From: Paul Davey <[email protected]>
On big endian architectures the mhi debugfs files which report pm state
give "Invalid State" for all states. This is caused by using
find_last_bit which takes an unsigned long* while the state is passed in
as an enum mhi_pm_state which will be of int size.
Fix by using __fls to pass the value of state instead of find_last_bit.
Fixes: a6e2e3522f29 ("bus: mhi: core: Add support for PM state transitions")
Signed-off-by: Paul Davey <[email protected]>
Reviewed-by: Manivannan Sadhasivam <[email protected]>
Reviewed-by: Hemant Kumar <[email protected]>
Cc: [email protected]
Signed-off-by: Manivannan Sadhasivam <[email protected]>
---
drivers/bus/mhi/core/init.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/bus/mhi/core/init.c b/drivers/bus/mhi/core/init.c
index 046f407dc5d6..af484b03558a 100644
--- a/drivers/bus/mhi/core/init.c
+++ b/drivers/bus/mhi/core/init.c
@@ -79,10 +79,12 @@ static const char * const mhi_pm_state_str[] = {
const char *to_mhi_pm_state_str(enum mhi_pm_state state)
{
- unsigned long pm_state = state;
- int index = find_last_bit(&pm_state, 32);
+ int index;
- if (index >= ARRAY_SIZE(mhi_pm_state_str))
+ if (state)
+ index = __fls(state);
+
+ if (!state || index >= ARRAY_SIZE(mhi_pm_state_str))
return "Invalid State";
return mhi_pm_state_str[index];
--
2.25.1
On 2/12/22 12:20 PM, Manivannan Sadhasivam wrote:
> From: Paul Davey <[email protected]>
>
> On big endian architectures the mhi debugfs files which report pm state
> give "Invalid State" for all states. This is caused by using
> find_last_bit which takes an unsigned long* while the state is passed in
> as an enum mhi_pm_state which will be of int size.
I think this would have fixed it too, but your fix is better.
int index = find_last_bit(&(unsigned long)state, 32);
> Fix by using __fls to pass the value of state instead of find_last_bit.
>
> Fixes: a6e2e3522f29 ("bus: mhi: core: Add support for PM state transitions")
> Signed-off-by: Paul Davey <[email protected]>
> Reviewed-by: Manivannan Sadhasivam <[email protected]>
> Reviewed-by: Hemant Kumar <[email protected]>
> Cc: [email protected]
> Signed-off-by: Manivannan Sadhasivam <[email protected]>
> ---
> drivers/bus/mhi/core/init.c | 8 +++++---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/bus/mhi/core/init.c b/drivers/bus/mhi/core/init.c
> index 046f407dc5d6..af484b03558a 100644
> --- a/drivers/bus/mhi/core/init.c
> +++ b/drivers/bus/mhi/core/init.c
> @@ -79,10 +79,12 @@ static const char * const mhi_pm_state_str[] = {
>
> const char *to_mhi_pm_state_str(enum mhi_pm_state state)
The mhi_pm_state enumerated type is an enumerated sequence, not
a bit mask. So knowing what the last (most significant) set bit
is not meaningful. Or normally it shouldn't be.
If mhi_pm_state really were a bit mask, then its values should
be defined that way, i.e.,
MHI_PM_STATE_DISABLE = 1 << 0,
MHI_PM_STATE_DISABLE = 1 << 1,
. . .
What's really going on is that the state value passed here
*is* a bitmask, whose bit positions are those mhi_pm_state
values. So the state argument should have type u32.
This is a *separate* bug/issue. It could be fixed separately
(before this patch), but I'd be OK with just explaining why
this change would occur as part of this modified patch.
> {
> - unsigned long pm_state = state;
> - int index = find_last_bit(&pm_state, 32);
> + int index;
>
> - if (index >= ARRAY_SIZE(mhi_pm_state_str))
> + if (state)
> + index = __fls(state);
> +
> + if (!state || index >= ARRAY_SIZE(mhi_pm_state_str))
> return "Invalid State";
Do this test and return first, and skip the additional
check for "if (state)".
-Alex
> return mhi_pm_state_str[index];
On Tue, Feb 15, 2022 at 02:01:54PM -0600, Alex Elder wrote:
> On 2/12/22 12:20 PM, Manivannan Sadhasivam wrote:
> > From: Paul Davey <[email protected]>
> >
> > On big endian architectures the mhi debugfs files which report pm state
> > give "Invalid State" for all states. This is caused by using
> > find_last_bit which takes an unsigned long* while the state is passed in
> > as an enum mhi_pm_state which will be of int size.
>
> I think this would have fixed it too, but your fix is better.
>
> int index = find_last_bit(&(unsigned long)state, 32);
>
> > Fix by using __fls to pass the value of state instead of find_last_bit.
> >
> > Fixes: a6e2e3522f29 ("bus: mhi: core: Add support for PM state transitions")
> > Signed-off-by: Paul Davey <[email protected]>
> > Reviewed-by: Manivannan Sadhasivam <[email protected]>
> > Reviewed-by: Hemant Kumar <[email protected]>
> > Cc: [email protected]
> > Signed-off-by: Manivannan Sadhasivam <[email protected]>
> > ---
> > drivers/bus/mhi/core/init.c | 8 +++++---
> > 1 file changed, 5 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/bus/mhi/core/init.c b/drivers/bus/mhi/core/init.c
> > index 046f407dc5d6..af484b03558a 100644
> > --- a/drivers/bus/mhi/core/init.c
> > +++ b/drivers/bus/mhi/core/init.c
> > @@ -79,10 +79,12 @@ static const char * const mhi_pm_state_str[] = {
> > const char *to_mhi_pm_state_str(enum mhi_pm_state state)
>
> The mhi_pm_state enumerated type is an enumerated sequence, not
> a bit mask. So knowing what the last (most significant) set bit
> is not meaningful. Or normally it shouldn't be.
>
> If mhi_pm_state really were a bit mask, then its values should
> be defined that way, i.e.,
>
> MHI_PM_STATE_DISABLE = 1 << 0,
> MHI_PM_STATE_DISABLE = 1 << 1,
> . . .
>
> What's really going on is that the state value passed here
> *is* a bitmask, whose bit positions are those mhi_pm_state
> values. So the state argument should have type u32.
>
I agree with you. It should be u32.
> This is a *separate* bug/issue. It could be fixed separately
> (before this patch), but I'd be OK with just explaining why
> this change would occur as part of this modified patch.
>
It makes sense to do it in the same patch itself as the change is
minimal and moreover this patch will also get backported to stable.
> > {
> > - unsigned long pm_state = state;
> > - int index = find_last_bit(&pm_state, 32);
> > + int index;
> > - if (index >= ARRAY_SIZE(mhi_pm_state_str))
> > + if (state)
> > + index = __fls(state);
> > +
> > + if (!state || index >= ARRAY_SIZE(mhi_pm_state_str))
> > return "Invalid State";
>
> Do this test and return first, and skip the additional
> check for "if (state)".
>
We need to calculate index for the second check, so I guess the current
code is fine.
Thanks,
Mani
> -Alex
>
> > return mhi_pm_state_str[index];
>
On 2/16/22 5:33 AM, Manivannan Sadhasivam wrote:
> On Tue, Feb 15, 2022 at 02:01:54PM -0600, Alex Elder wrote:
>> On 2/12/22 12:20 PM, Manivannan Sadhasivam wrote:
>>> From: Paul Davey <[email protected]>
>>>
>>> On big endian architectures the mhi debugfs files which report pm state
>>> give "Invalid State" for all states. This is caused by using
>>> find_last_bit which takes an unsigned long* while the state is passed in
>>> as an enum mhi_pm_state which will be of int size.
>>
>> I think this would have fixed it too, but your fix is better.
>>
>> int index = find_last_bit(&(unsigned long)state, 32);
>>
>>> Fix by using __fls to pass the value of state instead of find_last_bit.
>>>
>>> Fixes: a6e2e3522f29 ("bus: mhi: core: Add support for PM state transitions")
>>> Signed-off-by: Paul Davey <[email protected]>
>>> Reviewed-by: Manivannan Sadhasivam <[email protected]>
>>> Reviewed-by: Hemant Kumar <[email protected]>
>>> Cc: [email protected]
>>> Signed-off-by: Manivannan Sadhasivam <[email protected]>
>>> ---
>>> drivers/bus/mhi/core/init.c | 8 +++++---
>>> 1 file changed, 5 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/bus/mhi/core/init.c b/drivers/bus/mhi/core/init.c
>>> index 046f407dc5d6..af484b03558a 100644
>>> --- a/drivers/bus/mhi/core/init.c
>>> +++ b/drivers/bus/mhi/core/init.c
>>> @@ -79,10 +79,12 @@ static const char * const mhi_pm_state_str[] = {
>>> const char *to_mhi_pm_state_str(enum mhi_pm_state state)
>>
>> The mhi_pm_state enumerated type is an enumerated sequence, not
>> a bit mask. So knowing what the last (most significant) set bit
>> is not meaningful. Or normally it shouldn't be.
>>
>> If mhi_pm_state really were a bit mask, then its values should
>> be defined that way, i.e.,
>>
>> MHI_PM_STATE_DISABLE = 1 << 0,
>> MHI_PM_STATE_DISABLE = 1 << 1,
>> . . .
>>
>> What's really going on is that the state value passed here
>> *is* a bitmask, whose bit positions are those mhi_pm_state
>> values. So the state argument should have type u32.
>>
>
> I agree with you. It should be u32.
>
>> This is a *separate* bug/issue. It could be fixed separately
>> (before this patch), but I'd be OK with just explaining why
>> this change would occur as part of this modified patch.
>>
>
> It makes sense to do it in the same patch itself as the change is
> minimal and moreover this patch will also get backported to stable.
Sounds good to me. -Alex
>>> {
>>> - unsigned long pm_state = state;
>>> - int index = find_last_bit(&pm_state, 32);
>>> + int index;
>>> - if (index >= ARRAY_SIZE(mhi_pm_state_str))
>>> + if (state)
>>> + index = __fls(state);
>>> +
>>> + if (!state || index >= ARRAY_SIZE(mhi_pm_state_str))
>>> return "Invalid State";
>>
>> Do this test and return first, and skip the additional
>> check for "if (state)".
>>
>
> We need to calculate index for the second check, so I guess the current
> code is fine.
>
> Thanks,
> Mani
>
>> -Alex
>>
>>> return mhi_pm_state_str[index];
>>