From 470c66239ef0336429b35345f3f615d47341e13b Mon Sep 17 00:00:00 2001 From: David Brownell Date: Mon, 1 Dec 2008 14:31:37 -0800 Subject: genirq: warn when IRQF_DISABLED may be ignored Impact: emit new warning We periodically waste time tracking down problems from the genirq framework not respecting IRQF_DISABLED for some shared IRQ cases. Linus views this as "will not fix", but we're still left with the bugs caused by this misbehavior. This patch adds a nag message in request_irq(), so that drivers can fix their IRQ handlers to avoid this problem. Note that developers will never see the relevant bugs when they run with LOCKDEP, so it's no wonder these bugs are hard to find. (That also means LOCKDEP is overlooking some IRQ-related bugs involving IRQ handlers that don't set IRQF_DISABLED...) Signed-off-by: David Brownell Signed-off-by: Andrew Morton Acked-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- kernel/irq/manage.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'kernel/irq') diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index c498a1b8c621..7fd891c3a33d 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -635,6 +635,18 @@ int request_irq(unsigned int irq, irq_handler_t handler, struct irq_desc *desc; int retval; + /* + * handle_IRQ_event() always ignores IRQF_DISABLED except for + * the _first_ irqaction (sigh). That can cause oopsing, but + * the behavior is classified as "will not fix" so we need to + * start nudging drivers away from using that idiom. + */ + if ((irqflags & (IRQF_SHARED|IRQF_DISABLED)) + == (IRQF_SHARED|IRQF_DISABLED)) + pr_warning("IRQ %d/%s: IRQF_DISABLED is not " + "guaranteed on shared IRQs\n", + irq, devname); + #ifdef CONFIG_LOCKDEP /* * Lockdep wants atomic interrupt handlers: -- cgit v1.2.3-55-g7522 From 29c0177e6a4ac094302bed54a1d4bbb6b740a9ef Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 13 Dec 2008 21:20:25 +1030 Subject: cpumask: change cpumask_scnprintf, cpumask_parse_user, cpulist_parse, and cpulist_scnprintf to take pointers. Impact: change calling convention of existing cpumask APIs Most cpumask functions started with cpus_: these have been replaced by cpumask_ ones which take struct cpumask pointers as expected. These four functions don't have good replacement names; fortunately they're rarely used, so we just change them over. Signed-off-by: Rusty Russell Signed-off-by: Mike Travis Acked-by: Ingo Molnar Cc: paulus@samba.org Cc: mingo@redhat.com Cc: tony.luck@intel.com Cc: ralf@linux-mips.org Cc: Greg Kroah-Hartman Cc: cl@linux-foundation.org Cc: srostedt@redhat.com --- arch/ia64/kernel/topology.c | 2 +- arch/mips/kernel/smp-cmp.c | 4 +- arch/powerpc/platforms/pseries/xics.c | 2 +- arch/x86/kernel/cpu/intel_cacheinfo.c | 4 +- arch/x86/kernel/setup_percpu.c | 2 +- drivers/base/cpu.c | 2 +- drivers/base/node.c | 4 +- drivers/base/topology.c | 4 +- drivers/pci/pci-sysfs.c | 4 +- drivers/pci/probe.c | 4 +- include/linux/cpumask.h | 87 +++++++++++++++++++++++------------ kernel/cpuset.c | 4 +- kernel/irq/proc.c | 4 +- kernel/profile.c | 4 +- kernel/sched.c | 4 +- kernel/sched_stats.h | 2 +- kernel/taskstats.c | 2 +- kernel/trace/trace.c | 4 +- mm/slub.c | 2 +- 19 files changed, 86 insertions(+), 59 deletions(-) (limited to 'kernel/irq') diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index c75b914f2d6b..a8d61a3e9a94 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -219,7 +219,7 @@ static ssize_t show_shared_cpu_map(struct cache_info *this_leaf, char *buf) cpumask_t shared_cpu_map; cpus_and(shared_cpu_map, this_leaf->shared_cpu_map, cpu_online_map); - len = cpumask_scnprintf(buf, NR_CPUS+1, shared_cpu_map); + len = cpumask_scnprintf(buf, NR_CPUS+1, &shared_cpu_map); len += sprintf(buf+len, "\n"); return len; } diff --git a/arch/mips/kernel/smp-cmp.c b/arch/mips/kernel/smp-cmp.c index 6789c1a12120..f27beca4b26d 100644 --- a/arch/mips/kernel/smp-cmp.c +++ b/arch/mips/kernel/smp-cmp.c @@ -51,10 +51,10 @@ static int __init allowcpus(char *str) int len; cpus_clear(cpu_allow_map); - if (cpulist_parse(str, cpu_allow_map) == 0) { + if (cpulist_parse(str, &cpu_allow_map) == 0) { cpu_set(0, cpu_allow_map); cpus_and(cpu_possible_map, cpu_possible_map, cpu_allow_map); - len = cpulist_scnprintf(buf, sizeof(buf)-1, cpu_possible_map); + len = cpulist_scnprintf(buf, sizeof(buf)-1, &cpu_possible_map); buf[len] = '\0'; pr_debug("Allowable CPUs: %s\n", buf); return 1; diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c index e1904774a70f..64d24310ce7e 100644 --- a/arch/powerpc/platforms/pseries/xics.c +++ b/arch/powerpc/platforms/pseries/xics.c @@ -358,7 +358,7 @@ static void xics_set_affinity(unsigned int virq, cpumask_t cpumask) irq_server = get_irq_server(virq, 1); if (irq_server == -1) { char cpulist[128]; - cpumask_scnprintf(cpulist, sizeof(cpulist), cpumask); + cpumask_scnprintf(cpulist, sizeof(cpulist), &cpumask); printk(KERN_WARNING "%s: No online cpus in the mask %s for irq %d\n", __func__, cpulist, virq); diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 3f46afbb1cf1..43ea612d3e9d 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -626,8 +626,8 @@ static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf, cpumask_t *mask = &this_leaf->shared_cpu_map; n = type? - cpulist_scnprintf(buf, len-2, *mask): - cpumask_scnprintf(buf, len-2, *mask); + cpulist_scnprintf(buf, len-2, mask) : + cpumask_scnprintf(buf, len-2, mask); buf[n++] = '\n'; buf[n] = '\0'; } diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index ae0c0d3bb770..1c2084291f97 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -282,7 +282,7 @@ static void __cpuinit numa_set_cpumask(int cpu, int enable) else cpu_clear(cpu, *mask); - cpulist_scnprintf(buf, sizeof(buf), *mask); + cpulist_scnprintf(buf, sizeof(buf), mask); printk(KERN_DEBUG "%s cpu %d node %d: mask now %s\n", enable? "numa_add_cpu":"numa_remove_cpu", cpu, node, buf); } diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 64f5d54f7edc..4259072f5bd0 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -109,7 +109,7 @@ static SYSDEV_ATTR(crash_notes, 0400, show_crash_notes, NULL); */ static ssize_t print_cpus_map(char *buf, cpumask_t *map) { - int n = cpulist_scnprintf(buf, PAGE_SIZE-2, *map); + int n = cpulist_scnprintf(buf, PAGE_SIZE-2, map); buf[n++] = '\n'; buf[n] = '\0'; diff --git a/drivers/base/node.c b/drivers/base/node.c index f5207090885a..91636cd8b6c9 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -30,8 +30,8 @@ static ssize_t node_read_cpumap(struct sys_device *dev, int type, char *buf) BUILD_BUG_ON((NR_CPUS/32 * 9) > (PAGE_SIZE-1)); len = type? - cpulist_scnprintf(buf, PAGE_SIZE-2, *mask): - cpumask_scnprintf(buf, PAGE_SIZE-2, *mask); + cpulist_scnprintf(buf, PAGE_SIZE-2, mask) : + cpumask_scnprintf(buf, PAGE_SIZE-2, mask); buf[len++] = '\n'; buf[len] = '\0'; return len; diff --git a/drivers/base/topology.c b/drivers/base/topology.c index 199cd97e32e6..a8bc1cbcfa7c 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -49,8 +49,8 @@ static ssize_t show_cpumap(int type, cpumask_t *mask, char *buf) if (len > 1) { n = type? - cpulist_scnprintf(buf, len-2, *mask): - cpumask_scnprintf(buf, len-2, *mask); + cpulist_scnprintf(buf, len-2, mask) : + cpumask_scnprintf(buf, len-2, mask); buf[n++] = '\n'; buf[n] = '\0'; } diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 5d72866897a8..c88485860a0a 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -74,7 +74,7 @@ static ssize_t local_cpus_show(struct device *dev, int len; mask = pcibus_to_cpumask(to_pci_dev(dev)->bus); - len = cpumask_scnprintf(buf, PAGE_SIZE-2, mask); + len = cpumask_scnprintf(buf, PAGE_SIZE-2, &mask); buf[len++] = '\n'; buf[len] = '\0'; return len; @@ -88,7 +88,7 @@ static ssize_t local_cpulist_show(struct device *dev, int len; mask = pcibus_to_cpumask(to_pci_dev(dev)->bus); - len = cpulist_scnprintf(buf, PAGE_SIZE-2, mask); + len = cpulist_scnprintf(buf, PAGE_SIZE-2, &mask); buf[len++] = '\n'; buf[len] = '\0'; return len; diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 003a9b3c293f..5b3f5937ecf5 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -55,8 +55,8 @@ static ssize_t pci_bus_show_cpuaffinity(struct device *dev, cpumask = pcibus_to_cpumask(to_pci_bus(dev)); ret = type? - cpulist_scnprintf(buf, PAGE_SIZE-2, cpumask): - cpumask_scnprintf(buf, PAGE_SIZE-2, cpumask); + cpulist_scnprintf(buf, PAGE_SIZE-2, &cpumask) : + cpumask_scnprintf(buf, PAGE_SIZE-2, &cpumask); buf[ret++] = '\n'; buf[ret] = '\0'; return ret; diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h index 21e1dd43e52a..94a2ab88ae85 100644 --- a/include/linux/cpumask.h +++ b/include/linux/cpumask.h @@ -339,36 +339,6 @@ extern cpumask_t cpu_mask_all; #endif #define CPUMASK_PTR(v, m) cpumask_t *v = &(m->v) -#define cpumask_scnprintf(buf, len, src) \ - __cpumask_scnprintf((buf), (len), &(src), NR_CPUS) -static inline int __cpumask_scnprintf(char *buf, int len, - const cpumask_t *srcp, int nbits) -{ - return bitmap_scnprintf(buf, len, srcp->bits, nbits); -} - -#define cpumask_parse_user(ubuf, ulen, dst) \ - __cpumask_parse_user((ubuf), (ulen), &(dst), NR_CPUS) -static inline int __cpumask_parse_user(const char __user *buf, int len, - cpumask_t *dstp, int nbits) -{ - return bitmap_parse_user(buf, len, dstp->bits, nbits); -} - -#define cpulist_scnprintf(buf, len, src) \ - __cpulist_scnprintf((buf), (len), &(src), NR_CPUS) -static inline int __cpulist_scnprintf(char *buf, int len, - const cpumask_t *srcp, int nbits) -{ - return bitmap_scnlistprintf(buf, len, srcp->bits, nbits); -} - -#define cpulist_parse(buf, dst) __cpulist_parse((buf), &(dst), NR_CPUS) -static inline int __cpulist_parse(const char *buf, cpumask_t *dstp, int nbits) -{ - return bitmap_parselist(buf, dstp->bits, nbits); -} - #define cpu_remap(oldbit, old, new) \ __cpu_remap((oldbit), &(old), &(new), NR_CPUS) static inline int __cpu_remap(int oldbit, @@ -945,6 +915,63 @@ static inline void cpumask_copy(struct cpumask *dstp, */ #define cpumask_of(cpu) (get_cpu_mask(cpu)) +/** + * cpumask_scnprintf - print a cpumask into a string as comma-separated hex + * @buf: the buffer to sprintf into + * @len: the length of the buffer + * @srcp: the cpumask to print + * + * If len is zero, returns zero. Otherwise returns the length of the + * (nul-terminated) @buf string. + */ +static inline int cpumask_scnprintf(char *buf, int len, + const struct cpumask *srcp) +{ + return bitmap_scnprintf(buf, len, srcp->bits, nr_cpumask_bits); +} + +/** + * cpumask_parse_user - extract a cpumask from a user string + * @buf: the buffer to extract from + * @len: the length of the buffer + * @dstp: the cpumask to set. + * + * Returns -errno, or 0 for success. + */ +static inline int cpumask_parse_user(const char __user *buf, int len, + struct cpumask *dstp) +{ + return bitmap_parse_user(buf, len, dstp->bits, nr_cpumask_bits); +} + +/** + * cpulist_scnprintf - print a cpumask into a string as comma-separated list + * @buf: the buffer to sprintf into + * @len: the length of the buffer + * @srcp: the cpumask to print + * + * If len is zero, returns zero. Otherwise returns the length of the + * (nul-terminated) @buf string. + */ +static inline int cpulist_scnprintf(char *buf, int len, + const struct cpumask *srcp) +{ + return bitmap_scnlistprintf(buf, len, srcp->bits, nr_cpumask_bits); +} + +/** + * cpulist_parse_user - extract a cpumask from a user string of ranges + * @buf: the buffer to extract from + * @len: the length of the buffer + * @dstp: the cpumask to set. + * + * Returns -errno, or 0 for success. + */ +static inline int cpulist_parse(const char *buf, struct cpumask *dstp) +{ + return bitmap_parselist(buf, dstp->bits, nr_cpumask_bits); +} + /** * to_cpumask - convert an NR_CPUS bitmap to a struct cpumask * * @bitmap: the bitmap diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 96c0ba13b8cd..39c1a4c1c5a9 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -896,7 +896,7 @@ static int update_cpumask(struct cpuset *cs, const char *buf) if (!*buf) { cpus_clear(trialcs.cpus_allowed); } else { - retval = cpulist_parse(buf, trialcs.cpus_allowed); + retval = cpulist_parse(buf, &trialcs.cpus_allowed); if (retval < 0) return retval; @@ -1482,7 +1482,7 @@ static int cpuset_sprintf_cpulist(char *page, struct cpuset *cs) mask = cs->cpus_allowed; mutex_unlock(&callback_mutex); - return cpulist_scnprintf(page, PAGE_SIZE, mask); + return cpulist_scnprintf(page, PAGE_SIZE, &mask); } static int cpuset_sprintf_memlist(char *page, struct cpuset *cs) diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index d257e7d6a8a4..f293349d49d0 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -47,7 +47,7 @@ static ssize_t irq_affinity_proc_write(struct file *file, irq_balancing_disabled(irq)) return -EIO; - err = cpumask_parse_user(buffer, count, new_value); + err = cpumask_parse_user(buffer, count, &new_value); if (err) return err; @@ -95,7 +95,7 @@ static ssize_t default_affinity_write(struct file *file, cpumask_t new_value; int err; - err = cpumask_parse_user(buffer, count, new_value); + err = cpumask_parse_user(buffer, count, &new_value); if (err) return err; diff --git a/kernel/profile.c b/kernel/profile.c index dc41827fbfee..7d620dfdde59 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -442,7 +442,7 @@ void profile_tick(int type) static int prof_cpu_mask_read_proc(char *page, char **start, off_t off, int count, int *eof, void *data) { - int len = cpumask_scnprintf(page, count, *(cpumask_t *)data); + int len = cpumask_scnprintf(page, count, (cpumask_t *)data); if (count - len < 2) return -EINVAL; len += sprintf(page + len, "\n"); @@ -456,7 +456,7 @@ static int prof_cpu_mask_write_proc(struct file *file, unsigned long full_count = count, err; cpumask_t new_value; - err = cpumask_parse_user(buffer, count, new_value); + err = cpumask_parse_user(buffer, count, &new_value); if (err) return err; diff --git a/kernel/sched.c b/kernel/sched.c index e4bb1dd7b308..d2d16d1273b1 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6666,7 +6666,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, struct sched_group *group = sd->groups; char str[256]; - cpulist_scnprintf(str, sizeof(str), sd->span); + cpulist_scnprintf(str, sizeof(str), &sd->span); cpus_clear(*groupmask); printk(KERN_DEBUG "%*s domain %d: ", level, "", level); @@ -6720,7 +6720,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, cpus_or(*groupmask, *groupmask, group->cpumask); - cpulist_scnprintf(str, sizeof(str), group->cpumask); + cpulist_scnprintf(str, sizeof(str), &group->cpumask); printk(KERN_CONT " %s", str); group = group->next; diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h index 7dbf72a2b02c..6beff1e4eeae 100644 --- a/kernel/sched_stats.h +++ b/kernel/sched_stats.h @@ -42,7 +42,7 @@ static int show_schedstat(struct seq_file *seq, void *v) for_each_domain(cpu, sd) { enum cpu_idle_type itype; - cpumask_scnprintf(mask_str, mask_len, sd->span); + cpumask_scnprintf(mask_str, mask_len, &sd->span); seq_printf(seq, "domain%d %s", dcount++, mask_str); for (itype = CPU_IDLE; itype < CPU_MAX_IDLE_TYPES; itype++) { diff --git a/kernel/taskstats.c b/kernel/taskstats.c index bd6be76303cf..6d7dc4ec4aa5 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -352,7 +352,7 @@ static int parse(struct nlattr *na, cpumask_t *mask) if (!data) return -ENOMEM; nla_strlcpy(data, na, len); - ret = cpulist_parse(data, *mask); + ret = cpulist_parse(data, mask); kfree(data); return ret; } diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index d86e3252f300..d2e75479dc50 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2126,7 +2126,7 @@ tracing_cpumask_read(struct file *filp, char __user *ubuf, mutex_lock(&tracing_cpumask_update_lock); - len = cpumask_scnprintf(mask_str, count, tracing_cpumask); + len = cpumask_scnprintf(mask_str, count, &tracing_cpumask); if (count - len < 2) { count = -EINVAL; goto out_err; @@ -2147,7 +2147,7 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf, int err, cpu; mutex_lock(&tracing_cpumask_update_lock); - err = cpumask_parse_user(ubuf, count, tracing_cpumask_new); + err = cpumask_parse_user(ubuf, count, &tracing_cpumask_new); if (err) goto err_unlock; diff --git a/mm/slub.c b/mm/slub.c index a2cd47d89e0a..8e516e29f989 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -3626,7 +3626,7 @@ static int list_locations(struct kmem_cache *s, char *buf, len < PAGE_SIZE - 60) { len += sprintf(buf + len, " cpus="); len += cpulist_scnprintf(buf + len, PAGE_SIZE - len - 50, - l->cpus); + &l->cpus); } if (num_online_nodes() > 1 && !nodes_empty(l->nodes) && -- cgit v1.2.3-55-g7522 From 0de26520c7cabf36e1de090ea8092f011a6106ce Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 13 Dec 2008 21:20:26 +1030 Subject: cpumask: make irq_set_affinity() take a const struct cpumask Impact: change existing irq_chip API Not much point with gentle transition here: the struct irq_chip's setaffinity method signature needs to change. Fortunately, not widely used code, but hits a few architectures. Note: In irq_select_affinity() I save a temporary in by mangling irq_desc[irq].affinity directly. Ingo, does this break anything? (Folded in fix from KOSAKI Motohiro) Signed-off-by: Rusty Russell Signed-off-by: Mike Travis Reviewed-by: Grant Grundler Acked-by: Ingo Molnar Cc: ralf@linux-mips.org Cc: grundler@parisc-linux.org Cc: jeremy@xensource.com Cc: KOSAKI Motohiro --- arch/alpha/kernel/irq.c | 2 +- arch/alpha/kernel/sys_dp264.c | 8 ++-- arch/alpha/kernel/sys_titan.c | 4 +- arch/arm/common/gic.c | 4 +- arch/arm/kernel/irq.c | 2 +- arch/arm/oprofile/op_model_mpcore.c | 4 +- arch/cris/arch-v32/kernel/irq.c | 4 +- arch/ia64/hp/sim/hpsim_irq.c | 2 +- arch/ia64/kernel/iosapic.c | 12 +++--- arch/ia64/kernel/irq.c | 9 ++-- arch/ia64/kernel/msi_ia64.c | 12 +++--- arch/ia64/kernel/smpboot.c | 4 +- arch/ia64/sn/kernel/irq.c | 6 +-- arch/ia64/sn/kernel/msi_sn.c | 7 +-- arch/mips/include/asm/irq.h | 3 +- arch/mips/kernel/cevt-bcm1480.c | 2 +- arch/mips/kernel/cevt-sb1250.c | 2 +- arch/mips/kernel/irq-gic.c | 6 +-- arch/mips/mti-malta/malta-smtc.c | 6 +-- arch/mips/sibyte/bcm1480/irq.c | 8 ++-- arch/mips/sibyte/sb1250/irq.c | 8 ++-- arch/parisc/kernel/irq.c | 6 +-- arch/powerpc/kernel/irq.c | 2 +- arch/powerpc/platforms/pseries/xics.c | 6 +-- arch/powerpc/sysdev/mpic.c | 4 +- arch/powerpc/sysdev/mpic.h | 2 +- arch/sparc64/kernel/irq.c | 11 +++-- arch/sparc64/kernel/of_device.c | 2 +- arch/sparc64/kernel/pci_msi.c | 2 +- arch/x86/kernel/hpet.c | 4 +- arch/x86/kernel/io_apic.c | 81 +++++++++++++++++------------------ arch/x86/kernel/irq_32.c | 2 +- arch/x86/kernel/irq_64.c | 2 +- drivers/parisc/iosapic.c | 7 +-- drivers/xen/events.c | 6 +-- include/linux/interrupt.h | 4 +- include/linux/irq.h | 3 +- kernel/irq/chip.c | 2 +- kernel/irq/manage.c | 22 +++++----- kernel/irq/migration.c | 14 +++--- kernel/irq/proc.c | 29 ++++++++----- kernel/time/tick-common.c | 6 +-- 42 files changed, 171 insertions(+), 161 deletions(-) (limited to 'kernel/irq') diff --git a/arch/alpha/kernel/irq.c b/arch/alpha/kernel/irq.c index c626a821cdcb..d0f1620007f7 100644 --- a/arch/alpha/kernel/irq.c +++ b/arch/alpha/kernel/irq.c @@ -55,7 +55,7 @@ int irq_select_affinity(unsigned int irq) last_cpu = cpu; irq_desc[irq].affinity = cpumask_of_cpu(cpu); - irq_desc[irq].chip->set_affinity(irq, cpumask_of_cpu(cpu)); + irq_desc[irq].chip->set_affinity(irq, cpumask_of(cpu)); return 0; } #endif /* CONFIG_SMP */ diff --git a/arch/alpha/kernel/sys_dp264.c b/arch/alpha/kernel/sys_dp264.c index c71b0fd7a61f..ab44c164d9d4 100644 --- a/arch/alpha/kernel/sys_dp264.c +++ b/arch/alpha/kernel/sys_dp264.c @@ -177,19 +177,19 @@ cpu_set_irq_affinity(unsigned int irq, cpumask_t affinity) } static void -dp264_set_affinity(unsigned int irq, cpumask_t affinity) +dp264_set_affinity(unsigned int irq, const struct cpumask *affinity) { spin_lock(&dp264_irq_lock); - cpu_set_irq_affinity(irq, affinity); + cpu_set_irq_affinity(irq, *affinity); tsunami_update_irq_hw(cached_irq_mask); spin_unlock(&dp264_irq_lock); } static void -clipper_set_affinity(unsigned int irq, cpumask_t affinity) +clipper_set_affinity(unsigned int irq, const struct cpumask *affinity) { spin_lock(&dp264_irq_lock); - cpu_set_irq_affinity(irq - 16, affinity); + cpu_set_irq_affinity(irq - 16, *affinity); tsunami_update_irq_hw(cached_irq_mask); spin_unlock(&dp264_irq_lock); } diff --git a/arch/alpha/kernel/sys_titan.c b/arch/alpha/kernel/sys_titan.c index 52c91ccc1648..27f840a4ad3d 100644 --- a/arch/alpha/kernel/sys_titan.c +++ b/arch/alpha/kernel/sys_titan.c @@ -158,10 +158,10 @@ titan_cpu_set_irq_affinity(unsigned int irq, cpumask_t affinity) } static void -titan_set_irq_affinity(unsigned int irq, cpumask_t affinity) +titan_set_irq_affinity(unsigned int irq, const struct cpumask *affinity) { spin_lock(&titan_irq_lock); - titan_cpu_set_irq_affinity(irq - 16, affinity); + titan_cpu_set_irq_affinity(irq - 16, *affinity); titan_update_irq_hw(titan_cached_irq_mask); spin_unlock(&titan_irq_lock); } diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c index 7fc9860a97d7..c6884ba1d5ed 100644 --- a/arch/arm/common/gic.c +++ b/arch/arm/common/gic.c @@ -109,11 +109,11 @@ static void gic_unmask_irq(unsigned int irq) } #ifdef CONFIG_SMP -static void gic_set_cpu(unsigned int irq, cpumask_t mask_val) +static void gic_set_cpu(unsigned int irq, const struct cpumask *mask_val) { void __iomem *reg = gic_dist_base(irq) + GIC_DIST_TARGET + (gic_irq(irq) & ~3); unsigned int shift = (irq % 4) * 8; - unsigned int cpu = first_cpu(mask_val); + unsigned int cpu = cpumask_first(mask_val); u32 val; spin_lock(&irq_controller_lock); diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c index 2f3eb795fa6e..7141cee1fab7 100644 --- a/arch/arm/kernel/irq.c +++ b/arch/arm/kernel/irq.c @@ -174,7 +174,7 @@ static void route_irq(struct irq_desc *desc, unsigned int irq, unsigned int cpu) pr_debug("IRQ%u: moving from cpu%u to cpu%u\n", irq, desc->cpu, cpu); spin_lock_irq(&desc->lock); - desc->chip->set_affinity(irq, cpumask_of_cpu(cpu)); + desc->chip->set_affinity(irq, cpumask_of(cpu)); spin_unlock_irq(&desc->lock); } diff --git a/arch/arm/oprofile/op_model_mpcore.c b/arch/arm/oprofile/op_model_mpcore.c index 4de366e8b4c5..6d6bd5899240 100644 --- a/arch/arm/oprofile/op_model_mpcore.c +++ b/arch/arm/oprofile/op_model_mpcore.c @@ -260,10 +260,10 @@ static void em_stop(void) static void em_route_irq(int irq, unsigned int cpu) { struct irq_desc *desc = irq_desc + irq; - cpumask_t mask = cpumask_of_cpu(cpu); + const struct cpumask *mask = cpumask_of(cpu); spin_lock_irq(&desc->lock); - desc->affinity = mask; + desc->affinity = *mask; desc->chip->set_affinity(irq, mask); spin_unlock_irq(&desc->lock); } diff --git a/arch/cris/arch-v32/kernel/irq.c b/arch/cris/arch-v32/kernel/irq.c index 173c141ac9ba..295131fee710 100644 --- a/arch/cris/arch-v32/kernel/irq.c +++ b/arch/cris/arch-v32/kernel/irq.c @@ -325,11 +325,11 @@ static void end_crisv32_irq(unsigned int irq) { } -void set_affinity_crisv32_irq(unsigned int irq, cpumask_t dest) +void set_affinity_crisv32_irq(unsigned int irq, const struct cpumask *dest) { unsigned long flags; spin_lock_irqsave(&irq_lock, flags); - irq_allocations[irq - FIRST_IRQ].mask = dest; + irq_allocations[irq - FIRST_IRQ].mask = *dest; spin_unlock_irqrestore(&irq_lock, flags); } diff --git a/arch/ia64/hp/sim/hpsim_irq.c b/arch/ia64/hp/sim/hpsim_irq.c index c2f58ff364e7..cc0a3182db3c 100644 --- a/arch/ia64/hp/sim/hpsim_irq.c +++ b/arch/ia64/hp/sim/hpsim_irq.c @@ -22,7 +22,7 @@ hpsim_irq_noop (unsigned int irq) } static void -hpsim_set_affinity_noop (unsigned int a, cpumask_t b) +hpsim_set_affinity_noop(unsigned int a, const struct cpumask *b) { } diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index 5c4674ae8aea..c8adecd5b416 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -330,25 +330,25 @@ unmask_irq (unsigned int irq) static void -iosapic_set_affinity (unsigned int irq, cpumask_t mask) +iosapic_set_affinity(unsigned int irq, const struct cpumask *mask) { #ifdef CONFIG_SMP u32 high32, low32; - int dest, rte_index; + int cpu, dest, rte_index; int redir = (irq & IA64_IRQ_REDIRECTED) ? 1 : 0; struct iosapic_rte_info *rte; struct iosapic *iosapic; irq &= (~IA64_IRQ_REDIRECTED); - cpus_and(mask, mask, cpu_online_map); - if (cpus_empty(mask)) + cpu = cpumask_first_and(cpu_online_mask, mask); + if (cpu >= nr_cpu_ids) return; - if (irq_prepare_move(irq, first_cpu(mask))) + if (irq_prepare_move(irq, cpu)) return; - dest = cpu_physical_id(first_cpu(mask)); + dest = cpu_physical_id(cpu); if (!iosapic_intr_info[irq].count) return; /* not an IOSAPIC interrupt */ diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c index 7fd18f54c056..0b6db53fedcf 100644 --- a/arch/ia64/kernel/irq.c +++ b/arch/ia64/kernel/irq.c @@ -133,7 +133,6 @@ unsigned int vectors_in_migration[NR_IRQS]; */ static void migrate_irqs(void) { - cpumask_t mask; irq_desc_t *desc; int irq, new_cpu; @@ -152,15 +151,14 @@ static void migrate_irqs(void) if (desc->status == IRQ_PER_CPU) continue; - cpus_and(mask, irq_desc[irq].affinity, cpu_online_map); - if (any_online_cpu(mask) == NR_CPUS) { + if (cpumask_any_and(&irq_desc[irq].affinity, cpu_online_mask) + >= nr_cpu_ids) { /* * Save it for phase 2 processing */ vectors_in_migration[irq] = irq; new_cpu = any_online_cpu(cpu_online_map); - mask = cpumask_of_cpu(new_cpu); /* * Al three are essential, currently WARN_ON.. maybe panic? @@ -168,7 +166,8 @@ static void migrate_irqs(void) if (desc->chip && desc->chip->disable && desc->chip->enable && desc->chip->set_affinity) { desc->chip->disable(irq); - desc->chip->set_affinity(irq, mask); + desc->chip->set_affinity(irq, + cpumask_of(new_cpu)); desc->chip->enable(irq); } else { WARN_ON((!(desc->chip) || !(desc->chip->disable) || diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c index 702a09c13238..890339339035 100644 --- a/arch/ia64/kernel/msi_ia64.c +++ b/arch/ia64/kernel/msi_ia64.c @@ -49,11 +49,12 @@ static struct irq_chip ia64_msi_chip; #ifdef CONFIG_SMP -static void ia64_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask) +static void ia64_set_msi_irq_affinity(unsigned int irq, + const cpumask_t *cpu_mask) { struct msi_msg msg; u32 addr, data; - int cpu = first_cpu(cpu_mask); + int cpu = first_cpu(*cpu_mask); if (!cpu_online(cpu)) return; @@ -166,12 +167,11 @@ void arch_teardown_msi_irq(unsigned int irq) #ifdef CONFIG_DMAR #ifdef CONFIG_SMP -static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) +static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) { struct irq_cfg *cfg = irq_cfg + irq; struct msi_msg msg; - int cpu = first_cpu(mask); - + int cpu = cpumask_first(mask); if (!cpu_online(cpu)) return; @@ -187,7 +187,7 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) msg.address_lo |= MSI_ADDR_DESTID_CPU(cpu_physical_id(cpu)); dmar_msi_write(irq, &msg); - irq_desc[irq].affinity = mask; + irq_desc[irq].affinity = *mask; } #endif /* CONFIG_SMP */ diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c index 4ede6e571c38..11463994a7d5 100644 --- a/arch/ia64/kernel/smpboot.c +++ b/arch/ia64/kernel/smpboot.c @@ -682,7 +682,7 @@ int migrate_platform_irqs(unsigned int cpu) { int new_cpei_cpu; irq_desc_t *desc = NULL; - cpumask_t mask; + const struct cpumask *mask; int retval = 0; /* @@ -695,7 +695,7 @@ int migrate_platform_irqs(unsigned int cpu) * Now re-target the CPEI to a different processor */ new_cpei_cpu = any_online_cpu(cpu_online_map); - mask = cpumask_of_cpu(new_cpei_cpu); + mask = cpumask_of(new_cpei_cpu); set_cpei_target_cpu(new_cpei_cpu); desc = irq_desc + ia64_cpe_irq; /* diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c index 0c66dbdd1d72..66fd705e82c0 100644 --- a/arch/ia64/sn/kernel/irq.c +++ b/arch/ia64/sn/kernel/irq.c @@ -227,14 +227,14 @@ finish_up: return new_irq_info; } -static void sn_set_affinity_irq(unsigned int irq, cpumask_t mask) +static void sn_set_affinity_irq(unsigned int irq, const struct cpumask *mask) { struct sn_irq_info *sn_irq_info, *sn_irq_info_safe; nasid_t nasid; int slice; - nasid = cpuid_to_nasid(first_cpu(mask)); - slice = cpuid_to_slice(first_cpu(mask)); + nasid = cpuid_to_nasid(cpumask_first(mask)); + slice = cpuid_to_slice(cpumask_first(mask)); list_for_each_entry_safe(sn_irq_info, sn_irq_info_safe, sn_irq_lh[irq], list) diff --git a/arch/ia64/sn/kernel/msi_sn.c b/arch/ia64/sn/kernel/msi_sn.c index 83f190ffe350..ca553b0429ce 100644 --- a/arch/ia64/sn/kernel/msi_sn.c +++ b/arch/ia64/sn/kernel/msi_sn.c @@ -151,7 +151,8 @@ int sn_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *entry) } #ifdef CONFIG_SMP -static void sn_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask) +static void sn_set_msi_irq_affinity(unsigned int irq, + const struct cpumask *cpu_mask) { struct msi_msg msg; int slice; @@ -164,7 +165,7 @@ static void sn_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask) struct sn_pcibus_provider *provider; unsigned int cpu; - cpu = first_cpu(cpu_mask); + cpu = cpumask_first(cpu_mask); sn_irq_info = sn_msi_info[irq].sn_irq_info; if (sn_irq_info == NULL || sn_irq_info->irq_int_bit >= 0) return; @@ -204,7 +205,7 @@ static void sn_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask) msg.address_lo = (u32)(bus_addr & 0x00000000ffffffff); write_msi_msg(irq, &msg); - irq_desc[irq].affinity = cpu_mask; + irq_desc[irq].affinity = *cpu_mask; } #endif /* CONFIG_SMP */ diff --git a/arch/mips/include/asm/irq.h b/arch/mips/include/asm/irq.h index a58f0eecc68f..abc62aa744ac 100644 --- a/arch/mips/include/asm/irq.h +++ b/arch/mips/include/asm/irq.h @@ -49,7 +49,8 @@ static inline void smtc_im_ack_irq(unsigned int irq) #ifdef CONFIG_MIPS_MT_SMTC_IRQAFF #include -extern void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity); +extern void plat_set_irq_affinity(unsigned int irq, + const struct cpumask *affinity); extern void smtc_forward_irq(unsigned int irq); /* diff --git a/arch/mips/kernel/cevt-bcm1480.c b/arch/mips/kernel/cevt-bcm1480.c index 0a57f86945f1..d7e21bc8cd21 100644 --- a/arch/mips/kernel/cevt-bcm1480.c +++ b/arch/mips/kernel/cevt-bcm1480.c @@ -148,6 +148,6 @@ void __cpuinit sb1480_clockevent_init(void) action->name = name; action->dev_id = cd; - irq_set_affinity(irq, cpumask_of_cpu(cpu)); + irq_set_affinity(irq, cpumask_of(cpu)); setup_irq(irq, action); } diff --git a/arch/mips/kernel/cevt-sb1250.c b/arch/mips/kernel/cevt-sb1250.c index 63ac3ad462bc..0f188cd46e03 100644 --- a/arch/mips/kernel/cevt-sb1250.c +++ b/arch/mips/kernel/cevt-sb1250.c @@ -147,6 +147,6 @@ void __cpuinit sb1250_clockevent_init(void) action->name = name; action->dev_id = cd; - irq_set_affinity(irq, cpumask_of_cpu(cpu)); + irq_set_affinity(irq, cpumask_of(cpu)); setup_irq(irq, action); } diff --git a/arch/mips/kernel/irq-gic.c b/arch/mips/kernel/irq-gic.c index f0a4bb19e096..494a49a317e9 100644 --- a/arch/mips/kernel/irq-gic.c +++ b/arch/mips/kernel/irq-gic.c @@ -155,7 +155,7 @@ static void gic_unmask_irq(unsigned int irq) static DEFINE_SPINLOCK(gic_lock); -static void gic_set_affinity(unsigned int irq, cpumask_t cpumask) +static void gic_set_affinity(unsigned int irq, const struct cpumask *cpumask) { cpumask_t tmp = CPU_MASK_NONE; unsigned long flags; @@ -164,7 +164,7 @@ static void gic_set_affinity(unsigned int irq, cpumask_t cpumask) pr_debug(KERN_DEBUG "%s called\n", __func__); irq -= _irqbase; - cpus_and(tmp, cpumask, cpu_online_map); + cpumask_and(&tmp, cpumask, cpu_online_mask); if (cpus_empty(tmp)) return; @@ -187,7 +187,7 @@ static void gic_set_affinity(unsigned int irq, cpumask_t cpumask) set_bit(irq, pcpu_masks[first_cpu(tmp)].pcpu_mask); } - irq_desc[irq].affinity = cpumask; + irq_desc[irq].affinity = *cpumask; spin_unlock_irqrestore(&gic_lock, flags); } diff --git a/arch/mips/mti-malta/malta-smtc.c b/arch/mips/mti-malta/malta-smtc.c index f84a46a8ae6e..aabd7274507b 100644 --- a/arch/mips/mti-malta/malta-smtc.c +++ b/arch/mips/mti-malta/malta-smtc.c @@ -114,9 +114,9 @@ struct plat_smp_ops msmtc_smp_ops = { */ -void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity) +void plat_set_irq_affinity(unsigned int irq, const struct cpumask *affinity) { - cpumask_t tmask = affinity; + cpumask_t tmask = *affinity; int cpu = 0; void smtc_set_irq_affinity(unsigned int irq, cpumask_t aff); @@ -139,7 +139,7 @@ void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity) * be made to forward to an offline "CPU". */ - for_each_cpu_mask(cpu, affinity) { + for_each_cpu(cpu, affinity) { if ((cpu_data[cpu].vpe_id != 0) || !cpu_online(cpu)) cpu_clear(cpu, tmask); } diff --git a/arch/mips/sibyte/bcm1480/irq.c b/arch/mips/sibyte/bcm1480/irq.c index a35818ed4263..12b465d404df 100644 --- a/arch/mips/sibyte/bcm1480/irq.c +++ b/arch/mips/sibyte/bcm1480/irq.c @@ -50,7 +50,7 @@ static void enable_bcm1480_irq(unsigned int irq); static void disable_bcm1480_irq(unsigned int irq); static void ack_bcm1480_irq(unsigned int irq); #ifdef CONFIG_SMP -static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask); +static void bcm1480_set_affinity(unsigned int irq, const struct cpumask *mask); #endif #ifdef CONFIG_PCI @@ -109,7 +109,7 @@ void bcm1480_unmask_irq(int cpu, int irq) } #ifdef CONFIG_SMP -static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask) +static void bcm1480_set_affinity(unsigned int irq, const struct cpumask *mask) { int i = 0, old_cpu, cpu, int_on, k; u64 cur_ints; @@ -117,11 +117,11 @@ static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask) unsigned long flags; unsigned int irq_dirty; - if (cpus_weight(mask) != 1) { + if (cpumask_weight(mask) != 1) { printk("attempted to set irq affinity for irq %d to multiple CPUs\n", irq); return; } - i = first_cpu(mask); + i = cpumask_first(mask); /* Convert logical CPU to physical CPU */ cpu = cpu_logical_map(i); diff --git a/arch/mips/sibyte/sb1250/irq.c b/arch/mips/sibyte/sb1250/irq.c index a5158483986e..808ac2959b8c 100644 --- a/arch/mips/sibyte/sb1250/irq.c +++ b/arch/mips/sibyte/sb1250/irq.c @@ -50,7 +50,7 @@ static void enable_sb1250_irq(unsigned int irq); static void disable_sb1250_irq(unsigned int irq); static void ack_sb1250_irq(unsigned int irq); #ifdef CONFIG_SMP -static void sb1250_set_affinity(unsigned int irq, cpumask_t mask); +static void sb1250_set_affinity(unsigned int irq, const struct cpumask *mask); #endif #ifdef CONFIG_SIBYTE_HAS_LDT @@ -103,16 +103,16 @@ void sb1250_unmask_irq(int cpu, int irq) } #ifdef CONFIG_SMP -static void sb1250_set_affinity(unsigned int irq, cpumask_t mask) +static void sb1250_set_affinity(unsigned int irq, const struct cpumask *mask) { int i = 0, old_cpu, cpu, int_on; u64 cur_ints; struct irq_desc *desc = irq_desc + irq; unsigned long flags; - i = first_cpu(mask); + i = cpumask_first(mask); - if (cpus_weight(mask) > 1) { + if (cpumask_weight(mask) > 1) { printk("attempted to set irq affinity for irq %d to multiple CPUs\n", irq); return; } diff --git a/arch/parisc/kernel/irq.c b/arch/parisc/kernel/irq.c index 23ef950df008..4cea935e2f99 100644 --- a/arch/parisc/kernel/irq.c +++ b/arch/parisc/kernel/irq.c @@ -131,12 +131,12 @@ int cpu_check_affinity(unsigned int irq, cpumask_t *dest) return 0; } -static void cpu_set_affinity_irq(unsigned int irq, cpumask_t dest) +static void cpu_set_affinity_irq(unsigned int irq, const struct cpumask *dest) { - if (cpu_check_affinity(irq, &dest)) + if (cpu_check_affinity(irq, dest)) return; - irq_desc[irq].affinity = dest; + irq_desc[irq].affinity = *dest; } #endif diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index ac222d0ab12e..23b8b5e36f98 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -237,7 +237,7 @@ void fixup_irqs(cpumask_t map) mask = map; } if (irq_desc[irq].chip->set_affinity) - irq_desc[irq].chip->set_affinity(irq, mask); + irq_desc[irq].chip->set_affinity(irq, &mask); else if (irq_desc[irq].action && !(warned++)) printk("Cannot set affinity for irq %i\n", irq); } diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c index 64d24310ce7e..424b335a71c8 100644 --- a/arch/powerpc/platforms/pseries/xics.c +++ b/arch/powerpc/platforms/pseries/xics.c @@ -332,7 +332,7 @@ static void xics_eoi_lpar(unsigned int virq) lpar_xirr_info_set((0xff << 24) | irq); } -static void xics_set_affinity(unsigned int virq, cpumask_t cpumask) +static void xics_set_affinity(unsigned int virq, const struct cpumask *cpumask) { unsigned int irq; int status; @@ -358,7 +358,7 @@ static void xics_set_affinity(unsigned int virq, cpumask_t cpumask) irq_server = get_irq_server(virq, 1); if (irq_server == -1) { char cpulist[128]; - cpumask_scnprintf(cpulist, sizeof(cpulist), &cpumask); + cpumask_scnprintf(cpulist, sizeof(cpulist), cpumask); printk(KERN_WARNING "%s: No online cpus in the mask %s for irq %d\n", __func__, cpulist, virq); @@ -845,7 +845,7 @@ void xics_migrate_irqs_away(void) /* Reset affinity to all cpus */ irq_desc[virq].affinity = CPU_MASK_ALL; - desc->chip->set_affinity(virq, CPU_MASK_ALL); + desc->chip->set_affinity(virq, cpu_all_mask); unlock: spin_unlock_irqrestore(&desc->lock, flags); } diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c index 1890fb085cde..5d7f9f0c93c3 100644 --- a/arch/powerpc/sysdev/mpic.c +++ b/arch/powerpc/sysdev/mpic.c @@ -817,7 +817,7 @@ static void mpic_end_ipi(unsigned int irq) #endif /* CONFIG_SMP */ -void mpic_set_affinity(unsigned int irq, cpumask_t cpumask) +void mpic_set_affinity(unsigned int irq, const struct cpumask *cpumask) { struct mpic *mpic = mpic_from_irq(irq); unsigned int src = mpic_irq_to_hw(irq); @@ -829,7 +829,7 @@ void mpic_set_affinity(unsigned int irq, cpumask_t cpumask) } else { cpumask_t tmp; - cpus_and(tmp, cpumask, cpu_online_map); + cpumask_and(&tmp, cpumask, cpu_online_mask); mpic_irq_write(src, MPIC_INFO(IRQ_DESTINATION), mpic_physmask(cpus_addr(tmp)[0])); diff --git a/arch/powerpc/sysdev/mpic.h b/arch/powerpc/sysdev/mpic.h index 6209c62a426d..3cef2af10f42 100644 --- a/arch/powerpc/sysdev/mpic.h +++ b/arch/powerpc/sysdev/mpic.h @@ -36,6 +36,6 @@ static inline int mpic_pasemi_msi_init(struct mpic *mpic) extern int mpic_set_irq_type(unsigned int virq, unsigned int flow_type); extern void mpic_set_vector(unsigned int virq, unsigned int vector); -extern void mpic_set_affinity(unsigned int irq, cpumask_t cpumask); +extern void mpic_set_affinity(unsigned int irq, const struct cpumask *cpumask); #endif /* _POWERPC_SYSDEV_MPIC_H */ diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c index 52fc836f464d..4aaf18e83c8c 100644 --- a/arch/sparc64/kernel/irq.c +++ b/arch/sparc64/kernel/irq.c @@ -312,7 +312,8 @@ static void sun4u_irq_enable(unsigned int virt_irq) } } -static void sun4u_set_affinity(unsigned int virt_irq, cpumask_t mask) +static void sun4u_set_affinity(unsigned int virt_irq, + const struct cpumask *mask) { sun4u_irq_enable(virt_irq); } @@ -362,7 +363,8 @@ static void sun4v_irq_enable(unsigned int virt_irq) ino, err); } -static void sun4v_set_affinity(unsigned int virt_irq, cpumask_t mask) +static void sun4v_set_affinity(unsigned int virt_irq, + const struct cpumask *mask) { unsigned int ino = virt_irq_table[virt_irq].dev_ino; unsigned long cpuid = irq_choose_cpu(virt_irq); @@ -429,7 +431,8 @@ static void sun4v_virq_enable(unsigned int virt_irq) dev_handle, dev_ino, err); } -static void sun4v_virt_set_affinity(unsigned int virt_irq, cpumask_t mask) +static void sun4v_virt_set_affinity(unsigned int virt_irq, + const struct cpumask *mask) { unsigned long cpuid, dev_handle, dev_ino; int err; @@ -788,7 +791,7 @@ void fixup_irqs(void) !(irq_desc[irq].status & IRQ_PER_CPU)) { if (irq_desc[irq].chip->set_affinity) irq_desc[irq].chip->set_affinity(irq, - irq_desc[irq].affinity); + &irq_desc[irq].affinity); } spin_unlock_irqrestore(&irq_desc[irq].lock, flags); } diff --git a/arch/sparc64/kernel/of_device.c b/arch/sparc64/kernel/of_device.c index 0f616ae3246c..df2efb7fc14c 100644 --- a/arch/sparc64/kernel/of_device.c +++ b/arch/sparc64/kernel/of_device.c @@ -780,7 +780,7 @@ out: if (nid != -1) { cpumask_t numa_mask = node_to_cpumask(nid); - irq_set_affinity(irq, numa_mask); + irq_set_affinity(irq, &numa_mask); } return irq; diff --git a/arch/sparc64/kernel/pci_msi.c b/arch/sparc64/kernel/pci_msi.c index 2e680f34f727..0d0cd815e83e 100644 --- a/arch/sparc64/kernel/pci_msi.c +++ b/arch/sparc64/kernel/pci_msi.c @@ -288,7 +288,7 @@ static int bringup_one_msi_queue(struct pci_pbm_info *pbm, if (nid != -1) { cpumask_t numa_mask = node_to_cpumask(nid); - irq_set_affinity(irq, numa_mask); + irq_set_affinity(irq, &numa_mask); } err = request_irq(irq, sparc64_msiq_interrupt, 0, "MSIQ", diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 067d8de913f6..940f25851e1e 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -301,7 +301,7 @@ static void hpet_set_mode(enum clock_event_mode mode, struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt); hpet_setup_msi_irq(hdev->irq); disable_irq(hdev->irq); - irq_set_affinity(hdev->irq, cpumask_of_cpu(hdev->cpu)); + irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu)); enable_irq(hdev->irq); } break; @@ -449,7 +449,7 @@ static int hpet_setup_irq(struct hpet_dev *dev) return -1; disable_irq(dev->irq); - irq_set_affinity(dev->irq, cpumask_of_cpu(dev->cpu)); + irq_set_affinity(dev->irq, cpumask_of(dev->cpu)); enable_irq(dev->irq); printk(KERN_DEBUG "hpet: %s irq %d for MSI\n", diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c index 9043251210fb..1184210e6d0c 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/io_apic.c @@ -361,7 +361,8 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, u8 vector) static int assign_irq_vector(int irq, cpumask_t mask); -static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) +static void set_ioapic_affinity_irq(unsigned int irq, + const struct cpumask *mask) { struct irq_cfg *cfg; unsigned long flags; @@ -369,15 +370,14 @@ static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) cpumask_t tmp; struct irq_desc *desc; - cpus_and(tmp, mask, cpu_online_map); - if (cpus_empty(tmp)) + if (!cpumask_intersects(mask, cpu_online_mask)) return; cfg = irq_cfg(irq); - if (assign_irq_vector(irq, mask)) + if (assign_irq_vector(irq, *mask)) return; - cpus_and(tmp, cfg->domain, mask); + cpumask_and(&tmp, &cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); /* * Only the high 8 bits are valid. @@ -387,7 +387,7 @@ static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) desc = irq_to_desc(irq); spin_lock_irqsave(&ioapic_lock, flags); __target_IO_APIC_irq(irq, dest, cfg->vector); - desc->affinity = mask; + cpumask_copy(&desc->affinity, mask); spin_unlock_irqrestore(&ioapic_lock, flags); } #endif /* CONFIG_SMP */ @@ -2189,7 +2189,7 @@ static void ir_irq_migration(struct work_struct *work) continue; } - desc->chip->set_affinity(irq, desc->pending_mask); + desc->chip->set_affinity(irq, &desc->pending_mask); spin_unlock_irqrestore(&desc->lock, flags); } } @@ -2198,18 +2198,19 @@ static void ir_irq_migration(struct work_struct *work) /* * Migrates the IRQ destination in the process context. */ -static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask) +static void set_ir_ioapic_affinity_irq(unsigned int irq, + const struct cpumask *mask) { struct irq_desc *desc = irq_to_desc(irq); if (desc->status & IRQ_LEVEL) { desc->status |= IRQ_MOVE_PENDING; - desc->pending_mask = mask; + cpumask_copy(&desc->pending_mask, mask); migrate_irq_remapped_level(irq); return; } - migrate_ioapic_irq(irq, mask); + migrate_ioapic_irq(irq, *mask); } #endif @@ -3027,7 +3028,7 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms } #ifdef CONFIG_SMP -static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) +static void set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) { struct irq_cfg *cfg; struct msi_msg msg; @@ -3035,15 +3036,14 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) cpumask_t tmp; struct irq_desc *desc; - cpus_and(tmp, mask, cpu_online_map); - if (cpus_empty(tmp)) + if (!cpumask_intersects(mask, cpu_online_mask)) return; - if (assign_irq_vector(irq, mask)) + if (assign_irq_vector(irq, *mask)) return; cfg = irq_cfg(irq); - cpus_and(tmp, cfg->domain, mask); + cpumask_and(&tmp, &cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); read_msi_msg(irq, &msg); @@ -3055,7 +3055,7 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) write_msi_msg(irq, &msg); desc = irq_to_desc(irq); - desc->affinity = mask; + cpumask_copy(&desc->affinity, mask); } #ifdef CONFIG_INTR_REMAP @@ -3063,7 +3063,8 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask) * Migrate the MSI irq to another cpumask. This migration is * done in the process context using interrupt-remapping hardware. */ -static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask) +static void ir_set_msi_irq_affinity(unsigned int irq, + const struct cpumask *mask) { struct irq_cfg *cfg; unsigned int dest; @@ -3071,18 +3072,17 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask) struct irte irte; struct irq_desc *desc; - cpus_and(tmp, mask, cpu_online_map); - if (cpus_empty(tmp)) + if (!cpumask_intersects(mask, cpu_online_mask)) return; if (get_irte(irq, &irte)) return; - if (assign_irq_vector(irq, mask)) + if (assign_irq_vector(irq, *mask)) return; cfg = irq_cfg(irq); - cpus_and(tmp, cfg->domain, mask); + cpumask_and(&tmp, &cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); irte.vector = cfg->vector; @@ -3106,7 +3106,7 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask) } desc = irq_to_desc(irq); - desc->affinity = mask; + cpumask_copy(&desc->affinity, mask); } #endif #endif /* CONFIG_SMP */ @@ -3308,7 +3308,7 @@ void arch_teardown_msi_irq(unsigned int irq) #ifdef CONFIG_DMAR #ifdef CONFIG_SMP -static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) +static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) { struct irq_cfg *cfg; struct msi_msg msg; @@ -3316,15 +3316,14 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) cpumask_t tmp; struct irq_desc *desc; - cpus_and(tmp, mask, cpu_online_map); - if (cpus_empty(tmp)) + if (!cpumask_intersects(mask, cpu_online_mask)) return; - if (assign_irq_vector(irq, mask)) + if (assign_irq_vector(irq, *mask)) return; cfg = irq_cfg(irq); - cpus_and(tmp, cfg->domain, mask); + cpumask_and(&tmp, &cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); dmar_msi_read(irq, &msg); @@ -3336,7 +3335,7 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask) dmar_msi_write(irq, &msg); desc = irq_to_desc(irq); - desc->affinity = mask; + cpumask_copy(&desc->affinity, mask); } #endif /* CONFIG_SMP */ @@ -3369,7 +3368,7 @@ int arch_setup_dmar_msi(unsigned int irq) #ifdef CONFIG_HPET_TIMER #ifdef CONFIG_SMP -static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask) +static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask) { struct irq_cfg *cfg; struct irq_desc *desc; @@ -3377,15 +3376,14 @@ static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask) unsigned int dest; cpumask_t tmp; - cpus_and(tmp, mask, cpu_online_map); - if (cpus_empty(tmp)) + if (!cpumask_intersects(mask, cpu_online_mask)) return; - if (assign_irq_vector(irq, mask)) + if (assign_irq_vector(irq, *mask)) return; cfg = irq_cfg(irq); - cpus_and(tmp, cfg->domain, mask); + cpumask_and(&tmp, &cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); hpet_msi_read(irq, &msg); @@ -3397,7 +3395,7 @@ static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask) hpet_msi_write(irq, &msg); desc = irq_to_desc(irq); - desc->affinity = mask; + cpumask_copy(&desc->affinity, mask); } #endif /* CONFIG_SMP */ @@ -3451,27 +3449,26 @@ static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector) write_ht_irq_msg(irq, &msg); } -static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) +static void set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask) { struct irq_cfg *cfg; unsigned int dest; cpumask_t tmp; struct irq_desc *desc; - cpus_and(tmp, mask, cpu_online_map); - if (cpus_empty(tmp)) + if (!cpumask_intersects(mask, cpu_online_mask)) return; - if (assign_irq_vector(irq, mask)) + if (assign_irq_vector(irq, *mask)) return; cfg = irq_cfg(irq); - cpus_and(tmp, cfg->domain, mask); + cpumask_and(&tmp, &cfg->domain, mask); dest = cpu_mask_to_apicid(tmp); target_ht_irq(irq, dest, cfg->vector); desc = irq_to_desc(irq); - desc->affinity = mask; + cpumask_copy(&desc->affinity, mask); } #endif @@ -3794,10 +3791,10 @@ void __init setup_ioapic_dest(void) #ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) - set_ir_ioapic_affinity_irq(irq, mask); + set_ir_ioapic_affinity_irq(irq, &mask); else #endif - set_ioapic_affinity_irq(irq, mask); + set_ioapic_affinity_irq(irq, &mask); } } diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index a51382672de0..87870a49be4e 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -251,7 +251,7 @@ void fixup_irqs(cpumask_t map) mask = map; } if (desc->chip->set_affinity) - desc->chip->set_affinity(irq, mask); + desc->chip->set_affinity(irq, &mask); else if (desc->action && !(warned++)) printk("Cannot set affinity for irq %i\n", irq); } diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c index 60eb84eb77a0..7d37f847544d 100644 --- a/arch/x86/kernel/irq_64.c +++ b/arch/x86/kernel/irq_64.c @@ -116,7 +116,7 @@ void fixup_irqs(cpumask_t map) desc->chip->mask(irq); if (desc->chip->set_affinity) - desc->chip->set_affinity(irq, mask); + desc->chip->set_affinity(irq, &mask); else if (!(warned++)) set_affinity = 0; diff --git a/drivers/parisc/iosapic.c b/drivers/parisc/iosapic.c index 7beffcab2745..9dedbbd218c3 100644 --- a/drivers/parisc/iosapic.c +++ b/drivers/parisc/iosapic.c @@ -704,16 +704,17 @@ static unsigned int iosapic_startup_irq(unsigned int irq) } #ifdef CONFIG_SMP -static void iosapic_set_affinity_irq(unsigned int irq, cpumask_t dest) +static void iosapic_set_affinity_irq(unsigned int irq, + const struct cpumask *dest) { struct vector_info *vi = iosapic_get_vector(irq); u32 d0, d1, dummy_d0; unsigned long flags; - if (cpu_check_affinity(irq, &dest)) + if (cpu_check_affinity(irq, dest)) return; - vi->txn_addr = txn_affinity_addr(irq, first_cpu(dest)); + vi->txn_addr = txn_affinity_addr(irq, cpumask_first(dest)); spin_lock_irqsave(&iosapic_lock, flags); /* d1 contains the destination CPU, so only want to set that diff --git a/drivers/xen/events.c b/drivers/xen/events.c index 1e3b934a4cf7..eba5ec5b020e 100644 --- a/drivers/xen/events.c +++ b/drivers/xen/events.c @@ -579,7 +579,7 @@ void rebind_evtchn_irq(int evtchn, int irq) spin_unlock(&irq_mapping_update_lock); /* new event channels are always bound to cpu 0 */ - irq_set_affinity(irq, cpumask_of_cpu(0)); + irq_set_affinity(irq, cpumask_of(0)); /* Unmask the event channel. */ enable_irq(irq); @@ -608,9 +608,9 @@ static void rebind_irq_to_cpu(unsigned irq, unsigned tcpu) } -static void set_affinity_irq(unsigned irq, cpumask_t dest) +static void set_affinity_irq(unsigned irq, const struct cpumask *dest) { - unsigned tcpu = first_cpu(dest); + unsigned tcpu = cpumask_first(dest); rebind_irq_to_cpu(irq, tcpu); } diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index f58a0cf8929a..48e63934fabe 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -109,13 +109,13 @@ extern void enable_irq(unsigned int irq); extern cpumask_t irq_default_affinity; -extern int irq_set_affinity(unsigned int irq, cpumask_t cpumask); +extern int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask); extern int irq_can_set_affinity(unsigned int irq); extern int irq_select_affinity(unsigned int irq); #else /* CONFIG_SMP */ -static inline int irq_set_affinity(unsigned int irq, cpumask_t cpumask) +static inline int irq_set_affinity(unsigned int irq, const struct cpumask *m) { return -EINVAL; } diff --git a/include/linux/irq.h b/include/linux/irq.h index 3dddfa703ebd..ab70fd604d3a 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -113,7 +113,8 @@ struct irq_chip { void (*eoi)(unsigned int irq); void (*end)(unsigned int irq); - void (*set_affinity)(unsigned int irq, cpumask_t dest); + void (*set_affinity)(unsigned int irq, + const struct cpumask *dest); int (*retrigger)(unsigned int irq); int (*set_type)(unsigned int irq, unsigned int flow_type); int (*set_wake)(unsigned int irq, unsigned int on); diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index 10b5092e9bfe..58d8e31daa49 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -45,7 +45,7 @@ void dynamic_irq_init(unsigned int irq) desc->irq_count = 0; desc->irqs_unhandled = 0; #ifdef CONFIG_SMP - cpus_setall(desc->affinity); + cpumask_setall(&desc->affinity); #endif spin_unlock_irqrestore(&desc->lock, flags); } diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 801addda3c43..10ad2f87ed9a 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -79,7 +79,7 @@ int irq_can_set_affinity(unsigned int irq) * @cpumask: cpumask * */ -int irq_set_affinity(unsigned int irq, cpumask_t cpumask) +int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask) { struct irq_desc *desc = irq_to_desc(irq); unsigned long flags; @@ -91,14 +91,14 @@ int irq_set_affinity(unsigned int irq, cpumask_t cpumask) #ifdef CONFIG_GENERIC_PENDING_IRQ if (desc->status & IRQ_MOVE_PCNTXT || desc->status & IRQ_DISABLED) { - desc->affinity = cpumask; + cpumask_copy(&desc->affinity, cpumask); desc->chip->set_affinity(irq, cpumask); } else { desc->status |= IRQ_MOVE_PENDING; - desc->pending_mask = cpumask; + cpumask_copy(&desc->pending_mask, cpumask); } #else - desc->affinity = cpumask; + cpumask_copy(&desc->affinity, cpumask); desc->chip->set_affinity(irq, cpumask); #endif desc->status |= IRQ_AFFINITY_SET; @@ -112,26 +112,24 @@ int irq_set_affinity(unsigned int irq, cpumask_t cpumask) */ int do_irq_select_affinity(unsigned int irq, struct irq_desc *desc) { - cpumask_t mask; - if (!irq_can_set_affinity(irq)) return 0; - cpus_and(mask, cpu_online_map, irq_default_affinity); - /* * Preserve an userspace affinity setup, but make sure that * one of the targets is online. */ if (desc->status & (IRQ_AFFINITY_SET | IRQ_NO_BALANCING)) { - if (cpus_intersects(desc->affinity, cpu_online_map)) - mask = desc->affinity; + if (cpumask_any_and(&desc->affinity, cpu_online_mask) + < nr_cpu_ids) + goto set_affinity; else desc->status &= ~IRQ_AFFINITY_SET; } - desc->affinity = mask; - desc->chip->set_affinity(irq, mask); + cpumask_and(&desc->affinity, cpu_online_mask, &irq_default_affinity); +set_affinity: + desc->chip->set_affinity(irq, &desc->affinity); return 0; } diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c index 9db681d95814..bd72329e630c 100644 --- a/kernel/irq/migration.c +++ b/kernel/irq/migration.c @@ -4,7 +4,6 @@ void move_masked_irq(int irq) { struct irq_desc *desc = irq_to_desc(irq); - cpumask_t tmp; if (likely(!(desc->status & IRQ_MOVE_PENDING))) return; @@ -19,7 +18,7 @@ void move_masked_irq(int irq) desc->status &= ~IRQ_MOVE_PENDING; - if (unlikely(cpus_empty(desc->pending_mask))) + if (unlikely(cpumask_empty(&desc->pending_mask))) return; if (!desc->chip->set_affinity) @@ -27,8 +26,6 @@ void move_masked_irq(int irq) assert_spin_locked(&desc->lock); - cpus_and(tmp, desc->pending_mask, cpu_online_map); - /* * If there was a valid mask to work with, please * do the disable, re-program, enable sequence. @@ -41,10 +38,13 @@ void move_masked_irq(int irq) * For correct operation this depends on the caller * masking the irqs. */ - if (likely(!cpus_empty(tmp))) { - desc->chip->set_affinity(irq,tmp); + if (likely(cpumask_any_and(&desc->pending_mask, cpu_online_mask) + < nr_cpu_ids)) { + cpumask_and(&desc->affinity, + &desc->pending_mask, cpu_online_mask); + desc->chip->set_affinity(irq, &desc->affinity); } - cpus_clear(desc->pending_mask); + cpumask_clear(&desc->pending_mask); } void move_native_irq(int irq) diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index f293349d49d0..8e91c9762520 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -40,33 +40,42 @@ static ssize_t irq_affinity_proc_write(struct file *file, const char __user *buffer, size_t count, loff_t *pos) { unsigned int irq = (int)(long)PDE(file->f_path.dentry->d_inode)->data; - cpumask_t new_value; + cpumask_var_t new_value; int err; if (!irq_to_desc(irq)->chip->set_affinity || no_irq_affinity || irq_balancing_disabled(irq)) return -EIO; - err = cpumask_parse_user(buffer, count, &new_value); + if (!alloc_cpumask_var(&new_value, GFP_KERNEL)) + return -ENOMEM; + + err = cpumask_parse_user(buffer, count, new_value); if (err) - return err; + goto free_cpumask; - if (!is_affinity_mask_valid(new_value)) - return -EINVAL; + if (!is_affinity_mask_valid(*new_value)) { + err = -EINVAL; + goto free_cpumask; + } /* * Do not allow disabling IRQs completely - it's a too easy * way to make the system unusable accidentally :-) At least * one online CPU still has to be targeted. */ - if (!cpus_intersects(new_value, cpu_online_map)) + if (!cpumask_intersects(new_value, cpu_online_mask)) { /* Special case for empty set - allow the architecture code to set default SMP affinity. */ - return irq_select_affinity_usr(irq) ? -EINVAL : count; - - irq_set_affinity(irq, new_value); + err = irq_select_affinity_usr(irq) ? -EINVAL : count; + } else { + irq_set_affinity(irq, new_value); + err = count; + } - return count; +free_cpumask: + free_cpumask_var(new_value); + return err; } static int irq_affinity_proc_open(struct inode *inode, struct file *file) diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index df12434b43ca..ab65d217583f 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -136,7 +136,7 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast) */ static void tick_setup_device(struct tick_device *td, struct clock_event_device *newdev, int cpu, - const cpumask_t *cpumask) + const struct cpumask *cpumask) { ktime_t next_event; void (*handler)(struct clock_event_device *) = NULL; @@ -171,8 +171,8 @@ static void tick_setup_device(struct tick_device *td, * When the device is not per cpu, pin the interrupt to the * current cpu: */ - if (!cpus_equal(newdev->cpumask, *cpumask)) - irq_set_affinity(newdev->irq, *cpumask); + if (!cpumask_equal(&newdev->cpumask, cpumask)) + irq_set_affinity(newdev->irq, cpumask); /* * When global broadcasting is active, check if the current -- cgit v1.2.3-55-g7522 From f9af0e70911e9d6cc9a68f784dca86415486084d Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Fri, 26 Dec 2008 12:24:24 +0900 Subject: irq: for_each_irq_desc() move to irqnr.h Impact: cleanup before CONFIG_SPARSE_IRQ age, for_each_irq_desc() sat in irqnr.h and could be called from generic code. CONFIG_SPARSE_IRQ breaks this assumption, but SPARSE_IRQ version for_each_irq_desc() also can move into irqnr.h easily. Also, this patch unifies CONFIG_SPARSE_IRQ and !CONFIG_SPARSE_IRQ for_each_irq_desc(). Signed-off-by: KOSAKI Motohiro Signed-off-by: Ingo Molnar --- include/linux/irq.h | 24 ++++-------------------- include/linux/irqnr.h | 19 +++++++++---------- kernel/irq/handle.c | 13 +++++++++++-- 3 files changed, 24 insertions(+), 32 deletions(-) (limited to 'kernel/irq') diff --git a/include/linux/irq.h b/include/linux/irq.h index 98564dc64476..69da275c0ebd 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -202,33 +202,17 @@ extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc #ifndef CONFIG_SPARSE_IRQ extern struct irq_desc irq_desc[NR_IRQS]; - -static inline struct irq_desc *irq_to_desc(unsigned int irq) -{ - return (irq < NR_IRQS) ? irq_desc + irq : NULL; -} -static inline struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu) -{ - return irq_to_desc(irq); -} - -#else - -extern struct irq_desc *irq_to_desc(unsigned int irq); -extern struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu); +#else /* CONFIG_SPARSE_IRQ */ extern struct irq_desc *move_irq_desc(struct irq_desc *old_desc, int cpu); -# define for_each_irq_desc(irq, desc) \ - for (irq = 0, desc = irq_to_desc(irq); irq < nr_irqs; irq++, desc = irq_to_desc(irq)) -# define for_each_irq_desc_reverse(irq, desc) \ - for (irq = nr_irqs - 1, desc = irq_to_desc(irq); irq >= 0; irq--, desc = irq_to_desc(irq)) - #define kstat_irqs_this_cpu(DESC) \ ((DESC)->kstat_irqs[smp_processor_id()]) #define kstat_incr_irqs_this_cpu(irqno, DESC) \ ((DESC)->kstat_irqs[smp_processor_id()]++) -#endif +#endif /* CONFIG_SPARSE_IRQ */ + +extern struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu); static inline struct irq_desc * irq_remap_to_desc(unsigned int irq, struct irq_desc *desc) diff --git a/include/linux/irqnr.h b/include/linux/irqnr.h index 95d2b74641f5..c4a59c7a478b 100644 --- a/include/linux/irqnr.h +++ b/include/linux/irqnr.h @@ -15,20 +15,19 @@ # define for_each_irq_desc_reverse(irq, desc) \ for (irq = nr_irqs - 1; irq >= 0; irq--) -#else +#else /* CONFIG_GENERIC_HARDIRQS */ extern int nr_irqs; +extern struct irq_desc *irq_to_desc(unsigned int irq); -#ifndef CONFIG_SPARSE_IRQ +# define for_each_irq_desc(irq, desc) \ + for (irq = 0, desc = irq_to_desc(irq); irq < nr_irqs; \ + irq++, desc = irq_to_desc(irq)) +# define for_each_irq_desc_reverse(irq, desc) \ + for (irq = nr_irqs - 1, desc = irq_to_desc(irq); irq >= 0; \ + irq--, desc = irq_to_desc(irq)) -struct irq_desc; -# define for_each_irq_desc(irq, desc) \ - for (irq = 0, desc = irq_desc; irq < nr_irqs; irq++, desc++) -# define for_each_irq_desc_reverse(irq, desc) \ - for (irq = nr_irqs - 1, desc = irq_desc + (nr_irqs - 1); \ - irq >= 0; irq--, desc--) -#endif -#endif +#endif /* CONFIG_GENERIC_HARDIRQS */ #define for_each_irq_nr(irq) \ for (irq = 0; irq < nr_irqs; irq++) diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 6492400cb50d..4db7d2df86b6 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -203,7 +203,7 @@ out_unlock: return desc; } -#else +#else /* !CONFIG_SPARSE_IRQ */ struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = { [0 ... NR_IRQS-1] = { @@ -218,7 +218,16 @@ struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = { } }; -#endif +struct irq_desc *irq_to_desc(unsigned int irq) +{ + return (irq < NR_IRQS) ? irq_desc + irq : NULL; +} + +struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu) +{ + return irq_to_desc(irq); +} +#endif /* !CONFIG_SPARSE_IRQ */ /* * What should we do if we get a hw irq event on an illegal vector? -- cgit v1.2.3-55-g7522 From 26ddd8d5cac8a563953d5febe8c6e40909f7bce1 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Fri, 26 Dec 2008 14:24:10 +0900 Subject: proc: remove ifdef CONFIG_SPARSE_IRQ from stat.c Impact: cleanup irq_desc can be NULL when CONFIG_SPARSE_IRQ=y only. therefore, NULL checking can move into kstat_irqs_cpu() of SPARSE_IRQ version. Signed-off-by: KOSAKI Motohiro Acked-by: "Yinghai Lu" Signed-off-by: Ingo Molnar --- fs/proc/stat.c | 11 +---------- kernel/irq/handle.c | 2 +- 2 files changed, 2 insertions(+), 11 deletions(-) (limited to 'kernel/irq') diff --git a/fs/proc/stat.c b/fs/proc/stat.c index 3bb1cf1e7425..f75efa22df5e 100644 --- a/fs/proc/stat.c +++ b/fs/proc/stat.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #ifndef arch_irq_stat_cpu @@ -45,10 +46,6 @@ static int show_stat(struct seq_file *p, void *v) steal = cputime64_add(steal, kstat_cpu(i).cpustat.steal); guest = cputime64_add(guest, kstat_cpu(i).cpustat.guest); for_each_irq_nr(j) { -#ifdef CONFIG_SPARSE_IRQ - if (!irq_to_desc(j)) - continue; -#endif sum += kstat_irqs_cpu(j, i); } sum += arch_irq_stat_cpu(i); @@ -95,12 +92,6 @@ static int show_stat(struct seq_file *p, void *v) /* sum again ? it could be updated? */ for_each_irq_nr(j) { per_irq_sum = 0; -#ifdef CONFIG_SPARSE_IRQ - if (!irq_to_desc(j)) { - seq_printf(p, " %u", per_irq_sum); - continue; - } -#endif for_each_possible_cpu(i) per_irq_sum += kstat_irqs_cpu(j, i); diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 4db7d2df86b6..03479dfdebb8 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -448,7 +448,7 @@ void early_init_irq_lock_class(void) unsigned int kstat_irqs_cpu(unsigned int irq, int cpu) { struct irq_desc *desc = irq_to_desc(irq); - return desc->kstat_irqs[cpu]; + return desc ? desc->kstat_irqs[cpu] : 0; } #endif EXPORT_SYMBOL(kstat_irqs_cpu); -- cgit v1.2.3-55-g7522 From 18eefedfe8ad33e8fc7614c13359e29a9fab4644 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Fri, 26 Dec 2008 12:29:48 +0900 Subject: irq: simplify for_each_irq_desc() usage Impact: cleanup all for_each_irq_desc() usage point have !desc check. then its check can move into for_each_irq_desc() macro. Signed-off-by: KOSAKI Motohiro Acked-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/io_apic.c | 10 ---------- drivers/xen/events.c | 3 --- include/linux/irqnr.h | 8 ++++++-- kernel/irq/autoprobe.c | 15 --------------- kernel/irq/handle.c | 3 --- kernel/irq/spurious.c | 5 ----- 6 files changed, 6 insertions(+), 38 deletions(-) (limited to 'kernel/irq') diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c index a74887b416cc..2fe543f58ac8 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/io_apic.c @@ -1345,8 +1345,6 @@ void __setup_vector_irq(int cpu) /* Mark the inuse vectors */ for_each_irq_desc(irq, desc) { - if (!desc) - continue; cfg = desc->chip_data; if (!cpu_isset(cpu, cfg->domain)) continue; @@ -1730,8 +1728,6 @@ __apicdebuginit(void) print_IO_APIC(void) for_each_irq_desc(irq, desc) { struct irq_pin_list *entry; - if (!desc) - continue; cfg = desc->chip_data; entry = cfg->irq_2_pin; if (!entry) @@ -2378,9 +2374,6 @@ static void ir_irq_migration(struct work_struct *work) struct irq_desc *desc; for_each_irq_desc(irq, desc) { - if (!desc) - continue; - if (desc->status & IRQ_MOVE_PENDING) { unsigned long flags; @@ -2671,9 +2664,6 @@ static inline void init_IO_APIC_traps(void) * 0x80, because int 0x80 is hm, kind of importantish. ;) */ for_each_irq_desc(irq, desc) { - if (!desc) - continue; - cfg = desc->chip_data; if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) { /* diff --git a/drivers/xen/events.c b/drivers/xen/events.c index 46625cd38743..e26733a9df21 100644 --- a/drivers/xen/events.c +++ b/drivers/xen/events.c @@ -142,9 +142,6 @@ static void init_evtchn_cpu_bindings(void) /* By default all event channels notify CPU#0. */ for_each_irq_desc(i, desc) { - if (!desc) - continue; - desc->affinity = cpumask_of_cpu(0); } #endif diff --git a/include/linux/irqnr.h b/include/linux/irqnr.h index c4a59c7a478b..5504a5c97836 100644 --- a/include/linux/irqnr.h +++ b/include/linux/irqnr.h @@ -22,10 +22,14 @@ extern struct irq_desc *irq_to_desc(unsigned int irq); # define for_each_irq_desc(irq, desc) \ for (irq = 0, desc = irq_to_desc(irq); irq < nr_irqs; \ - irq++, desc = irq_to_desc(irq)) + irq++, desc = irq_to_desc(irq)) \ + if (desc) + + # define for_each_irq_desc_reverse(irq, desc) \ for (irq = nr_irqs - 1, desc = irq_to_desc(irq); irq >= 0; \ - irq--, desc = irq_to_desc(irq)) + irq--, desc = irq_to_desc(irq)) \ + if (desc) #endif /* CONFIG_GENERIC_HARDIRQS */ diff --git a/kernel/irq/autoprobe.c b/kernel/irq/autoprobe.c index 650ce4102a63..cc0f7321b8ce 100644 --- a/kernel/irq/autoprobe.c +++ b/kernel/irq/autoprobe.c @@ -40,9 +40,6 @@ unsigned long probe_irq_on(void) * flush such a longstanding irq before considering it as spurious. */ for_each_irq_desc_reverse(i, desc) { - if (!desc) - continue; - spin_lock_irq(&desc->lock); if (!desc->action && !(desc->status & IRQ_NOPROBE)) { /* @@ -71,9 +68,6 @@ unsigned long probe_irq_on(void) * happened in the previous stage, it may have masked itself) */ for_each_irq_desc_reverse(i, desc) { - if (!desc) - continue; - spin_lock_irq(&desc->lock); if (!desc->action && !(desc->status & IRQ_NOPROBE)) { desc->status |= IRQ_AUTODETECT | IRQ_WAITING; @@ -92,9 +86,6 @@ unsigned long probe_irq_on(void) * Now filter out any obviously spurious interrupts */ for_each_irq_desc(i, desc) { - if (!desc) - continue; - spin_lock_irq(&desc->lock); status = desc->status; @@ -133,9 +124,6 @@ unsigned int probe_irq_mask(unsigned long val) int i; for_each_irq_desc(i, desc) { - if (!desc) - continue; - spin_lock_irq(&desc->lock); status = desc->status; @@ -178,9 +166,6 @@ int probe_irq_off(unsigned long val) unsigned int status; for_each_irq_desc(i, desc) { - if (!desc) - continue; - spin_lock_irq(&desc->lock); status = desc->status; diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 03479dfdebb8..7dbdfe524693 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -437,9 +437,6 @@ void early_init_irq_lock_class(void) int i; for_each_irq_desc(i, desc) { - if (!desc) - continue; - lockdep_set_class(&desc->lock, &irq_desc_lock_class); } } diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index 3738107531fd..dd364c11e56e 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -91,9 +91,6 @@ static int misrouted_irq(int irq) int i, ok = 0; for_each_irq_desc(i, desc) { - if (!desc) - continue; - if (!i) continue; @@ -115,8 +112,6 @@ static void poll_spurious_irqs(unsigned long dummy) for_each_irq_desc(i, desc) { unsigned int status; - if (!desc) - continue; if (!i) continue; -- cgit v1.2.3-55-g7522 From 00c23634879062d1c38d60128bf150c394a359e8 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Tue, 23 Dec 2008 17:29:00 -0800 Subject: sparseirq: remove duplicated arch_early_irq_init() Impact: clean up We already have a weak copy of this function in init/main.c Signed-off-by: Yinghai Signed-off-by: Ingo Molnar --- kernel/irq/handle.c | 4 ---- 1 file changed, 4 deletions(-) (limited to 'kernel/irq') diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 7dbdfe524693..06b05a4d3007 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -56,10 +56,6 @@ void handle_bad_irq(unsigned int irq, struct irq_desc *desc) int nr_irqs = NR_IRQS; EXPORT_SYMBOL_GPL(nr_irqs); -void __init __attribute__((weak)) arch_early_irq_init(void) -{ -} - #ifdef CONFIG_SPARSE_IRQ static struct irq_desc irq_desc_init = { .irq = -1, -- cgit v1.2.3-55-g7522 From 8b07cd44511f3aa78dd912cca6493275a6787dc5 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 26 Dec 2008 19:10:04 +0100 Subject: sparseirq: do not printk when migrating IRQ descriptors Impact: reduce printk noise There were a couple of leftover KERN_DEBUG debugging printks, remove them. Also clarify an error message. Signed-off-by: Ingo Molnar --- kernel/irq/numa_migrate.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'kernel/irq') diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c index 089c3746358a..a565ce3a4fb5 100644 --- a/kernel/irq/numa_migrate.c +++ b/kernel/irq/numa_migrate.c @@ -74,10 +74,8 @@ static struct irq_desc *__real_move_irq_desc(struct irq_desc *old_desc, node = cpu_to_node(cpu); desc = kzalloc_node(sizeof(*desc), GFP_ATOMIC, node); - printk(KERN_DEBUG " move irq_desc for %d to cpu %d node %d\n", - irq, cpu, node); if (!desc) { - printk(KERN_ERR "can not get new irq_desc for moving\n"); + printk(KERN_ERR "irq %d: can not get new irq_desc for migration.\n", irq); /* still use old one */ desc = old_desc; goto out_unlock; @@ -106,8 +104,6 @@ struct irq_desc *move_irq_desc(struct irq_desc *desc, int cpu) return desc; old_cpu = desc->cpu; - printk(KERN_DEBUG - "try to move irq_desc from cpu %d to %d\n", old_cpu, cpu); if (old_cpu != cpu) { node = cpu_to_node(cpu); old_node = cpu_to_node(old_cpu); -- cgit v1.2.3-55-g7522 From 793f7b12a0c95e7bfec1badf9628043fb78fd440 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 26 Dec 2008 19:02:20 +0100 Subject: sparseirq: fix desc->lock init Impact: cleanup init_one_irq_desc() does not initialize the desc->lock properly - you cannot init a lock by memcpying some other lock on it. This happens to work right now (because irq_desc_init is never in use), but it's a dangerous construct nevertheless, so fix it. Signed-off-by: Ingo Molnar --- kernel/irq/handle.c | 2 ++ kernel/irq/numa_migrate.c | 1 + 2 files changed, 3 insertions(+) (limited to 'kernel/irq') diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 06b05a4d3007..893da67b7781 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -93,6 +93,8 @@ void __attribute__((weak)) arch_init_chip_data(struct irq_desc *desc, int cpu) static void init_one_irq_desc(int irq, struct irq_desc *desc, int cpu) { memcpy(desc, &irq_desc_init, sizeof(struct irq_desc)); + + spin_lock_init(&desc->lock); desc->irq = irq; #ifdef CONFIG_SMP desc->cpu = cpu; diff --git a/kernel/irq/numa_migrate.c b/kernel/irq/numa_migrate.c index a565ce3a4fb5..ecf765c6a77a 100644 --- a/kernel/irq/numa_migrate.c +++ b/kernel/irq/numa_migrate.c @@ -42,6 +42,7 @@ static void init_copy_one_irq_desc(int irq, struct irq_desc *old_desc, struct irq_desc *desc, int cpu) { memcpy(desc, old_desc, sizeof(struct irq_desc)); + spin_lock_init(&desc->lock); desc->cpu = cpu; lockdep_set_class(&desc->lock, &irq_desc_lock_class); init_copy_kstat_irqs(old_desc, desc, cpu, nr_cpu_ids); -- cgit v1.2.3-55-g7522 From 13a0c3c269b223f60abfac8a9811d77111a8b4ba Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 26 Dec 2008 02:05:47 -0800 Subject: sparseirq: work around compiler optimizing away __weak functions Impact: fix panic on null pointer with sparseirq Some GCC versions seem to inline the weak global function, when that function is empty. Work it around, by making the functions return a (dummy) integer. Signed-off-by: Yinghai Signed-off-by: Ingo Molnar --- arch/x86/kernel/io_apic.c | 8 ++++++-- include/linux/irq.h | 6 +++--- init/main.c | 7 ++++--- kernel/irq/handle.c | 7 ++++--- 4 files changed, 17 insertions(+), 11 deletions(-) (limited to 'kernel/irq') diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c index 2fe543f58ac8..976039377846 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/io_apic.c @@ -170,7 +170,7 @@ static struct irq_cfg irq_cfgx[NR_IRQS] = { [15] = { .domain = CPU_MASK_ALL, .vector = IRQ15_VECTOR, }, }; -void __init arch_early_irq_init(void) +int __init arch_early_irq_init(void) { struct irq_cfg *cfg; struct irq_desc *desc; @@ -184,6 +184,8 @@ void __init arch_early_irq_init(void) desc = irq_to_desc(i); desc->chip_data = &cfg[i]; } + + return 0; } #ifdef CONFIG_SPARSE_IRQ @@ -212,7 +214,7 @@ static struct irq_cfg *get_one_free_irq_cfg(int cpu) return cfg; } -void arch_init_chip_data(struct irq_desc *desc, int cpu) +int arch_init_chip_data(struct irq_desc *desc, int cpu) { struct irq_cfg *cfg; @@ -224,6 +226,8 @@ void arch_init_chip_data(struct irq_desc *desc, int cpu) BUG_ON(1); } } + + return 0; } #ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC diff --git a/include/linux/irq.h b/include/linux/irq.h index 69da275c0ebd..0e40af4bac40 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -193,9 +193,9 @@ struct irq_desc { const char *name; } ____cacheline_internodealigned_in_smp; -extern void early_irq_init(void); -extern void arch_early_irq_init(void); -extern void arch_init_chip_data(struct irq_desc *desc, int cpu); +extern int early_irq_init(void); +extern int arch_early_irq_init(void); +extern int arch_init_chip_data(struct irq_desc *desc, int cpu); extern void arch_init_copy_chip_data(struct irq_desc *old_desc, struct irq_desc *desc, int cpu); extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc); diff --git a/init/main.c b/init/main.c index c1f999a3cf31..c314aa15370e 100644 --- a/init/main.c +++ b/init/main.c @@ -539,13 +539,14 @@ void __init __weak thread_info_cache_init(void) { } -void __init __weak arch_early_irq_init(void) +int __init __weak arch_early_irq_init(void) { + return 0; } -void __init __weak early_irq_init(void) +int __init __weak early_irq_init(void) { - arch_early_irq_init(); + return arch_early_irq_init(); } asmlinkage void __init start_kernel(void) diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 893da67b7781..0bef3ecb7a0e 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -86,8 +86,9 @@ void init_kstat_irqs(struct irq_desc *desc, int cpu, int nr) desc->kstat_irqs = (unsigned int *)ptr; } -void __attribute__((weak)) arch_init_chip_data(struct irq_desc *desc, int cpu) +int __weak arch_init_chip_data(struct irq_desc *desc, int cpu) { + return 0; } static void init_one_irq_desc(int irq, struct irq_desc *desc, int cpu) @@ -132,7 +133,7 @@ static struct irq_desc irq_desc_legacy[NR_IRQS_LEGACY] __cacheline_aligned_in_sm /* FIXME: use bootmem alloc ...*/ static unsigned int kstat_irqs_legacy[NR_IRQS_LEGACY][NR_CPUS]; -void __init early_irq_init(void) +int __init early_irq_init(void) { struct irq_desc *desc; int legacy_count; @@ -151,7 +152,7 @@ void __init early_irq_init(void) for (i = legacy_count; i < NR_IRQS; i++) irq_desc_ptrs[i] = NULL; - arch_early_irq_init(); + return arch_early_irq_init(); } struct irq_desc *irq_to_desc(unsigned int irq) -- cgit v1.2.3-55-g7522 From fa6beb37b0d9bc00f90f11154eeed9502d8b0a37 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Mon, 22 Dec 2008 20:24:09 -0800 Subject: sparseirq: set lock_class for legacy irq when sparse_irq is selected Impact: add lockdep annotation to legacy IRQ descs Warnings resulting out of this were not seen in practice, but it's prudent to initialize the legacy descriptors to the lock class as well, symmetric to how we do it with other descriptors. Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- kernel/irq/handle.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/irq') diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 0bef3ecb7a0e..e1cf4e391cae 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -145,6 +145,7 @@ int __init early_irq_init(void) for (i = 0; i < legacy_count; i++) { desc[i].irq = i; desc[i].kstat_irqs = kstat_irqs_legacy[i]; + lockdep_set_class(&desc[i].lock, &irq_desc_lock_class); irq_desc_ptrs[i] = desc + i; } -- cgit v1.2.3-55-g7522 From 12026ea16a618b289fcf457661aed24f57323a20 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Fri, 26 Dec 2008 22:38:15 -0800 Subject: sparseirq: fix hang with !SPARSE_IRQ Impact: fix hang Suresh report his two sockets system only works with SPARSE_IRQ enable it turns out we miss the setting desc->irq so provide early_irq_init() even !SPARSE_IRQ to set desc->irq Reported-by: "Siddha, Suresh B" Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- kernel/irq/handle.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'kernel/irq') diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index e1cf4e391cae..157c04c3b158 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -218,6 +218,21 @@ struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = { } }; +int __init early_irq_init(void) +{ + struct irq_desc *desc; + int count; + int i; + + desc = irq_desc; + count = ARRAY_SIZE(irq_desc); + + for (i = 0; i < count; i++) + desc[i].irq = i; + + return arch_early_irq_init(); +} + struct irq_desc *irq_to_desc(unsigned int irq) { return (irq < NR_IRQS) ? irq_desc + irq : NULL; -- cgit v1.2.3-55-g7522 From b2e2fe99628c4f944c3075258e536197b5a4f3f8 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 29 Dec 2008 00:16:45 +0100 Subject: sparseirq: work around __weak alias bug Impact: fix boot crash if the kernel is built with certain GCC versions GCC has a bug with __weak alias functions: if the functions are in the same compilation unit as their call site, GCC can decide to inline them - and thus rob the linker of the opportunity to override the weak alias with the real thing. This can lead to the boot crash reported by Kamalesh Babulal: ACPI: Core revision 20080926 Setting APIC routing to flat BUG: unable to handle kernel NULL pointer dereference at 0000000000000000 IP: [] add_pin_to_irq_cpu+0x14/0x74 PGD 0 Oops: 0000 [#1] SMP [...] So move the arch_init_chip_data() function from handle.c to manage.c. Reported-by: Kamalesh Babulal Signed-off-by: Ingo Molnar --- kernel/irq/handle.c | 5 ----- kernel/irq/manage.c | 9 +++++++++ 2 files changed, 9 insertions(+), 5 deletions(-) (limited to 'kernel/irq') diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 157c04c3b158..c20db0be9173 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -86,11 +86,6 @@ void init_kstat_irqs(struct irq_desc *desc, int cpu, int nr) desc->kstat_irqs = (unsigned int *)ptr; } -int __weak arch_init_chip_data(struct irq_desc *desc, int cpu) -{ - return 0; -} - static void init_one_irq_desc(int irq, struct irq_desc *desc, int cpu) { memcpy(desc, &irq_desc_init, sizeof(struct irq_desc)); diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 46953a06f4a8..c2741b02ad38 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -261,6 +261,15 @@ void enable_irq(unsigned int irq) } EXPORT_SYMBOL(enable_irq); +/* + * [ Not in kernel/irq/handle.c, so that GCC does not + * inline the __weak alias: ] + */ +int __weak arch_init_chip_data(struct irq_desc *desc, int cpu) +{ + return 0; +} + static int set_irq_wake_real(unsigned int irq, unsigned int on) { struct irq_desc *desc = irq_to_desc(irq); -- cgit v1.2.3-55-g7522 From 43a256322ac1fc105c181b3cade3b9bfc0b63ca1 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 28 Dec 2008 16:01:13 -0800 Subject: sparseirq: move __weak symbols into separate compilation unit GCC has a bug with __weak alias functions: if the functions are in the same compilation unit as their call site, GCC can decide to inline them - and thus rob the linker of the opportunity to override the weak alias with the real thing. So move all the IRQ handling related __weak symbols to kernel/irq/chip.c. Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- include/linux/interrupt.h | 6 ++++++ include/linux/irq.h | 3 --- init/main.c | 10 ---------- kernel/irq/manage.c | 9 --------- kernel/softirq.c | 20 ++++++++++++++++++++ 5 files changed, 26 insertions(+), 22 deletions(-) (limited to 'kernel/irq') diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 777f89e00b4a..d9a370325ae2 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -467,4 +467,10 @@ static inline void init_irq_proc(void) int show_interrupts(struct seq_file *p, void *v); +struct irq_desc; + +extern int early_irq_init(void); +extern int arch_early_irq_init(void); +extern int arch_init_chip_data(struct irq_desc *desc, int cpu); + #endif diff --git a/include/linux/irq.h b/include/linux/irq.h index 0e40af4bac40..d64a6d49bdef 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -193,9 +193,6 @@ struct irq_desc { const char *name; } ____cacheline_internodealigned_in_smp; -extern int early_irq_init(void); -extern int arch_early_irq_init(void); -extern int arch_init_chip_data(struct irq_desc *desc, int cpu); extern void arch_init_copy_chip_data(struct irq_desc *old_desc, struct irq_desc *desc, int cpu); extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc); diff --git a/init/main.c b/init/main.c index c314aa15370e..2c183abbf61c 100644 --- a/init/main.c +++ b/init/main.c @@ -539,16 +539,6 @@ void __init __weak thread_info_cache_init(void) { } -int __init __weak arch_early_irq_init(void) -{ - return 0; -} - -int __init __weak early_irq_init(void) -{ - return arch_early_irq_init(); -} - asmlinkage void __init start_kernel(void) { char * command_line; diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index c2741b02ad38..46953a06f4a8 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -261,15 +261,6 @@ void enable_irq(unsigned int irq) } EXPORT_SYMBOL(enable_irq); -/* - * [ Not in kernel/irq/handle.c, so that GCC does not - * inline the __weak alias: ] - */ -int __weak arch_init_chip_data(struct irq_desc *desc, int cpu) -{ - return 0; -} - static int set_irq_wake_real(unsigned int irq, unsigned int on) { struct irq_desc *desc = irq_to_desc(irq); diff --git a/kernel/softirq.c b/kernel/softirq.c index e7c69a720d69..daf46358d2dd 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -797,3 +797,23 @@ int on_each_cpu(void (*func) (void *info), void *info, int wait) } EXPORT_SYMBOL(on_each_cpu); #endif + +/* + * [ These __weak aliases are kept in a separate compilation unit, so that + * GCC does not inline them incorrectly. ] + */ + +int __init __weak early_irq_init(void) +{ + return 0; +} + +int __init __weak arch_early_irq_init(void) +{ + return 0; +} + +int __weak arch_init_chip_data(struct irq_desc *desc, int cpu) +{ + return 0; +} -- cgit v1.2.3-55-g7522 From d036e67b40f52bdd95392390108defbac7e53837 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 1 Jan 2009 10:12:26 +1030 Subject: cpumask: convert kernel/irq Impact: Reduce stack usage, use new cpumask API. ALPHA mod! Main change is that irq_default_affinity becomes a cpumask_var_t, so treat it as a pointer (this effects alpha). Signed-off-by: Rusty Russell --- arch/alpha/kernel/irq.c | 3 ++- include/linux/interrupt.h | 2 +- kernel/irq/manage.c | 11 +++++++++-- kernel/irq/proc.c | 32 +++++++++++++++++++++----------- 4 files changed, 33 insertions(+), 15 deletions(-) (limited to 'kernel/irq') diff --git a/arch/alpha/kernel/irq.c b/arch/alpha/kernel/irq.c index d0f1620007f7..703731accda6 100644 --- a/arch/alpha/kernel/irq.c +++ b/arch/alpha/kernel/irq.c @@ -50,7 +50,8 @@ int irq_select_affinity(unsigned int irq) if (!irq_desc[irq].chip->set_affinity || irq_user_affinity[irq]) return 1; - while (!cpu_possible(cpu) || !cpu_isset(cpu, irq_default_affinity)) + while (!cpu_possible(cpu) || + !cpumask_test_cpu(cpu, irq_default_affinity)) cpu = (cpu < (NR_CPUS-1) ? cpu + 1 : 0); last_cpu = cpu; diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index dfaee6bd265b..91f1ef8e5810 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -109,7 +109,7 @@ extern void enable_irq(unsigned int irq); #if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS) -extern cpumask_t irq_default_affinity; +extern cpumask_var_t irq_default_affinity; extern int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask); extern int irq_can_set_affinity(unsigned int irq); diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 61c4a9b62165..cd0cd8dcb345 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -16,8 +16,15 @@ #include "internals.h" #ifdef CONFIG_SMP +cpumask_var_t irq_default_affinity; -cpumask_t irq_default_affinity = CPU_MASK_ALL; +static int init_irq_default_affinity(void) +{ + alloc_cpumask_var(&irq_default_affinity, GFP_KERNEL); + cpumask_setall(irq_default_affinity); + return 0; +} +core_initcall(init_irq_default_affinity); /** * synchronize_irq - wait for pending IRQ handlers (on other CPUs) @@ -127,7 +134,7 @@ int do_irq_select_affinity(unsigned int irq, struct irq_desc *desc) desc->status &= ~IRQ_AFFINITY_SET; } - cpumask_and(&desc->affinity, cpu_online_mask, &irq_default_affinity); + cpumask_and(&desc->affinity, cpu_online_mask, irq_default_affinity); set_affinity: desc->chip->set_affinity(irq, &desc->affinity); diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index d2c0e5ee53c5..2abd3a7716ed 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -20,7 +20,7 @@ static struct proc_dir_entry *root_irq_dir; static int irq_affinity_proc_show(struct seq_file *m, void *v) { struct irq_desc *desc = irq_to_desc((long)m->private); - cpumask_t *mask = &desc->affinity; + const struct cpumask *mask = &desc->affinity; #ifdef CONFIG_GENERIC_PENDING_IRQ if (desc->status & IRQ_MOVE_PENDING) @@ -93,7 +93,7 @@ static const struct file_operations irq_affinity_proc_fops = { static int default_affinity_show(struct seq_file *m, void *v) { - seq_cpumask(m, &irq_default_affinity); + seq_cpumask(m, irq_default_affinity); seq_putc(m, '\n'); return 0; } @@ -101,27 +101,37 @@ static int default_affinity_show(struct seq_file *m, void *v) static ssize_t default_affinity_write(struct file *file, const char __user *buffer, size_t count, loff_t *ppos) { - cpumask_t new_value; + cpumask_var_t new_value; int err; - err = cpumask_parse_user(buffer, count, &new_value); + if (!alloc_cpumask_var(&new_value, GFP_KERNEL)) + return -ENOMEM; + + err = cpumask_parse_user(buffer, count, new_value); if (err) - return err; + goto out; - if (!is_affinity_mask_valid(new_value)) - return -EINVAL; + if (!is_affinity_mask_valid(new_value)) { + err = -EINVAL; + goto out; + } /* * Do not allow disabling IRQs completely - it's a too easy * way to make the system unusable accidentally :-) At least * one online CPU still has to be targeted. */ - if (!cpus_intersects(new_value, cpu_online_map)) - return -EINVAL; + if (!cpumask_intersects(new_value, cpu_online_mask)) { + err = -EINVAL; + goto out; + } - irq_default_affinity = new_value; + cpumask_copy(irq_default_affinity, new_value); + err = count; - return count; +out: + free_cpumask_var(new_value); + return err; } static int default_affinity_open(struct inode *inode, struct file *file) -- cgit v1.2.3-55-g7522 From 6bdf197b04b3ae7c85785bc5a9576f1bcb0ac7c0 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sat, 3 Jan 2009 12:50:46 +0100 Subject: ia64: cpumask fix for is_affinity_mask_valid() Impact: build fix on ia64 ia64's default_affinity_write() still had old cpumask_t usage: /home/mingo/tip/kernel/irq/proc.c: In function `default_affinity_write': /home/mingo/tip/kernel/irq/proc.c:114: error: incompatible type for argument 1 of `is_affinity_mask_valid' make[3]: *** [kernel/irq/proc.o] Error 1 make[3]: *** Waiting for unfinished jobs.... update it to cpumask_var_t. Signed-off-by: Ingo Molnar --- arch/ia64/include/asm/irq.h | 2 +- arch/ia64/kernel/irq.c | 4 ++-- kernel/irq/proc.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel/irq') diff --git a/arch/ia64/include/asm/irq.h b/arch/ia64/include/asm/irq.h index 3627116fb0e2..36429a532630 100644 --- a/arch/ia64/include/asm/irq.h +++ b/arch/ia64/include/asm/irq.h @@ -27,7 +27,7 @@ irq_canonicalize (int irq) } extern void set_irq_affinity_info (unsigned int irq, int dest, int redir); -bool is_affinity_mask_valid(cpumask_t cpumask); +bool is_affinity_mask_valid(cpumask_var_t cpumask); #define is_affinity_mask_valid is_affinity_mask_valid diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c index 0b6db53fedcf..95ff16cb05d8 100644 --- a/arch/ia64/kernel/irq.c +++ b/arch/ia64/kernel/irq.c @@ -112,11 +112,11 @@ void set_irq_affinity_info (unsigned int irq, int hwid, int redir) } } -bool is_affinity_mask_valid(cpumask_t cpumask) +bool is_affinity_mask_valid(cpumask_var_t cpumask) { if (ia64_platform_is("sn2")) { /* Only allow one CPU to be specified in the smp_affinity mask */ - if (cpus_weight(cpumask) != 1) + if (cpumask_weight(cpumask) != 1) return false; } return true; diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index 2abd3a7716ed..aae3f742bcec 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -54,7 +54,7 @@ static ssize_t irq_affinity_proc_write(struct file *file, if (err) goto free_cpumask; - if (!is_affinity_mask_valid(*new_value)) { + if (!is_affinity_mask_valid(new_value)) { err = -EINVAL; goto free_cpumask; } -- cgit v1.2.3-55-g7522 From 22a9d645677feefd402befd02edd59b122289ef1 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Wed, 7 Jan 2009 08:45:46 -0800 Subject: async: Asynchronous function calls to speed up kernel boot Right now, most of the kernel boot is strictly synchronous, such that various hardware delays are done sequentially. In order to make the kernel boot faster, this patch introduces infrastructure to allow doing some of the initialization steps asynchronously, which will hide significant portions of the hardware delays in practice. In order to not change device order and other similar observables, this patch does NOT do full parallel initialization. Rather, it operates more in the way an out of order CPU does; the work may be done out of order and asynchronous, but the observable effects (instruction retiring for the CPU) are still done in the original sequence. Signed-off-by: Arjan van de Ven --- include/linux/async.h | 25 ++++ init/do_mounts.c | 2 + init/main.c | 5 +- kernel/Makefile | 3 +- kernel/async.c | 321 +++++++++++++++++++++++++++++++++++++++++++++++++ kernel/irq/autoprobe.c | 5 + kernel/module.c | 2 + 7 files changed, 361 insertions(+), 2 deletions(-) create mode 100644 include/linux/async.h create mode 100644 kernel/async.c (limited to 'kernel/irq') diff --git a/include/linux/async.h b/include/linux/async.h new file mode 100644 index 000000000000..c4ecacd0b327 --- /dev/null +++ b/include/linux/async.h @@ -0,0 +1,25 @@ +/* + * async.h: Asynchronous function calls for boot performance + * + * (C) Copyright 2009 Intel Corporation + * Author: Arjan van de Ven + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 + * of the License. + */ + +#include +#include + +typedef u64 async_cookie_t; +typedef void (async_func_ptr) (void *data, async_cookie_t cookie); + +extern async_cookie_t async_schedule(async_func_ptr *ptr, void *data); +extern async_cookie_t async_schedule_special(async_func_ptr *ptr, void *data, struct list_head *list); +extern void async_synchronize_full(void); +extern void async_synchronize_full_special(struct list_head *list); +extern void async_synchronize_cookie(async_cookie_t cookie); +extern void async_synchronize_cookie_special(async_cookie_t cookie, struct list_head *list); + diff --git a/init/do_mounts.c b/init/do_mounts.c index 5efca73b39f9..708105e163df 100644 --- a/init/do_mounts.c +++ b/init/do_mounts.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include @@ -372,6 +373,7 @@ void __init prepare_namespace(void) /* wait for the known devices to complete their probing */ while (driver_probe_done() != 0) msleep(100); + async_synchronize_full(); md_run_setup(); diff --git a/init/main.c b/init/main.c index b5a892c68375..f66715d8a853 100644 --- a/init/main.c +++ b/init/main.c @@ -62,6 +62,7 @@ #include #include #include +#include #include #include @@ -684,7 +685,7 @@ asmlinkage void __init start_kernel(void) rest_init(); } -static int initcall_debug; +int initcall_debug; core_param(initcall_debug, initcall_debug, bool, 0644); int do_one_initcall(initcall_t fn) @@ -785,6 +786,8 @@ static void run_init_process(char *init_filename) */ static noinline int init_post(void) { + /* need to finish all async __init code before freeing the memory */ + async_synchronize_full(); free_initmem(); unlock_kernel(); mark_rodata_ro(); diff --git a/kernel/Makefile b/kernel/Makefile index e1c5bf3365c0..2921d90ce32f 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -9,7 +9,8 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.o \ rcupdate.o extable.o params.o posix-timers.o \ kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \ hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \ - notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o + notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o \ + async.o ifdef CONFIG_FUNCTION_TRACER # Do not trace debug files and internal ftrace files diff --git a/kernel/async.c b/kernel/async.c new file mode 100644 index 000000000000..afaa8a653d5a --- /dev/null +++ b/kernel/async.c @@ -0,0 +1,321 @@ +/* + * async.c: Asynchronous function calls for boot performance + * + * (C) Copyright 2009 Intel Corporation + * Author: Arjan van de Ven + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 + * of the License. + */ + + +/* + +Goals and Theory of Operation + +The primary goal of this feature is to reduce the kernel boot time, +by doing various independent hardware delays and discovery operations +decoupled and not strictly serialized. + +More specifically, the asynchronous function call concept allows +certain operations (primarily during system boot) to happen +asynchronously, out of order, while these operations still +have their externally visible parts happen sequentially and in-order. +(not unlike how out-of-order CPUs retire their instructions in order) + +Key to the asynchronous function call implementation is the concept of +a "sequence cookie" (which, although it has an abstracted type, can be +thought of as a monotonically incrementing number). + +The async core will assign each scheduled event such a sequence cookie and +pass this to the called functions. + +The asynchronously called function should before doing a globally visible +operation, such as registering device numbers, call the +async_synchronize_cookie() function and pass in its own cookie. The +async_synchronize_cookie() function will make sure that all asynchronous +operations that were scheduled prior to the operation corresponding with the +cookie have completed. + +Subsystem/driver initialization code that scheduled asynchronous probe +functions, but which shares global resources with other drivers/subsystems +that do not use the asynchronous call feature, need to do a full +synchronization with the async_synchronize_full() function, before returning +from their init function. This is to maintain strict ordering between the +asynchronous and synchronous parts of the kernel. + +*/ + +#include +#include +#include +#include +#include +#include +#include + +static async_cookie_t next_cookie = 1; + +#define MAX_THREADS 256 +#define MAX_WORK 32768 + +static LIST_HEAD(async_pending); +static LIST_HEAD(async_running); +static DEFINE_SPINLOCK(async_lock); + +struct async_entry { + struct list_head list; + async_cookie_t cookie; + async_func_ptr *func; + void *data; + struct list_head *running; +}; + +static DECLARE_WAIT_QUEUE_HEAD(async_done); +static DECLARE_WAIT_QUEUE_HEAD(async_new); + +static atomic_t entry_count; +static atomic_t thread_count; + +extern int initcall_debug; + + +/* + * MUST be called with the lock held! + */ +static async_cookie_t __lowest_in_progress(struct list_head *running) +{ + struct async_entry *entry; + if (!list_empty(&async_pending)) { + entry = list_first_entry(&async_pending, + struct async_entry, list); + return entry->cookie; + } else if (!list_empty(running)) { + entry = list_first_entry(running, + struct async_entry, list); + return entry->cookie; + } else { + /* nothing in progress... next_cookie is "infinity" */ + return next_cookie; + } + +} +/* + * pick the first pending entry and run it + */ +static void run_one_entry(void) +{ + unsigned long flags; + struct async_entry *entry; + ktime_t calltime, delta, rettime; + + /* 1) pick one task from the pending queue */ + + spin_lock_irqsave(&async_lock, flags); + if (list_empty(&async_pending)) + goto out; + entry = list_first_entry(&async_pending, struct async_entry, list); + + /* 2) move it to the running queue */ + list_del(&entry->list); + list_add_tail(&entry->list, &async_running); + spin_unlock_irqrestore(&async_lock, flags); + + /* 3) run it (and print duration)*/ + if (initcall_debug) { + printk("calling %lli_%pF @ %i\n", entry->cookie, entry->func, task_pid_nr(current)); + calltime = ktime_get(); + } + entry->func(entry->data, entry->cookie); + if (initcall_debug) { + rettime = ktime_get(); + delta = ktime_sub(rettime, calltime); + printk("initcall %lli_%pF returned 0 after %lld usecs\n", entry->cookie, + entry->func, ktime_to_ns(delta) >> 10); + } + + /* 4) remove it from the running queue */ + spin_lock_irqsave(&async_lock, flags); + list_del(&entry->list); + + /* 5) free the entry */ + kfree(entry); + atomic_dec(&entry_count); + + spin_unlock_irqrestore(&async_lock, flags); + + /* 6) wake up any waiters. */ + wake_up(&async_done); + return; + +out: + spin_unlock_irqrestore(&async_lock, flags); +} + + +static async_cookie_t __async_schedule(async_func_ptr *ptr, void *data, struct list_head *running) +{ + struct async_entry *entry; + unsigned long flags; + async_cookie_t newcookie; + + + /* allow irq-off callers */ + entry = kzalloc(sizeof(struct async_entry), GFP_ATOMIC); + + /* + * If we're out of memory or if there's too much work + * pending already, we execute synchronously. + */ + if (!entry || atomic_read(&entry_count) > MAX_WORK) { + kfree(entry); + spin_lock_irqsave(&async_lock, flags); + newcookie = next_cookie++; + spin_unlock_irqrestore(&async_lock, flags); + + /* low on memory.. run synchronously */ + ptr(data, newcookie); + return newcookie; + } + entry->func = ptr; + entry->data = data; + entry->running = running; + + spin_lock_irqsave(&async_lock, flags); + newcookie = entry->cookie = next_cookie++; + list_add_tail(&entry->list, &async_pending); + atomic_inc(&entry_count); + spin_unlock_irqrestore(&async_lock, flags); + wake_up(&async_new); + return newcookie; +} + +async_cookie_t async_schedule(async_func_ptr *ptr, void *data) +{ + return __async_schedule(ptr, data, &async_pending); +} +EXPORT_SYMBOL_GPL(async_schedule); + +async_cookie_t async_schedule_special(async_func_ptr *ptr, void *data, struct list_head *running) +{ + return __async_schedule(ptr, data, running); +} +EXPORT_SYMBOL_GPL(async_schedule_special); + +void async_synchronize_full(void) +{ + async_synchronize_cookie(next_cookie); +} +EXPORT_SYMBOL_GPL(async_synchronize_full); + +void async_synchronize_full_special(struct list_head *list) +{ + async_synchronize_cookie_special(next_cookie, list); +} +EXPORT_SYMBOL_GPL(async_synchronize_full_special); + +void async_synchronize_cookie_special(async_cookie_t cookie, struct list_head *running) +{ + ktime_t starttime, delta, endtime; + + if (initcall_debug) { + printk("async_waiting @ %i\n", task_pid_nr(current)); + starttime = ktime_get(); + } + + wait_event(async_done, __lowest_in_progress(running) >= cookie); + + if (initcall_debug) { + endtime = ktime_get(); + delta = ktime_sub(endtime, starttime); + + printk("async_continuing @ %i after %lli usec\n", + task_pid_nr(current), ktime_to_ns(delta) >> 10); + } +} +EXPORT_SYMBOL_GPL(async_synchronize_cookie_special); + +void async_synchronize_cookie(async_cookie_t cookie) +{ + async_synchronize_cookie_special(cookie, &async_running); +} +EXPORT_SYMBOL_GPL(async_synchronize_cookie); + + +static int async_thread(void *unused) +{ + DECLARE_WAITQUEUE(wq, current); + add_wait_queue(&async_new, &wq); + + while (!kthread_should_stop()) { + int ret = HZ; + set_current_state(TASK_INTERRUPTIBLE); + /* + * check the list head without lock.. false positives + * are dealt with inside run_one_entry() while holding + * the lock. + */ + rmb(); + if (!list_empty(&async_pending)) + run_one_entry(); + else + ret = schedule_timeout(HZ); + + if (ret == 0) { + /* + * we timed out, this means we as thread are redundant. + * we sign off and die, but we to avoid any races there + * is a last-straw check to see if work snuck in. + */ + atomic_dec(&thread_count); + wmb(); /* manager must see our departure first */ + if (list_empty(&async_pending)) + break; + /* + * woops work came in between us timing out and us + * signing off; we need to stay alive and keep working. + */ + atomic_inc(&thread_count); + } + } + remove_wait_queue(&async_new, &wq); + + return 0; +} + +static int async_manager_thread(void *unused) +{ + DECLARE_WAITQUEUE(wq, current); + add_wait_queue(&async_new, &wq); + + while (!kthread_should_stop()) { + int tc, ec; + + set_current_state(TASK_INTERRUPTIBLE); + + tc = atomic_read(&thread_count); + rmb(); + ec = atomic_read(&entry_count); + + while (tc < ec && tc < MAX_THREADS) { + kthread_run(async_thread, NULL, "async/%i", tc); + atomic_inc(&thread_count); + tc++; + } + + schedule(); + } + remove_wait_queue(&async_new, &wq); + + return 0; +} + +static int __init async_init(void) +{ + kthread_run(async_manager_thread, NULL, "async/mgr"); + return 0; +} + +core_initcall(async_init); diff --git a/kernel/irq/autoprobe.c b/kernel/irq/autoprobe.c index cc0f7321b8ce..1de9700f416e 100644 --- a/kernel/irq/autoprobe.c +++ b/kernel/irq/autoprobe.c @@ -10,6 +10,7 @@ #include #include #include +#include #include "internals.h" @@ -34,6 +35,10 @@ unsigned long probe_irq_on(void) unsigned int status; int i; + /* + * quiesce the kernel, or at least the asynchronous portion + */ + async_synchronize_full(); mutex_lock(&probing_active); /* * something may have generated an irq long ago and we want to diff --git a/kernel/module.c b/kernel/module.c index 496dcb57b608..c9332c90d5a0 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -50,6 +50,7 @@ #include #include #include +#include #if 0 #define DEBUGP printk @@ -816,6 +817,7 @@ sys_delete_module(const char __user *name_user, unsigned int flags) mod->exit(); blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_GOING, mod); + async_synchronize_full(); mutex_lock(&module_mutex); /* Store the name of the last unloaded module for diagnostic purposes */ strlcpy(last_unloaded_module, mod->name, sizeof(last_unloaded_module)); -- cgit v1.2.3-55-g7522