irq: correct CPUMASKS_OFFSTACK typo -v2
Subsystems (7)
PATCH 1/2 irq: correct CPUMASKS_OFFSTACK typo -v2
Impact: fix smp_affinity copying when moving irq_desc
CPUMASKS_OFFSTACK is not defined anywhere. it is a typo
and init_allocate_desc_masks called before it set affinity to all cpus...
split init_alloc_desc_masks() into all_desc_masks() and init_desc_masks() so in the init_copy_desc_masks could use CPUMASK_OFFSTACK there. aka copy path will not calling init_desc_masks anymore.
also could use that CPUMASK_OFFSTACK in alloc_desc_masks()
--- include/linux/irq.h | 27 ++++++++++++++++++--------- kernel/irq/handle.c | 9 ++++++--- kernel/irq/numa_migrate.c | 2 +- 3 files changed, 25 insertions(+), 13 deletions(-)
include/linux/irq.h +18 -9
@@ -424,27 +424,25 @@ extern int set_irq_msi(unsigned int irq,#ifdef CONFIG_SMP /** -* init_alloc_desc_masks - allocate cpumasks for irq_desc +* alloc_desc_masks - allocate cpumasks for irq_desc * @desc: pointer to irq_desc struct * @cpu: cpu which will be handling the cpumasks * @boot: true if need bootmem * * Allocates affinity and pending_mask cpumask if required. * Returns true if successful (or not required). -* Side effect: affinity has all bits set, pending_mask has all bits clear. */ -static inline bool init_alloc_desc_masks (struct irq_desc *desc,int cpu, +static inline bool alloc_desc_masks (struct irq_desc *desc,int cpu,bool boot) { +#ifdef CONFIG_CPUMASK_OFFSTACK int node;if (boot) {alloc_bootmem_cpumask_var (&desc->affinity); -cpumask_setall (desc->affinity);#ifdef CONFIG_GENERIC_PENDING_IRQ alloc_bootmem_cpumask_var (&desc->pending_mask); -cpumask_clear (desc->pending_mask);#endif return true; }
@@ -453,18 +451,25 @@ static inline bool init_alloc_desc_masksif (!alloc_cpumask_var_node (&desc->affinity,GFP_ATOMIC , node))return false; -cpumask_setall (desc->affinity);#ifdef CONFIG_GENERIC_PENDING_IRQ if (!alloc_cpumask_var_node (&desc->pending_mask,GFP_ATOMIC , node)) {free_cpumask_var (desc->affinity);return false; } -cpumask_clear (desc->pending_mask); +#endif #endif return true; } +static inline void init_desc_masks (struct irq_desc *desc) +{ +cpumask_setall (desc->affinity); +#ifdef CONFIG_GENERIC_PENDING_IRQ +cpumask_clear (desc->pending_mask); +#endif +} +/** * init_copy_desc_masks - copy cpumasks for irq_desc * @old_desc: pointer to old irq_desc struct
@@ -478,7 +483,7 @@ static inline bool init_alloc_desc_masksstatic inline void init_copy_desc_masks (struct irq_desc *old_desc,struct irq_desc *new_desc) { -#ifdef CONFIG_CPUMASKS_OFFSTACK +#ifdef CONFIG_CPUMASK_OFFSTACK cpumask_copy (new_desc->affinity, old_desc->affinity);#ifdef CONFIG_GENERIC_PENDING_IRQ
@@ -499,12 +504,16 @@ static inline void free_desc_masks(struc#else /* !CONFIG_SMP */ -static inline bool init_alloc_desc_masks (struct irq_desc *desc,int cpu, +static inline bool alloc_desc_masks (struct irq_desc *desc,int cpu,bool boot) {return true; } +static inline void init_desc_masks (struct irq_desc *desc) +{ +} +static inline void init_copy_desc_masks (struct irq_desc *old_desc,struct irq_desc *new_desc) {
kernel/irq/handle.c +6 -3
@@ -115,10 +115,11 @@ static void init_one_irq_desc(int irq, sprintk (KERN_ERR "can not alloc kstat_irqs\n" );BUG_ON (1 ); } -if (!init_alloc_desc_masks (desc, cpu, false)) { +if (!alloc_desc_masks (desc, cpu, false)) {printk (KERN_ERR "can not alloc irq_desc cpumasks\n" );BUG_ON (1 ); } +init_desc_masks (desc);arch_init_chip_data (desc, cpu); }
@@ -169,7 +170,8 @@ int __init early_irq_init(void) desc[i].irq = i; desc[i].kstat_irqs = kstat_irqs_legacy + i * nr_cpu_ids;lockdep_set_class (&desc[i].lock, &irq_desc_lock_class); -init_alloc_desc_masks (&desc[i],0 , true); +alloc_desc_masks (&desc[i],0 , true); +init_desc_masks (&desc[i]); irq_desc_ptrs[i] = desc + i; }
@@ -256,7 +258,8 @@ int __init early_irq_init(void)for (i =0 ; i < count; i++) { desc[i].irq = i; -init_alloc_desc_masks (&desc[i],0 , true); +alloc_desc_masks (&desc[i],0 , true); +init_desc_masks (&desc[i]); desc[i].kstat_irqs = kstat_irqs_all[i]; }return arch_early_irq_init ();
kernel/irq/numa_migrate.c +1 -1
@@ -37,7 +37,7 @@ static bool init_copy_one_irq_desc(int i struct irq_desc *desc, int cpu) { memcpy(desc, old_desc, sizeof(struct irq_desc)); -if (!init_alloc_desc_masks (desc, cpu, false)) { +if (!alloc_desc_masks (desc, cpu, false)) {printk (KERN_ERR "irq %d: can not get new irq_desc cpumask " "for migration.\n" , irq);return false;
PATCH 2/2 irq: only update affinity in chip set_affinity()
Impact: keep affinity consistent
irq_set_affinity() and move_masked_irq() try to assign affinity before calling chip set_affinity(). some archs are assigning again in set_affinity again.
something like:cpumask_cpy(desc->affinity, mask);desc->chip->set_affinity(mask);
in the failing path, affinity should not be touched.
also set_extra_move_desc() ( called by set_affinity) will rely on the old affinity to decide if need to move irq_desc to different node when logical flat apic mode is used.
So try remove those assignment, and make some missed arch to assign affinity in their set_affinity.
--
arch/alpha/kernel/sys_dp264.c | 6 ++++-- arch/alpha/kernel/sys_titan.c | 3 ++- arch/arm/common/gic.c | 1 + arch/cris/arch-v32/kernel/irq.c | 1 + arch/ia64/kernel/iosapic.c | 3 +++ arch/ia64/sn/kernel/irq.c | 3 +++ arch/mips/cavium-octeon/octeon-irq.c | 6 ++++++ arch/mips/sibyte/bcm1480/irq.c | 2 ++ arch/mips/sibyte/sb1250/irq.c | 2 ++ arch/powerpc/platforms/pseries/xics.c | 5 +++++ arch/powerpc/sysdev/mpic.c | 2 ++ arch/sparc/kernel/irq_64.c | 7 +++++++ drivers/xen/events.c | 2 ++ kernel/irq/manage.c | 6 ++---- kernel/irq/migration.c | 8 +++----- 15 files changed, 45 insertions(+), 12 deletions(-)
arch/alpha/kernel/sys_dp264.c +4 -2
@@ -178,7 +178,8 @@ cpu_set_irq_affinity(unsigned int irq, cstatic void dp264_set_affinity (unsigned int irq,const struct cpumask *affinity) -{ +{ +cpumask_copy (irq_desc[irq].affinity, affinity);spin_lock (&dp264_irq_lock);cpu_set_irq_affinity (irq, *affinity);tsunami_update_irq_hw (cached_irq_mask);
@@ -187,7 +188,8 @@ dp264_set_affinity(unsigned int irq, constatic void clipper_set_affinity (unsigned int irq,const struct cpumask *affinity) -{ +{ +cpumask_copy (irq_desc[irq].affinity, affinity);spin_lock (&dp264_irq_lock);cpu_set_irq_affinity (irq -16 , *affinity);tsunami_update_irq_hw (cached_irq_mask);
arch/alpha/kernel/sys_titan.c +2 -1
@@ -159,7 +159,8 @@ titan_cpu_set_irq_affinity(unsigned intstatic void titan_set_irq_affinity (unsigned int irq,const struct cpumask *affinity) -{ +{ +cpumask_copy (irq_desc[irq].affinity, affinity);spin_lock (&titan_irq_lock);titan_cpu_set_irq_affinity (irq -16 , *affinity);titan_update_irq_hw (titan_cached_irq_mask);
arch/arm/common/gic.c +1 -0
@@ -118,6 +118,7 @@ static void gic_set_cpu(unsigned int irqspin_lock (&irq_controller_lock); irq_desc[irq].cpu = cpu; +cpumask_copy (irq_desc[irq].affinity, mask_val); val =readl (reg) & ~(0xff << shift); val |=1 << (cpu + shift);writel (val, reg);
arch/cris/arch-v32/kernel/irq.c +1 -0
@@ -330,6 +330,7 @@ void set_affinity_crisv32_irq(unsigned iunsigned long flags;spin_lock_irqsave (&irq_lock, flags); irq_allocations[irq -FIRST_IRQ ].mask = *dest; +cpumask_copy (irq_desc[j].affinity, dest);spin_unlock_irqrestore (&irq_lock, flags); }
arch/ia64/kernel/iosapic.c +3 -0
@@ -338,6 +338,7 @@ iosapic_set_affinity(unsigned int irq, cint redir = (irq &IA64_IRQ_REDIRECTED ) ?1 :0 ;struct iosapic_rte_info *rte;struct iosapic *iosapic; +irq_desc_t *idesc = irq_desc + irq; irq &= (~IA64_IRQ_REDIRECTED );
@@ -353,6 +354,8 @@ iosapic_set_affinity(unsigned int irq, cif (!iosapic_intr_info[irq].count)return ;/* not an IOSAPIC interrupt */ +cpumask_copy (desc->affinity, mask); +set_irq_affinity_info (irq, dest, redir);/* dest contains both id and eid */
arch/ia64/sn/kernel/irq.c +3 -0
@@ -232,6 +232,9 @@ static void sn_set_affinity_irq(unsignedstruct sn_irq_info *sn_irq_info, *sn_irq_info_safe;nasid_t nasid;int slice; +struct irq_desc *desc =irq_to_desc (irq); + +cpumask_copy (desc->affinity, mask); nasid =cpuid_to_nasid (cpumask_first (mask)); slice =cpuid_to_slice (cpumask_first (mask));
arch/mips/cavium-octeon/octeon-irq.c +6 -0
@@ -181,6 +181,9 @@ static void octeon_irq_ciu0_set_affinity {int cpu;int bit = irq -OCTEON_IRQ_WORKQ0 ;/* Bit 0-63 of EN0 */ +irq_desc_t *desc = irq_desc + irq; + +cpumask_copy (desc->affinity, dest);write_lock (&octeon_irq_ciu0_rwlock);for_each_online_cpu (cpu) {
@@ -296,6 +299,9 @@ static void octeon_irq_ciu1_set_affinity {int cpu;int bit = irq -OCTEON_IRQ_WDOG0 ;/* Bit 0-63 of EN1 */ +irq_desc_t *desc = irq_desc + irq; + +cpumask_copy (desc->affinity, dest);write_lock (&octeon_irq_ciu1_rwlock);for_each_online_cpu (cpu) {
arch/mips/sibyte/bcm1480/irq.c +2 -0
@@ -123,6 +123,8 @@ static void bcm1480_set_affinity(unsigne } i =cpumask_first (mask); +cpumask_copy (desc->affinity, mask); +/* Convert logical CPU to physical CPU */ cpu =cpu_logical_map (i);
arch/mips/sibyte/sb1250/irq.c +2 -0
@@ -117,6 +117,8 @@ static void sb1250_set_affinity(unsignedreturn ; } +cpumask_copy (desc->affinity, mask); +/* Convert logical CPU to physical CPU */ cpu =cpu_logical_map (i);
arch/powerpc/platforms/pseries/xics.c +5 -0
@@ -339,6 +339,7 @@ static void xics_set_affinity(unsigned iint status;int xics_status[2 ];int irq_server; +struct irq_desc *desc; irq = (unsigned int )irq_map[virq].hwirq;if (irq ==XICS_IPI || irq ==XICS_IRQ_SPURIOUS )
@@ -374,6 +375,10 @@ static void xics_set_affinity(unsigned i __func__, irq, status);return ; } + + desc =get_irq_desc (virq); + +cpumask_copy (desc->affinity, cpumask); }static struct irq_chip xics_pic_direct = {
arch/powerpc/sysdev/mpic.c +2 -0
@@ -811,7 +811,9 @@ void mpic_set_affinity(unsigned int irq, {struct mpic *mpic =mpic_from_irq (irq);unsigned int src =mpic_irq_to_hw (irq); +struct irq_desc *desc =get_irq_desc (irq); +cpumask_copy (desc->affinity, cpumask);if (mpic->flags &MPIC_SINGLE_DEST_CPU ) {int cpuid =irq_choose_cpu (irq);
arch/sparc/kernel/irq_64.c +7 -0
@@ -321,6 +321,9 @@ static void sun4u_irq_enable(unsigned instatic void sun4u_set_affinity (unsigned int virt_irq,const struct cpumask *mask) { +struct irq_desc *desc = irq_desc + virt_irq; + +cpumask_copy (desc->affinity, mask);sun4u_irq_enable (virt_irq); }
@@ -382,8 +385,10 @@ static void sun4v_set_affinity(unsigned {unsigned int ino = virt_irq_table[virt_irq].dev_ino;unsigned long cpuid =irq_choose_cpu (virt_irq); +struct irq_desc *desc = irq_desc + virt_irq;int err; +cpumask_copy (desc->affinity, mask); err =sun4v_intr_settarget (ino, cpuid);if (err !=HV_EOK ) printk(KERN_ERR "sun4v_intr_settarget(%x,%lu): "
@@ -449,8 +454,10 @@ static void sun4v_virt_set_affinity(unsi const struct cpumask *mask) {unsigned long cpuid, dev_handle, dev_ino; +struct irq_desc *desc = irq_desc + virt_irq;int err; +cpumask_copy (desc->affinity, mask); cpuid =irq_choose_cpu (virt_irq); dev_handle = virt_irq_table[virt_irq].dev_handle;
drivers/xen/events.c +2 -0
@@ -713,6 +713,8 @@ static void rebind_irq_to_cpu(unsigned istatic void set_affinity_irq (unsigned irq,const struct cpumask *dest) {unsigned tcpu =cpumask_first (dest); + +cpumask_copy (irq_to_desc (irq)->affinity, mask);rebind_irq_to_cpu (irq, tcpu); }
kernel/irq/manage.c +2 -4
@@ -109,15 +109,13 @@ int irq_set_affinity(unsigned int irq, cspin_lock_irqsave (&desc->lock, flags);#ifdef CONFIG_GENERIC_PENDING_IRQ -if (desc->status &IRQ_MOVE_PCNTXT || desc->status &IRQ_DISABLED ) { -cpumask_copy (desc->affinity, cpumask); +if (desc->status &IRQ_MOVE_PCNTXT || desc->status &IRQ_DISABLED ) desc->chip->set_affinity (irq, cpumask); - }else { +else { desc->status |=IRQ_MOVE_PENDING ;cpumask_copy (desc->pending_mask, cpumask); }#else -cpumask_copy (desc->affinity, cpumask); desc->chip->set_affinity (irq, cpumask);#endif irq_set_thread_affinity (desc, cpumask);
kernel/irq/migration.c +3 -5
@@ -39,11 +39,9 @@ void move_masked_irq(int irq)* masking the irqs. */ 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); - } + < nr_cpu_ids)) + desc->chip->set_affinity (irq, desc->pending_mask); +cpumask_clear (desc->pending_mask); }
Re: [PATCH 2/2] irq: only update affinity in chip set_affinity()
Quoted 22 lines · Yinghai Lu
Impact: keep affinity consistent
irq_set_affinity()andmove_masked_irq()try to assign affinity before calling chipset_affinity(). some archs are assigning again inset_affinityagain.something like:
cpumask_cpy(desc->affinity, mask);desc->chip->set_affinity(mask);in the failing path, affinity should not be touched.
also
set_extra_move_desc()( called byset_affinity) will rely on the old affinity to decide if need to moveirq_descto different node when logical flat apic mode is used.So try remove those assignment, and make some missed arch to assign affinity in their
set_affinity.Signed-off-by Yinghai Lu <[email protected]>--
Quoted 16 lines
arch/alpha/kernel/sys_dp264.c | 6 ++++-- arch/alpha/kernel/sys_titan.c | 3 ++- arch/arm/common/gic.c | 1 + arch/cris/arch-v32/kernel/irq.c | 1 + arch/ia64/kernel/iosapic.c | 3 +++ arch/ia64/sn/kernel/irq.c | 3 +++ arch/mips/cavium-octeon/octeon-irq.c | 6 ++++++ arch/mips/sibyte/bcm1480/irq.c | 2 ++ arch/mips/sibyte/sb1250/irq.c | 2 ++ arch/powerpc/platforms/pseries/xics.c | 5 +++++ arch/powerpc/sysdev/mpic.c | 2 ++ arch/sparc/kernel/irq_64.c | 7 +++++++ drivers/xen/events.c | 2 ++ kernel/irq/manage.c | 6 ++---- kernel/irq/migration.c | 8 +++----- 15 files changed, 45 insertions(+), 12 deletions(-)
Hm, this spreads a lot of instances of identical lines:
cpumask_copy(irq_desc[irq].affinity, mask_val);
all around architectures. How is that an improvement?
Ingo
Re: [PATCH 2/2] irq: only update affinity in chip set_affinity()
Quoted 42 lines · Ingo Molnar
Quoted 37 lines · Yinghai Lu
Impact: keep affinity consistent
irq_set_affinity()andmove_masked_irq()try to assign affinity before calling chipset_affinity(). some archs are assigning again inset_affinityagain.something like:
cpumask_cpy(desc->affinity, mask);desc->chip->set_affinity(mask);in the failing path, affinity should not be touched.
also
set_extra_move_desc()( called byset_affinity) will rely on the old affinity to decide if need to moveirq_descto different node when logical flat apic mode is used.So try remove those assignment, and make some missed arch to assign affinity in their
set_affinity.Signed-off-by Yinghai Lu <[email protected]>--
arch/alpha/kernel/sys_dp264.c | 6 ++++--arch/alpha/kernel/sys_titan.c| 3 ++-arch/arm/common/gic.c| 1 +arch/cris/arch-v32/kernel/irq.c| 1 +arch/ia64/kernel/iosapic.c| 3 +++arch/ia64/sn/kernel/irq.c| 3 +++arch/mips/cavium-octeon/octeon-irq.c| 6 ++++++arch/mips/sibyte/bcm1480/irq.c| 2 ++arch/mips/sibyte/sb1250/irq.c| 2 ++arch/powerpc/platforms/pseries/xics.c| 5 +++++arch/powerpc/sysdev/mpic.c| 2 ++arch/sparc/kernel/irq_64.c| 7 +++++++drivers/xen/events.c| 2 ++kernel/irq/manage.c| 6 ++----kernel/irq/migration.c| 8 +++-----Quoted 1 line
15 files changed, 45 insertions(+), 12 deletions(-)
Quoted 6 lines
Hm, this spreads a lot of instances of identical lines:
cpumask_copy(irq_desc[irq].affinity, mask_val);all around architectures. How is that an improvement?
in failing path in set_affinity, for example it can not get vector in specified cpu,
then affinity should not be changed.
YH
Re: [PATCH 2/2] irq: only update affinity in chip set_affinity()
Quoted 45 lines · Yinghai Lu
Quoted 44 lines · Ingo Molnar
Quoted 37 lines · Yinghai Lu
Impact: keep affinity consistent
irq_set_affinity()andmove_masked_irq()try to assign affinity before calling chipset_affinity(). some archs are assigning again inset_affinityagain.something like:
cpumask_cpy(desc->affinity, mask);desc->chip->set_affinity(mask);in the failing path, affinity should not be touched.
also
set_extra_move_desc()( called byset_affinity) will rely on the old affinity to decide if need to moveirq_descto different node when logical flat apic mode is used.So try remove those assignment, and make some missed arch to assign affinity in their
set_affinity.Signed-off-by Yinghai Lu <[email protected]>--
arch/alpha/kernel/sys_dp264.c | 6 ++++--arch/alpha/kernel/sys_titan.c| 3 ++-arch/arm/common/gic.c| 1 +arch/cris/arch-v32/kernel/irq.c| 1 +arch/ia64/kernel/iosapic.c| 3 +++arch/ia64/sn/kernel/irq.c| 3 +++arch/mips/cavium-octeon/octeon-irq.c| 6 ++++++arch/mips/sibyte/bcm1480/irq.c| 2 ++arch/mips/sibyte/sb1250/irq.c| 2 ++arch/powerpc/platforms/pseries/xics.c| 5 +++++arch/powerpc/sysdev/mpic.c| 2 ++arch/sparc/kernel/irq_64.c| 7 +++++++drivers/xen/events.c| 2 ++kernel/irq/manage.c| 6 ++----kernel/irq/migration.c| 8 +++-----Quoted 1 line
15 files changed, 45 insertions(+), 12 deletions(-)Hm, this spreads a lot of instances of identical lines:
Quoted 7 lines
Quoted 4 lines
cpumask_copy(irq_desc[irq].affinity, mask_val);all around architectures. How is that an improvement?
in failing path in
set_affinity, for example it can not get vector in specified cpu, then affinity should not be changed.
isnt the right solution then to propagate the failure code back to the generic code?
Preferably via a new callback, and the patches only touching the core code plus maybe x86, so that other architectures can be converted/fixed more gradually.
Ingo