lkml.org 
[lkml]   [2017]   [Aug]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[PATCH] make "nr_cpu_ids" unsigned
First, number of CPUs can't be negative number.

Second, different signnnedness leads to suboptimal code in the following
cases:

1)
kmalloc(nr_cpu_ids * sizeof(X));

"int" has to be sign extended to size_t.

2)
while (loff_t *pos < nr_cpu_ids)

MOVSXD is 1 byte longed than the same MOV.

Other cases exist as well. Basically compiler is told that nr_cpu_ids
can't be negative which can't be deduced if it is "int".

Code savings on allyesconfig kernel: -3KB

add/remove: 0/0 grow/shrink: 25/264 up/down: 261/-3631 (-3370)
function old new delta
coretemp_cpu_online 450 512 +62
rcu_init_one 1234 1272 +38
pci_device_probe 374 399 +25

...

pgdat_reclaimable_pages 628 556 -72
select_fallback_rq 446 369 -77
task_numa_find_cpu 1923 1807 -116

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
---

arch/arm64/kernel/smp.c | 2 +-
arch/powerpc/kernel/paca.c | 2 +-
arch/powerpc/kernel/setup-common.c | 2 +-
arch/powerpc/sysdev/xive/native.c | 4 ++--
arch/tile/kernel/setup.c | 2 +-
arch/x86/kernel/apic/apic.c | 2 +-
arch/x86/kernel/setup_percpu.c | 2 +-
arch/x86/kernel/smpboot.c | 2 +-
drivers/base/cpu.c | 4 ++--
drivers/scsi/scsi_debug.c | 2 +-
include/linux/cpumask.h | 6 +++---
kernel/rcu/tree.c | 2 +-
kernel/rcu/tree_plugin.h | 2 +-
kernel/sched/topology.c | 2 +-
kernel/smp.c | 2 +-
kernel/trace/trace_functions_graph.c | 2 +-
mm/slub.c | 2 +-
17 files changed, 21 insertions(+), 21 deletions(-)

--- a/arch/arm64/kernel/smp.c
+++ b/arch/arm64/kernel/smp.c
@@ -690,7 +690,7 @@ void __init smp_init_cpus(void)
acpi_parse_gic_cpu_interface, 0);

if (cpu_count > nr_cpu_ids)
- pr_warn("Number of cores (%d) exceeds configured maximum of %d - clipping\n",
+ pr_warn("Number of cores (%d) exceeds configured maximum of %u - clipping\n",
cpu_count, nr_cpu_ids);

if (!bootcpu_valid) {
--- a/arch/powerpc/kernel/paca.c
+++ b/arch/powerpc/kernel/paca.c
@@ -215,7 +215,7 @@ void __init allocate_pacas(void)
paca = __va(memblock_alloc_base(paca_size, PAGE_SIZE, limit));
memset(paca, 0, paca_size);

- printk(KERN_DEBUG "Allocated %u bytes for %d pacas at %p\n",
+ printk(KERN_DEBUG "Allocated %u bytes for %u pacas at %p\n",
paca_size, nr_cpu_ids, paca);

allocate_lppacas(nr_cpu_ids, limit);
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -551,7 +551,7 @@ void __init smp_setup_cpu_maps(void)
if (maxcpus > nr_cpu_ids) {
printk(KERN_WARNING
"Partition configured for %d cpus, "
- "operating system maximum is %d.\n",
+ "operating system maximum is %u.\n",
maxcpus, nr_cpu_ids);
maxcpus = nr_cpu_ids;
} else
--- a/arch/powerpc/sysdev/xive/native.c
+++ b/arch/powerpc/sysdev/xive/native.c
@@ -515,13 +515,13 @@ static bool xive_parse_provisioning(struct device_node *np)
static void xive_native_setup_pools(void)
{
/* Allocate a pool big enough */
- pr_debug("XIVE: Allocating VP block for pool size %d\n", nr_cpu_ids);
+ pr_debug("XIVE: Allocating VP block for pool size %u\n", nr_cpu_ids);

xive_pool_vps = xive_native_alloc_vp_block(nr_cpu_ids);
if (WARN_ON(xive_pool_vps == XIVE_INVALID_VP))
pr_err("XIVE: Failed to allocate pool VP, KVM might not function\n");

- pr_debug("XIVE: Pool VPs allocated at 0x%x for %d max CPUs\n",
+ pr_debug("XIVE: Pool VPs allocated at 0x%x for %u max CPUs\n",
xive_pool_vps, nr_cpu_ids);
}

--- a/arch/tile/kernel/setup.c
+++ b/arch/tile/kernel/setup.c
@@ -1200,7 +1200,7 @@ static void __init validate_hv(void)
* We use a struct cpumask for this, so it must be big enough.
*/
if ((smp_height * smp_width) > nr_cpu_ids)
- early_panic("Hypervisor %d x %d grid too big for Linux NR_CPUS %d\n",
+ early_panic("Hypervisor %d x %d grid too big for Linux NR_CPUS %u\n",
smp_height, smp_width, nr_cpu_ids);
#endif

--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -2137,7 +2137,7 @@ static int allocate_logical_cpuid(int apicid)

/* Allocate a new cpuid. */
if (nr_logical_cpuids >= nr_cpu_ids) {
- WARN_ONCE(1, "APIC: NR_CPUS/possible_cpus limit of %i reached. "
+ WARN_ONCE(1, "APIC: NR_CPUS/possible_cpus limit of %u reached. "
"Processor %d/0x%x and the rest are ignored.\n",
nr_cpu_ids, nr_logical_cpuids, apicid);
return -EINVAL;
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -171,7 +171,7 @@ void __init setup_per_cpu_areas(void)
unsigned long delta;
int rc;

- pr_info("NR_CPUS:%d nr_cpumask_bits:%d nr_cpu_ids:%d nr_node_ids:%d\n",
+ pr_info("NR_CPUS:%d nr_cpumask_bits:%d nr_cpu_ids:%u nr_node_ids:%d\n",
NR_CPUS, nr_cpumask_bits, nr_cpu_ids, nr_node_ids);

/*
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -1457,7 +1457,7 @@ __init void prefill_possible_map(void)

/* nr_cpu_ids could be reduced via nr_cpus= */
if (possible > nr_cpu_ids) {
- pr_warn("%d Processors exceeds NR_CPUS limit of %d\n",
+ pr_warn("%d Processors exceeds NR_CPUS limit of %u\n",
possible, nr_cpu_ids);
possible = nr_cpu_ids;
}
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -256,9 +256,9 @@ static ssize_t print_cpus_offline(struct device *dev,
buf[n++] = ',';

if (nr_cpu_ids == total_cpus-1)
- n += snprintf(&buf[n], len - n, "%d", nr_cpu_ids);
+ n += snprintf(&buf[n], len - n, "%u", nr_cpu_ids);
else
- n += snprintf(&buf[n], len - n, "%d-%d",
+ n += snprintf(&buf[n], len - n, "%u-%d",
nr_cpu_ids, total_cpus-1);
}

--- a/drivers/scsi/scsi_debug.c
+++ b/drivers/scsi/scsi_debug.c
@@ -5466,7 +5466,7 @@ static int sdebug_driver_probe(struct device * dev)
return error;
}
if (submit_queues > nr_cpu_ids) {
- pr_warn("%s: trim submit_queues (was %d) to nr_cpu_ids=%d\n",
+ pr_warn("%s: trim submit_queues (was %d) to nr_cpu_ids=%u\n",
my_name, submit_queues, nr_cpu_ids);
submit_queues = nr_cpu_ids;
}
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -32,15 +32,15 @@ typedef struct cpumask { DECLARE_BITMAP(bits, NR_CPUS); } cpumask_t;
#define cpumask_pr_args(maskp) nr_cpu_ids, cpumask_bits(maskp)

#if NR_CPUS == 1
-#define nr_cpu_ids 1
+#define nr_cpu_ids 1U
#else
-extern int nr_cpu_ids;
+extern unsigned int nr_cpu_ids;
#endif

#ifdef CONFIG_CPUMASK_OFFSTACK
/* Assuming NR_CPUS is huge, a runtime limit is more efficient. Also,
* not all bits may be allocated. */
-#define nr_cpumask_bits ((unsigned int)nr_cpu_ids)
+#define nr_cpumask_bits nr_cpu_ids
#else
#define nr_cpumask_bits ((unsigned int)NR_CPUS)
#endif
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -4134,7 +4134,7 @@ static void __init rcu_init_geometry(void)
if (rcu_fanout_leaf == RCU_FANOUT_LEAF &&
nr_cpu_ids == NR_CPUS)
return;
- pr_info("RCU: Adjusting geometry for rcu_fanout_leaf=%d, nr_cpu_ids=%d\n",
+ pr_info("RCU: Adjusting geometry for rcu_fanout_leaf=%d, nr_cpu_ids=%u\n",
rcu_fanout_leaf, nr_cpu_ids);

/*
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -89,7 +89,7 @@ static void __init rcu_bootup_announce_oddness(void)
if (rcu_fanout_leaf != RCU_FANOUT_LEAF)
pr_info("\tBoot-time adjustment of leaf fanout to %d.\n", rcu_fanout_leaf);
if (nr_cpu_ids != NR_CPUS)
- pr_info("\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%d.\n", NR_CPUS, nr_cpu_ids);
+ pr_info("\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%u.\n", NR_CPUS, nr_cpu_ids);
#ifdef CONFIG_RCU_BOOST
pr_info("\tRCU priority boosting: priority %d delay %d ms.\n", kthread_prio, CONFIG_RCU_BOOST_DELAY);
#endif
--- a/kernel/sched/topology.c
+++ b/kernel/sched/topology.c
@@ -476,7 +476,7 @@ static int __init isolated_cpu_setup(char *str)
alloc_bootmem_cpumask_var(&cpu_isolated_map);
ret = cpulist_parse(str, cpu_isolated_map);
if (ret) {
- pr_err("sched: Error, all isolcpus= values must be between 0 and %d\n", nr_cpu_ids);
+ pr_err("sched: Error, all isolcpus= values must be between 0 and %u\n", nr_cpu_ids);
return 0;
}
return 1;
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -548,7 +548,7 @@ static int __init maxcpus(char *str)
early_param("maxcpus", maxcpus);

/* Setup number of possible processor ids */
-int nr_cpu_ids __read_mostly = NR_CPUS;
+unsigned int nr_cpu_ids __read_mostly = NR_CPUS;
EXPORT_SYMBOL(nr_cpu_ids);

/* An arch may set nr_cpu_ids earlier if needed, so this would be redundant */
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -1543,7 +1543,7 @@ fs_initcall(init_graph_tracefs);

static __init int init_graph_trace(void)
{
- max_bytes_for_cpu = snprintf(NULL, 0, "%d", nr_cpu_ids - 1);
+ max_bytes_for_cpu = snprintf(NULL, 0, "%u", nr_cpu_ids - 1);

if (!register_trace_event(&graph_trace_entry_event)) {
pr_warn("Warning: could not register graph trace events\n");
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -4196,7 +4196,7 @@ void __init kmem_cache_init(void)
cpuhp_setup_state_nocalls(CPUHP_SLUB_DEAD, "slub:dead", NULL,
slub_cpu_dead);

- pr_info("SLUB: HWalign=%d, Order=%d-%d, MinObjects=%d, CPUs=%d, Nodes=%d\n",
+ pr_info("SLUB: HWalign=%d, Order=%d-%d, MinObjects=%d, CPUs=%u, Nodes=%d\n",
cache_line_size(),
slub_min_order, slub_max_order, slub_min_objects,
nr_cpu_ids, nr_node_ids);
\
 
 \ /
  Last update: 2017-08-19 13:50    [W:0.026 / U:1.040 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site