Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

[PATCH] usb/core: skip unnecessary line IRQ request for USB3

33 views
Skip to first unread message

Alex Shi

unread,
Mar 15, 2012, 12:50:01 AM3/15/12
to

USB3 requests line IRQ here but will disable it
in later driver->start function and try MSI first.

xhci_hcd 0000:02:00.0: irq 18, io mem 0xfe500000
xhci_hcd 0000:02:00.0: irq 45 for MSI/MSI-X
xhci_hcd 0000:02:00.0: irq 46 for MSI/MSI-X

So it is better to remove the redundant request here. And
save a little time in booting.

Signed-off-by: Alex Shi <alex...@intel.com>
---
drivers/usb/core/hcd.c | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index e128232..5b09825 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -2450,7 +2450,8 @@ int usb_add_hcd(struct usb_hcd *hcd,
/* enable irqs just before we start the controller,
* if the BIOS provides legacy PCI irqs.
*/
- if (usb_hcd_is_primary_hcd(hcd) && irqnum) {
+ if (usb_hcd_is_primary_hcd(hcd) && irqnum
+ && (hcd->driver->flags & HCD_MASK) != HCD_USB3) {
retval = usb_hcd_request_irqs(hcd, irqnum, irqflags);
if (retval)
goto err_request_irq;
--
1.7.5.1



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Xu, Andiry

unread,
Mar 15, 2012, 2:30:01 AM3/15/12
to
> -----Original Message-----
> From: linux-u...@vger.kernel.org [mailto:linux-usb-
> ow...@vger.kernel.org] On Behalf Of Alex Shi
> Sent: Thursday, March 15, 2012 12:48 PM
> To: st...@rowland.harvard.edu; gregkh
> Cc: sarah....@linux.intel.com; linux-...@vger.kernel.org; linux-usb
> Subject: [PATCH] usb/core: skip unnecessary line IRQ request for USB3
>
>
> USB3 requests line IRQ here but will disable it
> in later driver->start function and try MSI first.
>

You may also remove the unregister the legacy interrupt part in
xhci_try_enable_msi().

In fact, I think all the MSI related codes should be moved to xhci-pci.c,
Since MSI is PCI-only.

Thanks,
Andiry
> To unsubscribe from this list: send the line "unsubscribe linux-usb" in

Alex Shi

unread,
Mar 15, 2012, 2:40:02 AM3/15/12
to

>
> You may also remove the unregister the legacy interrupt part in
> xhci_try_enable_msi().

Sure, thanks!

>From 6ae16651ae33ea1841984f3a5eb4a691ed730a19 Mon Sep 17 00:00:00 2001
From: Alex Shi <alex...@intel.com>
Date: Thu, 15 Mar 2012 08:12:47 +0800
Subject: [PATCH] usb/core: skip unnecessary line IRQ request for USB3

USB3(only xhci driver) requests line IRQ here but will disable it
in later driver->start function and try MSI first:

xhci_hcd 0000:02:00.0: irq 18, io mem 0xfe500000
xhci_hcd 0000:02:00.0: irq 45 for MSI/MSI-X
xhci_hcd 0000:02:00.0: irq 46 for MSI/MSI-X

So it is better to remove the redundant request here.

Signed-off-by: Alex Shi <alex...@intel.com>
---
drivers/usb/core/hcd.c | 3 ++-
drivers/usb/host/xhci.c | 3 ---
2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index e128232..5b09825 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -2450,7 +2450,8 @@ int usb_add_hcd(struct usb_hcd *hcd,
/* enable irqs just before we start the controller,
* if the BIOS provides legacy PCI irqs.
*/
- if (usb_hcd_is_primary_hcd(hcd) && irqnum) {
+ if (usb_hcd_is_primary_hcd(hcd) && irqnum
+ && (hcd->driver->flags & HCD_MASK) != HCD_USB3) {
retval = usb_hcd_request_irqs(hcd, irqnum, irqflags);
if (retval)
goto err_request_irq;
diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
index c939f5f..64979c2 100644
--- a/drivers/usb/host/xhci.c
+++ b/drivers/usb/host/xhci.c
@@ -338,9 +338,6 @@ static int xhci_try_enable_msi(struct usb_hcd *hcd)
if (xhci->quirks & XHCI_BROKEN_MSI)
return 0;

- /* unregister the legacy interrupt */
- if (hcd->irq)
- free_irq(hcd->irq, hcd);
hcd->irq = -1;

ret = xhci_setup_msix(xhci);
--
1.6.3.3


>
> In fact, I think all the MSI related codes should be moved to xhci-pci.c,
> Since MSI is PCI-only.

A patch moving MSI from xhci to USB core is in reviewing. That walks
far. :) https://lkml.org/lkml/2012/2/20/76


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

Alan Stern

unread,
Mar 15, 2012, 10:30:01 AM3/15/12
to
This is not a good idea. What happens if some USB-3 controller
(perhaps a non-xHCI USB-3 controller) doesn't use MSI?

The right way to avoid registering a legacy interrupt handler is to
call usb_add_hcd() with irqnum equal to 0.

Alan Stern

Alex Shi

unread,
Mar 15, 2012, 10:40:02 AM3/15/12
to

>
> This is not a good idea. What happens if some USB-3 controller
> (perhaps a non-xHCI USB-3 controller) doesn't use MSI?
>


current usb3 driver just xhci. but sure it need to modify if non-xhci
usb pops up.

> The right way to avoid registering a legacy interrupt handler is to
> call usb_add_hcd() with irqnum equal to 0.


a good trick! will try it tomorrow.

Alex Shi

unread,
Mar 16, 2012, 4:10:01 AM3/16/12
to
On Thu, 2012-03-15 at 22:39 +0800, Alex Shi wrote:
> >
> > This is not a good idea. What happens if some USB-3 controller
> > (perhaps a non-xHCI USB-3 controller) doesn't use MSI?
> >
>
>
> current usb3 driver just xhci. but sure it need to modify if non-xhci
> usb pops up.
>
> > The right way to avoid registering a legacy interrupt handler is to
> > call usb_add_hcd() with irqnum equal to 0.
>
>
> a good trick! will try it tomorrow.

Do you mean the following patch? It tested. :)
------------
>From 5ae9a6fe88752495af6da6182cad26620f1ab190 Mon Sep 17 00:00:00 2001
From: Alex Shi <alex...@intel.com>
Date: Thu, 15 Mar 2012 08:12:47 +0800
Subject: [PATCH] usb/core: skip unnecessary line IRQ request for USB3

USB3(only xhci driver) requests line IRQ here but will disable it
in later driver->start function and try MSI first:

xhci_hcd 0000:02:00.0: irq 18, io mem 0xfe500000
xhci_hcd 0000:02:00.0: irq 45 for MSI/MSI-X
xhci_hcd 0000:02:00.0: irq 46 for MSI/MSI-X

So it is better to remove the redundant request here.

Signed-off-by: Alex Shi <alex...@intel.com>
---
drivers/usb/core/hcd-pci.c | 5 ++++-
drivers/usb/host/xhci.c | 3 ---
2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/usb/core/hcd-pci.c b/drivers/usb/core/hcd-pci.c
index 81e2c0d..7b73c74 100644
--- a/drivers/usb/core/hcd-pci.c
+++ b/drivers/usb/core/hcd-pci.c
@@ -245,7 +245,10 @@ int usb_hcd_pci_probe(struct pci_dev *dev, const struct pci_device_id *id)

pci_set_master(dev);

- retval = usb_add_hcd(hcd, dev->irq, IRQF_SHARED);
+ if ((driver->flags & HCD_MASK) != HCD_USB3)
+ retval = usb_add_hcd(hcd, dev->irq, IRQF_SHARED);
+ else
+ retval = usb_add_hcd(hcd, 0, IRQF_SHARED);
if (retval != 0)
goto unmap_registers;
set_hs_companion(dev, hcd);
diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
index c939f5f..64979c2 100644
--- a/drivers/usb/host/xhci.c
+++ b/drivers/usb/host/xhci.c
@@ -338,9 +338,6 @@ static int xhci_try_enable_msi(struct usb_hcd *hcd)
if (xhci->quirks & XHCI_BROKEN_MSI)
return 0;

- /* unregister the legacy interrupt */
- if (hcd->irq)
- free_irq(hcd->irq, hcd);
hcd->irq = -1;

ret = xhci_setup_msix(xhci);
--
1.6.3.3



>
> >
0 new messages