2012/3/27 Andiry Xu <andiry.xu@xxxxxxx>:
> On 03/26/2012 08:08 PM, Elric Fu wrote:
>> Software have to abort command ring and cancel command
>> when a command is failed or hang. Otherwise, the command
>> ring will hang and can't handle another commands. An
>> example of a command that may hang is the Address Device
>> Command, because waiting for a SET_ADDRESS request to be
>> acknowledged by a USB device is outside of the xHC's
>> ability to control.
>>
>> According to xHCI spec section 4.6.1.1 and section 4.6.1.2,
>> after aborting a command on the command ring, xHC will
>> generate a command completion event with its completion code
>> set to Command Ring Stopped at least. If a command is
>> currently executing at the time of aborting a command, xHC
>> also generate a command completion event with its completion
>> code set to Command Abort. When the command ring is stopped,
>> software may remove, add, or rearrage Command Descriptors.
>>
>> The RFT patch use to cancel command when the above situation
>> occur.
>>
>> To cancel a command, software will initialize a command
>> descriptor for the cancel command, and add it into a
>> cancel_cd_list of xhci. When the command ring is stopped,
>> software will find the command trbs described by command
>> descriptors in cancel_cd_list and modify it to No Op
>> command. If software can't find the matched trbs, we can
>> think it had been finished.
>>
>> Signed-off-by: Elric Fu <elricfu1@xxxxxxxxx>
>> ---
>> drivers/usb/host/xhci-mem.c | 1 +
>> drivers/usb/host/xhci-ring.c | 238 ++++++++++++++++++++++++++++++++++++++++++
>> drivers/usb/host/xhci.c | 12 ++-
>> drivers/usb/host/xhci.h | 18 +++
>> 4 files changed, 266 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c
>> index 383fc85..84e7245 100644
>> --- a/drivers/usb/host/xhci-mem.c
>> +++ b/drivers/usb/host/xhci-mem.c
>> @@ -2238,6 +2238,7 @@ int xhci_mem_init(struct xhci_hcd *xhci, gfp_t flags)
>> xhci->cmd_ring = xhci_ring_alloc(xhci, 1, true, false, flags);
>> if (!xhci->cmd_ring)
>> goto fail;
>> + INIT_LIST_HEAD(&xhci->cancel_cd_list);
>
> What does cd mean? command descriptor? I think cancel_cmd_list is better.
Yes. Well, the "cancel_cmd_list" is easier to understand.
>
>> xhci_dbg(xhci, "Allocated command ring at %p\n", xhci->cmd_ring);
>> xhci_dbg(xhci, "First segment DMA is 0x%llx\n",
>> (unsigned long long)xhci->cmd_ring->first_seg->dma);
>> diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c
>> index b62037b..d4d7f8d 100644
>> --- a/drivers/usb/host/xhci-ring.c
>> +++ b/drivers/usb/host/xhci-ring.c
>> @@ -292,12 +292,119 @@ static int room_on_ring(struct xhci_hcd *xhci, struct xhci_ring *ring,
>> /* Ring the host controller doorbell after placing a command on the ring */
>> void xhci_ring_cmd_db(struct xhci_hcd *xhci)
>> {
>> + if (!(xhci->cmd_ring_state & CMD_RING_STATE_RUNNING))
>> + return;
>> +
>> xhci_dbg(xhci, "// Ding dong!\n");
>> xhci_writel(xhci, DB_VALUE_HOST, &xhci->dba->doorbell[0]);
>> /* Flush PCI posted writes */
>> xhci_readl(xhci, &xhci->dba->doorbell[0]);
>> }
>>
>> +int xhci_abort_cmd_ring(struct xhci_hcd *xhci)
>> +{
>> + u64 temp_64;
>> + int ret;
>> +
>> + xhci_dbg(xhci, "Abort command ring\n");
>> +
>> + if (!(xhci->cmd_ring_state & CMD_RING_STATE_RUNNING)) {
>> + xhci_dbg(xhci, "The command ring isn't running, "
>> + "Have the command ring been stopped?");
>> + return -EINVAL;
>> + }
>> +
>> + temp_64 = xhci_read_64(xhci, &xhci->op_regs->cmd_ring);
>> + if (!(temp_64 & CMD_RING_RUNNING)) {
>> + xhci_dbg(xhci, "Command ring had been stopped\n");
>> + return -EINVAL;
>> + }
>> + xhci->cmd_ring_state = CMD_RING_STATE_ABORTED;
>> + xhci_write_64(xhci, temp_64 | CMD_RING_ABORT,
>> + &xhci->op_regs->cmd_ring);
>> +
>> + /* Section 4.6.1.2 of xHCI 1.0 spec says software should
>> + * time the completion od all xHCI commands, including
>> + * the Command Abort operation. If software doesn't see
>> + * CRR negated in a timely manner (e.g. longer than 5
>> + * seconds), then it should assume that the there are
>> + * larger problems with the xHC and assert HCRST.
>> + */
>> + ret = handshake(xhci, &xhci->op_regs->cmd_ring,
>> + CMD_RING_RUNNING, 0, 5 * 1000 * 1000);
>> + if (ret < 0) {
>> + xhci_err(xhci, "Stopped the command ring failed, "
>> + "maybe the host is dead\n");
>> + xhci->xhc_state |= XHCI_STATE_DYING;
>> + xhci_quiesce(xhci);
>> + xhci_halt(xhci);
>> + return -ESHUTDOWN;
>> + }
>> +
>> + return 0;
>> +}
>> +
>> +static int queue_cd(struct xhci_hcd *xhci, int slot_id,
>> + u32 cmd_type, int ep_index)
>> +{
>> + struct xhci_cd *cd;
>> +
>> + while (1) {
>> + if (!(xhci->cmd_ring_state & CMD_RING_STATE_MODIFYING))
>> + break;
>> + }
>
> Who will modify the cmd_ring state and break the loop? You are holding
> the lock now.
It seems like a issue. I don't want to insert the cd into the list when
the list is being modified. You remind me that maybe the loop
can't keep the cancel_cd_list safe. The handle_cmd_completion()
doesn't hold any lock. It is possible that the command ring state
isn't CMD_RING_STATE_MODIFYING so that cd can be inserted.
But when cd is being inserted into list command ring state turn into
CMD_RING_STATE_MODIFYING and cancel_cd() is called. Maybe
we need a lock for protecting the list. What do you think?
>
>> +
>> + cd = kzalloc(sizeof(struct xhci_cd), GFP_ATOMIC);
>> + if (!cd)
>> + return -ENOMEM;
>> + INIT_LIST_HEAD(&cd->cancel_cd_list);
>> +
>> + cd->slot_id = slot_id;
>> + cd->cmd_type = cmd_type;
>> + cd->ep_index = ep_index;
>> + list_add_tail(&cd->cancel_cd_list, &xhci->cancel_cd_list);
>> +
>> + return 0;
>> +}
>> +
>> +int xhci_cancel_cmd(struct xhci_hcd *xhci, int slot_id,
>> + u32 cmd_type, int ep_index)
>> +{
>> + int retval;
>> + unsigned long flags;
>> +
>> + spin_lock_irqsave(&xhci->lock, flags);
>> +
>> + if (xhci->xhc_state & XHCI_STATE_DYING) {
>> + xhci_warn(xhci, "Abort the command ring,"
>> + " but the xHCI is dead.\n");
>> + spin_unlock_irqrestore(&xhci->lock, flags);
>> + return -ESHUTDOWN;
>> + }
>> +
>> + /* queue the cmd desriptor to cancelled_cd_list */
>> + retval = queue_cd(xhci, slot_id, cmd_type, ep_index);
>> + if (retval) {
>> + spin_unlock_irqrestore(&xhci->lock, flags);
>> + return retval;
>> + }
>> +
>> + /* abort command ring */
>> + retval = xhci_abort_cmd_ring(xhci);
>> + if (retval) {
>> + xhci_err(xhci, "Abort command ring failed\n");
>
> What happen if queue_cd() succeeds but xhci_abort_cmd_ring() fails?
> Do you want to remove the cd from the cancel_cd_list? Will there be any
> mismatch?
I don't want to remove cd from cancel_cd_list. Because I think we
can insert cd into the list when the command ring is stopped even
if the xhci_abort_cmd_ring() is failed. But you remind me that the
xhci_abort_cmd_ring is failed due to the CRR bit is always 1. In
this case I have to free the cd, or else the memory leak will occur.
What do you think about freeing all of command descriptors in
cancel_cd_list when the xhci_mem_cleanup() is called?
>
>> + spin_unlock_irqrestore(&xhci->lock, flags);
>> + if (retval == -ESHUTDOWN) {
>> + usb_hc_died(xhci_to_hcd(xhci)->primary_hcd);
>> + xhci_dbg(xhci, "xHCI host controller is dead.\n");
>> + }
>> + return retval;
>> + }
>> +
>> + spin_unlock_irqrestore(&xhci->lock, flags);
>> + return 0;
>> +}
>> +
>> void xhci_ring_ep_doorbell(struct xhci_hcd *xhci,
>> unsigned int slot_id,
>> unsigned int ep_index,
>> @@ -1050,6 +1157,130 @@ static int handle_cmd_in_cmd_wait_list(struct xhci_hcd *xhci,
>> return 1;
>> }
>>
>> +/*
>> + * Find the command trb need to be cancelled and modify it to NO OP
>> + * command.
>> + *
>> + * If we can't find the command trb, we think it had already been
>> + * executed.
>> + */
>> +static void cd_to_noop(struct xhci_hcd *xhci, struct xhci_cd *cur_cd)
>> +{
>> + struct xhci_segment *cur_seg;
>> + union xhci_trb *cmd_trb;
>> + int slot_id;
>> + u32 cmd_type;
>> + u32 cycle_state;
>> + int ep_index;
>> +
>> + if (xhci->cmd_ring->dequeue == xhci->cmd_ring->enqueue)
>> + return;
>> +
>> + /* find the current segmene of command ring */
>> + cur_seg = find_trb_seg(xhci->cmd_ring->first_seg,
>> + xhci->cmd_ring->dequeue, &cycle_state);
>> +
>> + /* find the command trb matched by CD from command ring */
>> + for (cmd_trb = xhci->cmd_ring->dequeue;
>> + cmd_trb != xhci->cmd_ring->enqueue;
>> + next_trb(xhci, xhci->cmd_ring, &cur_seg, &cmd_trb)) {
>> + if (TRB_TYPE_LINK_LE32(cmd_trb->generic.field[3]))
>> + continue;
>> +
>> + slot_id = TRB_TO_SLOT_ID(
>> + le32_to_cpu(cmd_trb->generic.field[3]));
>> + cmd_type = TRB_FIELD_TO_TYPE(
>> + le32_to_cpu(cmd_trb->generic.field[3]));
>> + if ((slot_id == cur_cd->slot_id) &&
>> + (cmd_type == cur_cd->cmd_type)) {
>> +
>> + /* The ep_index in the command desriptors of Stop
>> + * Endpoint command, Set TR Dequeue Pointer command
>> + * and Reset Endpoint command is vaild. so it should
>> + * be match the ep_index of cd.
>> + */
>> + if ((cmd_type == TRB_STOP_RING) ||
>> + (cmd_type == TRB_SET_DEQ) ||
>> + (cmd_type == TRB_RESET_EP)) {
>> + ep_index = TRB_TO_EP_INDEX(le32_to_cpu(
>> + cmd_trb->generic.field[3]));
>> + if (ep_index != cur_cd->ep_index)
>> + continue;
>> + }
>> +
>> + /* get cycle state from the origin command trb */
>> + cycle_state = le32_to_cpu(cmd_trb->generic.field[3])
>> + & TRB_CYCLE;
>> +
>> + /* modify the command trb to NO OP command */
>> + cmd_trb->generic.field[0] = 0;
>> + cmd_trb->generic.field[1] = 0;
>> + cmd_trb->generic.field[2] = 0;
>> + cmd_trb->generic.field[3] = cpu_to_le32(
>> + TRB_TYPE(TRB_CMD_NOOP) | cycle_state);
>> + break;
>> + }
>> + }
>> +}
>> +
>> +static void cancel_cd(struct xhci_hcd *xhci)
>> +{
>> + struct xhci_cd *cur_cd;
>> + struct list_head *entry;
>> + struct list_head *next;
>> +
>> + if (list_empty(&xhci->cancel_cd_list))
>> + return;
>> +
>> + list_for_each_safe(entry, next, &xhci->cancel_cd_list) {
>> + cur_cd = list_entry(entry, struct xhci_cd, cancel_cd_list);
>
> list_for_each_entry_safe()
You're right.
>
>> + xhci_dbg(xhci, "Cancel command, slot %d, cmd type %d, "
>> + "ep_index = %d\n", cur_cd->slot_id,
>> + cur_cd->cmd_type, cur_cd->ep_index);
>> + cd_to_noop(xhci, cur_cd);
>> +
>> + /* Whatever we find the matched command trb, we need to
>> + * release the command descriptor.
>> + */
>> + list_del(&cur_cd->cancel_cd_list);
>> + kfree(cur_cd);
>> + }
>> +}
>> +
>> +static void handle_stopped_cmd_ring(struct xhci_hcd *xhci,
>> + struct xhci_event_cmd *event)
>> +{
>> + struct xhci_virt_device *virt_dev;
>> + int slot_id = TRB_TO_SLOT_ID(
>> + le32_to_cpu(xhci->cmd_ring->dequeue->generic.field[3]));
>> + int cmd_trb_comp_code = GET_COMP_CODE(le32_to_cpu(event->status));
>> +
>> + virt_dev = xhci->devs[slot_id];
>> + if (virt_dev) {
>> + /* if the command is in cmd_list, handle it. */
>> + handle_cmd_in_cmd_wait_list(xhci, virt_dev, event);
>> + }
>> +
>> + /* advance the dequeue pointer to next cmd trb */
>> + inc_deq(xhci, xhci->cmd_ring, false);
>> +
>> + /* set the cmd ring state */
>> + if (cmd_trb_comp_code != COMP_CMD_STOP) {
>> + xhci->cmd_ring_state = CMD_RING_STATE_STOPPED;
>
> if cmd_trb_comp_code == COMP_CMD_ABORT, you just change the
> cmd_ring_state and return. Are you waiting for the COMP_CMD_STOP event
> and process them all in one time?
Yes. The host doesn't definitely generated the completion event with
the completion code set to COMP_CMD_ABORT. So I just wait for the
COMP_CMD_STOP event.
Best Regards,
Elric
>
>
> Thanks,
> Andiry
>
>> + } else {
>> + xhci->cmd_ring_state = CMD_RING_STATE_MODIFYING;
>> + cancel_cd(xhci);
>> + xhci->cmd_ring_state = CMD_RING_STATE_RUNNING;
>> +
>> + /*
>> + * ring command ring doorbell again to handle the waiting
>> + * command trbs due to aborting command ring
>> + */
>> + if (xhci->cmd_ring->dequeue != xhci->cmd_ring->enqueue)
>> + xhci_ring_cmd_db(xhci);
>> + }
>> +}
>> +
>> static void handle_cmd_completion(struct xhci_hcd *xhci,
>> struct xhci_event_cmd *event)
>> {
>> @@ -1075,6 +1306,13 @@ static void handle_cmd_completion(struct xhci_hcd *xhci,
>> xhci->error_bitmask |= 1 << 5;
>> return;
>> }
>> +
>> + if ((GET_COMP_CODE(le32_to_cpu(event->status)) == COMP_CMD_ABORT) ||
>> + (GET_COMP_CODE(le32_to_cpu(event->status)) == COMP_CMD_STOP)) {
>> + handle_stopped_cmd_ring(xhci, event);
>> + return;
>> + }
>> +
>> switch (le32_to_cpu(xhci->cmd_ring->dequeue->generic.field[3])
>> & TRB_TYPE_BITMASK) {
>> case TRB_TYPE(TRB_ENABLE_SLOT):
>> diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
>> index c939f5f..9434771 100644
>> --- a/drivers/usb/host/xhci.c
>> +++ b/drivers/usb/host/xhci.c
>> @@ -51,7 +51,7 @@ MODULE_PARM_DESC(link_quirk, "Don't clear the chain bit on a link TRB");
>> * handshake done). There are two failure modes: "usec" have passed (major
>> * hardware flakeout), or the register reads as all-ones (hardware removed).
>> */
>> -static int handshake(struct xhci_hcd *xhci, void __iomem *ptr,
>> +int handshake(struct xhci_hcd *xhci, void __iomem *ptr,
>> u32 mask, u32 done, int usec)
>> {
>> u32 result;
>> @@ -104,8 +104,10 @@ int xhci_halt(struct xhci_hcd *xhci)
>>
>> ret = handshake(xhci, &xhci->op_regs->status,
>> STS_HALT, STS_HALT, XHCI_MAX_HALT_USEC);
>> - if (!ret)
>> + if (!ret) {
>> xhci->xhc_state |= XHCI_STATE_HALTED;
>> + xhci->cmd_ring_state = CMD_RING_STATE_STOPPED;
>> + }
>> return ret;
>> }
>>
>> @@ -473,6 +475,7 @@ static int xhci_run_finished(struct xhci_hcd *xhci)
>> return -ENODEV;
>> }
>> xhci->shared_hcd->state = HC_STATE_RUNNING;
>> + xhci->cmd_ring_state = CMD_RING_STATE_RUNNING;
>>
>> if (xhci->quirks & XHCI_NEC_HOST)
>> xhci_ring_cmd_db(xhci);
>> @@ -3542,7 +3545,10 @@ int xhci_address_device(struct usb_hcd *hcd, struct usb_device *udev)
>> if (timeleft <= 0) {
>> xhci_warn(xhci, "%s while waiting for address device command\n",
>> timeleft == 0 ? "Timeout" : "Signal");
>> - /* FIXME cancel the address device command */
>> + /* cancel the address device command */
>> + ret = xhci_cancel_cmd(xhci, udev->slot_id, TRB_ADDR_DEV, 0);
>> + if (ret < 0)
>> + return ret;
>> return -ETIME;
>> }
>>
>> diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h
>> index 5b29d3c..8e59c42 100644
>> --- a/drivers/usb/host/xhci.h
>> +++ b/drivers/usb/host/xhci.h
>> @@ -1247,6 +1247,14 @@ struct xhci_td {
>> union xhci_trb *last_trb;
>> };
>>
>> +/* command descriptor */
>> +struct xhci_cd {
>> + struct list_head cancel_cd_list;
>> + int slot_id;
>> + u32 cmd_type;
>> + int ep_index;
>> +};
>> +
>> struct xhci_dequeue_state {
>> struct xhci_segment *new_deq_seg;
>> union xhci_trb *new_deq_ptr;
>> @@ -1394,6 +1402,12 @@ struct xhci_hcd {
>> /* data structures */
>> struct xhci_device_context_array *dcbaa;
>> struct xhci_ring *cmd_ring;
>> + unsigned int cmd_ring_state;
>> +#define CMD_RING_STATE_RUNNING (1 << 0)
>> +#define CMD_RING_STATE_ABORTED (1 << 1)
>> +#define CMD_RING_STATE_MODIFYING (1 << 2)
>> +#define CMD_RING_STATE_STOPPED (1 << 3)
>> + struct list_head cancel_cd_list;
>> unsigned int cmd_ring_reserved_trbs;
>> struct xhci_ring *event_ring;
>> struct xhci_erst erst;
>> @@ -1651,6 +1665,8 @@ static inline void xhci_unregister_pci(void) {}
>>
>> /* xHCI host controller glue */
>> typedef void (*xhci_get_quirks_t)(struct device *, struct xhci_hcd *);
>> +int handshake(struct xhci_hcd *xhci, void __iomem *ptr,
>> + u32 mask, u32 done, int usec);
>> void xhci_quiesce(struct xhci_hcd *xhci);
>> int xhci_halt(struct xhci_hcd *xhci);
>> int xhci_reset(struct xhci_hcd *xhci);
>> @@ -1741,6 +1757,8 @@ void xhci_queue_config_ep_quirk(struct xhci_hcd *xhci,
>> unsigned int slot_id, unsigned int ep_index,
>> struct xhci_dequeue_state *deq_state);
>> void xhci_stop_endpoint_command_watchdog(unsigned long arg);
>> +int xhci_cancel_cmd(struct xhci_hcd *xhci, int slot_id,
>> + u32 cmd_type, int ep_index);
>> void xhci_ring_ep_doorbell(struct xhci_hcd *xhci, unsigned int slot_id,
>> unsigned int ep_index, unsigned int stream_id);
>>
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-usb" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
