[U-Boot] [PATCH] usb: host: Print device name when scanning
Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com --- drivers/usb/host/usb-uclass.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c index 611ea97a72..0575f5393b 100644 --- a/drivers/usb/host/usb-uclass.c +++ b/drivers/usb/host/usb-uclass.c @@ -255,7 +255,7 @@ int usb_init(void)
uclass_foreach_dev(bus, uc) { /* init low_level USB */ - printf("USB%d: ", count); + printf("USB%d(%s): ", count, bus->name); count++;
#ifdef CONFIG_SANDBOX
On 18/Feb/2019 09:23, Ismael Luceno Cortes wrote:
Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com
drivers/usb/host/usb-uclass.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c index 611ea97a72..0575f5393b 100644 --- a/drivers/usb/host/usb-uclass.c +++ b/drivers/usb/host/usb-uclass.c @@ -255,7 +255,7 @@ int usb_init(void)
uclass_foreach_dev(bus, uc) { /* init low_level USB */
printf("USB%d: ", count);
printf("USB%d(%s): ", count, bus->name);#ifdef CONFIG_SANDBOX
2.19.1
Ping.
On 3/14/19 12:44 PM, Ismael Luceno Cortes wrote:
On 18/Feb/2019 09:23, Ismael Luceno Cortes wrote:
Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com
drivers/usb/host/usb-uclass.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c index 611ea97a72..0575f5393b 100644 --- a/drivers/usb/host/usb-uclass.c +++ b/drivers/usb/host/usb-uclass.c @@ -255,7 +255,7 @@ int usb_init(void)
uclass_foreach_dev(bus, uc) { /* init low_level USB */
printf("USB%d: ", count);
printf("USB%d(%s): ", count, bus->name);#ifdef CONFIG_SANDBOX
2.19.1
Ping.
What is this patch doing ? The commit description doesn't explain anything about it.
On 14/Mar/2019 12:55, Marek Vasut wrote:
On 3/14/19 12:44 PM, Ismael Luceno Cortes wrote:
On 18/Feb/2019 09:23, Ismael Luceno Cortes wrote:
Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com
drivers/usb/host/usb-uclass.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c index 611ea97a72..0575f5393b 100644 --- a/drivers/usb/host/usb-uclass.c +++ b/drivers/usb/host/usb-uclass.c @@ -255,7 +255,7 @@ int usb_init(void)
uclass_foreach_dev(bus, uc) { /* init low_level USB */
printf("USB%d: ", count);
printf("USB%d(%s): ", count, bus->name);#ifdef CONFIG_SANDBOX
2.19.1
Ping.
What is this patch doing ? The commit description doesn't explain anything about it.
It prints the host device name. I'm not sure the count is at all useful given there's a name...
On 3/14/19 1:57 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 12:55, Marek Vasut wrote:
On 3/14/19 12:44 PM, Ismael Luceno Cortes wrote:
On 18/Feb/2019 09:23, Ismael Luceno Cortes wrote:
Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com
drivers/usb/host/usb-uclass.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c index 611ea97a72..0575f5393b 100644 --- a/drivers/usb/host/usb-uclass.c +++ b/drivers/usb/host/usb-uclass.c @@ -255,7 +255,7 @@ int usb_init(void)
uclass_foreach_dev(bus, uc) { /* init low_level USB */
printf("USB%d: ", count);
printf("USB%d(%s): ", count, bus->name);#ifdef CONFIG_SANDBOX
2.19.1
Ping.
What is this patch doing ? The commit description doesn't explain anything about it.
It prints the host device name. I'm not sure the count is at all useful given there's a name...
If you could share the log before and after to better illustrate the difference, that'd be nice.
However, shouldn't the same approach be applied to 'usb tree' subcommand and possibly others ?
On 14/Mar/2019 16:09, Marek Vasut wrote:
On 3/14/19 1:57 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 12:55, Marek Vasut wrote:
On 3/14/19 12:44 PM, Ismael Luceno Cortes wrote:
On 18/Feb/2019 09:23, Ismael Luceno Cortes wrote:
Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com
drivers/usb/host/usb-uclass.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c index 611ea97a72..0575f5393b 100644 --- a/drivers/usb/host/usb-uclass.c +++ b/drivers/usb/host/usb-uclass.c @@ -255,7 +255,7 @@ int usb_init(void)
uclass_foreach_dev(bus, uc) { /* init low_level USB */
printf("USB%d: ", count);
printf("USB%d(%s): ", count, bus->name);#ifdef CONFIG_SANDBOX
2.19.1
Ping.
What is this patch doing ? The commit description doesn't explain anything about it.
It prints the host device name. I'm not sure the count is at all useful given there's a name...
If you could share the log before and after to better illustrate the difference, that'd be nice.
unpatched:
=> usb reset resetting USB... USB0: USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
patched:
=> usb reset resetting USB... USB0(usb@ee080100): USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
However, shouldn't the same approach be applied to 'usb tree' subcommand and possibly others ?
The number shown during usb scanning is not used nor saved anywhere else, so seems pretty useless and a special case.
OTOH the number used in the usb tree command is taken from struct usb_device, and is used for lookups.
The name is only relevant to non-discoverable devices at the moment.
dm tree shows: ... ehci_generic | |-- usb@ee080100 ... usb_hub | | `-- usb_hub ... usb_mass_storage | | `-- usb_mass_storage ... usb_storage_blk | | `-- usb_mass_storage.lun0
On 3/14/19 5:19 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 16:09, Marek Vasut wrote:
On 3/14/19 1:57 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 12:55, Marek Vasut wrote:
On 3/14/19 12:44 PM, Ismael Luceno Cortes wrote:
On 18/Feb/2019 09:23, Ismael Luceno Cortes wrote:
Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com
drivers/usb/host/usb-uclass.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c index 611ea97a72..0575f5393b 100644 --- a/drivers/usb/host/usb-uclass.c +++ b/drivers/usb/host/usb-uclass.c @@ -255,7 +255,7 @@ int usb_init(void)
uclass_foreach_dev(bus, uc) { /* init low_level USB */
printf("USB%d: ", count);
printf("USB%d(%s): ", count, bus->name);#ifdef CONFIG_SANDBOX
2.19.1
Ping.
What is this patch doing ? The commit description doesn't explain anything about it.
It prints the host device name. I'm not sure the count is at all useful given there's a name...
If you could share the log before and after to better illustrate the difference, that'd be nice.
unpatched:
=> usb reset resetting USB... USB0: USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
patched:
=> usb reset resetting USB... USB0(usb@ee080100): USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
However, shouldn't the same approach be applied to 'usb tree' subcommand and possibly others ?
The number shown during usb scanning is not used nor saved anywhere else, so seems pretty useless and a special case.
What about usb part ? That one uses the number somehow I think ?
OTOH the number used in the usb tree command is taken from struct usb_device, and is used for lookups.
Maybe it's time to clean that numbering mess up a bit , and make it consistent ?
The name is only relevant to non-discoverable devices at the moment.
dm tree shows: ... ehci_generic | |-- usb@ee080100 ... usb_hub | | `-- usb_hub ... usb_mass_storage | | `-- usb_mass_storage ... usb_storage_blk | | `-- usb_mass_storage.lun0
On 15/Mar/2019 18:34, Marek Vasut wrote:
On 3/14/19 5:19 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 16:09, Marek Vasut wrote:
On 3/14/19 1:57 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 12:55, Marek Vasut wrote:
On 3/14/19 12:44 PM, Ismael Luceno Cortes wrote:
On 18/Feb/2019 09:23, Ismael Luceno Cortes wrote: > Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com > --- > drivers/usb/host/usb-uclass.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c > index 611ea97a72..0575f5393b 100644 > --- a/drivers/usb/host/usb-uclass.c > +++ b/drivers/usb/host/usb-uclass.c > @@ -255,7 +255,7 @@ int usb_init(void) > > uclass_foreach_dev(bus, uc) { > /* init low_level USB */ > - printf("USB%d: ", count); > + printf("USB%d(%s): ", count, bus->name); > count++; > > #ifdef CONFIG_SANDBOX > -- > 2.19.1
Ping.
What is this patch doing ? The commit description doesn't explain anything about it.
It prints the host device name. I'm not sure the count is at all useful given there's a name...
If you could share the log before and after to better illustrate the difference, that'd be nice.
unpatched:
=> usb reset resetting USB... USB0: USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
patched:
=> usb reset resetting USB... USB0(usb@ee080100): USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
However, shouldn't the same approach be applied to 'usb tree' subcommand and possibly others ?
The number shown during usb scanning is not used nor saved anywhere else, so seems pretty useless and a special case.
What about usb part ? That one uses the number somehow I think ?
Not this number.
OTOH the number used in the usb tree command is taken from struct usb_device, and is used for lookups.
Maybe it's time to clean that numbering mess up a bit , and make it consistent ?
Maybe implement it like a vfs? It would force some consistency into the drivers and commands.
On 3/15/19 8:50 PM, Ismael Luceno Cortes wrote:
On 15/Mar/2019 18:34, Marek Vasut wrote:
On 3/14/19 5:19 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 16:09, Marek Vasut wrote:
On 3/14/19 1:57 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 12:55, Marek Vasut wrote:
On 3/14/19 12:44 PM, Ismael Luceno Cortes wrote: > On 18/Feb/2019 09:23, Ismael Luceno Cortes wrote: >> Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com >> --- >> drivers/usb/host/usb-uclass.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c >> index 611ea97a72..0575f5393b 100644 >> --- a/drivers/usb/host/usb-uclass.c >> +++ b/drivers/usb/host/usb-uclass.c >> @@ -255,7 +255,7 @@ int usb_init(void) >> >> uclass_foreach_dev(bus, uc) { >> /* init low_level USB */ >> - printf("USB%d: ", count); >> + printf("USB%d(%s): ", count, bus->name); >> count++; >> >> #ifdef CONFIG_SANDBOX >> -- >> 2.19.1 > > Ping.
What is this patch doing ? The commit description doesn't explain anything about it.
It prints the host device name. I'm not sure the count is at all useful given there's a name...
If you could share the log before and after to better illustrate the difference, that'd be nice.
unpatched:
=> usb reset resetting USB... USB0: USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
patched:
=> usb reset resetting USB... USB0(usb@ee080100): USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
However, shouldn't the same approach be applied to 'usb tree' subcommand and possibly others ?
The number shown during usb scanning is not used nor saved anywhere else, so seems pretty useless and a special case.
What about usb part ? That one uses the number somehow I think ?
Not this number.
Lovely.
Anyway, this looks good, can you repost this patch with proper commit message, ideally with the example output above so I can pick it for next ?
OTOH the number used in the usb tree command is taken from struct usb_device, and is used for lookups.
Maybe it's time to clean that numbering mess up a bit , and make it consistent ?
Maybe implement it like a vfs? It would force some consistency into the drivers and commands.
Do you want to take that one up ? :)
On 16/Mar/2019 02:41, Marek Vasut wrote:
On 3/15/19 8:50 PM, Ismael Luceno Cortes wrote:
On 15/Mar/2019 18:34, Marek Vasut wrote:
On 3/14/19 5:19 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 16:09, Marek Vasut wrote:
On 3/14/19 1:57 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 12:55, Marek Vasut wrote: > On 3/14/19 12:44 PM, Ismael Luceno Cortes wrote: >> On 18/Feb/2019 09:23, Ismael Luceno Cortes wrote: >>> Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com >>> --- >>> drivers/usb/host/usb-uclass.c | 2 +- >>> 1 file changed, 1 insertion(+), 1 deletion(-) >>> >>> diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c >>> index 611ea97a72..0575f5393b 100644 >>> --- a/drivers/usb/host/usb-uclass.c >>> +++ b/drivers/usb/host/usb-uclass.c >>> @@ -255,7 +255,7 @@ int usb_init(void) >>> >>> uclass_foreach_dev(bus, uc) { >>> /* init low_level USB */ >>> - printf("USB%d: ", count); >>> + printf("USB%d(%s): ", count, bus->name); >>> count++; >>> >>> #ifdef CONFIG_SANDBOX >>> -- >>> 2.19.1 >> >> Ping. > > What is this patch doing ? The commit description doesn't explain > anything about it.
It prints the host device name. I'm not sure the count is at all useful given there's a name...
If you could share the log before and after to better illustrate the difference, that'd be nice.
unpatched:
=> usb reset resetting USB... USB0: USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
patched:
=> usb reset resetting USB... USB0(usb@ee080100): USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
However, shouldn't the same approach be applied to 'usb tree' subcommand and possibly others ?
The number shown during usb scanning is not used nor saved anywhere else, so seems pretty useless and a special case.
What about usb part ? That one uses the number somehow I think ?
Not this number.
Lovely.
Anyway, this looks good, can you repost this patch with proper commit message, ideally with the example output above so I can pick it for next ?
Ok.
OTOH the number used in the usb tree command is taken from struct usb_device, and is used for lookups.
Maybe it's time to clean that numbering mess up a bit , and make it consistent ?
Maybe implement it like a vfs? It would force some consistency into the drivers and commands.
Do you want to take that one up ? :)
I would consider implementing it. Is there any preferences?
On 3/18/19 1:02 PM, Ismael Luceno Cortes wrote:
On 16/Mar/2019 02:41, Marek Vasut wrote:
On 3/15/19 8:50 PM, Ismael Luceno Cortes wrote:
On 15/Mar/2019 18:34, Marek Vasut wrote:
On 3/14/19 5:19 PM, Ismael Luceno Cortes wrote:
On 14/Mar/2019 16:09, Marek Vasut wrote:
On 3/14/19 1:57 PM, Ismael Luceno Cortes wrote: > On 14/Mar/2019 12:55, Marek Vasut wrote: >> On 3/14/19 12:44 PM, Ismael Luceno Cortes wrote: >>> On 18/Feb/2019 09:23, Ismael Luceno Cortes wrote: >>>> Signed-off-by: Ismael Luceno ismael.luceno@silicon-gears.com >>>> --- >>>> drivers/usb/host/usb-uclass.c | 2 +- >>>> 1 file changed, 1 insertion(+), 1 deletion(-) >>>> >>>> diff --git a/drivers/usb/host/usb-uclass.c b/drivers/usb/host/usb-uclass.c >>>> index 611ea97a72..0575f5393b 100644 >>>> --- a/drivers/usb/host/usb-uclass.c >>>> +++ b/drivers/usb/host/usb-uclass.c >>>> @@ -255,7 +255,7 @@ int usb_init(void) >>>> >>>> uclass_foreach_dev(bus, uc) { >>>> /* init low_level USB */ >>>> - printf("USB%d: ", count); >>>> + printf("USB%d(%s): ", count, bus->name); >>>> count++; >>>> >>>> #ifdef CONFIG_SANDBOX >>>> -- >>>> 2.19.1 >>> >>> Ping. >> >> What is this patch doing ? The commit description doesn't explain >> anything about it. > > It prints the host device name. I'm not sure the count is at all useful > given there's a name...
If you could share the log before and after to better illustrate the difference, that'd be nice.
unpatched:
=> usb reset resetting USB... USB0: USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
patched:
=> usb reset resetting USB... USB0(usb@ee080100): USB EHCI 1.10 scanning bus 0 for devices... 2 USB Device(s) found scanning usb for storage devices... 1 Storage Device(s) found
However, shouldn't the same approach be applied to 'usb tree' subcommand and possibly others ?
The number shown during usb scanning is not used nor saved anywhere else, so seems pretty useless and a special case.
What about usb part ? That one uses the number somehow I think ?
Not this number.
Lovely.
Anyway, this looks good, can you repost this patch with proper commit message, ideally with the example output above so I can pick it for next ?
Ok.
OTOH the number used in the usb tree command is taken from struct usb_device, and is used for lookups.
Maybe it's time to clean that numbering mess up a bit , and make it consistent ?
Maybe implement it like a vfs? It would force some consistency into the drivers and commands.
Do you want to take that one up ? :)
I would consider implementing it. Is there any preferences?
Prepare an RFC patchset and we'll see how that looks :)
participants (2)
-
Ismael Luceno Cortes -
Marek Vasut