Mailing List Archive

[PATCH REBASE 0/3] atomisp: Rebased fixes
Hi Mauro,

I've rebased the patches now, but there is a slight hiccup. For patches 2
and 3 of this series there will now be a conflict with commit 9289cdf39992
("staging: media: atomisp: Convert to GPIO descriptors") in Greg's tree.

I'm not sure what the best way to handle this is? The merge conflicts
will be trivial (due to a conversion between the gpio_* and gpiod_*
APIs), but I could alternatively send these last two patches in via
Greg's tree if that's easier for people. Let me know what works.

Best,
Alex
Re: [PATCH REBASE 0/3] atomisp: Rebased fixes [ In reply to ]
Em Tue, 22 Sep 2020 10:09:07 +0100
Alex Dewar <alex.dewar90@gmail.com> escreveu:

> Hi Mauro,
>
> I've rebased the patches now, but there is a slight hiccup. For patches 2
> and 3 of this series there will now be a conflict with commit 9289cdf39992
> ("staging: media: atomisp: Convert to GPIO descriptors") in Greg's tree.
>
> I'm not sure what the best way to handle this is? The merge conflicts
> will be trivial (due to a conversion between the gpio_* and gpiod_*
> APIs), but I could alternatively send these last two patches in via
> Greg's tree if that's easier for people. Let me know what works.

Maybe the best would be to re-send those after the merge window, when
both patches will arrive upstream.

Thanks,
Mauro
Re: [PATCH REBASE 0/3] atomisp: Rebased fixes [ In reply to ]
On 22/09/2020 10:27, Mauro Carvalho Chehab wrote:
> Em Tue, 22 Sep 2020 10:09:07 +0100
> Alex Dewar <alex.dewar90@gmail.com> escreveu:
>
>> Hi Mauro,
>>
>> I've rebased the patches now, but there is a slight hiccup. For patches 2
>> and 3 of this series there will now be a conflict with commit 9289cdf39992
>> ("staging: media: atomisp: Convert to GPIO descriptors") in Greg's tree.
>>
>> I'm not sure what the best way to handle this is? The merge conflicts
>> will be trivial (due to a conversion between the gpio_* and gpiod_*
>> APIs), but I could alternatively send these last two patches in via
>> Greg's tree if that's easier for people. Let me know what works.
> Maybe the best would be to re-send those after the merge window, when
> both patches will arrive upstream.
>
> Thanks,
> Mauro
That sounds more sensible. I've also just noticed that I introduced a
bug in the first patch when rebasing it :-/, so let's hold off on the
whole series and I'll do a proper tidy and resend after the next merge
window.

Best,
Alex
Re: [PATCH REBASE 0/3] atomisp: Rebased fixes [ In reply to ]
On Tue, Sep 22, 2020 at 12:02:33PM +0100, Alex Dewar wrote:
> On 22/09/2020 10:27, Mauro Carvalho Chehab wrote:
> > Em Tue, 22 Sep 2020 10:09:07 +0100
> > Alex Dewar <alex.dewar90@gmail.com> escreveu:
> >
> > > Hi Mauro,
> > >
> > > I've rebased the patches now, but there is a slight hiccup. For patches 2
> > > and 3 of this series there will now be a conflict with commit 9289cdf39992
> > > ("staging: media: atomisp: Convert to GPIO descriptors") in Greg's tree.
> > >
> > > I'm not sure what the best way to handle this is? The merge conflicts
> > > will be trivial (due to a conversion between the gpio_* and gpiod_*
> > > APIs), but I could alternatively send these last two patches in via
> > > Greg's tree if that's easier for people. Let me know what works.
> > Maybe the best would be to re-send those after the merge window, when
> > both patches will arrive upstream.
> >
> > Thanks,
> > Mauro
> That sounds more sensible. I've also just noticed that I introduced a bug in
> the first patch when rebasing it :-/, so let's hold off on the whole series
> and I'll do a proper tidy and resend after the next merge window.

Is the bug the memory leak if lm3554_platform_data_func() fails?

regards,
dan carpenter
Re: [PATCH REBASE 0/3] atomisp: Rebased fixes [ In reply to ]
[snip]
> > That sounds more sensible. I've also just noticed that I introduced a bug in
> > the first patch when rebasing it :-/, so let's hold off on the whole series
> > and I'll do a proper tidy and resend after the next merge window.
>
> Is the bug the memory leak if lm3554_platform_data_func() fails?
>
> regards,
> dan carpenter
>

>
Nope. I put a "return ret" for the last check instead of "goto err_del_timer"...

The version of this code in linux-next does the correct "if (PTR_ERR(...))"
check after calling lm3554_platform_data_func(), but this patch doesn't
seem to have made its way into linux-media yet. All the more reason to
resend my patches after the merge window, I suppose.

Best,
Alex