Skip to contentSkip to conversation

usb: xhci: Error enumerating USB TV Tuner

PATCH In review Al Cooper 2 participants 5 messages Mar 6, 2020 – Mar 11, 2020

Subsystems (2)
lore
Share & tools
lore

Al Cooper

author
#0 lore raw

PATCH 1/1 usb: xhci: Error enumerating USB TV Tuner

Unable to complete the enumeration of a USB TV Tuner device.

Per XHCI spec (4.6.5), the EP state field of the input context shall be cleared for a set address command. In the special case of an FS device that has "MaxPacketSize0 = 8", the Linux XHCI driver does not do this before evaluating the context. With an XHCI controller that checks the EP state field for parameter context error this causes a problem in cases such as the device getting reset again after enumeration.

When that field is cleared, the problem does not occur.

This was found and fixed by Sasi Kumar.

---
 drivers/usb/host/xhci.c | 1 +
 1 file changed, 1 insertion(+)
drivers/usb/host/xhci.c +1 -0
index dbac0fa9748d..5f034e143082 100644
@@ -1428,6 +1428,7 @@ static int xhci_check_maxpacket(struct xhci_hcd *xhci, unsigned int slot_id,
 				xhci->devs[slot_id]->out_ctx, ep_index);
 
 		ep_ctx = xhci_get_ep_ctx(xhci, command->in_ctx, ep_index);
+		ep_ctx->ep_info &= cpu_to_le32(~EP_STATE_MASK);/* must clear */
 		ep_ctx->ep_info2 &= cpu_to_le32(~MAX_PACKET_MASK);
 		ep_ctx->ep_info2 |= cpu_to_le32(MAX_PACKET(max_packet_size));
 
Signature · 1 line
2.17.1
#1 lore raw
Quoted 25 lines · Al Cooper

Unable to complete the enumeration of a USB TV Tuner device.

Per XHCI spec (4.6.5), the EP state field of the input context shall be cleared for a set address command. In the special case of an FS device that has "MaxPacketSize0 = 8", the Linux XHCI driver does not do this before evaluating the context. With an XHCI controller that checks the EP state field for parameter context error this causes a problem in cases such as the device getting reset again after enumeration.

When that field is cleared, the problem does not occur.

This was found and fixed by Sasi Kumar.

---
 drivers/usb/host/xhci.c | 1 +
 1 file changed, 1 insertion(+)
drivers/usb/host/xhci.c +0 -0
index dbac0fa9748d..5f034e143082 100644

drivers/usb/host/xhci.c static int xhci_check_maxpacket(struct xhci_hcd *xhci, unsigned int slot_id, · line 1434

Quoted 7 lines
drivers/usb/host/xhci.c +1 -0
@@ -1428,6 +1428,7 @@ static int xhci_check_maxpacket(struct xhci_hcd *xhci, unsigned int slot_id,
 				xhci->devs[slot_id]->out_ctx, ep_index);
 
 		ep_ctx = xhci_get_ep_ctx(xhci, command->in_ctx, ep_index);
+		ep_ctx->ep_info &= cpu_to_le32(~EP_STATE_MASK);/* must clear */
 		ep_ctx->ep_info2 &= cpu_to_le32(~MAX_PACKET_MASK);
 		ep_ctx->ep_info2 |= cpu_to_le32(MAX_PACKET(max_packet_size));
 

Thanks, nice catch.

If you agree I'd like to change the the subject of this patch to something like:
"xhci: Fix enumeration issue when setting actual max packet size for FS devices"

While looking at this it seems that the current EP_STATE_MASK is not correct either. It should be 0x7 instead of 0xf.

-Mathias

#2 lore raw

Agree on both points.
Should I re-send a v2?

Thanks
Al

Quoted 44 lines · Mathias Nyman
Quoted 25 lines · Al Cooper

Unable to complete the enumeration of a USB TV Tuner device.

Per XHCI spec (4.6.5), the EP state field of the input context shall be cleared for a set address command. In the special case of an FS device that has "MaxPacketSize0 = 8", the Linux XHCI driver does not do this before evaluating the context. With an XHCI controller that checks the EP state field for parameter context error this causes a problem in cases such as the device getting reset again after enumeration.

When that field is cleared, the problem does not occur.

This was found and fixed by Sasi Kumar.

---
 drivers/usb/host/xhci.c | 1 +
 1 file changed, 1 insertion(+)
drivers/usb/host/xhci.c +0 -0
index dbac0fa9748d..5f034e143082 100644

drivers/usb/host/xhci.c static int xhci_check_maxpacket(struct xhci_hcd *xhci, unsigned int slot_id, · line 1434

Quoted 7 lines
drivers/usb/host/xhci.c +1 -0
@@ -1428,6 +1428,7 @@ static int xhci_check_maxpacket(struct xhci_hcd *xhci, unsigned int slot_id,
                              xhci->devs[slot_id]->out_ctx, ep_index);
 
              ep_ctx = xhci_get_ep_ctx(xhci, command->in_ctx, ep_index);
+             ep_ctx->ep_info &= cpu_to_le32(~EP_STATE_MASK);/* must clear */
              ep_ctx->ep_info2 &= cpu_to_le32(~MAX_PACKET_MASK);
              ep_ctx->ep_info2 |= cpu_to_le32(MAX_PACKET(max_packet_size));
 

Thanks, nice catch.

If you agree I'd like to change the the subject of this patch to something like:
"xhci: Fix enumeration issue when setting actual max packet size for FS devices"

While looking at this it seems that the current EP_STATE_MASK is not correct either. It should be 0x7 instead of 0xf.

-Mathias

#3 lore raw
Quoted 42 lines · Alan Cooper
Quoted 40 lines · Mathias Nyman
Quoted 25 lines · Al Cooper

Unable to complete the enumeration of a USB TV Tuner device.

Per XHCI spec (4.6.5), the EP state field of the input context shall be cleared for a set address command. In the special case of an FS device that has "MaxPacketSize0 = 8", the Linux XHCI driver does not do this before evaluating the context. With an XHCI controller that checks the EP state field for parameter context error this causes a problem in cases such as the device getting reset again after enumeration.

When that field is cleared, the problem does not occur.

This was found and fixed by Sasi Kumar.

---
 drivers/usb/host/xhci.c | 1 +
 1 file changed, 1 insertion(+)
drivers/usb/host/xhci.c +0 -0
index dbac0fa9748d..5f034e143082 100644

drivers/usb/host/xhci.c static int xhci_check_maxpacket(struct xhci_hcd *xhci, unsigned int slot_id, · line 1434

Quoted 7 lines
drivers/usb/host/xhci.c +1 -0
@@ -1428,6 +1428,7 @@ static int xhci_check_maxpacket(struct xhci_hcd *xhci, unsigned int slot_id,
                              xhci->devs[slot_id]->out_ctx, ep_index);
 
              ep_ctx = xhci_get_ep_ctx(xhci, command->in_ctx, ep_index);
+             ep_ctx->ep_info &= cpu_to_le32(~EP_STATE_MASK);/* must clear */
              ep_ctx->ep_info2 &= cpu_to_le32(~MAX_PACKET_MASK);
              ep_ctx->ep_info2 |= cpu_to_le32(MAX_PACKET(max_packet_size));
 

Thanks, nice catch.

If you agree I'd like to change the the subject of this patch to something like:
"xhci: Fix enumeration issue when setting actual max packet size for FS devices"

Quoted 7 lines
Quoted 3 lines

While looking at this it seems that the current EP_STATE_MASK is not correct either. It should be 0x7 instead of 0xf.

Agree on both points.
Should I re-send a v2?

No need, I can make those changes

-Mathias

#4 lore raw

Thanks!

Al

Quoted 55 lines · Mathias Nyman
Quoted 42 lines · Alan Cooper
Quoted 40 lines · Mathias Nyman
Quoted 25 lines · Al Cooper

Unable to complete the enumeration of a USB TV Tuner device.

Per XHCI spec (4.6.5), the EP state field of the input context shall be cleared for a set address command. In the special case of an FS device that has "MaxPacketSize0 = 8", the Linux XHCI driver does not do this before evaluating the context. With an XHCI controller that checks the EP state field for parameter context error this causes a problem in cases such as the device getting reset again after enumeration.

When that field is cleared, the problem does not occur.

This was found and fixed by Sasi Kumar.

---
 drivers/usb/host/xhci.c | 1 +
 1 file changed, 1 insertion(+)
drivers/usb/host/xhci.c +0 -0
index dbac0fa9748d..5f034e143082 100644

drivers/usb/host/xhci.c static int xhci_check_maxpacket(struct xhci_hcd *xhci, unsigned int slot_id, · line 1434

Quoted 7 lines
drivers/usb/host/xhci.c +1 -0
@@ -1428,6 +1428,7 @@ static int xhci_check_maxpacket(struct xhci_hcd *xhci, unsigned int slot_id,
                              xhci->devs[slot_id]->out_ctx, ep_index);
 
              ep_ctx = xhci_get_ep_ctx(xhci, command->in_ctx, ep_index);
+             ep_ctx->ep_info &= cpu_to_le32(~EP_STATE_MASK);/* must clear */
              ep_ctx->ep_info2 &= cpu_to_le32(~MAX_PACKET_MASK);
              ep_ctx->ep_info2 |= cpu_to_le32(MAX_PACKET(max_packet_size));
 

Thanks, nice catch.

If you agree I'd like to change the the subject of this patch to something like:
"xhci: Fix enumeration issue when setting actual max packet size for FS devices"

Quoted 7 lines
Quoted 3 lines

While looking at this it seems that the current EP_STATE_MASK is not correct either. It should be 0x7 instead of 0xf.

Agree on both points.
Should I re-send a v2?

No need, I can make those changes

-Mathias