Re: [PATCH v2 2/5] HID: intel_ish-hid: clarify locking in client code

From: Srinivas Pandruvada
Date: Fri May 26 2017 - 21:34:55 EST


On Thu, 2017-05-18 at 22:21 +0200, Arnd Bergmann wrote:
> I was trying to understand this code while working on a warning
> fix and the locking made no sense: spin_lock_irqsave() is pointless
> when run inside of an interrupt handler or nested inside of another
> spin_lock_irq() or spin_lock_irqsave().
>
> Here it turned out that the comment above the function is wrong,
> as both recv_ishtp_cl_msg_dma() and recv_ishtp_cl_msg() can in fact
> be called from a work queue rather than an ISR, so we do have to
> use the irqsave() version once.
>
> This fixes the comments accordingly, removes the misleading
> 'dev_flags'
> variable and modifies the inner spinlock to not use 'irqsave'.
>
> No functional change is intended, this is just for readability and
> it slightly simplifies the object code.
>
> Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx>
ÂAcked-by: Srinivas Pandruvada <srinivas.pandruvada@xxxxxxxxxxxxxxx>

> ---
> Âdrivers/hid/intel-ish-hid/ishtp/client.c | 43 +++++++++++++---------
> ----------
> Â1 file changed, 17 insertions(+), 26 deletions(-)
>
> diff --git a/drivers/hid/intel-ish-hid/ishtp/client.c
> b/drivers/hid/intel-ish-hid/ishtp/client.c
> index 78d393e616a4..f54689ee67e1 100644
> --- a/drivers/hid/intel-ish-hid/ishtp/client.c
> +++ b/drivers/hid/intel-ish-hid/ishtp/client.c
> @@ -803,7 +803,7 @@ void ishtp_cl_send_msg(struct ishtp_device *dev,
> struct ishtp_cl *cl)
> Â * @ishtp_hdr: Pointer to message header
> Â *
> Â * Receive and dispatch ISHTP client messages. This function
> executes in ISR
> - * context
> + * or work queue context
> Â */
> Âvoid recv_ishtp_cl_msg(struct ishtp_device *dev,
> Â ÂÂÂÂÂÂÂstruct ishtp_msg_hdr *ishtp_hdr)
> @@ -813,7 +813,6 @@ void recv_ishtp_cl_msg(struct ishtp_device *dev,
> Â struct ishtp_cl_rb *new_rb;
> Â unsigned char *buffer = NULL;
> Â struct ishtp_cl_rb *complete_rb = NULL;
> - unsigned long dev_flags;
> Â unsigned long flags;
> Â int rb_count;
> Â
> @@ -828,7 +827,7 @@ void recv_ishtp_cl_msg(struct ishtp_device *dev,
> Â goto eoi;
> Â }
> Â
> - spin_lock_irqsave(&dev->read_list_spinlock, dev_flags);
> + spin_lock_irqsave(&dev->read_list_spinlock, flags);
> Â rb_count = -1;
> Â list_for_each_entry(rb, &dev->read_list.list, list) {
> Â ++rb_count;
> @@ -840,8 +839,7 @@ void recv_ishtp_cl_msg(struct ishtp_device *dev,
> Â
> Â Â/* If no Rx buffer is allocated, disband the rb */
> Â if (rb->buffer.size == 0 || rb->buffer.data == NULL)
> {
> - spin_unlock_irqrestore(&dev-
> >read_list_spinlock,
> - dev_flags);
> + spin_unlock_irqrestore(&dev-
> >read_list_spinlock, flags);
> Â dev_err(&cl->device->dev,
> Â "Rx buffer is not allocated.\n");
> Â list_del(&rb->list);
> @@ -857,8 +855,7 @@ void recv_ishtp_cl_msg(struct ishtp_device *dev,
> Â Â* back FC, so communication will be stuck anyway)
> Â Â*/
> Â if (rb->buffer.size < ishtp_hdr->length + rb-
> >buf_idx) {
> - spin_unlock_irqrestore(&dev-
> >read_list_spinlock,
> - dev_flags);
> + spin_unlock_irqrestore(&dev-
> >read_list_spinlock, flags);
> Â dev_err(&cl->device->dev,
> Â "message overflow. size %d len %d
> idx %ld\n",
> Â rb->buffer.size, ishtp_hdr->length,
> @@ -884,14 +881,13 @@ void recv_ishtp_cl_msg(struct ishtp_device
> *dev,
> Â Â* the whole msg arrived, send a new FC, and
> add a new
> Â Â* rb buffer for the next coming msg
> Â Â*/
> - spin_lock_irqsave(&cl->free_list_spinlock,
> flags);
> + spin_lock(&cl->free_list_spinlock);
> Â
> Â if (!list_empty(&cl->free_rb_list.list)) {
> Â new_rb = list_entry(cl-
> >free_rb_list.list.next,
> Â struct ishtp_cl_rb, list);
> Â list_del_init(&new_rb->list);
> - spin_unlock_irqrestore(&cl-
> >free_list_spinlock,
> - flags);
> + spin_unlock(&cl-
> >free_list_spinlock);
> Â new_rb->cl = cl;
> Â new_rb->buf_idx = 0;
> Â INIT_LIST_HEAD(&new_rb->list);
> @@ -900,8 +896,7 @@ void recv_ishtp_cl_msg(struct ishtp_device *dev,
> Â
> Â ishtp_hbm_cl_flow_control_req(dev,
> cl);
> Â } else {
> - spin_unlock_irqrestore(&cl-
> >free_list_spinlock,
> - flags);
> + spin_unlock(&cl-
> >free_list_spinlock);
> Â }
> Â }
> Â /* One more fragment in message (even if this was
> last) */
> @@ -914,7 +909,7 @@ void recv_ishtp_cl_msg(struct ishtp_device *dev,
> Â break;
> Â }
> Â
> - spin_unlock_irqrestore(&dev->read_list_spinlock, dev_flags);
> + spin_unlock_irqrestore(&dev->read_list_spinlock, flags);
> Â /* If it's nobody's message, just read and discard it */
> Â if (!buffer) {
> Â uint8_t rd_msg_buf[ISHTP_RD_MSG_BUF_SIZE];
> @@ -941,7 +936,7 @@ void recv_ishtp_cl_msg(struct ishtp_device *dev,
> Â * @hbm: hbm buffer
> Â *
> Â * Receive and dispatch ISHTP client messages using DMA. This
> function executes
> - * in ISR context
> + * in ISR or work queue context
> Â */
> Âvoid recv_ishtp_cl_msg_dma(struct ishtp_device *dev, void *msg,
> Â ÂÂÂstruct dma_xfer_hbm *hbm)
> @@ -951,10 +946,10 @@ void recv_ishtp_cl_msg_dma(struct ishtp_device
> *dev, void *msg,
> Â struct ishtp_cl_rb *new_rb;
> Â unsigned char *buffer = NULL;
> Â struct ishtp_cl_rb *complete_rb = NULL;
> - unsigned long dev_flags;
> Â unsigned long flags;
> Â
> - spin_lock_irqsave(&dev->read_list_spinlock, dev_flags);
> + spin_lock_irqsave(&dev->read_list_spinlock, flags);
> +
> Â list_for_each_entry(rb, &dev->read_list.list, list) {
> Â cl = rb->cl;
> Â if (!cl || !(cl->host_client_id == hbm-
> >host_client_id &&
> @@ -966,8 +961,7 @@ void recv_ishtp_cl_msg_dma(struct ishtp_device
> *dev, void *msg,
> Â Â* If no Rx buffer is allocated, disband the rb
> Â Â*/
> Â if (rb->buffer.size == 0 || rb->buffer.data == NULL)
> {
> - spin_unlock_irqrestore(&dev-
> >read_list_spinlock,
> - dev_flags);
> + spin_unlock_irqrestore(&dev-
> >read_list_spinlock, flags);
> Â dev_err(&cl->device->dev,
> Â "response buffer is not
> allocated.\n");
> Â list_del(&rb->list);
> @@ -983,8 +977,7 @@ void recv_ishtp_cl_msg_dma(struct ishtp_device
> *dev, void *msg,
> Â Â* back FC, so communication will be stuck anyway)
> Â Â*/
> Â if (rb->buffer.size < hbm->msg_length) {
> - spin_unlock_irqrestore(&dev-
> >read_list_spinlock,
> - dev_flags);
> + spin_unlock_irqrestore(&dev-
> >read_list_spinlock, flags);
> Â dev_err(&cl->device->dev,
> Â "message overflow. size %d len %d
> idx %ld\n",
> Â rb->buffer.size, hbm->msg_length,
> rb->buf_idx);
> @@ -1008,14 +1001,13 @@ void recv_ishtp_cl_msg_dma(struct
> ishtp_device *dev, void *msg,
> Â Â* the whole msg arrived, send a new FC, and add a
> new
> Â Â* rb buffer for the next coming msg
> Â Â*/
> - spin_lock_irqsave(&cl->free_list_spinlock, flags);
> + spin_lock(&cl->free_list_spinlock);
> Â
> Â if (!list_empty(&cl->free_rb_list.list)) {
> Â new_rb = list_entry(cl-
> >free_rb_list.list.next,
> Â struct ishtp_cl_rb, list);
> Â list_del_init(&new_rb->list);
> - spin_unlock_irqrestore(&cl-
> >free_list_spinlock,
> - flags);
> + spin_unlock(&cl->free_list_spinlock);
> Â new_rb->cl = cl;
> Â new_rb->buf_idx = 0;
> Â INIT_LIST_HEAD(&new_rb->list);
> @@ -1024,8 +1016,7 @@ void recv_ishtp_cl_msg_dma(struct ishtp_device
> *dev, void *msg,
> Â
> Â ishtp_hbm_cl_flow_control_req(dev, cl);
> Â } else {
> - spin_unlock_irqrestore(&cl-
> >free_list_spinlock,
> - flags);
> + spin_unlock(&cl->free_list_spinlock);
> Â }
> Â
> Â /* One more fragment in message (this is always
> last) */
> @@ -1038,7 +1029,7 @@ void recv_ishtp_cl_msg_dma(struct ishtp_device
> *dev, void *msg,
> Â break;
> Â }
> Â
> - spin_unlock_irqrestore(&dev->read_list_spinlock, dev_flags);
> + spin_unlock_irqrestore(&dev->read_list_spinlock, flags);
> Â /* If it's nobody's message, just read and discard it */
> Â if (!buffer) {
> Â dev_err(dev->devc, "Dropped Rx (DMA) msg - no
> request\n");