Re: [PATCH v2 2/2] PCI: dwc: rcar-gen4: Use .post_deinit to handle dw_pcie_ep_init() failures

On 8/20/26 10:11 AM, Koichiro Den wrote:

Hello Den-san,

> On Thu, Aug 20, 2026 at 05:57:56AM +0200, Marek Vasut wrote:
>> Implement .post_deinit() callback in R-Car Gen4 struct dw_pcie_ep_ops {}
>> which asserts reset and stops the clock. This undoes start of clock and
>> deassert of reset performed in .pre_init() in case dw_pcie_ep_init() fails
>> after successful call of .pre_init() callback.
>>
>> The use of .post_deinit() callback correctly handles the clock and reset
>> stop, unlike the call of rcar_gen4_pcie_ep_deinit() in dw_pcie_ep_init()
>> which could not discern at which point the dw_pcie_ep_init() failed and
>> might have attempted to stop clock and assert reset twice, remove it.
>>
>> Since dw_pcie_ep_deinit() also invokes the .post_deinit() callback, drop
>> calls to rcar_gen4_pcie_ep_deinit() in both rcar_gen4_add_dw_pcie_ep()
>> dw_pcie_ep_init_registers() fail path and rcar_gen4_remove_dw_pcie_ep()
>> to avoid duplicate stop of clock and assert of reset, and drop no longer
>> used rcar_gen4_pcie_ep_deinit() entirely.
>>
>> Initialize PCIEDMAINTSTSEN early in rcar_gen4_pcie_ep_pre_init() to 0,
>> to make sure that edma_int bits will never be set in case of failure
> 
> I think s/will never be set/will never be left set/, because .pre_init() writes
> 0xffff before later initialization can fail, and .post_deinit() then sets it
> back to 0. Otherwise LGTM.
> 
> (Again, I did some quick tests on an S4 Spider, covering the normal path and
> several injected failure paths. I don't have V4H hardware, though.)
> Please feel free to pick either or both tags, if that helps.
> 
> Reviewed-by: Koichiro Den <[email protected]>
> Tested-by: Koichiro Den <[email protected]>

I have updated the commit message and collected the tags, thank you !

-- 
Best regards,
Marek Vasut