All of lore.kernel.org
 help / color / mirror / Atom feed
From: Douglas Anderson <dianders@chromium.org>
To: John Youn <John.Youn@synopsys.com>,
	balbi@ti.com, kever.yang@rock-chips.com
Cc: william.wu@rock-chips.com, huangtao@rock-chips.com,
	heiko@sntech.de, stefan.wahren@i2se.com,
	linux-rockchip@lists.infradead.org,
	linux-rpi-kernel@lists.infradead.org,
	Julius Werner <jwerner@chromium.org>,
	gregory.herrero@intel.com, yousaf.kaukab@intel.com,
	dinguyen@opensource.altera.com, stern@rowland.harvard.edu,
	ming.lei@canonical.com, Douglas Anderson <dianders@chromium.org>,
	johnyoun@synopsys.com, gregkh@linuxfoundation.org,
	linux-usb@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH v6 22/22] usb: dwc2: host: If using uframe scheduler, end splits better
Date: Thu, 28 Jan 2016 18:20:13 -0800	[thread overview]
Message-ID: <1454034013-24657-23-git-send-email-dianders@chromium.org> (raw)
In-Reply-To: <1454034013-24657-1-git-send-email-dianders@chromium.org>

The microframe scheduler figured out exactly how many transfers we need
for a split transaction.  Let's use this knowledge to know when to end
things.

Without this I found that certain devices would just keep responding
with tons of NYET resonses on their INT_IN endpoint.  These would just
keep going and going and eventually we'd decide to terminate the
transfer (because the whole frame changed), but by that time the
scheduler would decide that we "missed" the start of the next transfer.
I can also imagine that if we blow past the end of our scheduled time we
may mess up other things that were scheduled to happen.

No known test cases are improved by this patch except that the scheduler
code doesn't yell about MISSES constantly anymore.

Signed-off-by: Douglas Anderson <dianders@chromium.org>
Tested-by: Heiko Stuebner <heiko@sntech.de>
Tested-by: Stefan Wahren <stefan.wahren@i2se.com>
---
Changes in v6:
- Add Heiko's Tested-by.
- Add Stefan's Tested-by.

Changes in v5: None
Changes in v4:
- If using uframe scheduler, end splits better new for v4.

Changes in v3: None
Changes in v2: None

 drivers/usb/dwc2/hcd_intr.c | 48 +++++++++++++++++++++++++++++++++++++++------
 1 file changed, 42 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/dwc2/hcd_intr.c b/drivers/usb/dwc2/hcd_intr.c
index dc285667233a..f7a325a8c741 100644
--- a/drivers/usb/dwc2/hcd_intr.c
+++ b/drivers/usb/dwc2/hcd_intr.c
@@ -1363,14 +1363,50 @@ static void dwc2_hc_nyet_intr(struct dwc2_hsotg *hsotg,
 
 		if (chan->ep_type == USB_ENDPOINT_XFER_INT ||
 		    chan->ep_type == USB_ENDPOINT_XFER_ISOC) {
-			int frnum = dwc2_hcd_get_frame_number(hsotg);
+			struct dwc2_qh *qh = chan->qh;
+			bool past_end;
+
+			if (hsotg->core_params->uframe_sched <= 0) {
+				int frnum = dwc2_hcd_get_frame_number(hsotg);
+
+				/* Don't have num_hs_transfers; simple logic */
+				past_end = dwc2_full_frame_num(frnum) !=
+				     dwc2_full_frame_num(qh->next_active_frame);
+			} else {
+				int end_frnum;
 
-			if (dwc2_full_frame_num(frnum) !=
-			    dwc2_full_frame_num(chan->qh->next_active_frame)) {
 				/*
-				 * No longer in the same full speed frame.
-				 * Treat this as a transaction error.
-				 */
+				* Figure out the end frame based on schedule.
+				*
+				* We don't want to go on trying again and again
+				* forever.  Let's stop when we've done all the
+				* transfers that were scheduled.
+				*
+				* We're going to be comparing start_active_frame
+				* and next_active_frame, both of which are 1
+				* before the time the packet goes on the wire,
+				* so that cancels out.  Basically if had 1
+				* transfer and we saw 1 NYET then we're done.
+				* We're getting a NYET here so if next >=
+				* (start + num_transfers) we're done. The
+				* complexity is that for all but ISOC_OUT we
+				* skip one slot.
+				*/
+				end_frnum = dwc2_frame_num_inc(
+					qh->start_active_frame,
+					qh->num_hs_transfers);
+
+				if (qh->ep_type != USB_ENDPOINT_XFER_ISOC ||
+				    qh->ep_is_in)
+					end_frnum =
+					       dwc2_frame_num_inc(end_frnum, 1);
+
+				past_end = dwc2_frame_num_le(
+					end_frnum, qh->next_active_frame);
+			}
+
+			if (past_end) {
+				/* Treat this as a transaction error. */
 #if 0
 				/*
 				 * Todo: Fix system performance so this can
-- 
2.7.0.rc3.207.g0ac5344

WARNING: multiple messages have this Message-ID (diff)
From: Douglas Anderson <dianders-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
To: John Youn <John.Youn-HKixBCOQz3hWk0Htik3J/w@public.gmane.org>,
	balbi-l0cyMroinI0@public.gmane.org,
	kever.yang-TNX95d0MmH7DzftRWevZcw@public.gmane.org
Cc: huangtao-TNX95d0MmH7DzftRWevZcw@public.gmane.org,
	stefan.wahren-eS4NqCHxEME@public.gmane.org,
	heiko-4mtYJXux2i+zQB+pC5nmwQ@public.gmane.org,
	johnyoun-HKixBCOQz3hWk0Htik3J/w@public.gmane.org,
	gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org,
	ming.lei-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org,
	linux-usb-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Douglas Anderson
	<dianders-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
	yousaf.kaukab-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org,
	stern-nwvwT67g6+6dFdvTe/nMLpVzexx5G7lz@public.gmane.org,
	linux-rpi-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
	gregory.herrero-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org,
	william.wu-TNX95d0MmH7DzftRWevZcw@public.gmane.org,
	Julius Werner <jwerner-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>,
	dinguyen-yzvPICuk2ABMcg4IHK0kFoH6Mc4MB0Vx@public.gmane.org
Subject: [PATCH v6 22/22] usb: dwc2: host: If using uframe scheduler, end splits better
Date: Thu, 28 Jan 2016 18:20:13 -0800	[thread overview]
Message-ID: <1454034013-24657-23-git-send-email-dianders@chromium.org> (raw)
In-Reply-To: <1454034013-24657-1-git-send-email-dianders-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>

The microframe scheduler figured out exactly how many transfers we need
for a split transaction.  Let's use this knowledge to know when to end
things.

Without this I found that certain devices would just keep responding
with tons of NYET resonses on their INT_IN endpoint.  These would just
keep going and going and eventually we'd decide to terminate the
transfer (because the whole frame changed), but by that time the
scheduler would decide that we "missed" the start of the next transfer.
I can also imagine that if we blow past the end of our scheduled time we
may mess up other things that were scheduled to happen.

No known test cases are improved by this patch except that the scheduler
code doesn't yell about MISSES constantly anymore.

Signed-off-by: Douglas Anderson <dianders-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>
Tested-by: Heiko Stuebner <heiko-4mtYJXux2i+zQB+pC5nmwQ@public.gmane.org>
Tested-by: Stefan Wahren <stefan.wahren-eS4NqCHxEME@public.gmane.org>
---
Changes in v6:
- Add Heiko's Tested-by.
- Add Stefan's Tested-by.

Changes in v5: None
Changes in v4:
- If using uframe scheduler, end splits better new for v4.

Changes in v3: None
Changes in v2: None

 drivers/usb/dwc2/hcd_intr.c | 48 +++++++++++++++++++++++++++++++++++++++------
 1 file changed, 42 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/dwc2/hcd_intr.c b/drivers/usb/dwc2/hcd_intr.c
index dc285667233a..f7a325a8c741 100644
--- a/drivers/usb/dwc2/hcd_intr.c
+++ b/drivers/usb/dwc2/hcd_intr.c
@@ -1363,14 +1363,50 @@ static void dwc2_hc_nyet_intr(struct dwc2_hsotg *hsotg,
 
 		if (chan->ep_type == USB_ENDPOINT_XFER_INT ||
 		    chan->ep_type == USB_ENDPOINT_XFER_ISOC) {
-			int frnum = dwc2_hcd_get_frame_number(hsotg);
+			struct dwc2_qh *qh = chan->qh;
+			bool past_end;
+
+			if (hsotg->core_params->uframe_sched <= 0) {
+				int frnum = dwc2_hcd_get_frame_number(hsotg);
+
+				/* Don't have num_hs_transfers; simple logic */
+				past_end = dwc2_full_frame_num(frnum) !=
+				     dwc2_full_frame_num(qh->next_active_frame);
+			} else {
+				int end_frnum;
 
-			if (dwc2_full_frame_num(frnum) !=
-			    dwc2_full_frame_num(chan->qh->next_active_frame)) {
 				/*
-				 * No longer in the same full speed frame.
-				 * Treat this as a transaction error.
-				 */
+				* Figure out the end frame based on schedule.
+				*
+				* We don't want to go on trying again and again
+				* forever.  Let's stop when we've done all the
+				* transfers that were scheduled.
+				*
+				* We're going to be comparing start_active_frame
+				* and next_active_frame, both of which are 1
+				* before the time the packet goes on the wire,
+				* so that cancels out.  Basically if had 1
+				* transfer and we saw 1 NYET then we're done.
+				* We're getting a NYET here so if next >=
+				* (start + num_transfers) we're done. The
+				* complexity is that for all but ISOC_OUT we
+				* skip one slot.
+				*/
+				end_frnum = dwc2_frame_num_inc(
+					qh->start_active_frame,
+					qh->num_hs_transfers);
+
+				if (qh->ep_type != USB_ENDPOINT_XFER_ISOC ||
+				    qh->ep_is_in)
+					end_frnum =
+					       dwc2_frame_num_inc(end_frnum, 1);
+
+				past_end = dwc2_frame_num_le(
+					end_frnum, qh->next_active_frame);
+			}
+
+			if (past_end) {
+				/* Treat this as a transaction error. */
 #if 0
 				/*
 				 * Todo: Fix system performance so this can
-- 
2.7.0.rc3.207.g0ac5344

  parent reply	other threads:[~2016-01-29  2:21 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-29  2:19 [PATCH v6 0/22] usb: dwc2: host: Fix and speed up all the stuff, especially with splits Douglas Anderson
2016-01-29  2:19 ` [PATCH v6 01/22] usb: dwc2: rockchip: Make the max_transfer_size automatic Douglas Anderson
2016-01-29  2:19   ` Douglas Anderson
2016-01-29  2:19 ` [PATCH v6 02/22] usb: dwc2: host: Get aligned DMA in a more supported way Douglas Anderson
2016-01-29  2:19   ` Douglas Anderson
2016-01-29  2:19 ` [PATCH v6 03/22] usb: dwc2: host: Set host_rx_fifo_size to 525 for rk3066 Douglas Anderson
2016-01-29  2:19   ` Douglas Anderson
2016-01-29  2:19 ` [PATCH v6 04/22] usb: dwc2: host: Avoid use of chan->qh after qh freed Douglas Anderson
2016-01-29  2:19   ` Douglas Anderson
2016-01-29  2:19 ` [PATCH v6 05/22] usb: dwc2: host: Always add to the tail of queues Douglas Anderson
2016-01-29  2:19   ` Douglas Anderson
2016-01-29  2:19 ` [PATCH v6 06/22] usb: dwc2: host: fix split transfer schedule sequence Douglas Anderson
2016-01-29  2:19   ` Douglas Anderson
2016-01-29  2:19 ` [PATCH v6 07/22] usb: dwc2: host: Add scheduler tracing Douglas Anderson
2016-01-29  2:19   ` Douglas Anderson
2016-01-29  2:19 ` [PATCH v6 08/22] usb: dwc2: host: Add a delay before releasing periodic bandwidth Douglas Anderson
2016-01-29  2:19   ` Douglas Anderson
2016-01-29  2:20 ` [PATCH v6 09/22] usb: dwc2: host: Giveback URB in tasklet context Douglas Anderson
2016-01-29  2:20 ` [PATCH v6 10/22] usb: dwc2: host: Properly set the HFIR Douglas Anderson
2016-01-31  9:23   ` Kever Yang
2016-01-31  9:23     ` Kever Yang
2016-01-31 22:19     ` Doug Anderson
2016-02-10  2:08       ` John Youn
2016-01-29  2:20 ` [PATCH v6 11/22] usb: dwc2: host: There's not really a TT for the root hub Douglas Anderson
2016-01-29  2:20   ` Douglas Anderson
2016-01-31  9:25   ` Kever Yang
2016-01-29  2:20 ` [PATCH v6 12/22] usb: dwc2: host: Use periodic interrupt even with DMA Douglas Anderson
2016-01-29  2:20   ` Douglas Anderson
2016-01-29  2:20 ` [PATCH v6 13/22] usb: dwc2: host: Rename some fields in struct dwc2_qh Douglas Anderson
2016-01-29  2:20 ` [PATCH v6 14/22] usb: dwc2: host: Reorder things in hcd_queue.c Douglas Anderson
2016-01-29  2:20   ` Douglas Anderson
2016-01-29  2:20 ` [PATCH v6 15/22] usb: dwc2: host: Split code out to make dwc2_do_reserve() Douglas Anderson
2016-01-29  2:20   ` Douglas Anderson
2016-01-29  2:20 ` [PATCH v6 16/22] usb: dwc2: host: Add scheduler logging for missed SOFs Douglas Anderson
2016-01-29  2:20 ` [PATCH v6 17/22] usb: dwc2: host: Manage frame nums better in scheduler Douglas Anderson
2016-01-29  2:20   ` Douglas Anderson
2016-02-03 20:29   ` Doug Anderson
2016-02-03 20:29     ` Doug Anderson
2016-02-09  9:53     ` Herrero, Gregory
2016-01-29  2:20 ` [PATCH v6 18/22] usb: dwc2: host: Schedule periodic right away if it's time Douglas Anderson
2016-01-31  9:36   ` Kever Yang
2016-01-31  9:36     ` Kever Yang
2016-01-31 22:09     ` Doug Anderson
2016-01-31 22:09       ` Doug Anderson
2016-02-01  3:32       ` Kever Yang
2016-02-01  4:36         ` Doug Anderson
2016-02-01  4:36           ` Doug Anderson
2016-02-02  0:36           ` Doug Anderson
2016-02-02  0:36             ` Doug Anderson
2016-02-02  7:04             ` Kever Yang
2016-02-02  7:04               ` Kever Yang
2016-02-02 23:28               ` Doug Anderson
2016-02-02 23:28                 ` Doug Anderson
2016-01-29  2:20 ` [PATCH v6 19/22] usb: dwc2: host: Add dwc2_hcd_get_future_frame_number() call Douglas Anderson
2016-01-29  2:20   ` Douglas Anderson
2016-01-29  2:20 ` [PATCH v6 20/22] usb: dwc2: host: Properly set even/odd frame Douglas Anderson
2016-02-02  7:46   ` Kever Yang
2016-02-02 22:47     ` Doug Anderson
2016-02-02 22:47       ` Doug Anderson
2016-02-03  7:47       ` Kever Yang
2016-02-03  7:47         ` Kever Yang
2016-01-29  2:20 ` [PATCH v6 21/22] usb: dwc2: host: Totally redo the microframe scheduler Douglas Anderson
2016-01-29  2:20   ` Douglas Anderson
2016-01-29  2:20 ` Douglas Anderson [this message]
2016-01-29  2:20   ` [PATCH v6 22/22] usb: dwc2: host: If using uframe scheduler, end splits better Douglas Anderson
2016-02-02 23:57 ` [PATCH v6 0/22] usb: dwc2: host: Fix and speed up all the stuff, especially with splits John Youn
2016-02-02 23:57   ` John Youn
2016-02-03 18:23   ` Doug Anderson
2016-02-03 18:23     ` Doug Anderson
2016-02-10  2:25     ` John Youn
2016-02-10  2:25       ` John Youn

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1454034013-24657-23-git-send-email-dianders@chromium.org \
    --to=dianders@chromium.org \
    --cc=John.Youn@synopsys.com \
    --cc=balbi@ti.com \
    --cc=dinguyen@opensource.altera.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=gregory.herrero@intel.com \
    --cc=heiko@sntech.de \
    --cc=huangtao@rock-chips.com \
    --cc=johnyoun@synopsys.com \
    --cc=jwerner@chromium.org \
    --cc=kever.yang@rock-chips.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rockchip@lists.infradead.org \
    --cc=linux-rpi-kernel@lists.infradead.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=ming.lei@canonical.com \
    --cc=stefan.wahren@i2se.com \
    --cc=stern@rowland.harvard.edu \
    --cc=william.wu@rock-chips.com \
    --cc=yousaf.kaukab@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.