From 3044646148cdfa83a311bf1c146a70e550280159 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Sun, 9 Nov 2008 10:07:58 -0800 Subject: x86: move iomap.h to the new include location a new file was accidentally added to include/asm-x86; move it to the new arch/x86/include/asm location Signed-off-by: Arjan van de Ven --- arch/x86/include/asm/iomap.h | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) create mode 100644 arch/x86/include/asm/iomap.h (limited to 'arch/x86') diff --git a/arch/x86/include/asm/iomap.h b/arch/x86/include/asm/iomap.h new file mode 100644 index 000000000000..c1f06289b14b --- /dev/null +++ b/arch/x86/include/asm/iomap.h @@ -0,0 +1,30 @@ +/* + * Copyright © 2008 Ingo Molnar + * + * 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; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA. + */ + +#include +#include +#include +#include +#include +#include + +void * +iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot); + +void +iounmap_atomic(void *kvaddr, enum km_type type); -- cgit v1.2.2 From 1de5b0854623d30d01d72cd4ea323eb5f39d1f16 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Sun, 2 Nov 2008 16:04:18 +0000 Subject: x86: HPET: convert WARN_ON to WARN_ON_ONCE It is possible to flood the console with call traces if the WARN_ON condition is true because of the frequency with which this function is called. Signed-off-by: Matt Fleming Cc: mingo@elte.hu Cc: venkatesh.pallipadi@intel.com Signed-off-by: Thomas Gleixner --- arch/x86/kernel/hpet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 77017e834cf7..f10f9461a43d 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -322,7 +322,7 @@ static int hpet_next_event(unsigned long delta, * what we wrote hit the chip before we compare it to the * counter. */ - WARN_ON((u32)hpet_readl(HPET_T0_CMP) != cnt); + WARN_ON_ONCE((u32)hpet_readl(HPET_T0_CMP) != cnt); return (s32)((u32)hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0; } -- cgit v1.2.2 From 89d77a1eb60be916d85d9394bedbfa2037af89c5 Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Sun, 2 Nov 2008 16:04:20 +0000 Subject: x86: HPET: read from HPET_Tn_CMP() not HPET_T0_CMP In hpet_next_event() we check that the value we just wrote to HPET_Tn_CMP(timer) has reached the chip. Currently, we're checking that the value we wrote to HPET_Tn_CMP(timer) is in HPET_T0_CMP, which, if timer is anything other than timer 0, is likely to fail. Signed-off-by: Matt Fleming Cc: mingo@elte.hu Cc: venkatesh.pallipadi@intel.com Signed-off-by: Thomas Gleixner --- arch/x86/kernel/hpet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index f10f9461a43d..cfe6aa56f71b 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -322,7 +322,7 @@ static int hpet_next_event(unsigned long delta, * what we wrote hit the chip before we compare it to the * counter. */ - WARN_ON_ONCE((u32)hpet_readl(HPET_T0_CMP) != cnt); + WARN_ON_ONCE((u32)hpet_readl(HPET_Tn_CMP(timer)) != cnt); return (s32)((u32)hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0; } -- cgit v1.2.2 From 5ceb1a04187553e08c6ab60d30cee7c454ee139a Mon Sep 17 00:00:00 2001 From: Matt Fleming Date: Sun, 2 Nov 2008 22:23:13 +0000 Subject: x86: HPET: enter hpet_interrupt_handler with interrupts disabled Some functions that may be called from this handler require that interrupts are disabled. Also, combining IRQF_DISABLED and IRQF_SHARED does not reliably disable interrupts in a handler, so remove IRQF_SHARED from the irq flags (this irq is not shared anyway). Signed-off-by: Matt Fleming Cc: mingo@elte.hu Cc: venkatesh.pallipadi@intel.com Cc: "Will Newton" Signed-off-by: Thomas Gleixner --- arch/x86/kernel/hpet.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index cfe6aa56f71b..067d8de913f6 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -445,7 +445,7 @@ static int hpet_setup_irq(struct hpet_dev *dev) { if (request_irq(dev->irq, hpet_interrupt_handler, - IRQF_SHARED|IRQF_NOBALANCING, dev->name, dev)) + IRQF_DISABLED|IRQF_NOBALANCING, dev->name, dev)) return -1; disable_irq(dev->irq); -- cgit v1.2.2 From 4694516d1987303dd83bfd0efdd36fa5b65d701b Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 10 Nov 2008 21:52:47 +0100 Subject: x86: Make NUMA on 32-bit depend on BROKEN While investigating the failure of hibernation on 32-bit x86 with CONFIG_NUMA set, as described in this message http://marc.info/?l=linux-kernel&m=122634118116226&w=4 I asked some people for help and I was told that it wasn't really worth the effort, because CONFIG_NUMA was generally broken on 32-bit x86 systems and it shouldn't be used in such configs. For this reason, make CONFIG_NUMA depend on BROKEN instead of EXPERIMENTAL on x86-32. Signed-off-by: Rafael J. Wysocki Cc: Andi Kleen Cc: Pavel Machek Cc: Peter Zijlstra Signed-off-by: Linus Torvalds --- arch/x86/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 4cf0ab13d187..93224b569187 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -957,7 +957,7 @@ config ARCH_PHYS_ADDR_T_64BIT config NUMA bool "Numa Memory Allocation and Scheduler Support (EXPERIMENTAL)" depends on SMP - depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || X86_BIGSMP || X86_SUMMIT && ACPI) && EXPERIMENTAL) + depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || X86_BIGSMP || X86_SUMMIT && ACPI) && BROKEN) default n if X86_PC default y if (X86_NUMAQ || X86_SUMMIT || X86_BIGSMP) help -- cgit v1.2.2 From 6cd10f8db385ba547811baa5b26f672fdff232e6 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Sun, 9 Nov 2008 11:53:14 -0600 Subject: x86, voyager: fix smp generic helper voyager breakage Impact: build/boot fix for x86/Voyager This change: | commit 3d4422332711ef48ef0f132f1fcbfcbd56c7f3d1 | Author: Jens Axboe | Date: Thu Jun 26 11:21:34 2008 +0200 | | Add generic helpers for arch IPI function calls didn't wire up the voyager smp call function correctly, so do that here. Also make CONFIG_USE_GENERIC_SMP_HELPERS a def_bool y again, since we now use the generic helpers for every x86 architecture. Signed-off-by: James Bottomley Cc: Jens Axboe Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 5 ++++- arch/x86/mach-voyager/voyager_smp.c | 16 ++++++++++++++-- 2 files changed, 18 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 4cf0ab13d187..ac22bb7719f7 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -167,9 +167,12 @@ config GENERIC_PENDING_IRQ config X86_SMP bool depends on SMP && ((X86_32 && !X86_VOYAGER) || X86_64) - select USE_GENERIC_SMP_HELPERS default y +config USE_GENERIC_SMP_HELPERS + def_bool y + depends on SMP + config X86_32_SMP def_bool y depends on X86_32 && SMP diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c index 0e331652681e..52145007bd7e 100644 --- a/arch/x86/mach-voyager/voyager_smp.c +++ b/arch/x86/mach-voyager/voyager_smp.c @@ -7,6 +7,7 @@ * This file provides all the same external entries as smp.c but uses * the voyager hal to provide the functionality */ +#include #include #include #include @@ -1790,6 +1791,17 @@ void __init smp_setup_processor_id(void) x86_write_percpu(cpu_number, hard_smp_processor_id()); } +static void voyager_send_call_func(cpumask_t callmask) +{ + __u32 mask = cpus_addr(callmask)[0] & ~(1 << smp_processor_id()); + send_CPI(mask, VIC_CALL_FUNCTION_CPI); +} + +static void voyager_send_call_func_single(int cpu) +{ + send_CPI(1 << cpu, VIC_CALL_FUNCTION_SINGLE_CPI); +} + struct smp_ops smp_ops = { .smp_prepare_boot_cpu = voyager_smp_prepare_boot_cpu, .smp_prepare_cpus = voyager_smp_prepare_cpus, @@ -1799,6 +1811,6 @@ struct smp_ops smp_ops = { .smp_send_stop = voyager_smp_send_stop, .smp_send_reschedule = voyager_smp_send_reschedule, - .send_call_func_ipi = native_send_call_func_ipi, - .send_call_func_single_ipi = native_send_call_func_single_ipi, + .send_call_func_ipi = voyager_send_call_func, + .send_call_func_single_ipi = voyager_send_call_func_single, }; -- cgit v1.2.2 From c41ef344de212bd918f7765af21b5008628c03e0 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 28 Oct 2008 18:16:58 -0200 Subject: KVM: MMU: increase per-vcpu rmap cache alloc size The page fault path can use two rmap_desc structures, if: - walk_addr's dirty pte update allocates one rmap_desc. - mmu_lock is dropped, sptes are zapped resulting in rmap_desc being freed. - fetch->mmu_set_spte allocates another rmap_desc. Increase to 4 for safety. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 2a5e64881d9b..f1983d9477cd 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -314,7 +314,7 @@ static int mmu_topup_memory_caches(struct kvm_vcpu *vcpu) if (r) goto out; r = mmu_topup_memory_cache(&vcpu->arch.mmu_rmap_desc_cache, - rmap_desc_cache, 1); + rmap_desc_cache, 4); if (r) goto out; r = mmu_topup_memory_cache_page(&vcpu->arch.mmu_page_cache, 8); -- cgit v1.2.2 From a29a2af378f3f6362b68e126e2541c8bde885ead Mon Sep 17 00:00:00 2001 From: Rakib Mullick Date: Wed, 29 Oct 2008 14:13:39 -0700 Subject: x86: KVM guest: fix section mismatch warning in kvmclock.c WARNING: arch/x86/kernel/built-in.o(.text+0x1722c): Section mismatch in reference from the function kvm_setup_secondary_clock() to the function .devinit.text:setup_secondary_APIC_clock() The function kvm_setup_secondary_clock() references the function __devinit setup_secondary_APIC_clock(). This is often because kvm_setup_secondary_clock lacks a __devinit annotation or the annotation of setup_secondary_APIC_clock is wrong. Signed-off-by: Md.Rakib H. Mullick Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Avi Kivity --- arch/x86/kernel/kvmclock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index 774ac4991568..1c9cc431ea4f 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -128,7 +128,7 @@ static int kvm_register_clock(char *txt) } #ifdef CONFIG_X86_LOCAL_APIC -static void kvm_setup_secondary_clock(void) +static void __devinit kvm_setup_secondary_clock(void) { /* * Now that the first cpu already had this clocksource initialized, -- cgit v1.2.2 From ca93e992fdfdc6569ac2845d7560eeb5de4a4e0b Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Tue, 4 Nov 2008 11:25:17 +0200 Subject: KVM: Require the PCI subsystem PCI device assignment makes calls to pci code, so require it to be built into the kernel. Signed-off-by: Avi Kivity --- arch/x86/kvm/Kconfig | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig index ce3251ce5504..b81125f0bdee 100644 --- a/arch/x86/kvm/Kconfig +++ b/arch/x86/kvm/Kconfig @@ -20,6 +20,8 @@ if VIRTUALIZATION config KVM tristate "Kernel-based Virtual Machine (KVM) support" depends on HAVE_KVM + # for device assignment: + depends on PCI select PREEMPT_NOTIFIERS select MMU_NOTIFIER select ANON_INODES -- cgit v1.2.2 From 928d4bf747e9c290b690ff515d8f81e8ee226d97 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Thu, 6 Nov 2008 14:55:45 +0800 Subject: KVM: VMX: Set IGMT bit in EPT entry There is a potential issue that, when guest using pagetable without vmexit when EPT enabled, guest would use PAT/PCD/PWT bits to index PAT msr for it's memory, which would be inconsistent with host side and would cause host MCE due to inconsistent cache attribute. The patch set IGMT bit in EPT entry to ignore guest PAT and use WB as default memory type to protect host (notice that all memory mapped by KVM should be WB). Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/kvm/vmx.c | 3 ++- arch/x86/kvm/vmx.h | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 2643b430d83a..d06b4dc0e2ea 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3564,7 +3564,8 @@ static int __init vmx_init(void) bypass_guest_pf = 0; kvm_mmu_set_base_ptes(VMX_EPT_READABLE_MASK | VMX_EPT_WRITABLE_MASK | - VMX_EPT_DEFAULT_MT << VMX_EPT_MT_EPTE_SHIFT); + VMX_EPT_DEFAULT_MT << VMX_EPT_MT_EPTE_SHIFT | + VMX_EPT_IGMT_BIT); kvm_mmu_set_mask_ptes(0ull, 0ull, 0ull, 0ull, VMX_EPT_EXECUTABLE_MASK); kvm_enable_tdp(); diff --git a/arch/x86/kvm/vmx.h b/arch/x86/kvm/vmx.h index 3e010d21fdd7..ec5edc339da6 100644 --- a/arch/x86/kvm/vmx.h +++ b/arch/x86/kvm/vmx.h @@ -352,6 +352,7 @@ enum vmcs_field { #define VMX_EPT_READABLE_MASK 0x1ull #define VMX_EPT_WRITABLE_MASK 0x2ull #define VMX_EPT_EXECUTABLE_MASK 0x4ull +#define VMX_EPT_IGMT_BIT (1ull << 6) #define VMX_EPT_IDENTITY_PAGETABLE_ADDR 0xfffbc000ul -- cgit v1.2.2 From e17d1dc0863767bab8fde4ba9be92c7f79e7fe50 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Tue, 11 Nov 2008 13:09:36 +0200 Subject: KVM: Fix pit memory leak if unable to allocate irq source id MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Reported-By: Daniel Marjamäki Signed-off-by: Avi Kivity --- arch/x86/kvm/i8254.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index 8772dc946823..59ebd37ad79e 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -548,8 +548,10 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm) mutex_lock(&kvm->lock); pit->irq_source_id = kvm_request_irq_source_id(kvm); mutex_unlock(&kvm->lock); - if (pit->irq_source_id < 0) + if (pit->irq_source_id < 0) { + kfree(pit); return NULL; + } mutex_init(&pit->pit_state.lock); mutex_lock(&pit->pit_state.lock); -- cgit v1.2.2 From 32836259ff25ce97010569706cd33ba94de81d62 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 5 Nov 2008 16:17:52 -0700 Subject: ACPI: pci_link: remove acpi_irq_balance_set() interface This removes the acpi_irq_balance_set() interface from the PCI interrupt link driver. x86 used acpi_irq_balance_set() to tell the PCI interrupt link driver to configure links to minimize IRQ sharing. But the link driver can easily figure out whether to turn on IRQ balancing based on the IRQ model (PIC/IOAPIC/etc), so we can get rid of that external interface. It's better for the driver to figure this out at init-time. If we set it externally via the x86 code, the interface reduces modularity, and we depend on the fact that acpi_process_madt() happens before we process the kernel command line. Signed-off-by: Bjorn Helgaas Signed-off-by: Len Brown --- arch/x86/include/asm/acpi.h | 1 - arch/x86/kernel/acpi/boot.c | 1 - 2 files changed, 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index 8d676d8ecde9..9830681446ad 100644 --- a/arch/x86/include/asm/acpi.h +++ b/arch/x86/include/asm/acpi.h @@ -113,7 +113,6 @@ static inline void acpi_disable_pci(void) acpi_pci_disabled = 1; acpi_noirq_set(); } -extern int acpi_irq_balance_set(char *str); /* routines for saving/restoring kernel state */ extern int acpi_save_state_mem(void); diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 8c1f76abae9e..4c51a2f8fd31 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -1343,7 +1343,6 @@ static void __init acpi_process_madt(void) error = acpi_parse_madt_ioapic_entries(); if (!error) { acpi_irq_model = ACPI_IRQ_MODEL_IOAPIC; - acpi_irq_balance_set(NULL); acpi_ioapic = 1; smp_found_config = 1; -- cgit v1.2.2 From 97a70e548bd97d5a46ae9d44f24aafcc013fd701 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 12 Nov 2008 23:22:35 +0100 Subject: x86, hibernate: fix breakage on x86_32 with CONFIG_NUMA set Impact: fix crash during hibernation on 32-bit NUMA The NUMA code on x86_32 creates special memory mapping that allows each node's pgdat to be located in this node's memory. For this purpose it allocates a memory area at the end of each node's memory and maps this area so that it is accessible with virtual addresses belonging to low memory. As a result, if there is high memory, these NUMA-allocated areas are physically located in high memory, although they are mapped to low memory addresses. Our hibernation code does not take that into account and for this reason hibernation fails on all x86_32 systems with CONFIG_NUMA=y and with high memory present. Fix this by adding a special mapping for the NUMA-allocated memory areas to the temporary page tables created during the last phase of resume. Signed-off-by: Rafael J. Wysocki Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mmzone_32.h | 4 ++++ arch/x86/mm/numa_32.c | 35 +++++++++++++++++++++++++++++++++++ arch/x86/power/hibernate_32.c | 4 ++++ 3 files changed, 43 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h index 485bdf059ffb..07f1af494ca5 100644 --- a/arch/x86/include/asm/mmzone_32.h +++ b/arch/x86/include/asm/mmzone_32.h @@ -34,10 +34,14 @@ static inline void get_memcfg_numa(void) extern int early_pfn_to_nid(unsigned long pfn); +extern void resume_map_numa_kva(pgd_t *pgd); + #else /* !CONFIG_NUMA */ #define get_memcfg_numa get_memcfg_numa_flat +static inline void resume_map_numa_kva(pgd_t *pgd) {} + #endif /* CONFIG_NUMA */ #ifdef CONFIG_DISCONTIGMEM diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index 847c164725f4..8518c678d83f 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -222,6 +222,41 @@ static void __init remap_numa_kva(void) } } +#ifdef CONFIG_HIBERNATION +/** + * resume_map_numa_kva - add KVA mapping to the temporary page tables created + * during resume from hibernation + * @pgd_base - temporary resume page directory + */ +void resume_map_numa_kva(pgd_t *pgd_base) +{ + int node; + + for_each_online_node(node) { + unsigned long start_va, start_pfn, size, pfn; + + start_va = (unsigned long)node_remap_start_vaddr[node]; + start_pfn = node_remap_start_pfn[node]; + size = node_remap_size[node]; + + printk(KERN_DEBUG "%s: node %d\n", __FUNCTION__, node); + + for (pfn = 0; pfn < size; pfn += PTRS_PER_PTE) { + unsigned long vaddr = start_va + (pfn << PAGE_SHIFT); + pgd_t *pgd = pgd_base + pgd_index(vaddr); + pud_t *pud = pud_offset(pgd, vaddr); + pmd_t *pmd = pmd_offset(pud, vaddr); + + set_pmd(pmd, pfn_pmd(start_pfn + pfn, + PAGE_KERNEL_LARGE_EXEC)); + + printk(KERN_DEBUG "%s: %08lx -> pfn %08lx\n", + __FUNCTION__, vaddr, start_pfn + pfn); + } + } +} +#endif + static unsigned long calculate_numa_remap_pages(void) { int nid; diff --git a/arch/x86/power/hibernate_32.c b/arch/x86/power/hibernate_32.c index f2b6e3f11bfc..81197c62d5b3 100644 --- a/arch/x86/power/hibernate_32.c +++ b/arch/x86/power/hibernate_32.c @@ -12,6 +12,7 @@ #include #include #include +#include /* Defined in hibernate_asm_32.S */ extern int restore_image(void); @@ -127,6 +128,9 @@ static int resume_physical_mapping_init(pgd_t *pgd_base) } } } + + resume_map_numa_kva(pgd_base); + return 0; } -- cgit v1.2.2 From 604d20554883cf03f888440d58ea7c6d36899839 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 12 Nov 2008 23:26:14 +0100 Subject: x86: make NUMA on 32-bit depend on EXPERIMENTAL again My previous patch to make CONFIG_NUMA on x86_32 depend on BROKEN turned out to be unnecessary, after all, since the source of the hibernation vs CONFIG_NUMA problem turned out to be the fact that we didn't take the NUMA KVA remapping into account in the hibernation code. Signed-off-by: Rafael J. Wysocki Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 93224b569187..4cf0ab13d187 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -957,7 +957,7 @@ config ARCH_PHYS_ADDR_T_64BIT config NUMA bool "Numa Memory Allocation and Scheduler Support (EXPERIMENTAL)" depends on SMP - depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || X86_BIGSMP || X86_SUMMIT && ACPI) && BROKEN) + depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || X86_BIGSMP || X86_SUMMIT && ACPI) && EXPERIMENTAL) default n if X86_PC default y if (X86_NUMAQ || X86_SUMMIT || X86_BIGSMP) help -- cgit v1.2.2 From 52168e60f7d86d83124903098ac8c2dba93cd1c4 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 14 Nov 2008 13:47:31 +0000 Subject: Revert "x86: blacklist DMAR on Intel G31/G33 chipsets" This reverts commit e51af6630848406fc97adbd71443818cdcda297b, which was wrongly hoovered up and submitted about a month after a better fix had already been merged. The better fix is commit cbda1ba898647aeb4ee770b803c922f595e97731 ("PCI/iommu: blacklist DMAR on Intel G31/G33 chipsets"), where we do this blacklisting based on the DMI identification for the offending motherboard, since sometimes this chipset (or at least a chipset with the same PCI ID) apparently _does_ actually have an IOMMU. Signed-off-by: David Woodhouse Signed-off-by: Linus Torvalds --- arch/x86/include/asm/iommu.h | 1 - arch/x86/kernel/early-quirks.c | 18 ------------------ 2 files changed, 19 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/iommu.h b/arch/x86/include/asm/iommu.h index e4a552d44465..0b500c5b6446 100644 --- a/arch/x86/include/asm/iommu.h +++ b/arch/x86/include/asm/iommu.h @@ -6,7 +6,6 @@ extern void no_iommu_init(void); extern struct dma_mapping_ops nommu_dma_ops; extern int force_iommu, no_iommu; extern int iommu_detected; -extern int dmar_disabled; extern unsigned long iommu_nr_pages(unsigned long addr, unsigned long len); diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c index 3ce029ffaa55..1b894b72c0f5 100644 --- a/arch/x86/kernel/early-quirks.c +++ b/arch/x86/kernel/early-quirks.c @@ -188,20 +188,6 @@ static void __init ati_bugs_contd(int num, int slot, int func) } #endif -#ifdef CONFIG_DMAR -static void __init intel_g33_dmar(int num, int slot, int func) -{ - struct acpi_table_header *dmar_tbl; - acpi_status status; - - status = acpi_get_table(ACPI_SIG_DMAR, 0, &dmar_tbl); - if (ACPI_SUCCESS(status)) { - printk(KERN_INFO "BIOS BUG: DMAR advertised on Intel G31/G33 chipset -- ignoring\n"); - dmar_disabled = 1; - } -} -#endif - #define QFLAG_APPLY_ONCE 0x1 #define QFLAG_APPLIED 0x2 #define QFLAG_DONE (QFLAG_APPLY_ONCE|QFLAG_APPLIED) @@ -225,10 +211,6 @@ static struct chipset early_qrk[] __initdata = { PCI_CLASS_SERIAL_SMBUS, PCI_ANY_ID, 0, ati_bugs }, { PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_SBX00_SMBUS, PCI_CLASS_SERIAL_SMBUS, PCI_ANY_ID, 0, ati_bugs_contd }, -#ifdef CONFIG_DMAR - { PCI_VENDOR_ID_INTEL, 0x29c0, - PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, intel_g33_dmar }, -#endif {} }; -- cgit v1.2.2 From d1f1e9c01006b4b050e090055c75278f80c2a5c5 Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Sat, 15 Nov 2008 11:00:17 +0100 Subject: x86, bts: fix unlock problem in ds.c Fix a problem where ds_request() returned an error without releasing the ds lock. Reported-by: Stephane Eranian Signed-off-by: Markus Metzger Cc: Signed-off-by: Ingo Molnar --- arch/x86/kernel/ds.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index 2b69994fd3a8..ac1d5b0586ba 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -384,8 +384,9 @@ static int ds_request(struct task_struct *task, void *base, size_t size, spin_lock(&ds_lock); + error = -EPERM; if (!check_tracer(task)) - return -EPERM; + goto out_unlock; error = -ENOMEM; context = ds_alloc_context(task); -- cgit v1.2.2 From d3c6aa1e69f705ac3ab64584101b1d38435b1353 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 16 Nov 2008 00:49:31 -0800 Subject: x86: fix es7000 compiling Impact: fix es7000 build CC arch/x86/kernel/es7000_32.o arch/x86/kernel/es7000_32.c: In function find_unisys_acpi_oem_table: arch/x86/kernel/es7000_32.c:255: error: implicit declaration of function acpi_get_table_with_size arch/x86/kernel/es7000_32.c:261: error: implicit declaration of function early_acpi_os_unmap_memory arch/x86/kernel/es7000_32.c: In function unmap_unisys_acpi_oem_table: arch/x86/kernel/es7000_32.c:277: error: implicit declaration of function __acpi_unmap_table make[1]: *** [arch/x86/kernel/es7000_32.o] Error 1 we applied one patch out of order... | commit a73aaedd95703bd49f4c3f9df06fb7b7373ba905 | Author: Yinghai Lu | Date: Sun Sep 14 02:33:14 2008 -0700 | | x86: check dsdt before find oem table for es7000, v2 | | v2: use __acpi_unmap_table() that patch need: x86: use early_ioremap in __acpi_map_table x86: always explicitly map acpi memory acpi: remove final __acpi_map_table mapping before setting acpi_gbl_permanent_mmap acpi/x86: introduce __apci_map_table, v4 submitted to the ACPI tree but not upstream yet. fix it until those patches applied, need to revert this one Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/es7000_32.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/es7000_32.c b/arch/x86/kernel/es7000_32.c index f454c78fcef6..0aa2c443d600 100644 --- a/arch/x86/kernel/es7000_32.c +++ b/arch/x86/kernel/es7000_32.c @@ -250,31 +250,24 @@ int __init find_unisys_acpi_oem_table(unsigned long *oem_addr) { struct acpi_table_header *header = NULL; int i = 0; - acpi_size tbl_size; - while (ACPI_SUCCESS(acpi_get_table_with_size("OEM1", i++, &header, &tbl_size))) { + while (ACPI_SUCCESS(acpi_get_table("OEM1", i++, &header))) { if (!memcmp((char *) &header->oem_id, "UNISYS", 6)) { struct oem_table *t = (struct oem_table *)header; oem_addrX = t->OEMTableAddr; oem_size = t->OEMTableSize; - early_acpi_os_unmap_memory(header, tbl_size); *oem_addr = (unsigned long)__acpi_map_table(oem_addrX, oem_size); return 0; } - early_acpi_os_unmap_memory(header, tbl_size); } return -1; } void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr) { - if (!oem_addr) - return; - - __acpi_unmap_table((char *)oem_addr, oem_size); } #endif -- cgit v1.2.2 From 93ce99e849433ede4ce8b410b749dc0cad1100b2 Mon Sep 17 00:00:00 2001 From: Venki Pallipadi Date: Mon, 17 Nov 2008 14:43:58 -0800 Subject: x86: add rdtsc barrier to TSC sync check Impact: fix incorrectly marked unstable TSC clock Patch (commit 0d12cdd "sched: improve sched_clock() performance") has a regression on one of the test systems here. With the patch, I see: checking TSC synchronization [CPU#0 -> CPU#1]: Measured 28 cycles TSC warp between CPUs, turning off TSC clock. Marking TSC unstable due to check_tsc_sync_source failed Whereas, without the patch syncs pass fine on all CPUs: checking TSC synchronization [CPU#0 -> CPU#1]: passed. Due to this, TSC is marked unstable, when it is not actually unstable. This is because syncs in check_tsc_wrap() goes away due to this commit. As per the discussion on this thread, correct way to fix this is to add explicit syncs as below? Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- arch/x86/kernel/tsc_sync.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c index 9ffb01c31c40..1c0dfbca87c1 100644 --- a/arch/x86/kernel/tsc_sync.c +++ b/arch/x86/kernel/tsc_sync.c @@ -46,7 +46,9 @@ static __cpuinit void check_tsc_warp(void) cycles_t start, now, prev, end; int i; + rdtsc_barrier(); start = get_cycles(); + rdtsc_barrier(); /* * The measurement runs for 20 msecs: */ @@ -61,7 +63,9 @@ static __cpuinit void check_tsc_warp(void) */ __raw_spin_lock(&sync_lock); prev = last_tsc; + rdtsc_barrier(); now = get_cycles(); + rdtsc_barrier(); last_tsc = now; __raw_spin_unlock(&sync_lock); -- cgit v1.2.2 From 10db4ef7b9a65b86e4d047671a1886f4c101a859 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 18 Nov 2008 15:23:08 +0100 Subject: x86, PEBS/DS: fix code flow in ds_request() this compiler warning: arch/x86/kernel/ds.c: In function 'ds_request': arch/x86/kernel/ds.c:368: warning: 'context' may be used uninitialized in this function Shows that the code flow in ds_request() is buggy - it goes into the unlock+release-context path even when the context is not allocated yet. First allocate the context, then do the other checks. Also, take care with GFP allocations under the ds_lock spinlock. Cc: Signed-off-by: Ingo Molnar --- arch/x86/kernel/ds.c | 26 +++++++++++++++++++++----- 1 file changed, 21 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index ac1d5b0586ba..d1a121443bde 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -236,17 +236,33 @@ static inline struct ds_context *ds_alloc_context(struct task_struct *task) struct ds_context *context = *p_context; if (!context) { + spin_unlock(&ds_lock); + context = kzalloc(sizeof(*context), GFP_KERNEL); - if (!context) + if (!context) { + spin_lock(&ds_lock); return NULL; + } context->ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL); if (!context->ds) { kfree(context); + spin_lock(&ds_lock); return NULL; } + spin_lock(&ds_lock); + /* + * Check for race - another CPU could have allocated + * it meanwhile: + */ + if (*p_context) { + kfree(context->ds); + kfree(context); + return *p_context; + } + *p_context = context; context->this = p_context; @@ -384,15 +400,15 @@ static int ds_request(struct task_struct *task, void *base, size_t size, spin_lock(&ds_lock); - error = -EPERM; - if (!check_tracer(task)) - goto out_unlock; - error = -ENOMEM; context = ds_alloc_context(task); if (!context) goto out_unlock; + error = -EPERM; + if (!check_tracer(task)) + goto out_unlock; + error = -EALREADY; if (context->owner[qual] == current) goto out_unlock; -- cgit v1.2.2 From e5e1f606ecbf67e52ebe2df5d14f8b94ec6544d0 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 17 Nov 2008 15:07:17 +0100 Subject: AMD IOMMU: add parameter to disable device isolation Impact: add a new AMD IOMMU kernel command line parameter Signed-off-by: Joerg Roedel --- arch/x86/kernel/amd_iommu_init.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 0cdcda35a05f..838a2e1d5bb2 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -1213,6 +1213,8 @@ static int __init parse_amd_iommu_options(char *str) for (; *str; ++str) { if (strncmp(str, "isolate", 7) == 0) amd_iommu_isolate = 1; + if (strncmp(str, "share", 5) == 0) + amd_iommu_isolate = 0; if (strncmp(str, "fullflush", 11) == 0) amd_iommu_unmap_flush = true; } -- cgit v1.2.2 From 3ce1f93c6d53c3f91c3846cf66b018276c8ac2e7 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 17 Nov 2008 15:09:20 +0100 Subject: AMD IOMMU: enable device isolation per default Impact: makes device isolation the default for AMD IOMMU Some device drivers showed double-free bugs of DMA memory while testing them with AMD IOMMU. If all devices share the same protection domain this can lead to data corruption and data loss. Prevent this by putting each device into its own protection domain per default. Signed-off-by: Joerg Roedel --- arch/x86/kernel/amd_iommu_init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 838a2e1d5bb2..595edd2befc6 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -121,7 +121,7 @@ u16 amd_iommu_last_bdf; /* largest PCI device id we have LIST_HEAD(amd_iommu_unity_map); /* a list of required unity mappings we find in ACPI */ unsigned amd_iommu_aperture_order = 26; /* size of aperture in power of 2 */ -int amd_iommu_isolate; /* if 1, device isolation is enabled */ +int amd_iommu_isolate = 1; /* if 1, device isolation is enabled */ bool amd_iommu_unmap_flush; /* if true, flush on every unmap */ LIST_HEAD(amd_iommu_list); /* list of all AMD IOMMUs in the -- cgit v1.2.2 From 695b5676c727d80921a2dc8737d5b3322222db85 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 17 Nov 2008 15:16:43 +0100 Subject: AMD IOMMU: fix fullflush comparison length Impact: fix comparison length for 'fullflush' Signed-off-by: Joerg Roedel --- arch/x86/kernel/amd_iommu_init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 595edd2befc6..30ae2701b3df 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -1215,7 +1215,7 @@ static int __init parse_amd_iommu_options(char *str) amd_iommu_isolate = 1; if (strncmp(str, "share", 5) == 0) amd_iommu_isolate = 0; - if (strncmp(str, "fullflush", 11) == 0) + if (strncmp(str, "fullflush", 9) == 0) amd_iommu_unmap_flush = true; } -- cgit v1.2.2 From 8501c45cc32c311ae755a2d5ac8c4a5f04908d42 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 17 Nov 2008 19:11:46 +0100 Subject: AMD IOMMU: check for next_bit also in unmapped area Impact: fix possible use of stale IO/TLB entries Signed-off-by: Joerg Roedel --- arch/x86/kernel/amd_iommu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 331b318304eb..e4899e0e8787 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -537,7 +537,7 @@ static void dma_ops_free_addresses(struct dma_ops_domain *dom, address >>= PAGE_SHIFT; iommu_area_free(dom->bitmap, address, pages); - if (address + pages >= dom->next_bit) + if (address >= dom->next_bit) dom->need_flush = true; } -- cgit v1.2.2 From 0af40a4b1050c050e62eb1dc30b82d5ab22bf221 Mon Sep 17 00:00:00 2001 From: Philipp Kohlbecher Date: Sun, 16 Nov 2008 12:11:01 +0100 Subject: x86: more general identifier for Phoenix BIOS Impact: widen the reach of the low-memory-protect DMI quirk Phoenix BIOSes variously identify their vendor as "Phoenix Technologies, LTD" or "Phoenix Technologies LTD" (without the comma.) This patch makes the identification string in the bad_bios_dmi_table more general (following a suggestion by Ingo Molnar), so that both versions are handled. Again, the patched file compiles cleanly and the patch has been tested successfully on my machine. Signed-off-by: Philipp Kohlbecher Cc: Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 0fa6790c1dd3..9d5674f7b6cc 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -764,7 +764,7 @@ static struct dmi_system_id __initdata bad_bios_dmi_table[] = { .callback = dmi_low_memory_corruption, .ident = "Phoenix BIOS", .matches = { - DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies, LTD"), + DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies"), }, }, #endif -- cgit v1.2.2 From 093bac154c142fa1fb31a3ac69ae1bc08930231b Mon Sep 17 00:00:00 2001 From: Steve Conklin Date: Fri, 14 Nov 2008 00:55:51 -0600 Subject: x86: quirk for reboot stalls on a Dell Optiplex 330 Dell Optiplex 330 appears to hang on reboot. This is resolved by adding a quirk to set bios reboot. Signed-off-by: Leann Ogasawara Signed-off-by: Steve Conklin Signed-off-by: Ingo Molnar --- arch/x86/kernel/reboot.c | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 724adfc63cb9..cc5a2545dd41 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -169,6 +169,15 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_BOARD_NAME, "0KW626"), }, }, + { /* Handle problems with rebooting on Dell Optiplex 330 with 0KP561 */ + .callback = set_bios_reboot, + .ident = "Dell OptiPlex 330", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 330"), + DMI_MATCH(DMI_BOARD_NAME, "0KP561"), + }, + }, { /* Handle problems with rebooting on Dell 2400's */ .callback = set_bios_reboot, .ident = "Dell PowerEdge 2400", -- cgit v1.2.2 From 20a4a236c7de5c915551cdc562482aa53eaff40e Mon Sep 17 00:00:00 2001 From: Hiroshi Shimamoto Date: Thu, 13 Nov 2008 18:06:04 -0800 Subject: x86: uaccess_64: fix return value in __copy_from_user() __copy_from_user() will return invalid value 16 when it fails to access user space and the size is 10. Signed-off-by: Hiroshi Shimamoto Signed-off-by: Ingo Molnar --- arch/x86/include/asm/uaccess_64.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h index 664f15280f14..f8cfd00db450 100644 --- a/arch/x86/include/asm/uaccess_64.h +++ b/arch/x86/include/asm/uaccess_64.h @@ -46,7 +46,7 @@ int __copy_from_user(void *dst, const void __user *src, unsigned size) return ret; case 10: __get_user_asm(*(u64 *)dst, (u64 __user *)src, - ret, "q", "", "=r", 16); + ret, "q", "", "=r", 10); if (unlikely(ret)) return ret; __get_user_asm(*(u16 *)(8 + (char *)dst), -- cgit v1.2.2 From de11defebf00007677fb7ee91d9b089b78786fbb Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 19 Nov 2008 15:36:14 -0800 Subject: reintroduce accept4 Introduce a new accept4() system call. The addition of this system call matches analogous changes in 2.6.27 (dup3(), evenfd2(), signalfd4(), inotify_init1(), epoll_create1(), pipe2()) which added new system calls that differed from analogous traditional system calls in adding a flags argument that can be used to access additional functionality. The accept4() system call is exactly the same as accept(), except that it adds a flags bit-mask argument. Two flags are initially implemented. (Most of the new system calls in 2.6.27 also had both of these flags.) SOCK_CLOEXEC causes the close-on-exec (FD_CLOEXEC) flag to be enabled for the new file descriptor returned by accept4(). This is a useful security feature to avoid leaking information in a multithreaded program where one thread is doing an accept() at the same time as another thread is doing a fork() plus exec(). More details here: http://udrepper.livejournal.com/20407.html "Secure File Descriptor Handling", Ulrich Drepper). The other flag is SOCK_NONBLOCK, which causes the O_NONBLOCK flag to be enabled on the new open file description created by accept4(). (This flag is merely a convenience, saving the use of additional calls fcntl(F_GETFL) and fcntl (F_SETFL) to achieve the same result. Here's a test program. Works on x86-32. Should work on x86-64, but I (mtk) don't have a system to hand to test with. It tests accept4() with each of the four possible combinations of SOCK_CLOEXEC and SOCK_NONBLOCK set/clear in 'flags', and verifies that the appropriate flags are set on the file descriptor/open file description returned by accept4(). I tested Ulrich's patch in this thread by applying against 2.6.28-rc2, and it passes according to my test program. /* test_accept4.c Copyright (C) 2008, Linux Foundation, written by Michael Kerrisk Licensed under the GNU GPLv2 or later. */ #define _GNU_SOURCE #include #include #include #include #include #include #include #include #define PORT_NUM 33333 #define die(msg) do { perror(msg); exit(EXIT_FAILURE); } while (0) /**********************************************************************/ /* The following is what we need until glibc gets a wrapper for accept4() */ /* Flags for socket(), socketpair(), accept4() */ #ifndef SOCK_CLOEXEC #define SOCK_CLOEXEC O_CLOEXEC #endif #ifndef SOCK_NONBLOCK #define SOCK_NONBLOCK O_NONBLOCK #endif #ifdef __x86_64__ #define SYS_accept4 288 #elif __i386__ #define USE_SOCKETCALL 1 #define SYS_ACCEPT4 18 #else #error "Sorry -- don't know the syscall # on this architecture" #endif static int accept4(int fd, struct sockaddr *sockaddr, socklen_t *addrlen, int flags) { printf("Calling accept4(): flags = %x", flags); if (flags != 0) { printf(" ("); if (flags & SOCK_CLOEXEC) printf("SOCK_CLOEXEC"); if ((flags & SOCK_CLOEXEC) && (flags & SOCK_NONBLOCK)) printf(" "); if (flags & SOCK_NONBLOCK) printf("SOCK_NONBLOCK"); printf(")"); } printf("\n"); #if USE_SOCKETCALL long args[6]; args[0] = fd; args[1] = (long) sockaddr; args[2] = (long) addrlen; args[3] = flags; return syscall(SYS_socketcall, SYS_ACCEPT4, args); #else return syscall(SYS_accept4, fd, sockaddr, addrlen, flags); #endif } /**********************************************************************/ static int do_test(int lfd, struct sockaddr_in *conn_addr, int closeonexec_flag, int nonblock_flag) { int connfd, acceptfd; int fdf, flf, fdf_pass, flf_pass; struct sockaddr_in claddr; socklen_t addrlen; printf("=======================================\n"); connfd = socket(AF_INET, SOCK_STREAM, 0); if (connfd == -1) die("socket"); if (connect(connfd, (struct sockaddr *) conn_addr, sizeof(struct sockaddr_in)) == -1) die("connect"); addrlen = sizeof(struct sockaddr_in); acceptfd = accept4(lfd, (struct sockaddr *) &claddr, &addrlen, closeonexec_flag | nonblock_flag); if (acceptfd == -1) { perror("accept4()"); close(connfd); return 0; } fdf = fcntl(acceptfd, F_GETFD); if (fdf == -1) die("fcntl:F_GETFD"); fdf_pass = ((fdf & FD_CLOEXEC) != 0) == ((closeonexec_flag & SOCK_CLOEXEC) != 0); printf("Close-on-exec flag is %sset (%s); ", (fdf & FD_CLOEXEC) ? "" : "not ", fdf_pass ? "OK" : "failed"); flf = fcntl(acceptfd, F_GETFL); if (flf == -1) die("fcntl:F_GETFD"); flf_pass = ((flf & O_NONBLOCK) != 0) == ((nonblock_flag & SOCK_NONBLOCK) !=0); printf("nonblock flag is %sset (%s)\n", (flf & O_NONBLOCK) ? "" : "not ", flf_pass ? "OK" : "failed"); close(acceptfd); close(connfd); printf("Test result: %s\n", (fdf_pass && flf_pass) ? "PASS" : "FAIL"); return fdf_pass && flf_pass; } static int create_listening_socket(int port_num) { struct sockaddr_in svaddr; int lfd; int optval; memset(&svaddr, 0, sizeof(struct sockaddr_in)); svaddr.sin_family = AF_INET; svaddr.sin_addr.s_addr = htonl(INADDR_ANY); svaddr.sin_port = htons(port_num); lfd = socket(AF_INET, SOCK_STREAM, 0); if (lfd == -1) die("socket"); optval = 1; if (setsockopt(lfd, SOL_SOCKET, SO_REUSEADDR, &optval, sizeof(optval)) == -1) die("setsockopt"); if (bind(lfd, (struct sockaddr *) &svaddr, sizeof(struct sockaddr_in)) == -1) die("bind"); if (listen(lfd, 5) == -1) die("listen"); return lfd; } int main(int argc, char *argv[]) { struct sockaddr_in conn_addr; int lfd; int port_num; int passed; passed = 1; port_num = (argc > 1) ? atoi(argv[1]) : PORT_NUM; memset(&conn_addr, 0, sizeof(struct sockaddr_in)); conn_addr.sin_family = AF_INET; conn_addr.sin_addr.s_addr = htonl(INADDR_LOOPBACK); conn_addr.sin_port = htons(port_num); lfd = create_listening_socket(port_num); if (!do_test(lfd, &conn_addr, 0, 0)) passed = 0; if (!do_test(lfd, &conn_addr, SOCK_CLOEXEC, 0)) passed = 0; if (!do_test(lfd, &conn_addr, 0, SOCK_NONBLOCK)) passed = 0; if (!do_test(lfd, &conn_addr, SOCK_CLOEXEC, SOCK_NONBLOCK)) passed = 0; close(lfd); exit(passed ? EXIT_SUCCESS : EXIT_FAILURE); } [mtk.manpages@gmail.com: rewrote changelog, updated test program] Signed-off-by: Ulrich Drepper Tested-by: Michael Kerrisk Acked-by: Michael Kerrisk Cc: Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/include/asm/unistd_64.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/unistd_64.h b/arch/x86/include/asm/unistd_64.h index 834b2c1d89fb..d2e415e6666f 100644 --- a/arch/x86/include/asm/unistd_64.h +++ b/arch/x86/include/asm/unistd_64.h @@ -639,8 +639,8 @@ __SYSCALL(__NR_fallocate, sys_fallocate) __SYSCALL(__NR_timerfd_settime, sys_timerfd_settime) #define __NR_timerfd_gettime 287 __SYSCALL(__NR_timerfd_gettime, sys_timerfd_gettime) -#define __NR_paccept 288 -__SYSCALL(__NR_paccept, sys_paccept) +#define __NR_accept4 288 +__SYSCALL(__NR_accept4, sys_accept4) #define __NR_signalfd4 289 __SYSCALL(__NR_signalfd4, sys_signalfd4) #define __NR_eventfd2 290 -- cgit v1.2.2 From 0ca4b6b00113b064c080d26d803d0d7c80fb5dc8 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Thu, 20 Nov 2008 14:09:33 -0700 Subject: x86: Fix interrupt leak due to migration When we migrate an interrupt from one CPU to another, we set the move_in_progress flag and clean up the vectors later once they're not being used. If you're unlucky and call destroy_irq() before the vectors become un-used, the move_in_progress flag is never cleared, which causes the interrupt to become unusable. This was discovered by Jesse Brandeburg for whom it manifested as an MSI-X device refusing to use MSI-X mode when the driver was unloaded and reloaded repeatedly. Signed-off-by: Matthew Wilcox Signed-off-by: Linus Torvalds --- arch/x86/kernel/io_apic.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c index 7a3f2028e2eb..c9513e1ff28d 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/io_apic.c @@ -1140,6 +1140,20 @@ static void __clear_irq_vector(int irq) cfg->vector = 0; cpus_clear(cfg->domain); + + if (likely(!cfg->move_in_progress)) + return; + cpus_and(mask, cfg->old_domain, cpu_online_map); + for_each_cpu_mask_nr(cpu, mask) { + for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; + vector++) { + if (per_cpu(vector_irq, cpu)[vector] != irq) + continue; + per_cpu(vector_irq, cpu)[vector] = -1; + break; + } + } + cfg->move_in_progress = 0; } void __setup_vector_irq(int cpu) -- cgit v1.2.2