Hi,
On Mon, May 14, 2012 at 01:48:16PM -0400, Alan Stern wrote:
> This patch (as1556) works around a bug in the Philips ISP1562 EHCI
> controller. Although the controller claims to support frame-list
> lengths smaller than the default of 1024 for its periodic schedule, in
> fact smaller values don't work. A new quirk flag is added to indicate
> when the bug is present, and if it is then the schedule size is left
> at the default value.
>
> Signed-off-by: Alan Stern <stern@xxxxxxxxxxxxxxxxxxx>
>
> ---
>
> drivers/usb/host/ehci-hcd.c | 2 +-
> drivers/usb/host/ehci-pci.c | 7 +++++++
> drivers/usb/host/ehci.h | 1 +
> 3 files changed, 9 insertions(+), 1 deletion(-)
>
> Index: usb-3.4/drivers/usb/host/ehci.h
> ===================================================================
> --- usb-3.4.orig/drivers/usb/host/ehci.h
> +++ usb-3.4/drivers/usb/host/ehci.h
> @@ -149,6 +149,7 @@ struct ehci_hcd { /* one per controlle
> unsigned use_dummy_qh:1; /* AMD Frame List table quirk*/
> unsigned has_synopsys_hc_bug:1; /* Synopsys HC */
> unsigned frame_index_bug:1; /* MosChip (AKA NetMos) */
> + unsigned sched_size_bug:1; /* Philips */
>
> /* required for usb32 quirk */
> #define OHCI_CTRL_HCFS (3 << 6)
> Index: usb-3.4/drivers/usb/host/ehci-hcd.c
> ===================================================================
> --- usb-3.4.orig/drivers/usb/host/ehci-hcd.c
> +++ usb-3.4/drivers/usb/host/ehci-hcd.c
> @@ -639,7 +639,7 @@ static int ehci_init(struct usb_hcd *hcd
> INIT_LIST_HEAD(&ehci->cached_itd_list);
> INIT_LIST_HEAD(&ehci->cached_sitd_list);
>
> - if (HCC_PGM_FRAMELISTLEN(hcc_params)) {
> + if (HCC_PGM_FRAMELISTLEN(hcc_params) && !ehci->sched_size_bug) {
> /* periodic schedule size can be smaller than default */
> switch (EHCI_TUNE_FLS) {
> case 0: ehci->periodic_size = 1024; break;
> Index: usb-3.4/drivers/usb/host/ehci-pci.c
> ===================================================================
> --- usb-3.4.orig/drivers/usb/host/ehci-pci.c
> +++ usb-3.4/drivers/usb/host/ehci-pci.c
> @@ -97,6 +97,13 @@ static int ehci_pci_setup(struct usb_hcd
> break;
> }
> break;
> +
> + case PCI_VENDOR_ID_PHILIPS:
> + /*
> + * Philips controllers set HCC_PGM_FRAMELISTLEN, but
> + * they don't implement schedule sizes shorter than 1024.
> + */
> + ehci->sched_size_bug = 1;
> }
>
> /* cache this readonly data; minimize chip reads */
Unrelated to $SUBJECT, but all those quirks could be passed through
driver_data from pci ID table. That will remove the need for the switch
statement. My 2 cents
--
balbi
Attachment:
signature.asc
Description: Digital signature
