[PATCH V3 4/8] nvme: add sysfs attribute to change IO timeout per nvme controller

[Maurizio Lombardi]

On Tue Apr 28, 2026 at 7:23 PM CEST, Mohamed Khalfella wrote:
> On Fri 2026-04-10 09:39:20 +0200, Maurizio Lombardi wrote:
>> Currently, there is no method to adjust the timeout values on a
>> per controller basis with nvme I/O queues.
>> Add an io_timeout attribute to nvme so that different nvme controllers
>> which may have different timeout requirements can have custom
>> I/O timeouts set.
>> 
>> Signed-off-by: Maurizio Lombardi <mlombard at redhat.com>
> Reviewed-by: Mohamed Khalfella <mkhalfella at purestorage.com>
>
>> diff --git a/drivers/nvme/host/sysfs.c b/drivers/nvme/host/sysfs.c
>> index 000c29bd1f1f..1e6a0eecb30e 100644
>> --- a/drivers/nvme/host/sysfs.c
>> +++ b/drivers/nvme/host/sysfs.c
>> @@ -664,6 +664,52 @@ static ssize_t nvme_admin_timeout_store(struct device *dev,
>>  static DEVICE_ATTR(admin_timeout, S_IRUGO | S_IWUSR,
>>  	nvme_admin_timeout_show, nvme_admin_timeout_store);
>>  
>> +static ssize_t nvme_io_timeout_show(struct device *dev,
>> +			struct device_attribute *attr, char *buf)
>> +{
>> +	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);
>> +
>> +	return sysfs_emit(buf, "%u\n", jiffies_to_msecs(ctrl->io_timeout));
>> +}
>> +
>> +static ssize_t nvme_io_timeout_store(struct device *dev,
>> +			struct device_attribute *attr,
>> +			const char *buf, size_t count)
>> +{
>> +	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);
>> +	struct nvme_ns *ns;
>> +	u32 timeout;
>> +	int err;
>> +
>> +	/*
>> +	 * Wait until the controller reaches the LIVE state
>> +	 * to be sure that connect_q is properly initialized.
>> +	 */
>> +	if (!test_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags))
>> +		return -EBUSY;
>> +
>> +	err = kstrtou32(buf, 10, &timeout);
>> +	if (err || !timeout)
>> +		return -EINVAL;
>> +
>> +	/* Take the namespaces_lock to avoid racing against nvme_alloc_ns() */
>> +	mutex_lock(&ctrl->namespaces_lock);
>> +
>> +	ctrl->io_timeout = msecs_to_jiffies(timeout);
>> +	list_for_each_entry(ns, &ctrl->namespaces, list)
>> +		blk_queue_rq_timeout(ns->queue, ctrl->io_timeout);
>> +
>> +	if (ctrl->connect_q)
>> +		blk_queue_rq_timeout(ctrl->connect_q, ctrl->io_timeout);
>> +
>> +	mutex_unlock(&ctrl->namespaces_lock);
>
> ctrl->namespaces_lock is not needed to set timeout for ctrl->connect_q.
> Maybe move mutex_unlock() up just after iterating on namespaces?

Yes, it can be unlocked just after scanning the namespaces' list.

Thanks,
Maurizio