* [PATCH] usb: dwc3: gadget: Workaround Mirosoft's BESL check
@ 2019-08-29 3:11 Thinh Nguyen
2019-08-29 4:28 ` Felipe Balbi
0 siblings, 1 reply; 2+ messages in thread
From: Thinh Nguyen @ 2019-08-29 3:11 UTC (permalink / raw)
To: Felipe Balbi, Greg Kroah-Hartman, linux-usb; +Cc: John Youn, Thinh Nguyen
While testing our host system using Microsoft's usb stack against our
gadget for various BESL values, we found an issue with their usb stack
when the recommended baseline BESL value is 0 (125us) or when the deep
BESL is 1 or less. The Windows host will issue a usb reset immediately
after it receives the extended BOS descriptor and the enumeration will
fail after a few attempts.
To keep compatibility with Microsoft's host usb stack, let's workaround
this issue by using the recommended baseline BESL of 1 (or 150us)
and clamp the deep BESL value within 2 to 15.
This was tested against Windows 10 build 18956.
Signed-off-by: Thinh Nguyen <thinhn@synopsys.com>
---
drivers/usb/dwc3/gadget.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c
index 7b58e0e1e438..9ef2259829ba 100644
--- a/drivers/usb/dwc3/gadget.c
+++ b/drivers/usb/dwc3/gadget.c
@@ -2083,9 +2083,10 @@ static void dwc3_gadget_config_params(struct usb_gadget *g,
/* Recommended BESL */
if (!dwc->dis_enblslpm_quirk) {
- params->besl_baseline = 0;
+ params->besl_baseline = 1;
if (dwc->is_utmi_l1_suspend)
- params->besl_deep = min_t(u8, dwc->hird_threshold, 15);
+ params->besl_deep =
+ clamp_t(u8, dwc->hird_threshold, 2, 15);
}
/* U1 Device exit Latency */
--
2.11.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] usb: dwc3: gadget: Workaround Mirosoft's BESL check
2019-08-29 3:11 [PATCH] usb: dwc3: gadget: Workaround Mirosoft's BESL check Thinh Nguyen
@ 2019-08-29 4:28 ` Felipe Balbi
0 siblings, 0 replies; 2+ messages in thread
From: Felipe Balbi @ 2019-08-29 4:28 UTC (permalink / raw)
To: Thinh Nguyen, Greg Kroah-Hartman, linux-usb; +Cc: John Youn, Thinh Nguyen
Hi,
Thinh Nguyen <Thinh.Nguyen@synopsys.com> writes:
> While testing our host system using Microsoft's usb stack against our
> gadget for various BESL values, we found an issue with their usb stack
> when the recommended baseline BESL value is 0 (125us) or when the deep
> BESL is 1 or less. The Windows host will issue a usb reset immediately
> after it receives the extended BOS descriptor and the enumeration will
> fail after a few attempts.
>
> To keep compatibility with Microsoft's host usb stack, let's workaround
> this issue by using the recommended baseline BESL of 1 (or 150us)
> and clamp the deep BESL value within 2 to 15.
>
> This was tested against Windows 10 build 18956.
>
> Signed-off-by: Thinh Nguyen <thinhn@synopsys.com>
> ---
> drivers/usb/dwc3/gadget.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c
> index 7b58e0e1e438..9ef2259829ba 100644
> --- a/drivers/usb/dwc3/gadget.c
> +++ b/drivers/usb/dwc3/gadget.c
> @@ -2083,9 +2083,10 @@ static void dwc3_gadget_config_params(struct usb_gadget *g,
>
> /* Recommended BESL */
> if (!dwc->dis_enblslpm_quirk) {
let's add a comment here similar to what's in commit log so folks won't
go about changing this willy-nilly.
> - params->besl_baseline = 0;
> + params->besl_baseline = 1;
> if (dwc->is_utmi_l1_suspend)
> - params->besl_deep = min_t(u8, dwc->hird_threshold, 15);
> + params->besl_deep =
> + clamp_t(u8, dwc->hird_threshold, 2, 15);
> }
>
> /* U1 Device exit Latency */
> --
> 2.11.0
>
--
balbi
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2019-08-29 4:28 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-08-29 3:11 [PATCH] usb: dwc3: gadget: Workaround Mirosoft's BESL check Thinh Nguyen
2019-08-29 4:28 ` Felipe Balbi
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).