[PATCH v1 3/8] PCI: brcmstb: Use bridge reset if available

[Stanimir Varbanov]

Hi Jim,

On 7/2/24 21:36, Jim Quinlan wrote:
> On Tue, Jul 2, 2024 at 8:59 AM Stanimir Varbanov <svarbanov at suse.de> wrote:
>>
>>
>>
>> On 6/28/24 23:54, Jim Quinlan wrote:
>>> The 7712 SOC has a bridge reset which can be described in the device tree.
>>> If it is present, use it. Otherwise, continue to use the legacy method to
>>> reset the bridge.
>>>
>>> Signed-off-by: Jim Quinlan <james.quinlan at broadcom.com>
>>> ---
>>>  drivers/pci/controller/pcie-brcmstb.c | 24 +++++++++++++++++++-----
>>>  1 file changed, 19 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/drivers/pci/controller/pcie-brcmstb.c b/drivers/pci/controller/pcie-brcmstb.c
>>> index c2eb29b886f7..4104c3668fdb 100644
>>> --- a/drivers/pci/controller/pcie-brcmstb.c
>>> +++ b/drivers/pci/controller/pcie-brcmstb.c
>>> @@ -265,6 +265,7 @@ struct brcm_pcie {
>>>       enum pcie_type          type;
>>>       struct reset_control    *rescal;
>>>       struct reset_control    *perst_reset;
>>> +     struct reset_control    *bridge;
>>>       int                     num_memc;
>>>       u64                     memc_size[PCIE_BRCM_MAX_MEMC];
>>>       u32                     hw_rev;
>>> @@ -732,12 +733,19 @@ static void __iomem *brcm7425_pcie_map_bus(struct pci_bus *bus,
>>>
>>>  static void brcm_pcie_bridge_sw_init_set_generic(struct brcm_pcie *pcie, u32 val)
>>>  {
>>> -     u32 tmp, mask =  RGR1_SW_INIT_1_INIT_GENERIC_MASK;
>>> -     u32 shift = RGR1_SW_INIT_1_INIT_GENERIC_SHIFT;
>>> +     if (pcie->bridge) {
>>> +             if (val)
>>> +                     reset_control_assert(pcie->bridge);
>>> +             else
>>> +                     reset_control_deassert(pcie->bridge);
>>
>> Please check reset_control_assert/deassert() calls for error. This might
>> need to change the definition of brcm_pcie_bridge_sw_init_set_generic()
>> to return error.
> 
> Hi Stan,
> 
> Do you really think this is necessary?  If you look at
> "drivers/reset/reset-brcmstb.c"  there is no way for either of these
> calls to fail and I don't see that changing because it is just writing
> a bit into a register.

yes, I think there are kernel rules which we have to follow. We use
generic reset-control interface in pcie driver and we cannot rely on the
low-level implementation of this particular reset-controller driver
(reset-brcmstb.c).

regards,
~Stan