@@ -342,7 +342,7 @@ static void idxd_cdev_evl_drain_pasid(struct idxd_wq *wq, u32 pasid)
if (!evl)
return;
- spin_lock(&evl->lock);
+ mutex_lock(&evl->lock);
status.bits = ioread64(idxd->reg_base + IDXD_EVLSTATUS_OFFSET);
t = status.tail;
h = evl->head;
@@ -354,9 +354,8 @@ static void idxd_cdev_evl_drain_pasid(struct idxd_wq *wq, u32 pasid)
set_bit(h, evl->bmap);
h = (h + 1) % size;
}
- spin_unlock(&evl->lock);
-
drain_workqueue(wq->wq);
+ mutex_unlock(&evl->lock);
}
static int idxd_cdev_release(struct inode *node, struct file *filep)
@@ -66,7 +66,7 @@ static int debugfs_evl_show(struct seq_file *s, void *d)
if (!evl || !evl->log)
return 0;
- spin_lock(&evl->lock);
+ mutex_lock(&evl->lock);
h = evl->head;
evl_status.bits = ioread64(idxd->reg_base + IDXD_EVLSTATUS_OFFSET);
@@ -87,7 +87,7 @@ static int debugfs_evl_show(struct seq_file *s, void *d)
dump_event_entry(idxd, s, i, &count, processed);
}
- spin_unlock(&evl->lock);
+ mutex_unlock(&evl->lock);
return 0;
}
@@ -775,7 +775,7 @@ static int idxd_device_evl_setup(struct idxd_device *idxd)
goto err_alloc;
}
- spin_lock(&evl->lock);
+ mutex_lock(&evl->lock);
evl->log = addr;
evl->dma = dma_addr;
evl->log_size = size;
@@ -796,7 +796,7 @@ static int idxd_device_evl_setup(struct idxd_device *idxd)
gencfg.evl_en = 1;
iowrite32(gencfg.bits, idxd->reg_base + IDXD_GENCFG_OFFSET);
- spin_unlock(&evl->lock);
+ mutex_unlock(&evl->lock);
return 0;
err_alloc:
@@ -819,7 +819,7 @@ static void idxd_device_evl_free(struct idxd_device *idxd)
if (!gencfg.evl_en)
return;
- spin_lock(&evl->lock);
+ mutex_lock(&evl->lock);
gencfg.evl_en = 0;
iowrite32(gencfg.bits, idxd->reg_base + IDXD_GENCFG_OFFSET);
@@ -836,7 +836,7 @@ static void idxd_device_evl_free(struct idxd_device *idxd)
evl_dma = evl->dma;
evl->log = NULL;
evl->size = IDXD_EVL_SIZE_MIN;
- spin_unlock(&evl->lock);
+ mutex_unlock(&evl->lock);
dma_free_coherent(dev, evl_log_size, evl_log, evl_dma);
}
@@ -293,7 +293,7 @@ struct idxd_driver_data {
struct idxd_evl {
/* Lock to protect event log access. */
- spinlock_t lock;
+ struct mutex lock;
void *log;
dma_addr_t dma;
/* Total size of event log = number of entries * entry size. */
@@ -352,7 +352,7 @@ static int idxd_init_evl(struct idxd_device *idxd)
if (!evl)
return -ENOMEM;
- spin_lock_init(&evl->lock);
+ mutex_init(&evl->lock);
evl->size = IDXD_EVL_SIZE_MIN;
idxd->evl_cache = kmem_cache_create(dev_name(idxd_confdev(idxd)),
@@ -363,7 +363,7 @@ static void process_evl_entries(struct idxd_device *idxd)
evl_status.bits = 0;
evl_status.int_pending = 1;
- spin_lock(&evl->lock);
+ mutex_lock(&evl->lock);
/* Clear interrupt pending bit */
iowrite32(evl_status.bits_upper32,
idxd->reg_base + IDXD_EVLSTATUS_OFFSET + sizeof(u32));
@@ -381,7 +381,7 @@ static void process_evl_entries(struct idxd_device *idxd)
evl->head = h;
evl_status.head = h;
iowrite32(evl_status.bits_lower32, idxd->reg_base + IDXD_EVLSTATUS_OFFSET);
- spin_unlock(&evl->lock);
+ mutex_unlock(&evl->lock);
}
irqreturn_t idxd_misc_thread(int vec, void *data)
The drain_workqueue() is not in a locked context. In the multi-task case, it's possible to call queue_work() when drain_workqueue() is ongoing, then it can cause Call Trace due to pushing a work into a draining workqueue: Call Trace: <TASK> ? __warn+0x7d/0x140 ? __queue_work+0x2b2/0x440 ? report_bug+0x1f8/0x200 ? handle_bug+0x3c/0x70 ? exc_invalid_op+0x18/0x70 ? asm_exc_invalid_op+0x1a/0x20 ? __queue_work+0x2b2/0x440 queue_work_on+0x28/0x30 idxd_misc_thread+0x303/0x5a0 [idxd] ? __schedule+0x369/0xb40 ? __pfx_irq_thread_fn+0x10/0x10 ? irq_thread+0xbc/0x1b0 irq_thread_fn+0x21/0x70 irq_thread+0x102/0x1b0 ? preempt_count_add+0x74/0xa0 ? __pfx_irq_thread_dtor+0x10/0x10 ? __pfx_irq_thread+0x10/0x10 kthread+0x103/0x140 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x31/0x50 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1b/0x30 </TASK> The original lock for event log was spinlock, drain_workqueue() can't be in a spinlocked context because it may cause task rescheduling. The spinlock was called in thread and irq thread context, the current usages does not require a spinlock for event log, so it's feasible to convert spinlock to mutex. For putting drain_workqueue() into a locked context, convert the spinlock to mutex, then lock drain_workqueue() by mutex. Fixes: c40bd7d9737b ("dmaengine: idxd: process user page faults for completion record") Signed-off-by: Rex Zhang <rex.zhang@intel.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Reviewed-by: Fenghua Yu <fenghua.yu@intel.com> Reviewed-by: Lijun Pan <lijun.pan@intel.com> --- Changelog: v2: - Update commit message. v3: - Rebase on v6.8-rc1 --- drivers/dma/idxd/cdev.c | 5 ++--- drivers/dma/idxd/debugfs.c | 4 ++-- drivers/dma/idxd/device.c | 8 ++++---- drivers/dma/idxd/idxd.h | 2 +- drivers/dma/idxd/init.c | 2 +- drivers/dma/idxd/irq.c | 4 ++-- 6 files changed, 12 insertions(+), 13 deletions(-)