On Tue, Aug 6, 2024 at 10:53 PM Manivannan Sadhasivam
<manivannan.sadhasivam@xxxxxxxxxx> wrote:
>
> On Wed, Jul 31, 2024 at 06:28:17PM -0400, Jim Quinlan wrote:
> > o Move the clk_prepare_enable() below the resource allocations.
> > o Move the clk_prepare_enable() out of __brcm_pcie_remove() but
> > add it to the end of brcm_pcie_remove().
> > o Add a jump target (clk_disable_unprepare) so that a bit of exception
> > handling can be better reused at the end of this function implementation.
> > o Use dev_err_probe() where it makes sense.
> >
> > Signed-off-by: Jim Quinlan <james.quinlan@xxxxxxxxxxxx>
> > ---
> > drivers/pci/controller/pcie-brcmstb.c | 34 ++++++++++++---------------
> > 1 file changed, 15 insertions(+), 19 deletions(-)
> >
> > diff --git a/drivers/pci/controller/pcie-brcmstb.c b/drivers/pci/controller/pcie-brcmstb.c
> > index c08683febdd4..7595e7009192 100644
> > --- a/drivers/pci/controller/pcie-brcmstb.c
> > +++ b/drivers/pci/controller/pcie-brcmstb.c
> > @@ -1473,7 +1473,6 @@ static void __brcm_pcie_remove(struct brcm_pcie *pcie)
> > dev_err(pcie->dev, "Could not stop phy\n");
> > if (reset_control_rearm(pcie->rescal))
> > dev_err(pcie->dev, "Could not rearm rescal reset\n");
> > - clk_disable_unprepare(pcie->clk);
> > }
> >
> > static void brcm_pcie_remove(struct platform_device *pdev)
> > @@ -1484,6 +1483,7 @@ static void brcm_pcie_remove(struct platform_device *pdev)
> > pci_stop_root_bus(bridge->bus);
> > pci_remove_root_bus(bridge->bus);
> > __brcm_pcie_remove(pcie);
> > + clk_disable_unprepare(pcie->clk);
> > }
> >
> > static const int pcie_offsets[] = {
> > @@ -1613,31 +1613,26 @@ static int brcm_pcie_probe(struct platform_device *pdev)
> >
> > pcie->ssc = of_property_read_bool(np, "brcm,enable-ssc");
> >
> > - ret = clk_prepare_enable(pcie->clk);
> > - if (ret) {
> > - dev_err(&pdev->dev, "could not enable clock\n");
> > - return ret;
> > - }
> > pcie->rescal = devm_reset_control_get_optional_shared(&pdev->dev, "rescal");
> > - if (IS_ERR(pcie->rescal)) {
> > - clk_disable_unprepare(pcie->clk);
> > + if (IS_ERR(pcie->rescal))
> > return PTR_ERR(pcie->rescal);
> > - }
> > +
> > pcie->perst_reset = devm_reset_control_get_optional_exclusive(&pdev->dev, "perst");
> > - if (IS_ERR(pcie->perst_reset)) {
> > - clk_disable_unprepare(pcie->clk);
> > + if (IS_ERR(pcie->perst_reset))
> > return PTR_ERR(pcie->perst_reset);
> > - }
> >
> > - ret = reset_control_reset(pcie->rescal);
> > + ret = clk_prepare_enable(pcie->clk);
> > if (ret)
> > - dev_err(&pdev->dev, "failed to deassert 'rescal'\n");
> > + return dev_err_probe(&pdev->dev, ret, "could not enable clock\n");
> > +
> > + ret = reset_control_reset(pcie->rescal);
> > + if (dev_err_probe(&pdev->dev, ret, "failed to deassert 'rescal'\n"))
> > + goto clk_disable_unprepare;
> >
> > ret = brcm_phy_start(pcie);
> > if (ret) {
> > reset_control_rearm(pcie->rescal);
> > - clk_disable_unprepare(pcie->clk);
> > - return ret;
> > + goto clk_disable_unprepare;
> > }
> >
> > ret = brcm_pcie_setup(pcie);
> > @@ -1654,10 +1649,8 @@ static int brcm_pcie_probe(struct platform_device *pdev)
> > msi_np = of_parse_phandle(pcie->np, "msi-parent", 0);
> > if (pci_msi_enabled() && msi_np == pcie->np) {
> > ret = brcm_pcie_enable_msi(pcie);
> > - if (ret) {
> > - dev_err(pcie->dev, "probe of internal MSI failed");
> > + if (dev_err_probe(pcie->dev, ret, "probe of internal MSI failed"))
> > goto fail;
> > - }
> > }
> >
> > bridge->ops = pcie->type == BCM7425 ? &brcm7425_pcie_ops : &brcm_pcie_ops;
> > @@ -1678,6 +1671,9 @@ static int brcm_pcie_probe(struct platform_device *pdev)
> >
> > fail:
> > __brcm_pcie_remove(pcie);
> > +clk_disable_unprepare:
> > + clk_disable_unprepare(pcie->clk);
> > +
>
> TBH, this is not improving the code readability. __brcm_pcie_remove() used to
> free all the resources and now you just moved clk_disable_unprepare() out of it
> to save 2 lines in probe(). And you ended up calling clk_disable_unprepare()
> separately in brcm_pcie_remove().
Actually it saves more lines than that; the "swinit reset" commit adds
two more "goto clk_disable_unprepare" instances.
Nonetheless I will make the change
Regards,
Jim Quinlan
Broadcom STB/CM
>
> So please remove the label and call clk_disable_unprepare() in the error path
> (just 2 instances) and continue to use __brcm_pcie_remove() to free all
> resources (I would've preferred to have separate error labels instead of calling
> __brcm_pcie_remove() though, but not this).
>
> - Mani
>
> --
> மணிவண்ணன் சதாசிவம்
Attachment:
smime.p7s
Description: S/MIME Cryptographic Signature
