For Aspeed, HCHalted status depends on not only Run/Stop but also
ASS/PSS status.
Handshake CMD_RUN on startup instead.
Signed-off-by: Neal Liu <[email protected]>
Acked-by: Alan Stern <[email protected]>
Reviewed-by: Tao Ren <[email protected]>
Tested-by: Tao Ren <[email protected]>
---
drivers/usb/host/ehci-hcd.c | 11 ++++++++++-
drivers/usb/host/ehci-platform.c | 6 ++++++
drivers/usb/host/ehci.h | 1 +
3 files changed, 17 insertions(+), 1 deletion(-)
---
Fix STS_HALT handshake failure for Aspeed 2500/2600 platform.
Change since v3:
- Add more description.
Change since v2:
- Use my "real" name for both the Signed-off-by: line and the From: line.
Change since v1:
- Handshake CMD_RUN status on startup instead of easily skip it.
---
diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c
index 6bdc6d6bf74d..55f92d25336b 100644
--- a/drivers/usb/host/ehci-hcd.c
+++ b/drivers/usb/host/ehci-hcd.c
@@ -634,7 +634,16 @@ static int ehci_run (struct usb_hcd *hcd)
/* Wait until HC become operational */
ehci_readl(ehci, &ehci->regs->command); /* unblock posted writes */
msleep(5);
- rc = ehci_handshake(ehci, &ehci->regs->status, STS_HALT, 0, 100 * 1000);
+
+ /* For Aspeed, STS_HALT also depends on ASS/PSS status.
+ * Check CMD_RUN instead.
+ */
+ if (ehci->is_aspeed)
+ rc = ehci_handshake(ehci, &ehci->regs->command, CMD_RUN,
+ 1, 100 * 1000);
+ else
+ rc = ehci_handshake(ehci, &ehci->regs->status, STS_HALT,
+ 0, 100 * 1000);
up_write(&ehci_cf_port_reset_rwsem);
diff --git a/drivers/usb/host/ehci-platform.c b/drivers/usb/host/ehci-platform.c
index c70f2d0b4aaf..c3dc906274d9 100644
--- a/drivers/usb/host/ehci-platform.c
+++ b/drivers/usb/host/ehci-platform.c
@@ -297,6 +297,12 @@ static int ehci_platform_probe(struct platform_device *dev)
"has-transaction-translator"))
hcd->has_tt = 1;
+ if (of_device_is_compatible(dev->dev.of_node,
+ "aspeed,ast2500-ehci") ||
+ of_device_is_compatible(dev->dev.of_node,
+ "aspeed,ast2600-ehci"))
+ ehci->is_aspeed = 1;
+
if (soc_device_match(quirk_poll_match))
priv->quirk_poll = true;
diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h
index 80bb823aa9fe..fdd073cc053b 100644
--- a/drivers/usb/host/ehci.h
+++ b/drivers/usb/host/ehci.h
@@ -219,6 +219,7 @@ struct ehci_hcd { /* one per controller */
unsigned need_oc_pp_cycle:1; /* MPC834X port power */
unsigned imx28_write_fix:1; /* For Freescale i.MX28 */
unsigned spurious_oc:1;
+ unsigned is_aspeed:1;
/* required for usb32 quirk */
#define OHCI_CTRL_HCFS (3 << 6)
--
2.25.1
Hi Greg,
On Fri, 10 Sept 2021 at 07:38, Neal Liu <[email protected]> wrote:
>
> For Aspeed, HCHalted status depends on not only Run/Stop but also
> ASS/PSS status.
> Handshake CMD_RUN on startup instead.
>
> Signed-off-by: Neal Liu <[email protected]>
> Acked-by: Alan Stern <[email protected]>
> Reviewed-by: Tao Ren <[email protected]>
> Tested-by: Tao Ren <[email protected]>
Has this one gone in? If not, here's some tags:
Reviewed-by: Joel Stanley <[email protected]>
It also needs to go to stable:
Cc: [email protected]
Fixes: 280a9045bb18 ("ehci: fix EHCI host controller initialization sequence")
Cheers,
Joel
> ---
> drivers/usb/host/ehci-hcd.c | 11 ++++++++++-
> drivers/usb/host/ehci-platform.c | 6 ++++++
> drivers/usb/host/ehci.h | 1 +
> 3 files changed, 17 insertions(+), 1 deletion(-)
>
> ---
> Fix STS_HALT handshake failure for Aspeed 2500/2600 platform.
>
> Change since v3:
> - Add more description.
>
> Change since v2:
> - Use my "real" name for both the Signed-off-by: line and the From: line.
>
> Change since v1:
> - Handshake CMD_RUN status on startup instead of easily skip it.
> ---
>
> diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c
> index 6bdc6d6bf74d..55f92d25336b 100644
> --- a/drivers/usb/host/ehci-hcd.c
> +++ b/drivers/usb/host/ehci-hcd.c
> @@ -634,7 +634,16 @@ static int ehci_run (struct usb_hcd *hcd)
> /* Wait until HC become operational */
> ehci_readl(ehci, &ehci->regs->command); /* unblock posted writes */
> msleep(5);
> - rc = ehci_handshake(ehci, &ehci->regs->status, STS_HALT, 0, 100 * 1000);
> +
> + /* For Aspeed, STS_HALT also depends on ASS/PSS status.
> + * Check CMD_RUN instead.
> + */
> + if (ehci->is_aspeed)
> + rc = ehci_handshake(ehci, &ehci->regs->command, CMD_RUN,
> + 1, 100 * 1000);
> + else
> + rc = ehci_handshake(ehci, &ehci->regs->status, STS_HALT,
> + 0, 100 * 1000);
>
> up_write(&ehci_cf_port_reset_rwsem);
>
> diff --git a/drivers/usb/host/ehci-platform.c b/drivers/usb/host/ehci-platform.c
> index c70f2d0b4aaf..c3dc906274d9 100644
> --- a/drivers/usb/host/ehci-platform.c
> +++ b/drivers/usb/host/ehci-platform.c
> @@ -297,6 +297,12 @@ static int ehci_platform_probe(struct platform_device *dev)
> "has-transaction-translator"))
> hcd->has_tt = 1;
>
> + if (of_device_is_compatible(dev->dev.of_node,
> + "aspeed,ast2500-ehci") ||
> + of_device_is_compatible(dev->dev.of_node,
> + "aspeed,ast2600-ehci"))
> + ehci->is_aspeed = 1;
> +
> if (soc_device_match(quirk_poll_match))
> priv->quirk_poll = true;
>
> diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h
> index 80bb823aa9fe..fdd073cc053b 100644
> --- a/drivers/usb/host/ehci.h
> +++ b/drivers/usb/host/ehci.h
> @@ -219,6 +219,7 @@ struct ehci_hcd { /* one per controller */
> unsigned need_oc_pp_cycle:1; /* MPC834X port power */
> unsigned imx28_write_fix:1; /* For Freescale i.MX28 */
> unsigned spurious_oc:1;
> + unsigned is_aspeed:1;
>
> /* required for usb32 quirk */
> #define OHCI_CTRL_HCFS (3 << 6)
> --
> 2.25.1
>
On Wed, Oct 06, 2021 at 01:27:02AM +0000, Joel Stanley wrote:
> Hi Greg,
>
> On Fri, 10 Sept 2021 at 07:38, Neal Liu <[email protected]> wrote:
> >
> > For Aspeed, HCHalted status depends on not only Run/Stop but also
> > ASS/PSS status.
> > Handshake CMD_RUN on startup instead.
> >
> > Signed-off-by: Neal Liu <[email protected]>
> > Acked-by: Alan Stern <[email protected]>
> > Reviewed-by: Tao Ren <[email protected]>
> > Tested-by: Tao Ren <[email protected]>
>
> Has this one gone in? If not, here's some tags:
>
> Reviewed-by: Joel Stanley <[email protected]>
>
> It also needs to go to stable:
>
> Cc: [email protected]
> Fixes: 280a9045bb18 ("ehci: fix EHCI host controller initialization sequence")
This is already in my tree, but waiting until 5.16-rc1, sorry. You can
see it at commit 7f2d73788d90 ("usb: ehci: handshake CMD_RUN instead of
STS_HALT") in linux-next. I'll try to remember to queue it up in stable
after it hits Linus's tree.
thanks,
greg k-h
On Wed, Oct 06, 2021 at 01:27:02AM +0000, Joel Stanley wrote:
> Hi Greg,
>
> On Fri, 10 Sept 2021 at 07:38, Neal Liu <[email protected]> wrote:
> >
> > For Aspeed, HCHalted status depends on not only Run/Stop but also
> > ASS/PSS status.
> > Handshake CMD_RUN on startup instead.
> >
> > Signed-off-by: Neal Liu <[email protected]>
> > Acked-by: Alan Stern <[email protected]>
> > Reviewed-by: Tao Ren <[email protected]>
> > Tested-by: Tao Ren <[email protected]>
>
> Has this one gone in? If not, here's some tags:
>
> Reviewed-by: Joel Stanley <[email protected]>
>
> It also needs to go to stable:
>
> Cc: [email protected]
> Fixes: 280a9045bb18 ("ehci: fix EHCI host controller initialization sequence")
This patch failed to apply to 4.4.y, 4.9.y, and 4.14.y, so could you
provide a working backport for those kernels for me to be able to queue
it up to the stable queues there?
thanks,
greg k-h