[libcamera-devel] [PATCH 5/5] ipa: ipu3: af: Remove v4l2 interaction from AF algorithm

Kate Hsuan hpa at redhat.com
Mon Nov 29 03:04:34 CET 2021


Hi Daniel,

On Mon, Nov 29, 2021 at 7:14 AM Daniel Scally <djrscally at gmail.com> wrote:
>
> Hi Laurent
>
> On 28/11/2021 22:24, Laurent Pinchart wrote:
> > Hi Daniel,
> >
> > Thank you for the patch.
> >
> > On Fri, Nov 26, 2021 at 12:31:18AM +0000, Daniel Scally wrote:
> >> Now that the actual interaction with the VCM's V4L2 subdev can be
> >> done through the CameraLens class, remove it from the auto focus
> >> algorithm code.
> >>
> >> Signed-off-by: Daniel Scally <djrscally at gmail.com>
> >> ---
> >>  src/ipa/ipu3/algorithms/af.cpp | 29 +----------------------------
> >>  src/ipa/ipu3/algorithms/af.h   |  3 ---
> >>  2 files changed, 1 insertion(+), 31 deletions(-)
> >>
> >> diff --git a/src/ipa/ipu3/algorithms/af.cpp b/src/ipa/ipu3/algorithms/af.cpp
> >> index 2859a6a8..1bbf64aa 100644
> >> --- a/src/ipa/ipu3/algorithms/af.cpp
> >> +++ b/src/ipa/ipu3/algorithms/af.cpp
> >> @@ -66,17 +66,10 @@ static constexpr double MaxChange_ = 0.8;
> >>  Af::Af()
> >>      : focus_(0), currentVariance_(0.0)
> >>  {
> >> -    /**
> >> -     * For surface Go 2 back camera VCM (dw9719)
> >> -     * \todo move to control class
> >> -    */
> >> -    vcmFd_ = open("/dev/v4l-subdev8", O_RDWR);
> >>  }
> >>
> >>  Af::~Af()
> >>  {
> >> -    if (vcmFd_ != -1)
> >> -            close(vcmFd_);
> >>  }
> >>
> >>  void Af::prepare(IPAContext &context, ipu3_uapi_params *params)
> >> @@ -169,25 +162,6 @@ int Af::configure(IPAContext &context, [[maybe_unused]] const IPAConfigInfo &con
> >>      return 0;
> >>  }
> >>
> >> -/**
> >> - * \brief Send focus step to the VCM.
> >> - * \param[in] value Set lens position.
> >> - * \todo It is hardcoded here for the dw9717 VCM and will be moved to the
> >> - * subdev control in the future.
> >> - */
> >> -int Af::vcmSet(int value)
> >> -{
> >> -    int ret;
> >> -    struct v4l2_control ctrl;
> >> -    if (vcmFd_ == -1)
> >> -            return -EINVAL;
> >> -    memset(&ctrl, 0, sizeof(struct v4l2_control));
> >> -    ctrl.id = V4L2_CID_FOCUS_ABSOLUTE;
> >> -    ctrl.value = value;
> >> -    ret = ioctl(vcmFd_, VIDIOC_S_CTRL, &ctrl);
> > We can probably drop some #include statements, for the headers used by
> > open() and ioctl().
> >
> > The patch looks fine overall, I expect Kate will squash it into her
> > work.
>
>
> That was my thought too, as long as Kate's ok with that
>
> >
> >> -    return ret;
> >> -}
> >> -
> >>  /**
> >>   * \brief Determine the max contrast image and lens position. y_table is the
> >>   * statictic data from IPU3 and is composed of low pass and high pass filtered
> >> @@ -263,10 +237,9 @@ void Af::process(IPAContext &context, const ipu3_uapi_stats_3a *stats)
> >>                      if (focus_ > MaxFocusSteps_) {
> >>                              /* If reach the max step, move lens to the position and set "focus stable". */
> >>                              context.frameContext.af.stable = true;
> >> -                            vcmSet(context.frameContext.af.focus);
> >>                      } else {
> >>                              focus_ += MinSearchStep_;
> >> -                            vcmSet(focus_);
> >> +                            context.frameContext.af.focus = focus_;
> >>                      }
> >>                      LOG(IPU3Af, Debug) << "Focus searching max variance is: "
> >>                                         << context.frameContext.af.maxVariance
> >> diff --git a/src/ipa/ipu3/algorithms/af.h b/src/ipa/ipu3/algorithms/af.h
> >> index d32d5184..b21084e9 100644
> >> --- a/src/ipa/ipu3/algorithms/af.h
> >> +++ b/src/ipa/ipu3/algorithms/af.h
> >> @@ -36,9 +36,6 @@ public:
> >>      void process(IPAContext &context, const ipu3_uapi_stats_3a *stats) override;
> >>
> >>  private:
> >> -    int vcmSet(int value);
> >> -
> >> -    int vcmFd_;
> >>      /* Used for focus scan. */
> >>      uint32_t focus_;
> >>      /* Recent AF statistic variance. */
>

It's good to me :)

Reviewed-by: Kate Hsuan <hpa at redhat.com>


-- 
BR,
Kate



More information about the libcamera-devel mailing list