From caaa4c804fae7bb654f7d00b35b8583280a9c52c Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sat, 16 Nov 2013 17:46:02 +1100 Subject: KVM: PPC: Book3S HV: Fix physical address calculations This fixes a bug in kvmppc_do_h_enter() where the physical address for a page can be calculated incorrectly if transparent huge pages (THP) are active. Until THP came along, it was true that if we encountered a large (16M) page in kvmppc_do_h_enter(), then the associated memslot must be 16M aligned for both its guest physical address and the userspace address, and the physical address calculations in kvmppc_do_h_enter() assumed that. With THP, that is no longer true. In the case where we are using MMU notifiers and the page size that we get from the Linux page tables is larger than the page being mapped by the guest, we need to fill in some low-order bits of the physical address. Without THP, these bits would be the same in the guest physical address (gpa) and the host virtual address (hva). With THP, they can be different, and we need to use the bits from hva rather than gpa. In the case where we are not using MMU notifiers, the host physical address we get from the memslot->arch.slot_phys[] array already includes the low-order bits down to the PAGE_SIZE level, even if we are using large pages. Thus we can simplify the calculation in this case to just add in the remaining bits in the case where PAGE_SIZE is 64k and the guest is mapping a 4k page. The same bug exists in kvmppc_book3s_hv_page_fault(). The basic fix is to use psize (the page size from the HPTE) rather than pte_size (the page size from the Linux PTE) when updating the HPTE low word in r. That means that pfn needs to be computed to PAGE_SIZE granularity even if the Linux PTE is a huge page PTE. That can be arranged simply by doing the page_to_pfn() before setting page to the head of the compound page. If psize is less than PAGE_SIZE, then we need to make sure we only update the bits from PAGE_SIZE upwards, in order not to lose any sub-page offset bits in r. On the other hand, if psize is greater than PAGE_SIZE, we need to make sure we don't bring in non-zero low order bits in pfn, hence we mask (pfn << PAGE_SHIFT) with ~(psize - 1). Signed-off-by: Paul Mackerras Signed-off-by: Alexander Graf --- arch/powerpc/kvm/book3s_64_mmu_hv.c | 12 +++++++++--- arch/powerpc/kvm/book3s_hv_rm_mmu.c | 4 ++-- 2 files changed, 11 insertions(+), 5 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c index f3ff587a8b7d..47bbeaf2d320 100644 --- a/arch/powerpc/kvm/book3s_64_mmu_hv.c +++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c @@ -665,6 +665,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu, return -EFAULT; } else { page = pages[0]; + pfn = page_to_pfn(page); if (PageHuge(page)) { page = compound_head(page); pte_size <<= compound_order(page); @@ -689,7 +690,6 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu, } rcu_read_unlock_sched(); } - pfn = page_to_pfn(page); } ret = -EFAULT; @@ -707,8 +707,14 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu, r = (r & ~(HPTE_R_W|HPTE_R_I|HPTE_R_G)) | HPTE_R_M; } - /* Set the HPTE to point to pfn */ - r = (r & ~(HPTE_R_PP0 - pte_size)) | (pfn << PAGE_SHIFT); + /* + * Set the HPTE to point to pfn. + * Since the pfn is at PAGE_SIZE granularity, make sure we + * don't mask out lower-order bits if psize < PAGE_SIZE. + */ + if (psize < PAGE_SIZE) + psize = PAGE_SIZE; + r = (r & ~(HPTE_R_PP0 - psize)) | ((pfn << PAGE_SHIFT) & ~(psize - 1)); if (hpte_is_writable(r) && !write_ok) r = hpte_make_readonly(r); ret = RESUME_GUEST; diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c index 9c515440ad1a..fddbf989f37e 100644 --- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c +++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c @@ -225,6 +225,7 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags, is_io = pa & (HPTE_R_I | HPTE_R_W); pte_size = PAGE_SIZE << (pa & KVMPPC_PAGE_ORDER_MASK); pa &= PAGE_MASK; + pa |= gpa & ~PAGE_MASK; } else { /* Translate to host virtual address */ hva = __gfn_to_hva_memslot(memslot, gfn); @@ -238,13 +239,12 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags, ptel = hpte_make_readonly(ptel); is_io = hpte_cache_bits(pte_val(pte)); pa = pte_pfn(pte) << PAGE_SHIFT; + pa |= hva & (pte_size - 1); } } if (pte_size < psize) return H_PARAMETER; - if (pa && pte_size > psize) - pa |= gpa & (pte_size - 1); ptel &= ~(HPTE_R_PP0 - psize); ptel |= pa; -- cgit v1.2.3 From f019b7ad76e6bdbc8462cbe17ad5b86a25fcdf24 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sat, 16 Nov 2013 17:46:03 +1100 Subject: KVM: PPC: Book3S HV: Refine barriers in guest entry/exit Some users have reported instances of the host hanging with secondary threads of a core waiting for the primary thread to exit the guest, and the primary thread stuck in nap mode. This prompted a review of the memory barriers in the guest entry/exit code, and this is the result. Most of these changes are the suggestions of Dean Burdick . The barriers between updating napping_threads and reading the entry_exit_count on the one hand, and updating entry_exit_count and reading napping_threads on the other, need to be isync not lwsync, since we need to ensure that either the napping_threads update or the entry_exit_count update get seen. It is not sufficient to order the load vs. lwarx, as lwsync does; we need to order the load vs. the stwcx., so we need isync. In addition, we need a full sync before sending IPIs to wake other threads from nap, to ensure that the write to the entry_exit_count is visible before the IPI occurs. Signed-off-by: Paul Mackerras Signed-off-by: Alexander Graf --- arch/powerpc/kvm/book3s_hv_rmhandlers.S | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S index bc8de75b1925..bde28da69610 100644 --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S @@ -153,7 +153,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_206) 13: b machine_check_fwnmi - /* * We come in here when wakened from nap mode on a secondary hw thread. * Relocation is off and most register values are lost. @@ -224,6 +223,11 @@ kvm_start_guest: /* Clear our vcpu pointer so we don't come back in early */ li r0, 0 std r0, HSTATE_KVM_VCPU(r13) + /* + * Make sure we clear HSTATE_KVM_VCPU(r13) before incrementing + * the nap_count, because once the increment to nap_count is + * visible we could be given another vcpu. + */ lwsync /* Clear any pending IPI - we're an offline thread */ ld r5, HSTATE_XICS_PHYS(r13) @@ -241,7 +245,6 @@ kvm_start_guest: /* increment the nap count and then go to nap mode */ ld r4, HSTATE_KVM_VCORE(r13) addi r4, r4, VCORE_NAP_COUNT - lwsync /* make previous updates visible */ 51: lwarx r3, 0, r4 addi r3, r3, 1 stwcx. r3, 0, r4 @@ -990,14 +993,13 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201) */ /* Increment the threads-exiting-guest count in the 0xff00 bits of vcore->entry_exit_count */ - lwsync ld r5,HSTATE_KVM_VCORE(r13) addi r6,r5,VCORE_ENTRY_EXIT 41: lwarx r3,0,r6 addi r0,r3,0x100 stwcx. r0,0,r6 bne 41b - lwsync + isync /* order stwcx. vs. reading napping_threads */ /* * At this point we have an interrupt that we have to pass @@ -1030,6 +1032,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_201) sld r0,r0,r4 andc. r3,r3,r0 /* no sense IPI'ing ourselves */ beq 43f + /* Order entry/exit update vs. IPIs */ + sync mulli r4,r4,PACA_SIZE /* get paca for thread 0 */ subf r6,r4,r13 42: andi. r0,r3,1 @@ -1638,10 +1642,10 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_206) bge kvm_cede_exit stwcx. r4,0,r6 bne 31b + /* order napping_threads update vs testing entry_exit_count */ + isync li r0,1 stb r0,HSTATE_NAPPING(r13) - /* order napping_threads update vs testing entry_exit_count */ - lwsync mr r4,r3 lwz r7,VCORE_ENTRY_EXIT(r5) cmpwi r7,0x100 -- cgit v1.2.3 From bf3d32e1156c36c88b75960fd2e5457d5d75620b Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sat, 16 Nov 2013 17:46:04 +1100 Subject: KVM: PPC: Book3S HV: Make tbacct_lock irq-safe Lockdep reported that there is a potential for deadlock because vcpu->arch.tbacct_lock is not irq-safe, and is sometimes taken inside the rq_lock (run-queue lock) in the scheduler, which is taken within interrupts. The lockdep splat looks like: ====================================================== [ INFO: HARDIRQ-safe -> HARDIRQ-unsafe lock order detected ] 3.12.0-rc5-kvm+ #8 Not tainted ------------------------------------------------------ qemu-system-ppc/4803 [HC0[0]:SC0[0]:HE0:SE1] is trying to acquire: (&(&vcpu->arch.tbacct_lock)->rlock){+.+...}, at: [] .kvmppc_core_vcpu_put_hv+0x2c/0xa0 and this task is already holding: (&rq->lock){-.-.-.}, at: [] .__schedule+0x180/0xaa0 which would create a new lock dependency: (&rq->lock){-.-.-.} -> (&(&vcpu->arch.tbacct_lock)->rlock){+.+...} but this new dependency connects a HARDIRQ-irq-safe lock: (&rq->lock){-.-.-.} ... which became HARDIRQ-irq-safe at: [] .lock_acquire+0xbc/0x190 [] ._raw_spin_lock+0x34/0x60 [] .scheduler_tick+0x54/0x180 [] .update_process_times+0x70/0xa0 [] .tick_periodic+0x3c/0xe0 [] .tick_handle_periodic+0x28/0xb0 [] .timer_interrupt+0x120/0x2e0 [] decrementer_common+0x168/0x180 [] .get_page_from_freelist+0x924/0xc10 [] .__alloc_pages_nodemask+0x200/0xba0 [] .alloc_pages_exact_nid+0x68/0x110 [] .page_cgroup_init+0x1e0/0x270 [] .start_kernel+0x3e0/0x4e4 [] .start_here_common+0x20/0x70 to a HARDIRQ-irq-unsafe lock: (&(&vcpu->arch.tbacct_lock)->rlock){+.+...} ... which became HARDIRQ-irq-unsafe at: ... [] .lock_acquire+0xbc/0x190 [] ._raw_spin_lock+0x34/0x60 [] .kvmppc_core_vcpu_load_hv+0x2c/0x100 [] .kvmppc_core_vcpu_load+0x2c/0x40 [] .kvm_arch_vcpu_load+0x10/0x30 [] .vcpu_load+0x64/0xd0 [] .kvm_vcpu_ioctl+0x68/0x730 [] .do_vfs_ioctl+0x4dc/0x7a0 [] .SyS_ioctl+0xc4/0xe0 [] syscall_exit+0x0/0x98 Some users have reported this deadlock occurring in practice, though the reports have been primarily on 3.10.x-based kernels. This fixes the problem by making tbacct_lock be irq-safe. Signed-off-by: Paul Mackerras Signed-off-by: Alexander Graf --- arch/powerpc/kvm/book3s_hv.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c index 072287f1c3bc..31d9cfb64a23 100644 --- a/arch/powerpc/kvm/book3s_hv.c +++ b/arch/powerpc/kvm/book3s_hv.c @@ -131,8 +131,9 @@ static void kvmppc_fast_vcpu_kick_hv(struct kvm_vcpu *vcpu) static void kvmppc_core_vcpu_load_hv(struct kvm_vcpu *vcpu, int cpu) { struct kvmppc_vcore *vc = vcpu->arch.vcore; + unsigned long flags; - spin_lock(&vcpu->arch.tbacct_lock); + spin_lock_irqsave(&vcpu->arch.tbacct_lock, flags); if (vc->runner == vcpu && vc->vcore_state != VCORE_INACTIVE && vc->preempt_tb != TB_NIL) { vc->stolen_tb += mftb() - vc->preempt_tb; @@ -143,19 +144,20 @@ static void kvmppc_core_vcpu_load_hv(struct kvm_vcpu *vcpu, int cpu) vcpu->arch.busy_stolen += mftb() - vcpu->arch.busy_preempt; vcpu->arch.busy_preempt = TB_NIL; } - spin_unlock(&vcpu->arch.tbacct_lock); + spin_unlock_irqrestore(&vcpu->arch.tbacct_lock, flags); } static void kvmppc_core_vcpu_put_hv(struct kvm_vcpu *vcpu) { struct kvmppc_vcore *vc = vcpu->arch.vcore; + unsigned long flags; - spin_lock(&vcpu->arch.tbacct_lock); + spin_lock_irqsave(&vcpu->arch.tbacct_lock, flags); if (vc->runner == vcpu && vc->vcore_state != VCORE_INACTIVE) vc->preempt_tb = mftb(); if (vcpu->arch.state == KVMPPC_VCPU_BUSY_IN_HOST) vcpu->arch.busy_preempt = mftb(); - spin_unlock(&vcpu->arch.tbacct_lock); + spin_unlock_irqrestore(&vcpu->arch.tbacct_lock, flags); } static void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr) @@ -486,11 +488,11 @@ static u64 vcore_stolen_time(struct kvmppc_vcore *vc, u64 now) */ if (vc->vcore_state != VCORE_INACTIVE && vc->runner->arch.run_task != current) { - spin_lock(&vc->runner->arch.tbacct_lock); + spin_lock_irq(&vc->runner->arch.tbacct_lock); p = vc->stolen_tb; if (vc->preempt_tb != TB_NIL) p += now - vc->preempt_tb; - spin_unlock(&vc->runner->arch.tbacct_lock); + spin_unlock_irq(&vc->runner->arch.tbacct_lock); } else { p = vc->stolen_tb; } @@ -512,10 +514,10 @@ static void kvmppc_create_dtl_entry(struct kvm_vcpu *vcpu, core_stolen = vcore_stolen_time(vc, now); stolen = core_stolen - vcpu->arch.stolen_logged; vcpu->arch.stolen_logged = core_stolen; - spin_lock(&vcpu->arch.tbacct_lock); + spin_lock_irq(&vcpu->arch.tbacct_lock); stolen += vcpu->arch.busy_stolen; vcpu->arch.busy_stolen = 0; - spin_unlock(&vcpu->arch.tbacct_lock); + spin_unlock_irq(&vcpu->arch.tbacct_lock); if (!dt || !vpa) return; memset(dt, 0, sizeof(struct dtl_entry)); @@ -1115,13 +1117,13 @@ static void kvmppc_remove_runnable(struct kvmppc_vcore *vc, if (vcpu->arch.state != KVMPPC_VCPU_RUNNABLE) return; - spin_lock(&vcpu->arch.tbacct_lock); + spin_lock_irq(&vcpu->arch.tbacct_lock); now = mftb(); vcpu->arch.busy_stolen += vcore_stolen_time(vc, now) - vcpu->arch.stolen_logged; vcpu->arch.busy_preempt = now; vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST; - spin_unlock(&vcpu->arch.tbacct_lock); + spin_unlock_irq(&vcpu->arch.tbacct_lock); --vc->n_runnable; list_del(&vcpu->arch.run_list); } -- cgit v1.2.3 From c9438092cae4a5bdbd146ca1385e85dcd6e847f8 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sat, 16 Nov 2013 17:46:05 +1100 Subject: KVM: PPC: Book3S HV: Take SRCU read lock around kvm_read_guest() call Running a kernel with CONFIG_PROVE_RCU=y yields the following diagnostic: =============================== [ INFO: suspicious RCU usage. ] 3.12.0-rc5-kvm+ #9 Not tainted ------------------------------- include/linux/kvm_host.h:473 suspicious rcu_dereference_check() usage! other info that might help us debug this: rcu_scheduler_active = 1, debug_locks = 0 1 lock held by qemu-system-ppc/4831: stack backtrace: CPU: 28 PID: 4831 Comm: qemu-system-ppc Not tainted 3.12.0-rc5-kvm+ #9 Call Trace: [c000000be462b2a0] [c00000000001644c] .show_stack+0x7c/0x1f0 (unreliable) [c000000be462b370] [c000000000ad57c0] .dump_stack+0x88/0xb4 [c000000be462b3f0] [c0000000001315e8] .lockdep_rcu_suspicious+0x138/0x180 [c000000be462b480] [c00000000007862c] .gfn_to_memslot+0x13c/0x170 [c000000be462b510] [c00000000007d384] .gfn_to_hva_prot+0x24/0x90 [c000000be462b5a0] [c00000000007d420] .kvm_read_guest_page+0x30/0xd0 [c000000be462b630] [c00000000007d528] .kvm_read_guest+0x68/0x110 [c000000be462b6e0] [c000000000084594] .kvmppc_rtas_hcall+0x34/0x180 [c000000be462b7d0] [c000000000097934] .kvmppc_pseries_do_hcall+0x74/0x830 [c000000be462b880] [c0000000000990e8] .kvmppc_vcpu_run_hv+0xff8/0x15a0 [c000000be462b9e0] [c0000000000839cc] .kvmppc_vcpu_run+0x2c/0x40 [c000000be462ba50] [c0000000000810b4] .kvm_arch_vcpu_ioctl_run+0x54/0x1b0 [c000000be462bae0] [c00000000007b508] .kvm_vcpu_ioctl+0x478/0x730 [c000000be462bca0] [c00000000025532c] .do_vfs_ioctl+0x4dc/0x7a0 [c000000be462bd80] [c0000000002556b4] .SyS_ioctl+0xc4/0xe0 [c000000be462be30] [c000000000009ee4] syscall_exit+0x0/0x98 To fix this, we take the SRCU read lock around the kvmppc_rtas_hcall() call. Signed-off-by: Paul Mackerras Signed-off-by: Alexander Graf --- arch/powerpc/kvm/book3s_hv.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch') diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c index 31d9cfb64a23..b51d5db78068 100644 --- a/arch/powerpc/kvm/book3s_hv.c +++ b/arch/powerpc/kvm/book3s_hv.c @@ -591,7 +591,9 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu) if (list_empty(&vcpu->kvm->arch.rtas_tokens)) return RESUME_HOST; + idx = srcu_read_lock(&vcpu->kvm->srcu); rc = kvmppc_rtas_hcall(vcpu); + srcu_read_unlock(&vcpu->kvm->srcu, idx); if (rc == -ENOENT) return RESUME_HOST; -- cgit v1.2.3 From 91648ec09c1ef69c4d840ab6dab391bfb452d554 Mon Sep 17 00:00:00 2001 From: pingfan liu Date: Fri, 15 Nov 2013 16:35:00 +0800 Subject: powerpc: kvm: fix rare but potential deadlock scene Since kvmppc_hv_find_lock_hpte() is called from both virtmode and realmode, so it can trigger the deadlock. Suppose the following scene: Two physical cpuM, cpuN, two VM instances A, B, each VM has a group of vcpus. If on cpuM, vcpu_A_1 holds bitlock X (HPTE_V_HVLOCK), then is switched out, and on cpuN, vcpu_A_2 try to lock X in realmode, then cpuN will be caught in realmode for a long time. What makes things even worse if the following happens, On cpuM, bitlockX is hold, on cpuN, Y is hold. vcpu_B_2 try to lock Y on cpuM in realmode vcpu_A_2 try to lock X on cpuN in realmode Oops! deadlock happens Signed-off-by: Liu Ping Fan Reviewed-by: Paul Mackerras CC: stable@vger.kernel.org Signed-off-by: Alexander Graf --- arch/powerpc/kvm/book3s_64_mmu_hv.c | 6 +++++- arch/powerpc/kvm/book3s_hv_rm_mmu.c | 4 ++++ 2 files changed, 9 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c index 47bbeaf2d320..c5d148434c08 100644 --- a/arch/powerpc/kvm/book3s_64_mmu_hv.c +++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c @@ -469,11 +469,14 @@ static int kvmppc_mmu_book3s_64_hv_xlate(struct kvm_vcpu *vcpu, gva_t eaddr, slb_v = vcpu->kvm->arch.vrma_slb_v; } + preempt_disable(); /* Find the HPTE in the hash table */ index = kvmppc_hv_find_lock_hpte(kvm, eaddr, slb_v, HPTE_V_VALID | HPTE_V_ABSENT); - if (index < 0) + if (index < 0) { + preempt_enable(); return -ENOENT; + } hptep = (unsigned long *)(kvm->arch.hpt_virt + (index << 4)); v = hptep[0] & ~HPTE_V_HVLOCK; gr = kvm->arch.revmap[index].guest_rpte; @@ -481,6 +484,7 @@ static int kvmppc_mmu_book3s_64_hv_xlate(struct kvm_vcpu *vcpu, gva_t eaddr, /* Unlock the HPTE */ asm volatile("lwsync" : : : "memory"); hptep[0] = v; + preempt_enable(); gpte->eaddr = eaddr; gpte->vpage = ((v & HPTE_V_AVPN) << 4) | ((eaddr >> 12) & 0xfff); diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c index fddbf989f37e..1931aa341a72 100644 --- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c +++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c @@ -749,6 +749,10 @@ static int slb_base_page_shift[4] = { 20, /* 1M, unsupported */ }; +/* When called from virtmode, this func should be protected by + * preempt_disable(), otherwise, the holding of HPTE_V_HVLOCK + * can trigger deadlock issue. + */ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v, unsigned long valid) { -- cgit v1.2.3 From c94cae53f9e564484f906a79be5639fc66e8cb02 Mon Sep 17 00:00:00 2001 From: Eric Trudeau Date: Wed, 4 Dec 2013 11:39:33 +0000 Subject: XEN: Grant table address, xen_hvm_resume_frames, is a phys_addr not a pfn From: Eric Trudeau xen_hvm_resume_frames stores the physical address of the grant table. englighten.c was incorrectly setting it as if it was a page frame number. This caused the table to be mapped into the guest at an unexpected physical address. Additionally, a warning is improved to include the grant table address which failed in xen_remap. Signed-off-by: Eric Trudeau Signed-off-by: Stefano Stabellini --- arch/arm/xen/enlighten.c | 4 ++-- drivers/xen/grant-table.c | 3 ++- 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'arch') diff --git a/arch/arm/xen/enlighten.c b/arch/arm/xen/enlighten.c index 83e4f959ee47..6a288c7be49f 100644 --- a/arch/arm/xen/enlighten.c +++ b/arch/arm/xen/enlighten.c @@ -224,10 +224,10 @@ static int __init xen_guest_init(void) } if (of_address_to_resource(node, GRANT_TABLE_PHYSADDR, &res)) return 0; - xen_hvm_resume_frames = res.start >> PAGE_SHIFT; + xen_hvm_resume_frames = res.start; xen_events_irq = irq_of_parse_and_map(node, 0); pr_info("Xen %s support found, events_irq=%d gnttab_frame_pfn=%lx\n", - version, xen_events_irq, xen_hvm_resume_frames); + version, xen_events_irq, (xen_hvm_resume_frames >> PAGE_SHIFT)); xen_domain_type = XEN_HVM_DOMAIN; xen_setup_features(); diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c index 028387192b60..aa846a48f400 100644 --- a/drivers/xen/grant-table.c +++ b/drivers/xen/grant-table.c @@ -1176,7 +1176,8 @@ static int gnttab_setup(void) gnttab_shared.addr = xen_remap(xen_hvm_resume_frames, PAGE_SIZE * max_nr_gframes); if (gnttab_shared.addr == NULL) { - pr_warn("Failed to ioremap gnttab share frames!\n"); + pr_warn("Failed to ioremap gnttab share frames (addr=0x%08lx)!\n", + xen_hvm_resume_frames); return -ENOMEM; } } -- cgit v1.2.3 From cf30d52e2d11523c42048ab89ed4215b5021526a Mon Sep 17 00:00:00 2001 From: Maria Dimakopoulou Date: Thu, 5 Dec 2013 01:24:37 +0200 Subject: perf/x86: Fix constraint table end marker bug The EVENT_CONSTRAINT_END() macro defines the end marker as a constraint with a weight of zero. This was all fine until we blacklisted the corrupting memory events on Intel IvyBridge. These events are blacklisted by using a counter bitmask of zero. Thus, they also get a constraint weight of zero. The iteration macro: for_each_constraint tests the weight==0. Therefore, it was stopping at the first blacklisted event, i.e., 0xd0. The corrupting events were therefore considered as unconstrained and were scheduled on any of the generic counters. This patch fixes the end marker to have a weight of -1. With this, the blacklisted events get an empty constraint and cannot be scheduled which is what we want for now. Signed-off-by: Maria Dimakopoulou Reviewed-by: Stephane Eranian Cc: peterz@infradead.org Cc: ak@linux.intel.com Cc: jolsa@redhat.com Cc: zheng.z.yan@intel.com Link: http://lkml.kernel.org/r/20131204232437.GA10689@starlight Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event.h | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'arch') diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h index fd00bb29425d..c1a861829d81 100644 --- a/arch/x86/kernel/cpu/perf_event.h +++ b/arch/x86/kernel/cpu/perf_event.h @@ -262,11 +262,20 @@ struct cpu_hw_events { __EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK, \ HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_ST_HSW) -#define EVENT_CONSTRAINT_END \ - EVENT_CONSTRAINT(0, 0, 0) +/* + * We define the end marker as having a weight of -1 + * to enable blacklisting of events using a counter bitmask + * of zero and thus a weight of zero. + * The end marker has a weight that cannot possibly be + * obtained from counting the bits in the bitmask. + */ +#define EVENT_CONSTRAINT_END { .weight = -1 } +/* + * Check for end marker with weight == -1 + */ #define for_each_event_constraint(e, c) \ - for ((e) = (c); (e)->weight; (e)++) + for ((e) = (c); (e)->weight != -1; (e)++) /* * Extra registers for specific events. -- cgit v1.2.3 From d825a04387ff4ce66117306f2862c7cedca5c597 Mon Sep 17 00:00:00 2001 From: Alexander Graf Date: Fri, 29 Nov 2013 02:24:18 +0100 Subject: KVM: PPC: Book3S: PR: Don't clobber our exit handler id We call a C helper to save all svcpu fields into our vcpu. The C ABI states that r12 is considered volatile. However, we keep our exit handler id in r12 currently. So we need to save it away into a non-volatile register instead that definitely does get preserved across the C call. This bug usually didn't hit anyone yet since gcc is smart enough to generate code that doesn't even need r12 which means it stayed identical throughout the call by sheer luck. But we can't rely on that. Signed-off-by: Alexander Graf --- arch/powerpc/kvm/book3s_interrupts.S | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/kvm/book3s_interrupts.S b/arch/powerpc/kvm/book3s_interrupts.S index f4dd041c14ea..5e7cb32ce4dc 100644 --- a/arch/powerpc/kvm/book3s_interrupts.S +++ b/arch/powerpc/kvm/book3s_interrupts.S @@ -132,9 +132,17 @@ kvm_start_lightweight: * */ + PPC_LL r3, GPR4(r1) /* vcpu pointer */ + + /* + * kvmppc_copy_from_svcpu can clobber volatile registers, save + * the exit handler id to the vcpu and restore it from there later. + */ + stw r12, VCPU_TRAP(r3) + /* Transfer reg values from shadow vcpu back to vcpu struct */ /* On 64-bit, interrupts are still off at this point */ - PPC_LL r3, GPR4(r1) /* vcpu pointer */ + GET_SHADOW_VCPU(r4) bl FUNC(kvmppc_copy_from_svcpu) nop @@ -151,7 +159,6 @@ kvm_start_lightweight: */ ld r3, PACA_SPRG3(r13) mtspr SPRN_SPRG3, r3 - #endif /* CONFIG_PPC_BOOK3S_64 */ /* R7 = vcpu */ @@ -177,7 +184,7 @@ kvm_start_lightweight: PPC_STL r31, VCPU_GPR(R31)(r7) /* Pass the exit number as 3rd argument to kvmppc_handle_exit */ - mr r5, r12 + lwz r5, VCPU_TRAP(r7) /* Restore r3 (kvm_run) and r4 (vcpu) */ REST_2GPRS(3, r1) -- cgit v1.2.3 From c9dad7f9db4ed42de37d3f0ef2b2c0e10d5b6f92 Mon Sep 17 00:00:00 2001 From: Alexander Graf Date: Fri, 29 Nov 2013 02:27:23 +0100 Subject: KVM: PPC: Book3S: PR: Export kvmppc_copy_to|from_svcpu The kvmppc_copy_{to,from}_svcpu functions are publically visible, so we should also export them in a header for others C files to consume. So far we didn't need this because we only called it from asm code. The next patch will introduce a C caller. Signed-off-by: Alexander Graf --- arch/powerpc/include/asm/kvm_book3s.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch') diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h index 4a594b76674d..bc23b1ba7980 100644 --- a/arch/powerpc/include/asm/kvm_book3s.h +++ b/arch/powerpc/include/asm/kvm_book3s.h @@ -192,6 +192,10 @@ extern void kvmppc_load_up_vsx(void); extern u32 kvmppc_alignment_dsisr(struct kvm_vcpu *vcpu, unsigned int inst); extern ulong kvmppc_alignment_dar(struct kvm_vcpu *vcpu, unsigned int inst); extern int kvmppc_h_pr(struct kvm_vcpu *vcpu, unsigned long cmd); +extern void kvmppc_copy_to_svcpu(struct kvmppc_book3s_shadow_vcpu *svcpu, + struct kvm_vcpu *vcpu); +extern void kvmppc_copy_from_svcpu(struct kvm_vcpu *vcpu, + struct kvmppc_book3s_shadow_vcpu *svcpu); static inline struct kvmppc_vcpu_book3s *to_book3s(struct kvm_vcpu *vcpu) { -- cgit v1.2.3 From 40fdd8c88c4a5e9b26bfbed2215ac661f24aef07 Mon Sep 17 00:00:00 2001 From: Alexander Graf Date: Fri, 29 Nov 2013 02:29:00 +0100 Subject: KVM: PPC: Book3S: PR: Make svcpu -> vcpu store preempt savvy As soon as we get back to our "highmem" handler in virtual address space we may get preempted. Today the reason we can get preempted is that we replay interrupts and all the lazy logic thinks we have interrupts enabled. However, it's not hard to make the code interruptible and that way we can enable and handle interrupts even earlier. This fixes random guest crashes that happened with CONFIG_PREEMPT=y for me. Signed-off-by: Alexander Graf --- arch/powerpc/include/asm/kvm_book3s_asm.h | 1 + arch/powerpc/kvm/book3s_pr.c | 22 ++++++++++++++++++++++ 2 files changed, 23 insertions(+) (limited to 'arch') diff --git a/arch/powerpc/include/asm/kvm_book3s_asm.h b/arch/powerpc/include/asm/kvm_book3s_asm.h index 0bd9348a4db9..412b2f389474 100644 --- a/arch/powerpc/include/asm/kvm_book3s_asm.h +++ b/arch/powerpc/include/asm/kvm_book3s_asm.h @@ -106,6 +106,7 @@ struct kvmppc_host_state { }; struct kvmppc_book3s_shadow_vcpu { + bool in_use; ulong gpr[14]; u32 cr; u32 xer; diff --git a/arch/powerpc/kvm/book3s_pr.c b/arch/powerpc/kvm/book3s_pr.c index fe14ca3dd171..5b9e9063cfaf 100644 --- a/arch/powerpc/kvm/book3s_pr.c +++ b/arch/powerpc/kvm/book3s_pr.c @@ -66,6 +66,7 @@ static void kvmppc_core_vcpu_load_pr(struct kvm_vcpu *vcpu, int cpu) struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu); memcpy(svcpu->slb, to_book3s(vcpu)->slb_shadow, sizeof(svcpu->slb)); svcpu->slb_max = to_book3s(vcpu)->slb_shadow_max; + svcpu->in_use = 0; svcpu_put(svcpu); #endif vcpu->cpu = smp_processor_id(); @@ -78,6 +79,9 @@ static void kvmppc_core_vcpu_put_pr(struct kvm_vcpu *vcpu) { #ifdef CONFIG_PPC_BOOK3S_64 struct kvmppc_book3s_shadow_vcpu *svcpu = svcpu_get(vcpu); + if (svcpu->in_use) { + kvmppc_copy_from_svcpu(vcpu, svcpu); + } memcpy(to_book3s(vcpu)->slb_shadow, svcpu->slb, sizeof(svcpu->slb)); to_book3s(vcpu)->slb_shadow_max = svcpu->slb_max; svcpu_put(svcpu); @@ -110,12 +114,26 @@ void kvmppc_copy_to_svcpu(struct kvmppc_book3s_shadow_vcpu *svcpu, svcpu->ctr = vcpu->arch.ctr; svcpu->lr = vcpu->arch.lr; svcpu->pc = vcpu->arch.pc; + svcpu->in_use = true; } /* Copy data touched by real-mode code from shadow vcpu back to vcpu */ void kvmppc_copy_from_svcpu(struct kvm_vcpu *vcpu, struct kvmppc_book3s_shadow_vcpu *svcpu) { + /* + * vcpu_put would just call us again because in_use hasn't + * been updated yet. + */ + preempt_disable(); + + /* + * Maybe we were already preempted and synced the svcpu from + * our preempt notifiers. Don't bother touching this svcpu then. + */ + if (!svcpu->in_use) + goto out; + vcpu->arch.gpr[0] = svcpu->gpr[0]; vcpu->arch.gpr[1] = svcpu->gpr[1]; vcpu->arch.gpr[2] = svcpu->gpr[2]; @@ -139,6 +157,10 @@ void kvmppc_copy_from_svcpu(struct kvm_vcpu *vcpu, vcpu->arch.fault_dar = svcpu->fault_dar; vcpu->arch.fault_dsisr = svcpu->fault_dsisr; vcpu->arch.last_inst = svcpu->last_inst; + svcpu->in_use = false; + +out: + preempt_enable(); } static int kvmppc_core_check_requests_pr(struct kvm_vcpu *vcpu) -- cgit v1.2.3 From 3d3319b45eea26df56c53aae1a65adf74c8ab12a Mon Sep 17 00:00:00 2001 From: Alexander Graf Date: Fri, 29 Nov 2013 02:32:31 +0100 Subject: KVM: PPC: Book3S: PR: Enable interrupts earlier Now that the svcpu sync is interrupt aware we can enable interrupts earlier in the exit code path again, moving 32bit and 64bit closer together. While at it, document the fact that we're always executing the exit path with interrupts enabled so that the next person doesn't trap over this. Signed-off-by: Alexander Graf --- arch/powerpc/kvm/book3s_interrupts.S | 6 +----- arch/powerpc/kvm/book3s_rmhandlers.S | 6 +----- 2 files changed, 2 insertions(+), 10 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/kvm/book3s_interrupts.S b/arch/powerpc/kvm/book3s_interrupts.S index 5e7cb32ce4dc..f779450cb07c 100644 --- a/arch/powerpc/kvm/book3s_interrupts.S +++ b/arch/powerpc/kvm/book3s_interrupts.S @@ -129,6 +129,7 @@ kvm_start_lightweight: * R12 = exit handler id * R13 = PACA * SVCPU.* = guest * + * MSR.EE = 1 * */ @@ -148,11 +149,6 @@ kvm_start_lightweight: nop #ifdef CONFIG_PPC_BOOK3S_64 - /* Re-enable interrupts */ - ld r3, HSTATE_HOST_MSR(r13) - ori r3, r3, MSR_EE - MTMSR_EERI(r3) - /* * Reload kernel SPRG3 value. * No need to save guest value as usermode can't modify SPRG3. diff --git a/arch/powerpc/kvm/book3s_rmhandlers.S b/arch/powerpc/kvm/book3s_rmhandlers.S index a38c4c9edab8..c3c5231adade 100644 --- a/arch/powerpc/kvm/book3s_rmhandlers.S +++ b/arch/powerpc/kvm/book3s_rmhandlers.S @@ -153,15 +153,11 @@ _GLOBAL(kvmppc_entry_trampoline) li r6, MSR_IR | MSR_DR andc r6, r5, r6 /* Clear DR and IR in MSR value */ -#ifdef CONFIG_PPC_BOOK3S_32 /* * Set EE in HOST_MSR so that it's enabled when we get into our - * C exit handler function. On 64-bit we delay enabling - * interrupts until we have finished transferring stuff - * to or from the PACA. + * C exit handler function. */ ori r5, r5, MSR_EE -#endif mtsrr0 r7 mtsrr1 r6 RFI -- cgit v1.2.3 From f5f972102d5c12729f0a35fce266b580aaa03f66 Mon Sep 17 00:00:00 2001 From: Scott Wood Date: Fri, 22 Nov 2013 15:52:29 -0600 Subject: powerpc/kvm/booke: Fix build break due to stack frame size warning Commit ce11e48b7fdd256ec68b932a89b397a790566031 ("KVM: PPC: E500: Add userspace debug stub support") added "struct thread_struct" to the stack of kvmppc_vcpu_run(). thread_struct is 1152 bytes on my build, compared to 48 bytes for the recently-introduced "struct debug_reg". Use the latter instead. This fixes the following error: cc1: warnings being treated as errors arch/powerpc/kvm/booke.c: In function 'kvmppc_vcpu_run': arch/powerpc/kvm/booke.c:760:1: error: the frame size of 1424 bytes is larger than 1024 bytes make[2]: *** [arch/powerpc/kvm/booke.o] Error 1 make[1]: *** [arch/powerpc/kvm] Error 2 make[1]: *** Waiting for unfinished jobs.... Signed-off-by: Scott Wood Signed-off-by: Alexander Graf --- arch/powerpc/include/asm/switch_to.h | 2 +- arch/powerpc/kernel/process.c | 32 ++++++++++++++++---------------- arch/powerpc/kvm/booke.c | 12 ++++++------ 3 files changed, 23 insertions(+), 23 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/include/asm/switch_to.h b/arch/powerpc/include/asm/switch_to.h index 9ee12610af02..aace90547614 100644 --- a/arch/powerpc/include/asm/switch_to.h +++ b/arch/powerpc/include/asm/switch_to.h @@ -35,7 +35,7 @@ extern void giveup_vsx(struct task_struct *); extern void enable_kernel_spe(void); extern void giveup_spe(struct task_struct *); extern void load_up_spe(struct task_struct *); -extern void switch_booke_debug_regs(struct thread_struct *new_thread); +extern void switch_booke_debug_regs(struct debug_reg *new_debug); #ifndef CONFIG_SMP extern void discard_lazy_cpu_state(void); diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 75c2d1009985..83530af028b8 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -339,7 +339,7 @@ static void set_debug_reg_defaults(struct thread_struct *thread) #endif } -static void prime_debug_regs(struct thread_struct *thread) +static void prime_debug_regs(struct debug_reg *debug) { /* * We could have inherited MSR_DE from userspace, since @@ -348,22 +348,22 @@ static void prime_debug_regs(struct thread_struct *thread) */ mtmsr(mfmsr() & ~MSR_DE); - mtspr(SPRN_IAC1, thread->debug.iac1); - mtspr(SPRN_IAC2, thread->debug.iac2); + mtspr(SPRN_IAC1, debug->iac1); + mtspr(SPRN_IAC2, debug->iac2); #if CONFIG_PPC_ADV_DEBUG_IACS > 2 - mtspr(SPRN_IAC3, thread->debug.iac3); - mtspr(SPRN_IAC4, thread->debug.iac4); + mtspr(SPRN_IAC3, debug->iac3); + mtspr(SPRN_IAC4, debug->iac4); #endif - mtspr(SPRN_DAC1, thread->debug.dac1); - mtspr(SPRN_DAC2, thread->debug.dac2); + mtspr(SPRN_DAC1, debug->dac1); + mtspr(SPRN_DAC2, debug->dac2); #if CONFIG_PPC_ADV_DEBUG_DVCS > 0 - mtspr(SPRN_DVC1, thread->debug.dvc1); - mtspr(SPRN_DVC2, thread->debug.dvc2); + mtspr(SPRN_DVC1, debug->dvc1); + mtspr(SPRN_DVC2, debug->dvc2); #endif - mtspr(SPRN_DBCR0, thread->debug.dbcr0); - mtspr(SPRN_DBCR1, thread->debug.dbcr1); + mtspr(SPRN_DBCR0, debug->dbcr0); + mtspr(SPRN_DBCR1, debug->dbcr1); #ifdef CONFIG_BOOKE - mtspr(SPRN_DBCR2, thread->debug.dbcr2); + mtspr(SPRN_DBCR2, debug->dbcr2); #endif } /* @@ -371,11 +371,11 @@ static void prime_debug_regs(struct thread_struct *thread) * debug registers, set the debug registers from the values * stored in the new thread. */ -void switch_booke_debug_regs(struct thread_struct *new_thread) +void switch_booke_debug_regs(struct debug_reg *new_debug) { if ((current->thread.debug.dbcr0 & DBCR0_IDM) - || (new_thread->debug.dbcr0 & DBCR0_IDM)) - prime_debug_regs(new_thread); + || (new_debug->dbcr0 & DBCR0_IDM)) + prime_debug_regs(new_debug); } EXPORT_SYMBOL_GPL(switch_booke_debug_regs); #else /* !CONFIG_PPC_ADV_DEBUG_REGS */ @@ -683,7 +683,7 @@ struct task_struct *__switch_to(struct task_struct *prev, #endif /* CONFIG_SMP */ #ifdef CONFIG_PPC_ADV_DEBUG_REGS - switch_booke_debug_regs(&new->thread); + switch_booke_debug_regs(&new->thread.debug); #else /* * For PPC_BOOK3S_64, we use the hw-breakpoint interfaces that would diff --git a/arch/powerpc/kvm/booke.c b/arch/powerpc/kvm/booke.c index 53e65a210b9a..0591e05db74b 100644 --- a/arch/powerpc/kvm/booke.c +++ b/arch/powerpc/kvm/booke.c @@ -681,7 +681,7 @@ int kvmppc_core_check_requests(struct kvm_vcpu *vcpu) int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) { int ret, s; - struct thread_struct thread; + struct debug_reg debug; #ifdef CONFIG_PPC_FPU struct thread_fp_state fp; int fpexc_mode; @@ -723,9 +723,9 @@ int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) #endif /* Switch to guest debug context */ - thread.debug = vcpu->arch.shadow_dbg_reg; - switch_booke_debug_regs(&thread); - thread.debug = current->thread.debug; + debug = vcpu->arch.shadow_dbg_reg; + switch_booke_debug_regs(&debug); + debug = current->thread.debug; current->thread.debug = vcpu->arch.shadow_dbg_reg; kvmppc_fix_ee_before_entry(); @@ -736,8 +736,8 @@ int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) We also get here with interrupts enabled. */ /* Switch back to user space debug context */ - switch_booke_debug_regs(&thread); - current->thread.debug = thread.debug; + switch_booke_debug_regs(&debug); + current->thread.debug = debug; #ifdef CONFIG_PPC_FPU kvmppc_save_guest_fp(vcpu); -- cgit v1.2.3 From ba1f14fbe70965ae0fb1655a5275a62723f65b77 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 28 Nov 2013 14:26:41 +0100 Subject: sched: Remove PREEMPT_NEED_RESCHED from generic code While hunting a preemption issue with Alexander, Ben noticed that the currently generic PREEMPT_NEED_RESCHED stuff is horribly broken for load-store architectures. We currently rely on the IPI to fold TIF_NEED_RESCHED into PREEMPT_NEED_RESCHED, but when this IPI lands while we already have a load for the preempt-count but before the store, the store will erase the PREEMPT_NEED_RESCHED change. The current preempt-count only works on load-store archs because interrupts are assumed to be completely balanced wrt their preempt_count fiddling; the previous preempt_count load will match the preempt_count state after the interrupt and therefore nothing gets lost. This patch removes the PREEMPT_NEED_RESCHED usage from generic code and pushes it into x86 arch code; the generic code goes back to relying on TIF_NEED_RESCHED. Boot tested on x86_64 and compile tested on ppc64. Reported-by: Benjamin Herrenschmidt Reported-and-Tested-by: Alexander Graf Signed-off-by: Peter Zijlstra Cc: Linus Torvalds Link: http://lkml.kernel.org/r/20131128132641.GP10022@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- arch/x86/include/asm/preempt.h | 11 +++++++++++ include/asm-generic/preempt.h | 35 +++++++++++------------------------ include/linux/sched.h | 2 -- 3 files changed, 22 insertions(+), 26 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/preempt.h b/arch/x86/include/asm/preempt.h index 8729723636fd..c8b051933b1b 100644 --- a/arch/x86/include/asm/preempt.h +++ b/arch/x86/include/asm/preempt.h @@ -7,6 +7,12 @@ DECLARE_PER_CPU(int, __preempt_count); +/* + * We use the PREEMPT_NEED_RESCHED bit as an inverted NEED_RESCHED such + * that a decrement hitting 0 means we can and should reschedule. + */ +#define PREEMPT_ENABLED (0 + PREEMPT_NEED_RESCHED) + /* * We mask the PREEMPT_NEED_RESCHED bit so as not to confuse all current users * that think a non-zero value indicates we cannot preempt. @@ -74,6 +80,11 @@ static __always_inline void __preempt_count_sub(int val) __this_cpu_add_4(__preempt_count, -val); } +/* + * Because we keep PREEMPT_NEED_RESCHED set when we do _not_ need to reschedule + * a decrement which hits zero means we have no preempt_count and should + * reschedule. + */ static __always_inline bool __preempt_count_dec_and_test(void) { GEN_UNARY_RMWcc("decl", __preempt_count, __percpu_arg(0), "e"); diff --git a/include/asm-generic/preempt.h b/include/asm-generic/preempt.h index ddf2b420ac8f..1cd3f5d767a8 100644 --- a/include/asm-generic/preempt.h +++ b/include/asm-generic/preempt.h @@ -3,13 +3,11 @@ #include -/* - * We mask the PREEMPT_NEED_RESCHED bit so as not to confuse all current users - * that think a non-zero value indicates we cannot preempt. - */ +#define PREEMPT_ENABLED (0) + static __always_inline int preempt_count(void) { - return current_thread_info()->preempt_count & ~PREEMPT_NEED_RESCHED; + return current_thread_info()->preempt_count; } static __always_inline int *preempt_count_ptr(void) @@ -17,11 +15,6 @@ static __always_inline int *preempt_count_ptr(void) return ¤t_thread_info()->preempt_count; } -/* - * We now loose PREEMPT_NEED_RESCHED and cause an extra reschedule; however the - * alternative is loosing a reschedule. Better schedule too often -- also this - * should be a very rare operation. - */ static __always_inline void preempt_count_set(int pc) { *preempt_count_ptr() = pc; @@ -41,28 +34,17 @@ static __always_inline void preempt_count_set(int pc) task_thread_info(p)->preempt_count = PREEMPT_ENABLED; \ } while (0) -/* - * We fold the NEED_RESCHED bit into the preempt count such that - * preempt_enable() can decrement and test for needing to reschedule with a - * single instruction. - * - * We invert the actual bit, so that when the decrement hits 0 we know we both - * need to resched (the bit is cleared) and can resched (no preempt count). - */ - static __always_inline void set_preempt_need_resched(void) { - *preempt_count_ptr() &= ~PREEMPT_NEED_RESCHED; } static __always_inline void clear_preempt_need_resched(void) { - *preempt_count_ptr() |= PREEMPT_NEED_RESCHED; } static __always_inline bool test_preempt_need_resched(void) { - return !(*preempt_count_ptr() & PREEMPT_NEED_RESCHED); + return false; } /* @@ -81,7 +63,12 @@ static __always_inline void __preempt_count_sub(int val) static __always_inline bool __preempt_count_dec_and_test(void) { - return !--*preempt_count_ptr(); + /* + * Because of load-store architectures cannot do per-cpu atomic + * operations; we cannot use PREEMPT_NEED_RESCHED because it might get + * lost. + */ + return !--*preempt_count_ptr() && tif_need_resched(); } /* @@ -89,7 +76,7 @@ static __always_inline bool __preempt_count_dec_and_test(void) */ static __always_inline bool should_resched(void) { - return unlikely(!*preempt_count_ptr()); + return unlikely(!preempt_count() && tif_need_resched()); } #ifdef CONFIG_PREEMPT diff --git a/include/linux/sched.h b/include/linux/sched.h index 768b037dfacb..96d674ba3876 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -440,8 +440,6 @@ struct task_cputime { .sum_exec_runtime = 0, \ } -#define PREEMPT_ENABLED (PREEMPT_NEED_RESCHED) - #ifdef CONFIG_PREEMPT_COUNT #define PREEMPT_DISABLED (1 + PREEMPT_ENABLED) #else -- cgit v1.2.3 From be5e610c0fd6ef772cafb9e0bd4128134804aef3 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 18 Nov 2013 18:27:06 +0100 Subject: math64: Add mul_u64_u32_shr() Introduce mul_u64_u32_shr() as proposed by Andy a while back; it allows using 64x64->128 muls on 64bit archs and recent GCC which defines __SIZEOF_INT128__ and __int128. (This new method will be used by the scheduler.) Signed-off-by: Peter Zijlstra Cc: fweisbec@gmail.com Cc: Andy Lutomirski Cc: Linus Torvalds Link: http://lkml.kernel.org/n/tip-hxjoeuzmrcaumR0uZwjpe2pv@git.kernel.org Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 1 + include/linux/math64.h | 30 ++++++++++++++++++++++++++++++ init/Kconfig | 6 ++++++ 3 files changed, 37 insertions(+) (limited to 'arch') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index e903c71f7e69..0952ecd60eca 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -26,6 +26,7 @@ config X86 select HAVE_AOUT if X86_32 select HAVE_UNSTABLE_SCHED_CLOCK select ARCH_SUPPORTS_NUMA_BALANCING + select ARCH_SUPPORTS_INT128 if X86_64 select ARCH_WANTS_PROT_NUMA_PROT_NONE select HAVE_IDE select HAVE_OPROFILE diff --git a/include/linux/math64.h b/include/linux/math64.h index 69ed5f5e9f6e..c45c089bfdac 100644 --- a/include/linux/math64.h +++ b/include/linux/math64.h @@ -133,4 +133,34 @@ __iter_div_u64_rem(u64 dividend, u32 divisor, u64 *remainder) return ret; } +#if defined(CONFIG_ARCH_SUPPORTS_INT128) && defined(__SIZEOF_INT128__) + +#ifndef mul_u64_u32_shr +static inline u64 mul_u64_u32_shr(u64 a, u32 mul, unsigned int shift) +{ + return (u64)(((unsigned __int128)a * mul) >> shift); +} +#endif /* mul_u64_u32_shr */ + +#else + +#ifndef mul_u64_u32_shr +static inline u64 mul_u64_u32_shr(u64 a, u32 mul, unsigned int shift) +{ + u32 ah, al; + u64 ret; + + al = a; + ah = a >> 32; + + ret = ((u64)al * mul) >> shift; + if (ah) + ret += ((u64)ah * mul) << (32 - shift); + + return ret; +} +#endif /* mul_u64_u32_shr */ + +#endif + #endif /* _LINUX_MATH64_H */ diff --git a/init/Kconfig b/init/Kconfig index 79383d3aa5dc..4e5d96ab2034 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -809,6 +809,12 @@ config GENERIC_SCHED_CLOCK config ARCH_SUPPORTS_NUMA_BALANCING bool +# +# For architectures that know their GCC __int128 support is sound +# +config ARCH_SUPPORTS_INT128 + bool + # For architectures that (ab)use NUMA to represent different memory regions # all cpu-local but of different latencies, such as SuperH. # -- cgit v1.2.3 From 02ab71cdae248533620abefa1d46097581457110 Mon Sep 17 00:00:00 2001 From: Stefano Stabellini Date: Mon, 9 Dec 2013 15:55:11 +0000 Subject: xen/arm64: do not call the swiotlb functions twice On arm64 the dma_map_ops implementation is based on the swiotlb. swiotlb-xen, used by default in dom0 on Xen, is also based on the swiotlb. Avoid calling into the default arm64 dma_map_ops functions from xen_dma_map_page, xen_dma_unmap_page, xen_dma_sync_single_for_cpu, and xen_dma_sync_single_for_device otherwise we end up calling into the swiotlb twice. When arm64 gets a non-swiotlb based implementation of dma_map_ops, we'll probably have to reintroduce dma_map_ops calls in page-coherent.h. Signed-off-by: Stefano Stabellini CC: catalin.marinas@arm.com CC: Will.Deacon@arm.com CC: Ian.Campbell@citrix.com --- arch/arm64/include/asm/xen/page-coherent.h | 4 ---- 1 file changed, 4 deletions(-) (limited to 'arch') diff --git a/arch/arm64/include/asm/xen/page-coherent.h b/arch/arm64/include/asm/xen/page-coherent.h index 2820f1a6eebe..dde3fc9c49f0 100644 --- a/arch/arm64/include/asm/xen/page-coherent.h +++ b/arch/arm64/include/asm/xen/page-coherent.h @@ -23,25 +23,21 @@ static inline void xen_dma_map_page(struct device *hwdev, struct page *page, unsigned long offset, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) { - __generic_dma_ops(hwdev)->map_page(hwdev, page, offset, size, dir, attrs); } static inline void xen_dma_unmap_page(struct device *hwdev, dma_addr_t handle, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) { - __generic_dma_ops(hwdev)->unmap_page(hwdev, handle, size, dir, attrs); } static inline void xen_dma_sync_single_for_cpu(struct device *hwdev, dma_addr_t handle, size_t size, enum dma_data_direction dir) { - __generic_dma_ops(hwdev)->sync_single_for_cpu(hwdev, handle, size, dir); } static inline void xen_dma_sync_single_for_device(struct device *hwdev, dma_addr_t handle, size_t size, enum dma_data_direction dir) { - __generic_dma_ops(hwdev)->sync_single_for_device(hwdev, handle, size, dir); } #endif /* _ASM_ARM64_XEN_PAGE_COHERENT_H */ -- cgit v1.2.3 From a7892f32cc3534d4cc0e64b245fbf47a8e364652 Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Wed, 11 Dec 2013 17:02:27 +0000 Subject: arm: xen: foreign mapping PTEs are special. These mappings are in fact special and require special handling in privcmd, which already exists. Failure to mark the PTE as special on arm64 causes all sorts of bad PTE fun. e.g. e.g.: BUG: Bad page map in process xl pte:e0004077b33f53 pmd:4079575003 page:ffffffbce1a2f328 count:1 mapcount:-1 mapping: (null) index:0x0 page flags: 0x4000000000000014(referenced|dirty) addr:0000007fb5259000 vm_flags:040644fa anon_vma: (null) mapping:ffffffc03a6fda58 index:0 vma->vm_ops->fault: privcmd_fault+0x0/0x38 vma->vm_file->f_op->mmap: privcmd_mmap+0x0/0x2c CPU: 0 PID: 2657 Comm: xl Not tainted 3.12.0+ #102 Call trace: [] dump_backtrace+0x0/0x12c [] show_stack+0x14/0x1c [] dump_stack+0x70/0x90 [] print_bad_pte+0x12c/0x1bc [] unmap_single_vma+0x4cc/0x700 [] unmap_vmas+0x68/0xb4 [] unmap_region+0xcc/0x1d4 [] do_munmap+0x218/0x314 [] vm_munmap+0x44/0x64 [] SyS_munmap+0x24/0x34 Where unmap_single_vma contains inlined -> unmap_page_range -> zap_pud_range -> zap_pmd_range -> zap_pte_range -> print_bad_pte. Or: BUG: Bad page state in process xl pfn:4077b4d page:ffffffbce1a2f8d8 count:0 mapcount:-1 mapping: (null) index:0x0 page flags: 0x4000000000000014(referenced|dirty) Modules linked in: CPU: 0 PID: 2657 Comm: xl Tainted: G B 3.12.0+ #102 Call trace: [] dump_backtrace+0x0/0x12c [] show_stack+0x14/0x1c [] dump_stack+0x70/0x90 [] bad_page+0xc4/0x110 [] free_pages_prepare+0xd0/0xd8 [] free_hot_cold_page+0x28/0x178 [] free_hot_cold_page_list+0x38/0x60 [] release_pages+0x190/0x1dc [] unmap_region+0x15c/0x1d4 [] do_munmap+0x218/0x314 [] vm_munmap+0x44/0x64 [] SyS_munmap+0x24/0x34 x86 already gets this correct. 32-bit arm gets away with this because there is not PTE_SPECIAL bit in the PTE there and the vm_normal_page fallback path does the right thing. Signed-off-by: Ian Campbell Signed-off-by: Stefano Stabellini --- arch/arm/xen/enlighten.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/arm/xen/enlighten.c b/arch/arm/xen/enlighten.c index 6a288c7be49f..85501238b425 100644 --- a/arch/arm/xen/enlighten.c +++ b/arch/arm/xen/enlighten.c @@ -96,7 +96,7 @@ static int remap_pte_fn(pte_t *ptep, pgtable_t token, unsigned long addr, struct remap_data *info = data; struct page *page = info->pages[info->index++]; unsigned long pfn = page_to_pfn(page); - pte_t pte = pfn_pte(pfn, info->prot); + pte_t pte = pte_mkspecial(pfn_pte(pfn, info->prot)); if (map_foreign_page(pfn, info->fgmfn, info->domid)) return -EFAULT; -- cgit v1.2.3 From 36aa1b180e7398e2bd27642760bfaa4ad8c65ab6 Mon Sep 17 00:00:00 2001 From: Ulrich Weigand Date: Thu, 12 Dec 2013 15:59:34 +1100 Subject: powerpc: PTRACE_PEEKUSR always returns FPR0 There is a bug in using ptrace to access FPRs via PTRACE_PEEKUSR / PTRACE_POKEUSR. In effect, trying to access any of the FPRs always really accesses FPR0, which does seriously break debugging :-) The problem seems to have been introduced by commit 3ad26e5c4459d (Merge branch 'for-kvm' into next). [ It is indeed a merge conflict between Paul's FPU/VSX state rework and my LE patches - Anton ] Signed-off-by: Ulrich Weigand Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/ptrace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/kernel/ptrace.c b/arch/powerpc/kernel/ptrace.c index 75fb40498b41..2e3d2bf536c5 100644 --- a/arch/powerpc/kernel/ptrace.c +++ b/arch/powerpc/kernel/ptrace.c @@ -1555,7 +1555,7 @@ long arch_ptrace(struct task_struct *child, long request, flush_fp_to_thread(child); if (fpidx < (PT_FPSCR - PT_FPR0)) - memcpy(&tmp, &child->thread.fp_state.fpr, + memcpy(&tmp, &child->thread.TS_FPR(fpidx), sizeof(long)); else tmp = child->thread.fp_state.fpscr; @@ -1588,7 +1588,7 @@ long arch_ptrace(struct task_struct *child, long request, flush_fp_to_thread(child); if (fpidx < (PT_FPSCR - PT_FPR0)) - memcpy(&child->thread.fp_state.fpr, &data, + memcpy(&child->thread.TS_FPR(fpidx), &data, sizeof(long)); else child->thread.fp_state.fpscr = data; -- cgit v1.2.3 From 01666c8ee2b6afcd31de2064fbb7c097a75e5089 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 12 Dec 2013 15:59:35 +1100 Subject: powerpc: Fix endian issue in setup-common.c During on LE boot we see: Partition configured for 1073741824 cpus, operating system maximum is 2048. Clearly missing a byteswap here. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/setup-common.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c index febc80445d25..bc76cc6b419c 100644 --- a/arch/powerpc/kernel/setup-common.c +++ b/arch/powerpc/kernel/setup-common.c @@ -479,7 +479,7 @@ void __init smp_setup_cpu_maps(void) if (machine_is(pseries) && firmware_has_feature(FW_FEATURE_LPAR) && (dn = of_find_node_by_path("/rtas"))) { int num_addr_cell, num_size_cell, maxcpus; - const unsigned int *ireg; + const __be32 *ireg; num_addr_cell = of_n_addr_cells(dn); num_size_cell = of_n_size_cells(dn); @@ -489,7 +489,7 @@ void __init smp_setup_cpu_maps(void) if (!ireg) goto out; - maxcpus = ireg[num_addr_cell + num_size_cell]; + maxcpus = be32_to_cpup(ireg + num_addr_cell + num_size_cell); /* Double maxcpus for processors which have SMT capability */ if (cpu_has_feature(CPU_FTR_SMT)) -- cgit v1.2.3 From f8a1883a833bbad8e6b5ec4f0918b7797e652d65 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 12 Dec 2013 15:59:36 +1100 Subject: powerpc: Fix topology core_id endian issue on LE builds cpu_to_core_id() is missing a byteswap: cat /sys/devices/system/cpu/cpu63/topology/core_id 201326592 Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/smp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index a3b64f3bf9a2..c1cf4a1522d9 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -580,7 +580,7 @@ int __cpu_up(unsigned int cpu, struct task_struct *tidle) int cpu_to_core_id(int cpu) { struct device_node *np; - const int *reg; + const __be32 *reg; int id = -1; np = of_get_cpu_node(cpu, NULL); @@ -591,7 +591,7 @@ int cpu_to_core_id(int cpu) if (!reg) goto out; - id = *reg; + id = be32_to_cpup(reg); out: of_node_put(np); return id; -- cgit v1.2.3 From ca5de4e652ea09c0c18ebbd12dd4c2149271245a Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 12 Dec 2013 15:59:37 +1100 Subject: powerpc/pseries: Fix endian issues in /proc/ppc64/lparcfg Some obvious issues: cat /proc/ppc64/lparcfg ... partition_id=16777216 ... partition_potential_processors=268435456 Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/pseries/lparcfg.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/pseries/lparcfg.c b/arch/powerpc/platforms/pseries/lparcfg.c index e738007eae64..c9fecf09b8fa 100644 --- a/arch/powerpc/platforms/pseries/lparcfg.c +++ b/arch/powerpc/platforms/pseries/lparcfg.c @@ -157,7 +157,7 @@ static void parse_ppp_data(struct seq_file *m) { struct hvcall_ppp_data ppp_data; struct device_node *root; - const int *perf_level; + const __be32 *perf_level; int rc; rc = h_get_ppp(&ppp_data); @@ -201,7 +201,7 @@ static void parse_ppp_data(struct seq_file *m) perf_level = of_get_property(root, "ibm,partition-performance-parameters-level", NULL); - if (perf_level && (*perf_level >= 1)) { + if (perf_level && (be32_to_cpup(perf_level) >= 1)) { seq_printf(m, "physical_procs_allocated_to_virtualization=%d\n", ppp_data.phys_platform_procs); @@ -435,7 +435,7 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v) int partition_potential_processors; int partition_active_processors; struct device_node *rtas_node; - const int *lrdrp = NULL; + const __be32 *lrdrp = NULL; rtas_node = of_find_node_by_path("/rtas"); if (rtas_node) @@ -444,7 +444,7 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v) if (lrdrp == NULL) { partition_potential_processors = vdso_data->processorCount; } else { - partition_potential_processors = *(lrdrp + 4); + partition_potential_processors = be32_to_cpup(lrdrp + 4); } of_node_put(rtas_node); @@ -654,7 +654,7 @@ static int lparcfg_data(struct seq_file *m, void *v) const char *model = ""; const char *system_id = ""; const char *tmp; - const unsigned int *lp_index_ptr; + const __be32 *lp_index_ptr; unsigned int lp_index = 0; seq_printf(m, "%s %s\n", MODULE_NAME, MODULE_VERS); @@ -670,7 +670,7 @@ static int lparcfg_data(struct seq_file *m, void *v) lp_index_ptr = of_get_property(rootdn, "ibm,partition-no", NULL); if (lp_index_ptr) - lp_index = *lp_index_ptr; + lp_index = be32_to_cpup(lp_index_ptr); of_node_put(rootdn); } seq_printf(m, "serial_number=%s\n", system_id); -- cgit v1.2.3 From 9fa2984d1b5d9736a88c813ae89c160a2f9d5308 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 12 Dec 2013 15:59:38 +1100 Subject: powerpc/pseries: Fix endian issues in nvram code The NVRAM code has a number of endian issues. I noticed a very confused error log count: RTAS: 100663330 -------- RTAS event begin -------- 100663330 == 0x06000022. 0x6 LE error logs and 0x22 BE error logs. The pstore code has similar issues - if we write an oops in one endian and attempt to read it in another we get junk. Make both of these formats big endian, and byteswap as required. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/pseries/nvram.c | 46 +++++++++++++++++----------------- 1 file changed, 23 insertions(+), 23 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/pseries/nvram.c b/arch/powerpc/platforms/pseries/nvram.c index 7bfaf58d4664..d7096f2f7751 100644 --- a/arch/powerpc/platforms/pseries/nvram.c +++ b/arch/powerpc/platforms/pseries/nvram.c @@ -43,8 +43,8 @@ static char nvram_buf[NVRW_CNT]; /* assume this is in the first 4GB */ static DEFINE_SPINLOCK(nvram_lock); struct err_log_info { - int error_type; - unsigned int seq_num; + __be32 error_type; + __be32 seq_num; }; struct nvram_os_partition { @@ -79,9 +79,9 @@ static const char *pseries_nvram_os_partitions[] = { }; struct oops_log_info { - u16 version; - u16 report_length; - u64 timestamp; + __be16 version; + __be16 report_length; + __be64 timestamp; } __attribute__((packed)); static void oops_to_nvram(struct kmsg_dumper *dumper, @@ -291,8 +291,8 @@ int nvram_write_os_partition(struct nvram_os_partition *part, char * buff, length = part->size; } - info.error_type = err_type; - info.seq_num = error_log_cnt; + info.error_type = cpu_to_be32(err_type); + info.seq_num = cpu_to_be32(error_log_cnt); tmp_index = part->index; @@ -364,8 +364,8 @@ int nvram_read_partition(struct nvram_os_partition *part, char *buff, } if (part->os_partition) { - *error_log_cnt = info.seq_num; - *err_type = info.error_type; + *error_log_cnt = be32_to_cpu(info.seq_num); + *err_type = be32_to_cpu(info.error_type); } return 0; @@ -529,9 +529,9 @@ static int zip_oops(size_t text_len) pr_err("nvram: logging uncompressed oops/panic report\n"); return -1; } - oops_hdr->version = OOPS_HDR_VERSION; - oops_hdr->report_length = (u16) zipped_len; - oops_hdr->timestamp = get_seconds(); + oops_hdr->version = cpu_to_be16(OOPS_HDR_VERSION); + oops_hdr->report_length = cpu_to_be16(zipped_len); + oops_hdr->timestamp = cpu_to_be64(get_seconds()); return 0; } @@ -574,9 +574,9 @@ static int nvram_pstore_write(enum pstore_type_id type, clobbering_unread_rtas_event()) return -1; - oops_hdr->version = OOPS_HDR_VERSION; - oops_hdr->report_length = (u16) size; - oops_hdr->timestamp = get_seconds(); + oops_hdr->version = cpu_to_be16(OOPS_HDR_VERSION); + oops_hdr->report_length = cpu_to_be16(size); + oops_hdr->timestamp = cpu_to_be64(get_seconds()); if (compressed) err_type = ERR_TYPE_KERNEL_PANIC_GZ; @@ -670,16 +670,16 @@ static ssize_t nvram_pstore_read(u64 *id, enum pstore_type_id *type, size_t length, hdr_size; oops_hdr = (struct oops_log_info *)buff; - if (oops_hdr->version < OOPS_HDR_VERSION) { + if (be16_to_cpu(oops_hdr->version) < OOPS_HDR_VERSION) { /* Old format oops header had 2-byte record size */ hdr_size = sizeof(u16); - length = oops_hdr->version; + length = be16_to_cpu(oops_hdr->version); time->tv_sec = 0; time->tv_nsec = 0; } else { hdr_size = sizeof(*oops_hdr); - length = oops_hdr->report_length; - time->tv_sec = oops_hdr->timestamp; + length = be16_to_cpu(oops_hdr->report_length); + time->tv_sec = be64_to_cpu(oops_hdr->timestamp); time->tv_nsec = 0; } *buf = kmalloc(length, GFP_KERNEL); @@ -889,13 +889,13 @@ static void oops_to_nvram(struct kmsg_dumper *dumper, kmsg_dump_get_buffer(dumper, false, oops_data, oops_data_sz, &text_len); err_type = ERR_TYPE_KERNEL_PANIC; - oops_hdr->version = OOPS_HDR_VERSION; - oops_hdr->report_length = (u16) text_len; - oops_hdr->timestamp = get_seconds(); + oops_hdr->version = cpu_to_be16(OOPS_HDR_VERSION); + oops_hdr->report_length = cpu_to_be16(text_len); + oops_hdr->timestamp = cpu_to_be64(get_seconds()); } (void) nvram_write_os_partition(&oops_log_partition, oops_buf, - (int) (sizeof(*oops_hdr) + oops_hdr->report_length), err_type, + (int) (sizeof(*oops_hdr) + text_len), err_type, ++oops_count); spin_unlock_irqrestore(&lock, flags); -- cgit v1.2.3 From 5091f0c969d2d0639db2efddce7352e7c48069ba Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 12 Dec 2013 15:59:39 +1100 Subject: powerpc/pseries: Fix PCIE link speed endian issue We need to byteswap ibm,pcie-link-speed-stats. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/pseries/pci.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/pseries/pci.c b/arch/powerpc/platforms/pseries/pci.c index 5f93856cdf47..70670a2d9cf2 100644 --- a/arch/powerpc/platforms/pseries/pci.c +++ b/arch/powerpc/platforms/pseries/pci.c @@ -113,7 +113,7 @@ int pseries_root_bridge_prepare(struct pci_host_bridge *bridge) { struct device_node *dn, *pdn; struct pci_bus *bus; - const uint32_t *pcie_link_speed_stats; + const __be32 *pcie_link_speed_stats; bus = bridge->bus; @@ -122,7 +122,7 @@ int pseries_root_bridge_prepare(struct pci_host_bridge *bridge) return 0; for (pdn = dn; pdn != NULL; pdn = of_get_next_parent(pdn)) { - pcie_link_speed_stats = (const uint32_t *) of_get_property(pdn, + pcie_link_speed_stats = of_get_property(pdn, "ibm,pcie-link-speed-stats", NULL); if (pcie_link_speed_stats) break; @@ -135,7 +135,7 @@ int pseries_root_bridge_prepare(struct pci_host_bridge *bridge) return 0; } - switch (pcie_link_speed_stats[0]) { + switch (be32_to_cpup(pcie_link_speed_stats)) { case 0x01: bus->max_bus_speed = PCIE_SPEED_2_5GT; break; @@ -147,7 +147,7 @@ int pseries_root_bridge_prepare(struct pci_host_bridge *bridge) break; } - switch (pcie_link_speed_stats[1]) { + switch (be32_to_cpup(pcie_link_speed_stats)) { case 0x01: bus->cur_bus_speed = PCIE_SPEED_2_5GT; break; -- cgit v1.2.3 From 8d15315537578a9ef5d35912d06127205dd82be8 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 12 Dec 2013 15:59:40 +1100 Subject: powerpc/pseries: Fix endian issues in MSI code The MSI code is miscalculating quotas in little endian mode. Add required byteswaps to fix this. Before we claimed a quota of 65536, after the patch we see the correct value of 256. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/pseries/msi.c | 28 +++++++++++++++------------- 1 file changed, 15 insertions(+), 13 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/pseries/msi.c b/arch/powerpc/platforms/pseries/msi.c index 6d2f0abce6fa..0c882e83c4ce 100644 --- a/arch/powerpc/platforms/pseries/msi.c +++ b/arch/powerpc/platforms/pseries/msi.c @@ -130,7 +130,8 @@ static int check_req(struct pci_dev *pdev, int nvec, char *prop_name) { struct device_node *dn; struct pci_dn *pdn; - const u32 *req_msi; + const __be32 *p; + u32 req_msi; pdn = pci_get_pdn(pdev); if (!pdn) @@ -138,19 +139,20 @@ static int check_req(struct pci_dev *pdev, int nvec, char *prop_name) dn = pdn->node; - req_msi = of_get_property(dn, prop_name, NULL); - if (!req_msi) { + p = of_get_property(dn, prop_name, NULL); + if (!p) { pr_debug("rtas_msi: No %s on %s\n", prop_name, dn->full_name); return -ENOENT; } - if (*req_msi < nvec) { + req_msi = be32_to_cpup(p); + if (req_msi < nvec) { pr_debug("rtas_msi: %s requests < %d MSIs\n", prop_name, nvec); - if (*req_msi == 0) /* Be paranoid */ + if (req_msi == 0) /* Be paranoid */ return -ENOSPC; - return *req_msi; + return req_msi; } return 0; @@ -171,7 +173,7 @@ static int check_req_msix(struct pci_dev *pdev, int nvec) static struct device_node *find_pe_total_msi(struct pci_dev *dev, int *total) { struct device_node *dn; - const u32 *p; + const __be32 *p; dn = of_node_get(pci_device_to_OF_node(dev)); while (dn) { @@ -179,7 +181,7 @@ static struct device_node *find_pe_total_msi(struct pci_dev *dev, int *total) if (p) { pr_debug("rtas_msi: found prop on dn %s\n", dn->full_name); - *total = *p; + *total = be32_to_cpup(p); return dn; } @@ -232,13 +234,13 @@ struct msi_counts { static void *count_non_bridge_devices(struct device_node *dn, void *data) { struct msi_counts *counts = data; - const u32 *p; + const __be32 *p; u32 class; pr_debug("rtas_msi: counting %s\n", dn->full_name); p = of_get_property(dn, "class-code", NULL); - class = p ? *p : 0; + class = p ? be32_to_cpup(p) : 0; if ((class >> 8) != PCI_CLASS_BRIDGE_PCI) counts->num_devices++; @@ -249,7 +251,7 @@ static void *count_non_bridge_devices(struct device_node *dn, void *data) static void *count_spare_msis(struct device_node *dn, void *data) { struct msi_counts *counts = data; - const u32 *p; + const __be32 *p; int req; if (dn == counts->requestor) @@ -260,11 +262,11 @@ static void *count_spare_msis(struct device_node *dn, void *data) req = 0; p = of_get_property(dn, "ibm,req#msi", NULL); if (p) - req = *p; + req = be32_to_cpup(p); p = of_get_property(dn, "ibm,req#msi-x", NULL); if (p) - req = max(req, (int)*p); + req = max(req, (int)be32_to_cpup(p)); } if (req < counts->quota) -- cgit v1.2.3 From a29e30efa3a1b253fd1252731aa774a3544c5c3b Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 12 Dec 2013 15:59:41 +1100 Subject: powerpc: Fix endian issues in crash dump code A couple more device tree properties that need byte swapping. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/crash_dump.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/kernel/crash_dump.c b/arch/powerpc/kernel/crash_dump.c index 779a78c26435..11c1d069d920 100644 --- a/arch/powerpc/kernel/crash_dump.c +++ b/arch/powerpc/kernel/crash_dump.c @@ -124,15 +124,15 @@ ssize_t copy_oldmem_page(unsigned long pfn, char *buf, void crash_free_reserved_phys_range(unsigned long begin, unsigned long end) { unsigned long addr; - const u32 *basep, *sizep; + const __be32 *basep, *sizep; unsigned int rtas_start = 0, rtas_end = 0; basep = of_get_property(rtas.dev, "linux,rtas-base", NULL); sizep = of_get_property(rtas.dev, "rtas-size", NULL); if (basep && sizep) { - rtas_start = *basep; - rtas_end = *basep + *sizep; + rtas_start = be32_to_cpup(basep); + rtas_end = rtas_start + be32_to_cpup(sizep); } for (addr = begin; addr < end; addr += PAGE_SIZE) { -- cgit v1.2.3 From 01a9dbccbdfc7a93950b791405e476c510f73ab0 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Fri, 13 Dec 2013 15:53:43 +1100 Subject: powerpc/powernv: Fix endian issue in opal_xscom_read opal_xscom_read uses a pointer to return the data so we need to byteswap it on LE builds. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/opal.h | 2 +- arch/powerpc/platforms/powernv/opal-xscom.c | 4 +++- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h index 033c06be1d84..522f50c6c146 100644 --- a/arch/powerpc/include/asm/opal.h +++ b/arch/powerpc/include/asm/opal.h @@ -720,7 +720,7 @@ int64_t opal_pci_next_error(uint64_t phb_id, uint64_t *first_frozen_pe, int64_t opal_pci_poll(uint64_t phb_id); int64_t opal_return_cpu(void); -int64_t opal_xscom_read(uint32_t gcid, uint32_t pcb_addr, uint64_t *val); +int64_t opal_xscom_read(uint32_t gcid, uint32_t pcb_addr, __be64 *val); int64_t opal_xscom_write(uint32_t gcid, uint32_t pcb_addr, uint64_t val); int64_t opal_lpc_write(uint32_t chip_id, enum OpalLPCAddressType addr_type, diff --git a/arch/powerpc/platforms/powernv/opal-xscom.c b/arch/powerpc/platforms/powernv/opal-xscom.c index 4d99a8fd55ac..4fbf276ac99e 100644 --- a/arch/powerpc/platforms/powernv/opal-xscom.c +++ b/arch/powerpc/platforms/powernv/opal-xscom.c @@ -96,9 +96,11 @@ static int opal_scom_read(scom_map_t map, u64 reg, u64 *value) { struct opal_scom_map *m = map; int64_t rc; + __be64 v; reg = opal_scom_unmangle(reg); - rc = opal_xscom_read(m->chip, m->addr + reg, (uint64_t *)__pa(value)); + rc = opal_xscom_read(m->chip, m->addr + reg, (__be64 *)__pa(&v)); + *value = be64_to_cpu(v); return opal_xscom_err_xlate(rc); } -- cgit v1.2.3 From 803c2d2f84da9dc2619449994af34d27148ab20d Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 13 Dec 2013 15:56:06 +1100 Subject: powerpc/powernv: Fix OPAL LPC access in Little Endian We are passing pointers to the firmware for reads, we need to properly convert the result as OPAL is always BE. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/opal.h | 2 +- arch/powerpc/platforms/powernv/opal-lpc.c | 12 ++++++------ 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h index 522f50c6c146..7bdcf340016c 100644 --- a/arch/powerpc/include/asm/opal.h +++ b/arch/powerpc/include/asm/opal.h @@ -726,7 +726,7 @@ int64_t opal_xscom_write(uint32_t gcid, uint32_t pcb_addr, uint64_t val); int64_t opal_lpc_write(uint32_t chip_id, enum OpalLPCAddressType addr_type, uint32_t addr, uint32_t data, uint32_t sz); int64_t opal_lpc_read(uint32_t chip_id, enum OpalLPCAddressType addr_type, - uint32_t addr, uint32_t *data, uint32_t sz); + uint32_t addr, __be32 *data, uint32_t sz); int64_t opal_validate_flash(uint64_t buffer, uint32_t *size, uint32_t *result); int64_t opal_manage_flash(uint8_t op); int64_t opal_update_flash(uint64_t blk_list); diff --git a/arch/powerpc/platforms/powernv/opal-lpc.c b/arch/powerpc/platforms/powernv/opal-lpc.c index e7e59e4f9892..79d83cad3d67 100644 --- a/arch/powerpc/platforms/powernv/opal-lpc.c +++ b/arch/powerpc/platforms/powernv/opal-lpc.c @@ -24,25 +24,25 @@ static int opal_lpc_chip_id = -1; static u8 opal_lpc_inb(unsigned long port) { int64_t rc; - uint32_t data; + __be32 data; if (opal_lpc_chip_id < 0 || port > 0xffff) return 0xff; rc = opal_lpc_read(opal_lpc_chip_id, OPAL_LPC_IO, port, &data, 1); - return rc ? 0xff : data; + return rc ? 0xff : be32_to_cpu(data); } static __le16 __opal_lpc_inw(unsigned long port) { int64_t rc; - uint32_t data; + __be32 data; if (opal_lpc_chip_id < 0 || port > 0xfffe) return 0xffff; if (port & 1) return (__le16)opal_lpc_inb(port) << 8 | opal_lpc_inb(port + 1); rc = opal_lpc_read(opal_lpc_chip_id, OPAL_LPC_IO, port, &data, 2); - return rc ? 0xffff : data; + return rc ? 0xffff : be32_to_cpu(data); } static u16 opal_lpc_inw(unsigned long port) { @@ -52,7 +52,7 @@ static u16 opal_lpc_inw(unsigned long port) static __le32 __opal_lpc_inl(unsigned long port) { int64_t rc; - uint32_t data; + __be32 data; if (opal_lpc_chip_id < 0 || port > 0xfffc) return 0xffffffff; @@ -62,7 +62,7 @@ static __le32 __opal_lpc_inl(unsigned long port) (__le32)opal_lpc_inb(port + 2) << 8 | opal_lpc_inb(port + 3); rc = opal_lpc_read(opal_lpc_chip_id, OPAL_LPC_IO, port, &data, 4); - return rc ? 0xffffffff : data; + return rc ? 0xffffffff : be32_to_cpu(data); } static u32 opal_lpc_inl(unsigned long port) -- cgit v1.2.3 From 36e7bb38028d3d812aa7749208249d600a30c22c Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Mon, 11 Nov 2013 19:29:47 +0530 Subject: powerpc: book3s: kvm: Don't abuse host r2 in exit path We don't use PACATOC for PR. Avoid updating HOST_R2 with PR KVM mode when both HV and PR are enabled in the kernel. Without this we get the below crash (qemu) Unable to handle kernel paging request for data at address 0xffffffffffff8310 Faulting instruction address: 0xc00000000001d5a4 cpu 0x2: Vector: 300 (Data Access) at [c0000001dc53aef0] pc: c00000000001d5a4: .vtime_delta.isra.1+0x34/0x1d0 lr: c00000000001d760: .vtime_account_system+0x20/0x60 sp: c0000001dc53b170 msr: 8000000000009032 dar: ffffffffffff8310 dsisr: 40000000 current = 0xc0000001d76c62d0 paca = 0xc00000000fef1100 softe: 0 irq_happened: 0x01 pid = 4472, comm = qemu-system-ppc enter ? for help [c0000001dc53b200] c00000000001d760 .vtime_account_system+0x20/0x60 [c0000001dc53b290] c00000000008d050 .kvmppc_handle_exit_pr+0x60/0xa50 [c0000001dc53b340] c00000000008f51c kvm_start_lightweight+0xb4/0xc4 [c0000001dc53b510] c00000000008cdf0 .kvmppc_vcpu_run_pr+0x150/0x2e0 [c0000001dc53b9e0] c00000000008341c .kvmppc_vcpu_run+0x2c/0x40 [c0000001dc53ba50] c000000000080af4 .kvm_arch_vcpu_ioctl_run+0x54/0x1b0 [c0000001dc53bae0] c00000000007b4c8 .kvm_vcpu_ioctl+0x478/0x730 [c0000001dc53bca0] c0000000002140cc .do_vfs_ioctl+0x4ac/0x770 [c0000001dc53bd80] c0000000002143e8 .SyS_ioctl+0x58/0xb0 [c0000001dc53be30] c000000000009e58 syscall_exit+0x0/0x98 Signed-off-by: Alexander Graf --- arch/powerpc/include/asm/kvm_book3s_asm.h | 1 + arch/powerpc/kernel/asm-offsets.c | 1 + arch/powerpc/kvm/book3s_hv_rmhandlers.S | 7 +++---- 3 files changed, 5 insertions(+), 4 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/include/asm/kvm_book3s_asm.h b/arch/powerpc/include/asm/kvm_book3s_asm.h index 412b2f389474..192917d2239c 100644 --- a/arch/powerpc/include/asm/kvm_book3s_asm.h +++ b/arch/powerpc/include/asm/kvm_book3s_asm.h @@ -79,6 +79,7 @@ struct kvmppc_host_state { ulong vmhandler; ulong scratch0; ulong scratch1; + ulong scratch2; u8 in_guest; u8 restore_hid5; u8 napping; diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c index 2ea5cc033ec8..d3de01066f7d 100644 --- a/arch/powerpc/kernel/asm-offsets.c +++ b/arch/powerpc/kernel/asm-offsets.c @@ -576,6 +576,7 @@ int main(void) HSTATE_FIELD(HSTATE_VMHANDLER, vmhandler); HSTATE_FIELD(HSTATE_SCRATCH0, scratch0); HSTATE_FIELD(HSTATE_SCRATCH1, scratch1); + HSTATE_FIELD(HSTATE_SCRATCH2, scratch2); HSTATE_FIELD(HSTATE_IN_GUEST, in_guest); HSTATE_FIELD(HSTATE_RESTORE_HID5, restore_hid5); HSTATE_FIELD(HSTATE_NAPPING, napping); diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S index bde28da69610..be4fa04a37c9 100644 --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S @@ -754,15 +754,14 @@ kvmppc_interrupt_hv: * guest CR, R12 saved in shadow VCPU SCRATCH1/0 * guest R13 saved in SPRN_SCRATCH0 */ - /* abuse host_r2 as third scratch area; we get r2 from PACATOC(r13) */ - std r9, HSTATE_HOST_R2(r13) + std r9, HSTATE_SCRATCH2(r13) lbz r9, HSTATE_IN_GUEST(r13) cmpwi r9, KVM_GUEST_MODE_HOST_HV beq kvmppc_bad_host_intr #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE cmpwi r9, KVM_GUEST_MODE_GUEST - ld r9, HSTATE_HOST_R2(r13) + ld r9, HSTATE_SCRATCH2(r13) beq kvmppc_interrupt_pr #endif /* We're now back in the host but in guest MMU context */ @@ -782,7 +781,7 @@ kvmppc_interrupt_hv: std r6, VCPU_GPR(R6)(r9) std r7, VCPU_GPR(R7)(r9) std r8, VCPU_GPR(R8)(r9) - ld r0, HSTATE_HOST_R2(r13) + ld r0, HSTATE_SCRATCH2(r13) std r0, VCPU_GPR(R9)(r9) std r10, VCPU_GPR(R10)(r9) std r11, VCPU_GPR(R11)(r9) -- cgit v1.2.3 From df9059bb64023da9f27e56a94a3e2b8f4b6336a9 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Mon, 16 Dec 2013 13:31:46 +1100 Subject: KVM: PPC: Book3S HV: Don't drop low-order page address bits Commit caaa4c804fae ("KVM: PPC: Book3S HV: Fix physical address calculations") unfortunately resulted in some low-order address bits getting dropped in the case where the guest is creating a 4k HPTE and the host page size is 64k. By getting the low-order bits from hva rather than gpa we miss out on bits 12 - 15 in this case, since hva is at page granularity. This puts the missing bits back in. Reported-by: Alexey Kardashevskiy Signed-off-by: Paul Mackerras Signed-off-by: Alexander Graf --- arch/powerpc/kvm/book3s_hv_rm_mmu.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch') diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c index 1931aa341a72..8689e2e30857 100644 --- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c +++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c @@ -240,6 +240,7 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags, is_io = hpte_cache_bits(pte_val(pte)); pa = pte_pfn(pte) << PAGE_SHIFT; pa |= hva & (pte_size - 1); + pa |= gpa & ~PAGE_MASK; } } -- cgit v1.2.3 From 2b4847e73004c10ae6666c2e27b5c5430aed8698 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 18 Dec 2013 17:08:32 -0800 Subject: mm: numa: serialise parallel get_user_page against THP migration Base pages are unmapped and flushed from cache and TLB during normal page migration and replaced with a migration entry that causes any parallel NUMA hinting fault or gup to block until migration completes. THP does not unmap pages due to a lack of support for migration entries at a PMD level. This allows races with get_user_pages and get_user_pages_fast which commit 3f926ab945b6 ("mm: Close races between THP migration and PMD numa clearing") made worse by introducing a pmd_clear_flush(). This patch forces get_user_page (fast and normal) on a pmd_numa page to go through the slow get_user_page path where it will serialise against THP migration and properly account for the NUMA hinting fault. On the migration side the page table lock is taken for each PTE update. Signed-off-by: Mel Gorman Reviewed-by: Rik van Riel Cc: Alex Thorlton Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/mm/gup.c | 13 +++++++++++++ mm/huge_memory.c | 24 ++++++++++++++++-------- mm/migrate.c | 38 +++++++++++++++++++++++++++++++------- 3 files changed, 60 insertions(+), 15 deletions(-) (limited to 'arch') diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c index dd74e46828c0..0596e8e0cc19 100644 --- a/arch/x86/mm/gup.c +++ b/arch/x86/mm/gup.c @@ -83,6 +83,12 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr, pte_t pte = gup_get_pte(ptep); struct page *page; + /* Similar to the PMD case, NUMA hinting must take slow path */ + if (pte_numa(pte)) { + pte_unmap(ptep); + return 0; + } + if ((pte_flags(pte) & (mask | _PAGE_SPECIAL)) != mask) { pte_unmap(ptep); return 0; @@ -167,6 +173,13 @@ static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end, if (pmd_none(pmd) || pmd_trans_splitting(pmd)) return 0; if (unlikely(pmd_large(pmd))) { + /* + * NUMA hinting faults need to be handled in the GUP + * slowpath for accounting purposes and so that they + * can be serialised against THP migration. + */ + if (pmd_numa(pmd)) + return 0; if (!gup_huge_pmd(pmd, addr, next, write, pages, nr)) return 0; } else { diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 33a5dc492810..51f069303ab9 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1243,6 +1243,10 @@ struct page *follow_trans_huge_pmd(struct vm_area_struct *vma, if ((flags & FOLL_DUMP) && is_huge_zero_pmd(*pmd)) return ERR_PTR(-EFAULT); + /* Full NUMA hinting faults to serialise migration in fault paths */ + if ((flags & FOLL_NUMA) && pmd_numa(*pmd)) + goto out; + page = pmd_page(*pmd); VM_BUG_ON(!PageHead(page)); if (flags & FOLL_TOUCH) { @@ -1323,23 +1327,27 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, /* If the page was locked, there are no parallel migrations */ if (page_locked) goto clear_pmdnuma; + } - /* - * Otherwise wait for potential migrations and retry. We do - * relock and check_same as the page may no longer be mapped. - * As the fault is being retried, do not account for it. - */ + /* + * If there are potential migrations, wait for completion and retry. We + * do not relock and check_same as the page may no longer be mapped. + * Furtermore, even if the page is currently misplaced, there is no + * guarantee it is still misplaced after the migration completes. + */ + if (!page_locked) { spin_unlock(ptl); wait_on_page_locked(page); page_nid = -1; goto out; } - /* Page is misplaced, serialise migrations and parallel THP splits */ + /* + * Page is misplaced. Page lock serialises migrations. Acquire anon_vma + * to serialises splits + */ get_page(page); spin_unlock(ptl); - if (!page_locked) - lock_page(page); anon_vma = page_lock_anon_vma_read(page); /* Confirm the PMD did not change while page_table_lock was released */ diff --git a/mm/migrate.c b/mm/migrate.c index bb940045fe85..2cabbd5fa5bf 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1722,6 +1722,7 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, struct page *new_page = NULL; struct mem_cgroup *memcg = NULL; int page_lru = page_is_file_cache(page); + pmd_t orig_entry; /* * Rate-limit the amount of data that is being migrated to a node. @@ -1756,7 +1757,8 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, /* Recheck the target PMD */ ptl = pmd_lock(mm, pmd); - if (unlikely(!pmd_same(*pmd, entry))) { + if (unlikely(!pmd_same(*pmd, entry) || page_count(page) != 2)) { +fail_putback: spin_unlock(ptl); /* Reverse changes made by migrate_page_copy() */ @@ -1786,16 +1788,34 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, */ mem_cgroup_prepare_migration(page, new_page, &memcg); + orig_entry = *pmd; entry = mk_pmd(new_page, vma->vm_page_prot); - entry = pmd_mknonnuma(entry); - entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma); entry = pmd_mkhuge(entry); + entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma); + /* + * Clear the old entry under pagetable lock and establish the new PTE. + * Any parallel GUP will either observe the old page blocking on the + * page lock, block on the page table lock or observe the new page. + * The SetPageUptodate on the new page and page_add_new_anon_rmap + * guarantee the copy is visible before the pagetable update. + */ + flush_cache_range(vma, haddr, haddr + HPAGE_PMD_SIZE); + page_add_new_anon_rmap(new_page, vma, haddr); pmdp_clear_flush(vma, haddr, pmd); set_pmd_at(mm, haddr, pmd, entry); - page_add_new_anon_rmap(new_page, vma, haddr); update_mmu_cache_pmd(vma, address, &entry); + + if (page_count(page) != 2) { + set_pmd_at(mm, haddr, pmd, orig_entry); + flush_tlb_range(vma, haddr, haddr + HPAGE_PMD_SIZE); + update_mmu_cache_pmd(vma, address, &entry); + page_remove_rmap(new_page); + goto fail_putback; + } + page_remove_rmap(page); + /* * Finish the charge transaction under the page table lock to * prevent split_huge_page() from dividing up the charge @@ -1820,9 +1840,13 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, out_fail: count_vm_events(PGMIGRATE_FAIL, HPAGE_PMD_NR); out_dropref: - entry = pmd_mknonnuma(entry); - set_pmd_at(mm, haddr, pmd, entry); - update_mmu_cache_pmd(vma, address, &entry); + ptl = pmd_lock(mm, pmd); + if (pmd_same(*pmd, entry)) { + entry = pmd_mknonnuma(entry); + set_pmd_at(mm, haddr, pmd, entry); + update_mmu_cache_pmd(vma, address, &entry); + } + spin_unlock(ptl); unlock_page(page); put_page(page); -- cgit v1.2.3 From 20841405940e7be0617612d521e206e4b6b325db Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Wed, 18 Dec 2013 17:08:44 -0800 Subject: mm: fix TLB flush race between migration, and change_protection_range There are a few subtle races, between change_protection_range (used by mprotect and change_prot_numa) on one side, and NUMA page migration and compaction on the other side. The basic race is that there is a time window between when the PTE gets made non-present (PROT_NONE or NUMA), and the TLB is flushed. During that time, a CPU may continue writing to the page. This is fine most of the time, however compaction or the NUMA migration code may come in, and migrate the page away. When that happens, the CPU may continue writing, through the cached translation, to what is no longer the current memory location of the process. This only affects x86, which has a somewhat optimistic pte_accessible. All other architectures appear to be safe, and will either always flush, or flush whenever there is a valid mapping, even with no permissions (SPARC). The basic race looks like this: CPU A CPU B CPU C load TLB entry make entry PTE/PMD_NUMA fault on entry read/write old page start migrating page change PTE/PMD to new page read/write old page [*] flush TLB reload TLB from new entry read/write new page lose data [*] the old page may belong to a new user at this point! The obvious fix is to flush remote TLB entries, by making sure that pte_accessible aware of the fact that PROT_NONE and PROT_NUMA memory may still be accessible if there is a TLB flush pending for the mm. This should fix both NUMA migration and compaction. [mgorman@suse.de: fix build] Signed-off-by: Rik van Riel Signed-off-by: Mel Gorman Cc: Alex Thorlton Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/sparc/include/asm/pgtable_64.h | 4 ++-- arch/x86/include/asm/pgtable.h | 11 ++++++++-- include/asm-generic/pgtable.h | 2 +- include/linux/mm_types.h | 44 +++++++++++++++++++++++++++++++++++++ kernel/fork.c | 1 + mm/huge_memory.c | 7 ++++++ mm/mprotect.c | 2 ++ mm/pgtable-generic.c | 5 +++-- 8 files changed, 69 insertions(+), 7 deletions(-) (limited to 'arch') diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index 8358dc144959..0f9e94537eee 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -619,7 +619,7 @@ static inline unsigned long pte_present(pte_t pte) } #define pte_accessible pte_accessible -static inline unsigned long pte_accessible(pte_t a) +static inline unsigned long pte_accessible(struct mm_struct *mm, pte_t a) { return pte_val(a) & _PAGE_VALID; } @@ -847,7 +847,7 @@ static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr, * SUN4V NOTE: _PAGE_VALID is the same value in both the SUN4U * and SUN4V pte layout, so this inline test is fine. */ - if (likely(mm != &init_mm) && pte_accessible(orig)) + if (likely(mm != &init_mm) && pte_accessible(mm, orig)) tlb_batch_add(mm, addr, ptep, orig, fullmm); } diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 3d1999458709..bbc8b12fa443 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -452,9 +452,16 @@ static inline int pte_present(pte_t a) } #define pte_accessible pte_accessible -static inline int pte_accessible(pte_t a) +static inline bool pte_accessible(struct mm_struct *mm, pte_t a) { - return pte_flags(a) & _PAGE_PRESENT; + if (pte_flags(a) & _PAGE_PRESENT) + return true; + + if ((pte_flags(a) & (_PAGE_PROTNONE | _PAGE_NUMA)) && + mm_tlb_flush_pending(mm)) + return true; + + return false; } static inline int pte_hidden(pte_t pte) diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h index f330d28e4d0e..b12079afbd5f 100644 --- a/include/asm-generic/pgtable.h +++ b/include/asm-generic/pgtable.h @@ -217,7 +217,7 @@ static inline int pmd_same(pmd_t pmd_a, pmd_t pmd_b) #endif #ifndef pte_accessible -# define pte_accessible(pte) ((void)(pte),1) +# define pte_accessible(mm, pte) ((void)(pte), 1) #endif #ifndef flush_tlb_fix_spurious_fault diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index bd299418a934..e5c49c30460f 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -442,6 +442,14 @@ struct mm_struct { /* numa_scan_seq prevents two threads setting pte_numa */ int numa_scan_seq; +#endif +#if defined(CONFIG_NUMA_BALANCING) || defined(CONFIG_COMPACTION) + /* + * An operation with batched TLB flushing is going on. Anything that + * can move process memory needs to flush the TLB when moving a + * PROT_NONE or PROT_NUMA mapped page. + */ + bool tlb_flush_pending; #endif struct uprobes_state uprobes_state; }; @@ -459,4 +467,40 @@ static inline cpumask_t *mm_cpumask(struct mm_struct *mm) return mm->cpu_vm_mask_var; } +#if defined(CONFIG_NUMA_BALANCING) || defined(CONFIG_COMPACTION) +/* + * Memory barriers to keep this state in sync are graciously provided by + * the page table locks, outside of which no page table modifications happen. + * The barriers below prevent the compiler from re-ordering the instructions + * around the memory barriers that are already present in the code. + */ +static inline bool mm_tlb_flush_pending(struct mm_struct *mm) +{ + barrier(); + return mm->tlb_flush_pending; +} +static inline void set_tlb_flush_pending(struct mm_struct *mm) +{ + mm->tlb_flush_pending = true; + barrier(); +} +/* Clearing is done after a TLB flush, which also provides a barrier. */ +static inline void clear_tlb_flush_pending(struct mm_struct *mm) +{ + barrier(); + mm->tlb_flush_pending = false; +} +#else +static inline bool mm_tlb_flush_pending(struct mm_struct *mm) +{ + return false; +} +static inline void set_tlb_flush_pending(struct mm_struct *mm) +{ +} +static inline void clear_tlb_flush_pending(struct mm_struct *mm) +{ +} +#endif + #endif /* _LINUX_MM_TYPES_H */ diff --git a/kernel/fork.c b/kernel/fork.c index 728d5be9548c..5721f0e3f2da 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -537,6 +537,7 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p) spin_lock_init(&mm->page_table_lock); mm_init_aio(mm); mm_init_owner(mm, p); + clear_tlb_flush_pending(mm); if (likely(!mm_alloc_pgd(mm))) { mm->def_flags = 0; diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 7de1bf85f683..3d2783e10596 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1376,6 +1376,13 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, goto clear_pmdnuma; } + /* + * The page_table_lock above provides a memory barrier + * with change_protection_range. + */ + if (mm_tlb_flush_pending(mm)) + flush_tlb_range(vma, haddr, haddr + HPAGE_PMD_SIZE); + /* * Migrate the THP to the requested node, returns with page unlocked * and pmd_numa cleared. diff --git a/mm/mprotect.c b/mm/mprotect.c index f8421722acb9..bb53a6591aea 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -188,6 +188,7 @@ static unsigned long change_protection_range(struct vm_area_struct *vma, BUG_ON(addr >= end); pgd = pgd_offset(mm, addr); flush_cache_range(vma, addr, end); + set_tlb_flush_pending(mm); do { next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(pgd)) @@ -199,6 +200,7 @@ static unsigned long change_protection_range(struct vm_area_struct *vma, /* Only flush the TLB if we actually modified any entries: */ if (pages) flush_tlb_range(vma, start, end); + clear_tlb_flush_pending(mm); return pages; } diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c index e84cad27a801..a8b919925934 100644 --- a/mm/pgtable-generic.c +++ b/mm/pgtable-generic.c @@ -110,9 +110,10 @@ int pmdp_clear_flush_young(struct vm_area_struct *vma, pte_t ptep_clear_flush(struct vm_area_struct *vma, unsigned long address, pte_t *ptep) { + struct mm_struct *mm = (vma)->vm_mm; pte_t pte; - pte = ptep_get_and_clear((vma)->vm_mm, address, ptep); - if (pte_accessible(pte)) + pte = ptep_get_and_clear(mm, address, ptep); + if (pte_accessible(mm, pte)) flush_tlb_page(vma, address); return pte; } -- cgit v1.2.3 From 84ed8a99058e61567f495cc43118344261641c5f Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Wed, 18 Dec 2013 17:08:48 -0800 Subject: sh: always link in helper functions extracted from libgcc E.g. landisk_defconfig, which has CONFIG_NTFS_FS=m: ERROR: "__ashrdi3" [fs/ntfs/ntfs.ko] undefined! For "lib-y", if no symbols in a compilation unit are referenced by other units, the compilation unit will not be included in vmlinux. This breaks modules that do reference those symbols. Use "obj-y" instead to fix this. http://kisskb.ellerman.id.au/kisskb/buildresult/8838077/ This doesn't fix all cases. There are others, e.g. udivsi3. This is also not limited to sh, many architectures handle this in the same way. A simple solution is to unconditionally include all helper functions. A more complex solution is to make the choice of "lib-y" or "obj-y" depend on CONFIG_MODULES: obj-$(CONFIG_MODULES) += ... lib-y($CONFIG_MODULES) += ... Signed-off-by: Geert Uytterhoeven Cc: Paul Mundt Tested-by: Nobuhiro Iwamatsu Reviewed-by: Nobuhiro Iwamatsu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/sh/lib/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/sh/lib/Makefile b/arch/sh/lib/Makefile index 7b95f29e3174..3baff31e58cf 100644 --- a/arch/sh/lib/Makefile +++ b/arch/sh/lib/Makefile @@ -6,7 +6,7 @@ lib-y = delay.o memmove.o memchr.o \ checksum.o strlen.o div64.o div64-generic.o # Extracted from libgcc -lib-y += movmem.o ashldi3.o ashrdi3.o lshrdi3.o \ +obj-y += movmem.o ashldi3.o ashrdi3.o lshrdi3.o \ ashlsi3.o ashrsi3.o ashiftrt.o lshrsi3.o \ udiv_qrnnd.o -- cgit v1.2.3 From 1e01c7eb7c431a74437d73fe54670398b4d2b222 Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Thu, 19 Dec 2013 18:55:58 +0530 Subject: ARC: Allow conditional multiple inclusion of uapi/asm/unistd.h Commit 97bc386fc12d "ARC: Add guard macro to uapi/asm/unistd.h" inhibited multiple inclusion of ARCH unistd.h. This however hosed the system since Generic syscall table generator relies on it being included twice, and in lack-of an empty table was emitted by C preprocessor. Fix that by allowing one exception to rule for the special case (just like Xtensa) Suggested-by: Chen Gang Signed-off-by: Vineet Gupta --- arch/arc/include/uapi/asm/unistd.h | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/arc/include/uapi/asm/unistd.h b/arch/arc/include/uapi/asm/unistd.h index 68125dd766c6..39e58d1cdf90 100644 --- a/arch/arc/include/uapi/asm/unistd.h +++ b/arch/arc/include/uapi/asm/unistd.h @@ -8,7 +8,11 @@ /******** no-legacy-syscalls-ABI *******/ -#ifndef _UAPI_ASM_ARC_UNISTD_H +/* + * Non-typical guard macro to enable inclusion twice in ARCH sys.c + * That is how the Generic syscall wrapper generator works + */ +#if !defined(_UAPI_ASM_ARC_UNISTD_H) || defined(__SYSCALL) #define _UAPI_ASM_ARC_UNISTD_H #define __ARCH_WANT_SYS_EXECVE @@ -36,4 +40,6 @@ __SYSCALL(__NR_arc_gettls, sys_arc_gettls) #define __NR_sysfs (__NR_arch_specific_syscall + 3) __SYSCALL(__NR_sysfs, sys_sysfs) +#undef __SYSCALL + #endif -- cgit v1.2.3 From cdc27c27843248ae7eb0df5fc261dd004eaa5670 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 17 Dec 2013 17:09:08 +0000 Subject: arm64: ptrace: avoid using HW_BREAKPOINT_EMPTY for disabled events Commit 8f34a1da35ae ("arm64: ptrace: use HW_BREAKPOINT_EMPTY type for disabled breakpoints") fixed an issue with GDB trying to zero breakpoint control registers. The problem there is that the arch hw_breakpoint code will attempt to create a (disabled), execute breakpoint of length 0. This will fail validation and report unexpected failure to GDB. To avoid this, we treated disabled breakpoints as HW_BREAKPOINT_EMPTY, but that seems to have broken with recent kernels, causing watchpoints to be treated as TYPE_INST in the core code and returning ENOSPC for any further breakpoints. This patch fixes the problem by prioritising the `enable' field of the breakpoint: if it is cleared, we simply update the perf_event_attr to indicate that the thing is disabled and don't bother changing either the type or the length. This reinforces the behaviour that the breakpoint control register is essentially read-only apart from the enable bit when disabling a breakpoint. Cc: Reported-by: Aaron Liu Signed-off-by: Will Deacon Signed-off-by: Catalin Marinas --- arch/arm64/kernel/ptrace.c | 38 ++++++++++++++++++-------------------- 1 file changed, 18 insertions(+), 20 deletions(-) (limited to 'arch') diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c index 6777a2192b83..6a8928bba03c 100644 --- a/arch/arm64/kernel/ptrace.c +++ b/arch/arm64/kernel/ptrace.c @@ -214,31 +214,29 @@ static int ptrace_hbp_fill_attr_ctrl(unsigned int note_type, { int err, len, type, disabled = !ctrl.enabled; - if (disabled) { - len = 0; - type = HW_BREAKPOINT_EMPTY; - } else { - err = arch_bp_generic_fields(ctrl, &len, &type); - if (err) - return err; - - switch (note_type) { - case NT_ARM_HW_BREAK: - if ((type & HW_BREAKPOINT_X) != type) - return -EINVAL; - break; - case NT_ARM_HW_WATCH: - if ((type & HW_BREAKPOINT_RW) != type) - return -EINVAL; - break; - default: + attr->disabled = disabled; + if (disabled) + return 0; + + err = arch_bp_generic_fields(ctrl, &len, &type); + if (err) + return err; + + switch (note_type) { + case NT_ARM_HW_BREAK: + if ((type & HW_BREAKPOINT_X) != type) return -EINVAL; - } + break; + case NT_ARM_HW_WATCH: + if ((type & HW_BREAKPOINT_RW) != type) + return -EINVAL; + break; + default: + return -EINVAL; } attr->bp_len = len; attr->bp_type = type; - attr->disabled = disabled; return 0; } -- cgit v1.2.3 From 38958c15dc640a9249e4f0cd0dfb0ddc7a23464d Mon Sep 17 00:00:00 2001 From: Rajendra Nayak Date: Thu, 12 Dec 2013 15:22:49 +0530 Subject: ARM: DRA7: hwmod: Fix boot crash with DEBUG_LL With commit '7dedd34: ARM: OMAP2+: hwmod: Fix a crash in _setup_reset() with DEBUG_LL' we moved from parsing cmdline to identify uart used for earlycon to using the requsite hwmod CONFIG_DEBUG_OMAPxUARTy FLAGS. On DRA7 though, we seem to be missing this flag, and atleast on the DRA7 EVM where we use uart1 for console, boot fails with DEBUG_LL enabled. Reported-by: Lokesh Vutla Tested-by: Lokesh Vutla # on a different base Signed-off-by: Rajendra Nayak Fixes: 7dedd346941d ("ARM: OMAP2+: hwmod: Fix a crash in _setup_reset() with DEBUG_LL") Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_7xx_data.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/arm/mach-omap2/omap_hwmod_7xx_data.c b/arch/arm/mach-omap2/omap_hwmod_7xx_data.c index db32d5380b11..18f333c440db 100644 --- a/arch/arm/mach-omap2/omap_hwmod_7xx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_7xx_data.c @@ -1637,7 +1637,7 @@ static struct omap_hwmod dra7xx_uart1_hwmod = { .class = &dra7xx_uart_hwmod_class, .clkdm_name = "l4per_clkdm", .main_clk = "uart1_gfclk_mux", - .flags = HWMOD_SWSUP_SIDLE_ACT, + .flags = HWMOD_SWSUP_SIDLE_ACT | DEBUG_OMAP2UART1_FLAGS, .prcm = { .omap4 = { .clkctrl_offs = DRA7XX_CM_L4PER_UART1_CLKCTRL_OFFSET, -- cgit v1.2.3 From 6d4c88304794442055eaea1c07f3c7b988b8c924 Mon Sep 17 00:00:00 2001 From: Suman Anna Date: Mon, 23 Dec 2013 16:53:11 -0600 Subject: ARM: OMAP2+: hwmod_data: fix missing OMAP_INTC_START in irq data Commit 7d7e1eb (ARM: OMAP2+: Prepare for irqs.h removal) and commit ec2c082 (ARM: OMAP2+: Remove hardcoded IRQs and enable SPARSE_IRQ) updated the way interrupts for OMAP2/3 devices are defined in the HWMOD data structures to being an index plus a fixed offset (defined by OMAP_INTC_START). Couple of irqs in the OMAP2/3 hwmod data were misconfigured completely as they were missing this OMAP_INTC_START relative offset. Add this offset back to fix the incorrect irq data for the following modules: OMAP2 - GPMC, RNG OMAP3 - GPMC, ISP MMU & IVA MMU Signed-off-by: Suman Anna Fixes: 7d7e1eba7e92 ("ARM: OMAP2+: Prepare for irqs.h removal") Fixes: ec2c0825ca31 ("ARM: OMAP2+: Remove hardcoded IRQs and enable SPARSE_IRQ") Cc: Tony Lindgren Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_2xxx_ipblock_data.c | 4 ++-- arch/arm/mach-omap2/omap_hwmod_3xxx_data.c | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'arch') diff --git a/arch/arm/mach-omap2/omap_hwmod_2xxx_ipblock_data.c b/arch/arm/mach-omap2/omap_hwmod_2xxx_ipblock_data.c index 56cebb05509e..d23c77fadb31 100644 --- a/arch/arm/mach-omap2/omap_hwmod_2xxx_ipblock_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_2xxx_ipblock_data.c @@ -796,7 +796,7 @@ struct omap_hwmod omap2xxx_counter_32k_hwmod = { /* gpmc */ static struct omap_hwmod_irq_info omap2xxx_gpmc_irqs[] = { - { .irq = 20 }, + { .irq = 20 + OMAP_INTC_START, }, { .irq = -1 } }; @@ -841,7 +841,7 @@ static struct omap_hwmod_class omap2_rng_hwmod_class = { }; static struct omap_hwmod_irq_info omap2_rng_mpu_irqs[] = { - { .irq = 52 }, + { .irq = 52 + OMAP_INTC_START, }, { .irq = -1 } }; diff --git a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c index 9e56fabd7fa3..3bfb2db674ae 100644 --- a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c @@ -2172,7 +2172,7 @@ static struct omap_hwmod_class omap3xxx_gpmc_hwmod_class = { }; static struct omap_hwmod_irq_info omap3xxx_gpmc_irqs[] = { - { .irq = 20 }, + { .irq = 20 + OMAP_INTC_START, }, { .irq = -1 } }; @@ -3006,7 +3006,7 @@ static struct omap_mmu_dev_attr mmu_isp_dev_attr = { static struct omap_hwmod omap3xxx_mmu_isp_hwmod; static struct omap_hwmod_irq_info omap3xxx_mmu_isp_irqs[] = { - { .irq = 24 }, + { .irq = 24 + OMAP_INTC_START, }, { .irq = -1 } }; @@ -3048,7 +3048,7 @@ static struct omap_mmu_dev_attr mmu_iva_dev_attr = { static struct omap_hwmod omap3xxx_mmu_iva_hwmod; static struct omap_hwmod_irq_info omap3xxx_mmu_iva_irqs[] = { - { .irq = 28 }, + { .irq = 28 + OMAP_INTC_START, }, { .irq = -1 } }; -- cgit v1.2.3 From 7e367c18c059c638bf6fb540f1decec18d64cb55 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Fri, 27 Dec 2013 09:33:27 -0800 Subject: ARM: OMAP2+: Fix LCD panel backlight regression for LDP legacy booting Looks like the LCD panel on LDP has been broken quite a while, and recently got fixed by commit 0b2aa8bed3e1 (gpio: twl4030: Fix regression for twl gpio output). However, there's still an issue left where the panel backlight does not come on if the LCD drivers are built into the kernel. Fix the issue by registering the DPI LCD panel only after the twl4030 GPIO has probed. Reported-by: Russell King Acked-by: Tomi Valkeinen [tony@atomide.com: updated per Tomi's comments] Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/board-ldp.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/arm/mach-omap2/board-ldp.c b/arch/arm/mach-omap2/board-ldp.c index 4ec8d82b0492..44a59c3abfb0 100644 --- a/arch/arm/mach-omap2/board-ldp.c +++ b/arch/arm/mach-omap2/board-ldp.c @@ -242,12 +242,18 @@ static void __init ldp_display_init(void) static int ldp_twl_gpio_setup(struct device *dev, unsigned gpio, unsigned ngpio) { + int res; + /* LCD enable GPIO */ ldp_lcd_pdata.enable_gpio = gpio + 7; /* Backlight enable GPIO */ ldp_lcd_pdata.backlight_gpio = gpio + 15; + res = platform_device_register(&ldp_lcd_device); + if (res) + pr_err("Unable to register LCD: %d\n", res); + return 0; } @@ -346,7 +352,6 @@ static struct omap2_hsmmc_info mmc[] __initdata = { static struct platform_device *ldp_devices[] __initdata = { &ldp_gpio_keys_device, - &ldp_lcd_device, }; #ifdef CONFIG_OMAP_MUX -- cgit v1.2.3