Skip to contentSkip to conversation

Question: issues for usb keyboard work with OHCI HCD

U-Boot Bo Shen 3 participants 9 messages May 3, 2013 – May 14, 2013

Subsystems (2)
lore
Share & tools
lore

Bo Shen

author
#0 lore raw

Hi All,
Now, I test usb host support with Atmel boards, for example,
at91sam9x5ek board.

When test OHCI USB host with usb keyboard. I meet the following issue.
--->8---
U-Boot 2013.04-dirty (May 03 2013 - 11:00:34)

CPU: AT91SAM9G35
Crystal frequency:       12 MHz
CPU clock        :      400 MHz
Master clock     :  133.333 MHz
DRAM:  128 MiB
WARNING: Caches not enabled
NAND:  256 MiB
MMC:   mci: 0
In:    serial
Out:   serial
Err:   serial
Net:   macb0
Hit any key to stop autoboot:  0

U-Boot> usb start
(Re)start USB...
USB0: scanning bus 0 for devices... 2 USB Device(s) found
scanning usb for storage devices... 0 Storage Device(s) found
U-Boot> setenv stdin usbkbd
U-Boot> ERROR: sohci_submit_job: ENOMEM
ERROR: sohci_submit_job failed
... ...
(repeat to print these two error line)
---<8---

After dig the usb ohci-hcd.c driver, I found every time it call
submit_int_msg -->
submit_common_msg --> sohci_submit_job, it will allocate memory for td (just call td_alloc),
however nowhere free the td. So, after allocate 64 times (#define NUM_TD 64),
all the ptd[i].usb_dev is not NULL. So, it will report: ENOMEM.

I don't know why in sohci_return_job it call td_submit_job again?
--->8---
static inline int sohci_return_job(struct ohci *hc, urb_priv_t *urb)

{
     struct ohci_regs *regs = hc->regs;

     switch (usb_pipetype(urb->pipe)) {
     case PIPE_INTERRUPT:
         /* implicitly requeued */
         if (urb->dev->irq_handle &&
                 (urb->dev->irq_act_len = urb->actual_length)) {
             ohci_writel(OHCI_INTR_WDH, &regs->intrenable);
             ohci_readl(&regs->intrenable); /* PCI posting flush */
             urb->dev->irq_handle(urb->dev);
             ohci_writel(OHCI_INTR_WDH, &regs->intrdisable);
             ohci_readl(&regs->intrdisable); /* PCI posting flush */
         }
         urb->actual_length = 0;
         td_submit_job(
                 urb->dev,
                 urb->pipe,
                 urb->transfer_buffer,
                 urb->transfer_buffer_length,
                 NULL,
                 urb,
                 urb->interval);
         break

---<8---

Add where should we free the allocated td to fix the issue?

Thanks.

Best Regards,
Bo Shen

#1 lore raw

Dear Bo Shen,

Quoted 21 lines

Hi All,
Now, I test usb host support with Atmel boards, for example,
at91sam9x5ek board.

When test OHCI USB host with usb keyboard. I meet the following issue.
--->8---
U-Boot 2013.04-dirty (May 03 2013 - 11:00:34)

CPU: AT91SAM9G35
Crystal frequency:       12 MHz
CPU clock        :      400 MHz
Master clock     :  133.333 MHz
DRAM:  128 MiB
WARNING: Caches not enabled
NAND:  256 MiB
MMC:   mci: 0
In:    serial
Out:   serial
Err:   serial
Net:   macb0
Hit any key to stop autoboot:  0
Quoted 9 lines

U-Boot> usb start
(Re)start USB...
USB0: scanning bus 0 for devices... 2 USB Device(s) found
scanning usb for storage devices... 0 Storage Device(s) found
U-Boot> setenv stdin usbkbd
U-Boot> ERROR: sohci_submit_job: ENOMEM
ERROR: sohci_submit_job failed
... ...
(repeat to print these two error line)

So the USB subsystem is leaking memory? Or do you only have too small MALLOC area?

Quoted 37 lines

---<8---

After dig the usb ohci-hcd.c driver, I found every time it call
submit_int_msg -->
submit_common_msg --> sohci_submit_job, it will allocate memory for td (just call td_alloc),
however nowhere free the td. So, after allocate 64 times (#define NUM_TD 64),
all the ptd[i].usb_dev is not NULL. So, it will report: ENOMEM.

I don't know why in sohci_return_job it call td_submit_job again?
--->8---
static inline int sohci_return_job(struct ohci *hc, urb_priv_t *urb)

{
     struct ohci_regs *regs = hc->regs;

     switch (usb_pipetype(urb->pipe)) {
     case PIPE_INTERRUPT:
         /* implicitly requeued */
         if (urb->dev->irq_handle &&
                 (urb->dev->irq_act_len = urb->actual_length)) {
             ohci_writel(OHCI_INTR_WDH, &regs->intrenable);
             ohci_readl(&regs->intrenable); /* PCI posting flush */
             urb->dev->irq_handle(urb->dev);
             ohci_writel(OHCI_INTR_WDH, &regs->intrdisable);
             ohci_readl(&regs->intrdisable); /* PCI posting flush */
         }
         urb->actual_length = 0;
         td_submit_job(
                 urb->dev,
                 urb->pipe,
                 urb->transfer_buffer,
                 urb->transfer_buffer_length,
                 NULL,
                 urb,
                 urb->interval);
         break
Quoted 8 lines

---<8---

Add where should we free the allocated td to fix the issue?

Thanks.

Best Regards,
Bo Shen

Best regards,
Marek Vasut

author
#2 lore raw

Hi Marek,

Quoted 33 lines · Marek Vasut

Dear Bo Shen,

Quoted 21 lines

Hi All,
Now, I test usb host support with Atmel boards, for example,
at91sam9x5ek board.

When test OHCI USB host with usb keyboard. I meet the following issue.
--->8---
U-Boot 2013.04-dirty (May 03 2013 - 11:00:34)

CPU: AT91SAM9G35
Crystal frequency:       12 MHz
CPU clock        :      400 MHz
Master clock     :  133.333 MHz
DRAM:  128 MiB
WARNING: Caches not enabled
NAND:  256 MiB
MMC:   mci: 0
In:    serial
Out:   serial
Err:   serial
Net:   macb0
Hit any key to stop autoboot:  0
Quoted 9 lines

U-Boot> usb start
(Re)start USB...
USB0: scanning bus 0 for devices... 2 USB Device(s) found
scanning usb for storage devices... 0 Storage Device(s) found
U-Boot> setenv stdin usbkbd
U-Boot> ERROR: sohci_submit_job: ENOMEM
ERROR: sohci_submit_job failed
... ...
(repeat to print these two error line)

Quoted 2 lines

So the USB subsystem is leaking memory? Or do you only have too small MALLOC area?

I am not sure whether USB subsystem is leaking memory. I am digging it.

This issue is not relative with MALLOC area.
This issue come out when all ptd[i].usb_dev (the maximum value of i is
64) is not NULL. Each time to call td_alloc, it will check whether ptd[i].usb_dev is NULL (i from 0 to 63), if not find one of ptd[i].usb_dev is NULL, then report ENOMEM.

Quoted 11 lines
Quoted 11 lines

---<8---

After dig the usb ohci-hcd.c driver, I found every time it call
submit_int_msg -->
submit_common_msg --> sohci_submit_job, it will allocate memory for td (just call td_alloc),
however nowhere free the td. So, after allocate 64 times (#define NUM_TD 64),
all the ptd[i].usb_dev is not NULL. So, it will report: ENOMEM.

I don't know why in sohci_return_job it call td_submit_job again?

Any advice for this question?

Quoted 20 lines
Quoted 20 lines

--->8---
static inline int sohci_return_job(struct ohci *hc, urb_priv_t *urb)

{
      struct ohci_regs *regs = hc->regs;

      switch (usb_pipetype(urb->pipe)) {
      case PIPE_INTERRUPT:
          /* implicitly requeued */
          if (urb->dev->irq_handle &&
                  (urb->dev->irq_act_len = urb->actual_length)) {
              ohci_writel(OHCI_INTR_WDH, &regs->intrenable);
              ohci_readl(&regs->intrenable); /* PCI posting flush */
              urb->dev->irq_handle(urb->dev);
              ohci_writel(OHCI_INTR_WDH, &regs->intrdisable);
              ohci_readl(&regs->intrdisable); /* PCI posting flush */
          }
          urb->actual_length = 0;
          td_submit_job(
                  urb->dev,
                  urb->pipe,
Quoted 8 lines
Quoted 8 lines
                  urb->transfer_buffer,
                  urb->transfer_buffer_length,
                  NULL,
                  urb,
                  urb->interval);
          break

---<8---
Add where should we free the allocated td to fix the issue?

If I comment the td_submit_job and add urb_free_priv(urb) here, then usb keyboard work properly. It looks like

--->8---

drivers/usb/host/ohci-hcd.c +0 -0
index 9f47351..cc60bc5 100644
@@ -516,6 +516,7 @@ static inline int sohci_return_job(struct ohci *hc, 

urb_priv_t *urb)

                         ohci_readl(&regs->intrdisable); /* PCI posting 
flush */
                 }
                 urb->actual_length = 0;
+#if 0
                 td_submit_job(
                                 urb->dev,
                                 urb->pipe,
@@ -524,6 +525,8 @@ static inline int sohci_return_job(struct ohci *hc, 
urb_priv_t *urb)
                                 NULL,
                                 urb,
                                 urb->interval);

+#endif
+ urb_free_priv(urb);
---<8---

Quoted 8 lines
Quoted 4 lines

Thanks.

Best Regards,
Bo Shen

Best regards,
Marek Vasut

Best Regards,
Bo Shen

author
#3 lore raw

Hi All,

Quoted 41 lines · Bo Shen

Hi Marek,

Quoted 34 lines · Marek Vasut

Dear Bo Shen,

Quoted 22 lines

Hi All,
Now, I test usb host support with Atmel boards, for example,
at91sam9x5ek board.

When test OHCI USB host with usb keyboard. I meet the following
issue.
--->8---
U-Boot 2013.04-dirty (May 03 2013 - 11:00:34)

CPU: AT91SAM9G35
Crystal frequency:       12 MHz
CPU clock        :      400 MHz
Master clock     :  133.333 MHz
DRAM:  128 MiB
WARNING: Caches not enabled
NAND:  256 MiB
MMC:   mci: 0
In:    serial
Out:   serial
Err:   serial
Net:   macb0
Hit any key to stop autoboot:  0
Quoted 9 lines

U-Boot> usb start
(Re)start USB...
USB0: scanning bus 0 for devices... 2 USB Device(s) found
scanning usb for storage devices... 0 Storage Device(s) found
U-Boot> setenv stdin usbkbd
U-Boot> ERROR: sohci_submit_job: ENOMEM
ERROR: sohci_submit_job failed
... ...
(repeat to print these two error line)

Quoted 3 lines

So the USB subsystem is leaking memory? Or do you only have too small MALLOC
area?

Quoted 7 lines

I am not sure whether USB subsystem is leaking memory. I am digging it.

This issue is not relative with MALLOC area.
This issue come out when all ptd[i].usb_dev (the maximum value of i is
64) is not NULL. Each time to call td_alloc, it will check whether ptd[i].usb_dev is NULL (i from 0 to 63), if not find one of ptd[i].usb_dev is NULL, then report ENOMEM.

All clue for this issue?

Quoted 10 lines
Quoted 10 lines
Quoted 10 lines

---<8---

After dig the usb ohci-hcd.c driver, I found every time it call
submit_int_msg -->
submit_common_msg --> sohci_submit_job, it will allocate memory for td (just call td_alloc),
however nowhere free the td. So, after allocate 64 times (#define NUM_TD 64),
all the ptd[i].usb_dev is not NULL. So, it will report: ENOMEM.

Quoted 3 lines
Quoted 1 line
Quoted 1 line

I don't know why in sohci_return_job it call td_submit_job again?

Any advice for this question?

Any clue for this question?

Quoted 48 lines
Quoted 20 lines
Quoted 20 lines

--->8---
static inline int sohci_return_job(struct ohci *hc, urb_priv_t *urb)

{
      struct ohci_regs *regs = hc->regs;

      switch (usb_pipetype(urb->pipe)) {
      case PIPE_INTERRUPT:
          /* implicitly requeued */
          if (urb->dev->irq_handle &&
                  (urb->dev->irq_act_len = urb->actual_length)) {
              ohci_writel(OHCI_INTR_WDH, &regs->intrenable);
              ohci_readl(&regs->intrenable); /* PCI posting flush */
              urb->dev->irq_handle(urb->dev);
              ohci_writel(OHCI_INTR_WDH, &regs->intrdisable);
              ohci_readl(&regs->intrdisable); /* PCI posting flush */
          }
          urb->actual_length = 0;
          td_submit_job(
                  urb->dev,
                  urb->pipe,
Quoted 8 lines
Quoted 8 lines
                  urb->transfer_buffer,
                  urb->transfer_buffer_length,
                  NULL,
                  urb,
                  urb->interval);
          break

---<8---
Add where should we free the allocated td to fix the issue?

If I comment the td_submit_job and add urb_free_priv(urb) here, then usb keyboard work properly. It looks like

--->8---

drivers/usb/host/ohci-hcd.c +0 -0
index 9f47351..cc60bc5 100644
@@ -516,6 +516,7 @@ static inline int sohci_return_job(struct ohci *hc,
 urb_priv_t *urb)
                         ohci_readl(&regs->intrdisable); /* PCI posting

flush */

                 }
                 urb->actual_length = 0;
drivers/usb/host/ohci-hcd.c +1 -0
+#if 0
                 td_submit_job(
                                 urb->dev,
                                 urb->pipe,

drivers/usb/host/ohci-hcd.c static inline int sohci_return_job(struct ohci *hc, · line 531

Quoted 8 lines
drivers/usb/host/ohci-hcd.c +2 -1
@@ -524,6 +525,8 @@ static inline int sohci_return_job(struct ohci *hc,
 urb_priv_t *urb)
                                 NULL,
                                 urb,
                                 urb->interval);
+#endif
+               urb_free_priv(urb);
---<8---
 

After digging more about the USB HCD driver, I don't find any better solution for this issue, except this one.

So, may I submit this as an RFC patch?

Best Regards,
Bo Shen

#4 lore raw

Dear Bo Shen,

Quoted 51 lines

Hi All,

Quoted 48 lines · Bo Shen

Hi Marek,

Quoted 39 lines · Marek Vasut

Dear Bo Shen,

Quoted 29 lines

Hi All,

Now, I test usb host support with Atmel boards, for example,

at91sam9x5ek board.

When test OHCI USB host with usb keyboard. I meet the following

issue.
--->8---
U-Boot 2013.04-dirty (May 03 2013 - 11:00:34)

CPU: AT91SAM9G35
Crystal frequency:       12 MHz
CPU clock        :      400 MHz
Master clock     :  133.333 MHz
DRAM:  128 MiB
WARNING: Caches not enabled
NAND:  256 MiB
MMC:   mci: 0
In:    serial
Out:   serial
Err:   serial
Net:   macb0
Hit any key to stop autoboot:  0

U-Boot> usb start
(Re)start USB...
USB0: scanning bus 0 for devices... 2 USB Device(s) found

Quoted 7 lines

scanning usb for storage devices... 0 Storage Device(s) found

U-Boot> setenv stdin usbkbd
U-Boot> ERROR: sohci_submit_job: ENOMEM
ERROR: sohci_submit_job failed
... ...
(repeat to print these two error line)

Quoted 3 lines

So the USB subsystem is leaking memory? Or do you only have too small MALLOC
area?

I am not sure whether USB subsystem is leaking memory. I am digging it.

Quoted 7 lines
Quoted 5 lines

This issue is not relative with MALLOC area.
This issue come out when all ptd[i].usb_dev (the maximum value of i is
64) is not NULL. Each time to call td_alloc, it will check whether ptd[i].usb_dev is NULL (i from 0 to 63), if not find one of ptd[i].usb_dev is NULL, then report ENOMEM.

All clue for this issue?

I assume you mean the TDs are all used up then? Are they not free'd ?

Best regards,
Marek Vasut

author
#5 lore raw

Hi Marek,

Quoted 59 lines · Marek Vasut

Dear Bo Shen,

Quoted 57 lines

Hi All,

Quoted 53 lines · Bo Shen

Hi Marek,

Quoted 39 lines · Marek Vasut

Dear Bo Shen,

Quoted 29 lines

Hi All,

Now, I test usb host support with Atmel boards, for example,

at91sam9x5ek board.

When test OHCI USB host with usb keyboard. I meet the following

issue.
--->8---
U-Boot 2013.04-dirty (May 03 2013 - 11:00:34)

CPU: AT91SAM9G35
Crystal frequency:       12 MHz
CPU clock        :      400 MHz
Master clock     :  133.333 MHz
DRAM:  128 MiB
WARNING: Caches not enabled
NAND:  256 MiB
MMC:   mci: 0
In:    serial
Out:   serial
Err:   serial
Net:   macb0
Hit any key to stop autoboot:  0

U-Boot> usb start
(Re)start USB...
USB0: scanning bus 0 for devices... 2 USB Device(s) found

Quoted 7 lines

scanning usb for storage devices... 0 Storage Device(s) found

U-Boot> setenv stdin usbkbd
U-Boot> ERROR: sohci_submit_job: ENOMEM
ERROR: sohci_submit_job failed
... ...
(repeat to print these two error line)

Quoted 3 lines

So the USB subsystem is leaking memory? Or do you only have too small MALLOC
area?

I am not sure whether USB subsystem is leaking memory. I am digging it.

This issue is not relative with MALLOC area.
This issue come out when all ptd[i].usb_dev (the maximum value of i is
64) is not NULL. Each time to call td_alloc, it will check whether ptd[i].usb_dev is NULL (i from 0 to 63), if not find one of ptd[i].usb_dev is NULL, then report ENOMEM.

Quoted 3 lines
Quoted 1 line

All clue for this issue?

I assume you mean the TDs are all used up then? Are they not free'd ?

Yes, that's true, all TDs are used. Nowhere free them, so I use urb_free_priv() to free urb in sohci_return_job() instead of td_submit_job(), then it works well.

Quoted 2 lines

Best regards,
Marek Vasut

Best Regards,
Bo Shen

#6 lore raw

Dear Bo Shen,

Quoted 63 lines

Hi Marek,

Quoted 60 lines · Marek Vasut

Dear Bo Shen,

Quoted 58 lines

Hi All,

Quoted 54 lines · Bo Shen

Hi Marek,

Quoted 40 lines · Marek Vasut

Dear Bo Shen,

Quoted 29 lines

Hi All,

Now, I test usb host support with Atmel boards, for example,

at91sam9x5ek board.

When test OHCI USB host with usb keyboard. I meet the following

issue.
--->8---
U-Boot 2013.04-dirty (May 03 2013 - 11:00:34)

CPU: AT91SAM9G35
Crystal frequency:       12 MHz
CPU clock        :      400 MHz
Master clock     :  133.333 MHz
DRAM:  128 MiB
WARNING: Caches not enabled
NAND:  256 MiB
MMC:   mci: 0
In:    serial
Out:   serial
Err:   serial
Net:   macb0
Hit any key to stop autoboot:  0

U-Boot> usb start
(Re)start USB...
USB0: scanning bus 0 for devices... 2 USB Device(s) found

Quoted 8 lines

scanning usb for storage devices... 0 Storage Device(s) found

U-Boot> setenv stdin usbkbd
U-Boot> ERROR: sohci_submit_job: ENOMEM
ERROR: sohci_submit_job failed
... ...
(repeat to print these two error line)

Quoted 3 lines

So the USB subsystem is leaking memory? Or do you only have too small MALLOC
area?

I am not sure whether USB subsystem is leaking memory. I am digging it.

This issue is not relative with MALLOC area.
This issue come out when all ptd[i].usb_dev (the maximum value of i is
64) is not NULL. Each time to call td_alloc, it will check whether ptd[i].usb_dev is NULL (i from 0 to 63), if not find one of ptd[i].usb_dev is NULL, then report ENOMEM.

Quoted 7 lines
Quoted 3 lines
Quoted 1 line

All clue for this issue?

I assume you mean the TDs are all used up then? Are they not free'd ?

Yes, that's true, all TDs are used. Nowhere free them, so I use urb_free_priv() to free urb in sohci_return_job() instead of td_submit_job(), then it works well.

Benoit, you were the last one to plumb in it, right?

Best regards,
Marek Vasut

#7 lore raw

Hi Marek,

Quoted 67 lines · Marek Vasut

Dear Bo Shen,

Quoted 65 lines

Hi Marek,

Quoted 62 lines · Marek Vasut

Dear Bo Shen,

Quoted 59 lines

Hi All,

Quoted 47 lines · Bo Shen

Hi Marek,

Quoted 40 lines · Marek Vasut

Dear Bo Shen,

Quoted 29 lines

Hi All,

Now, I test usb host support with Atmel boards, for example,

at91sam9x5ek board.

When test OHCI USB host with usb keyboard. I meet the following

issue.
--->8---
U-Boot 2013.04-dirty (May 03 2013 - 11:00:34)

CPU: AT91SAM9G35
Crystal frequency:       12 MHz
CPU clock        :      400 MHz
Master clock     :  133.333 MHz
DRAM:  128 MiB
WARNING: Caches not enabled
NAND:  256 MiB
MMC:   mci: 0
In:    serial
Out:   serial
Err:   serial
Net:   macb0
Hit any key to stop autoboot:  0

U-Boot> usb start
(Re)start USB...
USB0: scanning bus 0 for devices... 2 USB Device(s) found

Quoted 8 lines

scanning usb for storage devices... 0 Storage Device(s) found

U-Boot> setenv stdin usbkbd
U-Boot> ERROR: sohci_submit_job: ENOMEM
ERROR: sohci_submit_job failed
... ...
(repeat to print these two error line)

Quoted 3 lines

So the USB subsystem is leaking memory? Or do you only have too small MALLOC
area?

Quoted 7 lines

I am not sure whether USB subsystem is leaking memory. I am digging it.

This issue is not relative with MALLOC area.
This issue come out when all ptd[i].usb_dev (the maximum value of i is
64) is not NULL. Each time to call td_alloc, it will check whether ptd[i].usb_dev is NULL (i from 0 to 63), if not find one of ptd[i].usb_dev is NULL, then report ENOMEM.

All clue for this issue?

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

I assume you mean the TDs are all used up then? Are they not free'd ?

Yes, that's true, all TDs are used. Nowhere free them, so I use urb_free_priv() to free urb in sohci_return_job() instead of td_submit_job(), then it works well.

Benoit, you were the last one to plumb in it, right?

No. That was for EHCI, not OHCI.

Best regards,
Beno?t

#8 lore raw

Dear Beno?t Th?baudeau,

Quoted 74 lines

Hi Marek,

Quoted 71 lines · Marek Vasut

Dear Bo Shen,

Quoted 69 lines

Hi Marek,

Quoted 65 lines · Marek Vasut

Dear Bo Shen,

Quoted 54 lines

Hi All,

Quoted 51 lines · Bo Shen

Hi Marek,

Quoted 41 lines · Marek Vasut

Dear Bo Shen,

Quoted 30 lines

Hi All,

Now, I test usb host support with Atmel boards, for example,

at91sam9x5ek board.

When test OHCI USB host with usb keyboard. I meet the following

issue.
--->8---
U-Boot 2013.04-dirty (May 03 2013 - 11:00:34)

CPU: AT91SAM9G35
Crystal frequency:       12 MHz
CPU clock        :      400 MHz
Master clock     :  133.333 MHz
DRAM:  128 MiB
WARNING: Caches not enabled
NAND:  256 MiB
MMC:   mci: 0
In:    serial
Out:   serial
Err:   serial
Net:   macb0
Hit any key to stop autoboot:  0

U-Boot> usb start
(Re)start USB...
USB0: scanning bus 0 for devices... 2 USB Device(s) found

Quoted 8 lines

scanning usb for storage devices... 0 Storage Device(s) found

U-Boot> setenv stdin usbkbd
U-Boot> ERROR: sohci_submit_job: ENOMEM
ERROR: sohci_submit_job failed
... ...
(repeat to print these two error line)

Quoted 3 lines

So the USB subsystem is leaking memory? Or do you only have too small MALLOC
area?

I am not sure whether USB subsystem is leaking memory. I am digging it.

Quoted 7 lines
Quoted 5 lines

This issue is not relative with MALLOC area.
This issue come out when all ptd[i].usb_dev (the maximum value of i is 64) is not NULL. Each time to call td_alloc, it will check whether ptd[i].usb_dev is NULL (i from 0 to 63), if not find one of ptd[i].usb_dev is NULL, then report ENOMEM.

All clue for this issue?

I assume you mean the TDs are all used up then? Are they not free'd ?

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines

Yes, that's true, all TDs are used. Nowhere free them, so I use urb_free_priv() to free urb in sohci_return_job() instead of td_submit_job(), then it works well.

Benoit, you were the last one to plumb in it, right?

No. That was for EHCI, not OHCI.

Hmmm ... I don't have any OHCI device. Bo, maybe you can try tracing it and fixing it?

Best regards,
Marek Vasut