On Sat, 2016-12-03 at 12:20 -0500, Alan Stern wrote:
> On Fri, 2 Dec 2016, Todd Brandt wrote:
>
> > Add a module parameter that replaces the USB_RESUME_TIMEOUT
> > and other hardcoded delay numbers with the USB spec minimums.
> > By default the patch retains the current values.
> >
> > The USB subsystem currently uses heavily padded values for TDRSMDN
> > and TRSTRCY. This patch keeps the current values by default, but if
> > the kernel is booted with usbcore.padding=0 they are set to the
> > spec minimums with no padding. The result is significant performance
> > improvement in usb device resume.
> >
> > Signed-off-by: Todd Brandt <todd.e.brandt@xxxxxxxxxxxxxxx>
>
> > --- a/Documentation/kernel-parameters.txt
> > +++ b/Documentation/kernel-parameters.txt
> > @@ -4237,6 +4237,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
> > usbcore.blinkenlights=
> > [USB] Set to cycle leds on hubs (default 0 = off).
> >
> > + usbcore.padding=
> > + [USB] Pad the USB 2.0 spec timing values (default 1 = on).
> > +
>
> Whether or not you take Greg's suggestion and change this to a debugfs
> entry, please don't call it "padding". The word is horribly imprecise
> and will usually be interpreted differently from what you want.
ok, I'll change it.
>
> You missed a spot on line 385 in uhci-hub.c; it should be tdrsmdn.
>
> What happened to ohci-hub.c?
ugh, totally missed that, thanks for the catch. So many hardcoded timing
values :P
>
> > --- a/include/linux/usb.h
> > +++ b/include/linux/usb.h
> > @@ -233,8 +233,30 @@ void usb_put_intf(struct usb_interface *intf);
> > * In order to avoid both conditions, we're using a 40 ms resume timeout, which
> > * should cope with both LPJ calibration errors and devices not following every
> > * detail of the USB Specification.
> > + *
> > + * struct _usb_timing_config - USB timing value settings
> > + * @tdrsmdn: TDRSMDN resume signal time 7.1.7.7
> > + * @trsmrcy; TRSMRCY resume recovery time 7.1.7.7
> > + * @trstrcy; TRSTRCY reset recovery time 7.1.7.5
> > + * @name: timing config name
>
> @name? What's that?
a remnant of a previous implementation, *blush*, thanks for the catch :)
>
> > + *
> > + * These timing values are defined in the USB 2.0 spec sec 7.3.2 table 7-13
> > + * Thir default values have been padded for various reasons and this config
> > + * allows the system to use different values.
> > */
> > -#define USB_RESUME_TIMEOUT 40 /* ms */
> > +#define USB_TIMING_TDRSMDN_MIN 20
> > +#define USB_TIMING_TRSMRCY_MIN 10
> > +#define USB_TIMING_TRSTRCY_MIN 0
> > +#define USB_TIMING_TDRSMDN_DEF 40
> > +#define USB_TIMING_TRSMRCY_DEF 10
> > +#define USB_TIMING_TRSTRCY_DEF 50
> > +
> > +struct _usb_timing_config {
> > + unsigned int tdrsmdn; /* resume signal time 20ms - infinity */
> > + unsigned int trsmrcy; /* resume recovery time 0ms - 10ms */
> > + unsigned int trstrcy; /* reset recovery time 0ms - infinity*/
> > +};
> > +extern struct _usb_timing_config usb_timing;
> >
> > /**
> > * struct usb_interface_cache - long-term representation of a device interface
>
> Alan Stern
>
--
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
