iommu/vt-d: queue_iova() boot crash backport
Subsystems (3)
cover iommu/vt-d: queue_iova() boot crash backport
Backport commits from master that fix boot failure on some intel machines.
I have only boot tested this in a VM. Functional testing for v4.14 is out of my scope as patches differ only on a trivial conflict from v4.19, where I discovered/debugged the issue. While testing v4.14 stable on affected nodes would require porting some other [local] patches, which is out of my timelimit for the backport.
Hopefully, that's fine.
Dmitry Safonov (1):
iommu/vt-d: Don't queue_iova() if there is no flush queue
Joerg Roedel (1):
iommu/iova: Fix compilation error with !CONFIG_IOMMU_IOVA
drivers/iommu/intel-iommu.c | 2 +- drivers/iommu/iova.c | 18 ++++++++++++++---- include/linux/iova.h | 6 ++++++ 3 files changed, 21 insertions(+), 5 deletions(-)
Signature · 1 line
2.22.0
PATCH 2/2 iommu/iova: Fix compilation error with !CONFIG_IOMMU_IOVA
From: Joerg Roedel <[email protected]>
[ Upstream commit 201c1db90cd643282185a00770f12f95da330eca ]
The stub function for !CONFIG_IOMMU_IOVA needs to be 'static inline'.
--- include/linux/iova.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/iova.h +1 -1
@@ -235,7 +235,7 @@ static inline void init_iova_domain(struct iova_domain *iovad, { } -bool has_iova_flush_queue (struct iova_domain *iovad) +static inline bool has_iova_flush_queue (struct iova_domain *iovad) {return false; }
Signature · 1 line
2.22.0
PATCH 1/2 iommu/vt-d: Don't queue_iova() if there is no flush queue
In reply to Dmitry Safonov via iommu[ Upstream commit effa467870c7612012885df4e246bdb8ffd8e44c ]
Intel VT-d driver was reworked to use common deferred flushing implementation. Previously there was one global per-cpu flush queue, afterwards - one per domain.
Before deferring a flush, the queue should be allocated and initialized.
Currently only domains with IOMMU_DOMAIN_DMA type initialize their flush queue. It's probably worth to init it for static or unmanaged domains too, but it may be arguable - I'm leaving it to iommu folks.
Prevent queuing an iova flush if the domain doesn't have a queue. The defensive check seems to be worth to keep even if queue would be initialized for all kinds of domains. And is easy backportable.
On 4.19.43 stable kernel it has a user-visible effect: previously for devices in si domain there were crashes, on sata devices:
BUG: spinlock bad magic on CPU#6, swapper/0/1 lock: 0xffff88844f582008, .magic: 00000000, .owner: <none>/-1, .owner_cpu: 0 CPU: 6 PID: 1 Comm: swapper/0 Not tainted 4.19.43 #1 Call Trace: <IRQ> dump_stack+0x61/0x7e spin_bug+0x9d/0xa3 do_raw_spin_lock+0x22/0x8e _raw_spin_lock_irqsave+0x32/0x3a queue_iova+0x45/0x115 intel_unmap+0x107/0x113 intel_unmap_sg+0x6b/0x76 __ata_qc_complete+0x7f/0x103 ata_qc_complete+0x9b/0x26a ata_qc_complete_multiple+0xd0/0xe3 ahci_handle_port_interrupt+0x3ee/0x48a ahci_handle_port_intr+0x73/0xa9 ahci_single_level_irq_intr+0x40/0x60 __handle_irq_event_percpu+0x7f/0x19a handle_irq_event_percpu+0x32/0x72 handle_irq_event+0x38/0x56 handle_edge_irq+0x102/0x121 handle_irq+0x147/0x15c do_IRQ+0x66/0xf2 common_interrupt+0xf/0xf RIP: 0010:__do_softirq+0x8c/0x2df
The same for usb devices that use ehci-pci:
BUG: spinlock bad magic on CPU#0, swapper/0/1 lock: 0xffff88844f402008, .magic: 00000000, .owner: <none>/-1, .owner_cpu: 0 CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.19.43 #4 Call Trace: <IRQ> dump_stack+0x61/0x7e spin_bug+0x9d/0xa3 do_raw_spin_lock+0x22/0x8e _raw_spin_lock_irqsave+0x32/0x3a queue_iova+0x77/0x145 intel_unmap+0x107/0x113 intel_unmap_page+0xe/0x10 usb_hcd_unmap_urb_setup_for_dma+0x53/0x9d usb_hcd_unmap_urb_for_dma+0x17/0x100 unmap_urb_for_dma+0x22/0x24 __usb_hcd_giveback_urb+0x51/0xc3 usb_giveback_urb_bh+0x97/0xde tasklet_action_common.isra.4+0x5f/0xa1 tasklet_action+0x2d/0x30 __do_softirq+0x138/0x2df irq_exit+0x7d/0x8b smp_apic_timer_interrupt+0x10f/0x151 apic_timer_interrupt+0xf/0x20 </IRQ> RIP: 0010:_raw_spin_unlock_irqrestore+0x17/0x39
[v4.14-port notes:
o minor conflict with untrusted IOMMU devices check under if-condition
o setup_timer() near one chunk is timer_setup() in v5.3]
--- drivers/iommu/intel-iommu.c | 2 +- drivers/iommu/iova.c | 18 ++++++++++++++---- include/linux/iova.h | 6 ++++++ 3 files changed, 21 insertions(+), 5 deletions(-)
drivers/iommu/intel-iommu.c +1 -1
@@ -3702,7 +3702,7 @@ static void intel_unmap(struct device *dev, dma_addr_t dev_addr, size_t size) freelist =domain_unmap (domain, start_pfn, last_pfn); -if (intel_iommu_strict) { +if (intel_iommu_strict || !has_iova_flush_queue (&domain->iovad)) {iommu_flush_iotlb_psi (iommu, domain, start_pfn, nrpages, !freelist,0 );/* free iova */
drivers/iommu/iova.c +14 -4
@@ -58,9 +58,14 @@ init_iova_domain(struct iova_domain *iovad, unsigned long granule, }EXPORT_SYMBOL_GPL (init_iova_domain); +bool has_iova_flush_queue (struct iova_domain *iovad) +{ +return !!iovad->fq; +} +static void free_iova_flush_queue (struct iova_domain *iovad) { -if (!iovad->fq) +if (!has_iova_flush_queue (iovad))return ;if (timer_pending (&iovad->fq_timer))
@@ -78,13 +83,14 @@ static void free_iova_flush_queue(struct iova_domain *iovad)int init_iova_flush_queue (struct iova_domain *iovad,iova_flush_cb flush_cb,iova_entry_dtor entry_dtor) { +struct iova_fq __percpu *queue;int cpu;atomic64_set (&iovad->fq_flush_start_cnt,0 );atomic64_set (&iovad->fq_flush_finish_cnt,0 ); - iovad->fq = alloc_percpu(struct iova_fq); -if (!iovad->fq) + queue = alloc_percpu(struct iova_fq); +if (!queue)return -ENOMEM ; iovad->flush_cb = flush_cb;
@@ -93,13 +99,17 @@ int init_iova_flush_queue(struct iova_domain *iovad,for_each_possible_cpu (cpu) {struct iova_fq *fq; - fq =per_cpu_ptr (iovad->fq, cpu); + fq =per_cpu_ptr (queue, cpu); fq->head =0 ; fq->tail =0 ;spin_lock_init (&fq->lock); } +smp_wmb (); + + iovad->fq = queue; +setup_timer (&iovad->fq_timer, fq_flush_timeout, (unsigned long )iovad);atomic_set (&iovad->fq_timer_on,0 );
include/linux/iova.h +6 -0
@@ -154,6 +154,7 @@ struct iova *reserve_iova(struct iova_domain *iovad, unsigned long pfn_lo,void copy_reserved_iova (struct iova_domain *from,struct iova_domain *to);void init_iova_domain (struct iova_domain *iovad,unsigned long granule,unsigned long start_pfn,unsigned long pfn_32bit); +bool has_iova_flush_queue (struct iova_domain *iovad);int init_iova_flush_queue (struct iova_domain *iovad,iova_flush_cb flush_cb,iova_entry_dtor entry_dtor);struct iova *find_iova (struct iova_domain *iovad,unsigned long pfn);
@@ -234,6 +235,11 @@ static inline void init_iova_domain(struct iova_domain *iovad, { } +bool has_iova_flush_queue (struct iova_domain *iovad) +{ +return false; +} +static inline int init_iova_flush_queue (struct iova_domain *iovad,iova_flush_cb flush_cb,iova_entry_dtor entry_dtor)
Signature · 1 line
2.22.0