" [--cpu_bind=...] [--mem_bind=...] [--reservation=name]\n"
+" [--mem_bind=...] [--reservation=name]\n"
" [--switches=max-switches{@max-time-to-wait}]\n"
" [--core-spec=cores] [--reboot]\n"
" [--array=index_values] [--profile=...] [--ignore-pbs]\n"
@@ -3071,8 +3072,6 @@ static void _help(void)
if (conf->task_plugin != NULL
&& strcasecmp(conf->task_plugin, "task/affinity") == 0) {
printf(
-" --cpu_bind= Bind tasks to CPUs\n"
-" (see \"--cpu_bind=help\" for options)\n"
" --hint= Bind tasks according to application hints\n"
" (see \"--hint=help\" for options)\n"
" --mem_bind= Bind memory to locality domains (ldom)\n"