Linux PCI subsystem development
 help / color / mirror / Atom feed
From: Hans Zhang <18255117159@163.com>
To: "Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>
Cc: lpieralisi@kernel.org, kw@linux.com,
	manivannan.sadhasivam@linaro.org, robh@kernel.org,
	bhelgaas@google.com, jingoohan1@gmail.com,
	thomas.richard@bootlin.com, linux-pci@vger.kernel.org,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [v6 3/5] PCI: cadence: Use common PCI host bridge APIs for finding the capabilities
Date: Mon, 24 Mar 2025 22:29:09 +0800	[thread overview]
Message-ID: <d370b69a-3b70-4e3b-94a3-43e0bc1305cd@163.com> (raw)
In-Reply-To: <f467056d-8d4a-9dab-2f0a-ca589adfde53@linux.intel.com>



On 2025/3/24 21:44, Ilpo Järvinen wrote:
> On Mon, 24 Mar 2025, Hans Zhang wrote:
> 
>> Since the PCI core is now exposing generic APIs for the host bridges to
> 
> No need to say "since ... is now exposing". Just say "Use ..." as if the
> API has always existed even if you just added it.
> 

Hi Ilpo,

Thanks your for reply. Will change.


>> search for the PCIe capabilities, make use of them in the CDNS driver.
>>
>> Signed-off-by: Hans Zhang <18255117159@163.com>
>> ---
>> Changes since v5:
>> https://lore.kernel.org/linux-pci/20250321163803.391056-4-18255117159@163.com
>>
>> - Kconfig add "select PCI_HOST_HELPERS"
>> ---
>>   drivers/pci/controller/cadence/Kconfig        |  1 +
>>   drivers/pci/controller/cadence/pcie-cadence.c | 25 +++++++++++++++++++
>>   drivers/pci/controller/cadence/pcie-cadence.h |  3 +++
>>   3 files changed, 29 insertions(+)
>>
>> diff --git a/drivers/pci/controller/cadence/Kconfig b/drivers/pci/controller/cadence/Kconfig
>> index 8a0044bb3989..0a4f245bbeb0 100644
>> --- a/drivers/pci/controller/cadence/Kconfig
>> +++ b/drivers/pci/controller/cadence/Kconfig
>> @@ -5,6 +5,7 @@ menu "Cadence-based PCIe controllers"
>>   
>>   config PCIE_CADENCE
>>   	bool
>> +	select PCI_HOST_HELPERS
>>   
>>   config PCIE_CADENCE_HOST
>>   	bool
>> diff --git a/drivers/pci/controller/cadence/pcie-cadence.c b/drivers/pci/controller/cadence/pcie-cadence.c
>> index 204e045aed8c..329dab4ff813 100644
>> --- a/drivers/pci/controller/cadence/pcie-cadence.c
>> +++ b/drivers/pci/controller/cadence/pcie-cadence.c
>> @@ -8,6 +8,31 @@
>>   
>>   #include "pcie-cadence.h"
>>   
>> +static u32 cdns_pcie_read_cfg(void *priv, int where, int size)
>> +{
>> +	struct cdns_pcie *pcie = priv;
>> +	u32 val;
>> +
>> +	if (size == 4)
>> +		val = readl(pcie->reg_base + where);
> 
> Should this use cdns_pcie_readl() ?

pci_host_bridge_find_*capability required to read two or four bytes.

reg = read_cfg(priv, cap_ptr, 2);
or
header = read_cfg(priv, pos, 4);

Here I mainly want to write it the same way as size == 2 and size == 1.
Or size == 4 should I write it as cdns_pcie_readl() ?

> 
>> +	else if (size == 2)
>> +		val = readw(pcie->reg_base + where);
>> +	else if (size == 1)
>> +		val = readb(pcie->reg_base + where);
>> +
>> +	return val;
>> +}
>> +
>> +u8 cdns_pcie_find_capability(struct cdns_pcie *pcie, u8 cap)
>> +{
>> +	return pci_host_bridge_find_capability(pcie, cdns_pcie_read_cfg, cap);
>> +}
>> +
>> +u16 cdns_pcie_find_ext_capability(struct cdns_pcie *pcie, u8 cap)
>> +{
>> +	return pci_host_bridge_find_ext_capability(pcie, cdns_pcie_read_cfg, cap);
>> +}
> 
> I'm really wondering why the read config function is provided directly as
> an argument. Shouldn't struct pci_host_bridge have some ops that can read
> config so wouldn't it make much more sense to pass it and use the func
> from there? There seems to ops in pci_host_bridge that has read(), does
> that work? If not, why?
> 

No effect. Because we need to get the offset of the capability before 
PCIe enumerates the device. I originally added a separate find 
capability related function for CDNS in the following patch. It's also 
copied directly from DWC. Mani felt there was too much duplicate code 
and also suggested passing a callback function that could manipulate the 
registers of the root port of DWC or CDNS.

https://patchwork.kernel.org/project/linux-pci/patch/20250308133903.322216-1-18255117159@163.com/

The original function is in the following file:
drivers/pci/controller/dwc/pcie-designware.c
u8 dw_pcie_find_capability(struct dw_pcie *pci, u8 cap)
u16 dw_pcie_find_ext_capability(struct dw_pcie *pci, u8 cap)

CDNS has the same need to find the offset of the capability.

>> +
>>   void cdns_pcie_detect_quiet_min_delay_set(struct cdns_pcie *pcie)
>>   {
>>   	u32 delay = 0x3;
>> diff --git a/drivers/pci/controller/cadence/pcie-cadence.h b/drivers/pci/controller/cadence/pcie-cadence.h
>> index f5eeff834ec1..6f4981fccb94 100644
>> --- a/drivers/pci/controller/cadence/pcie-cadence.h
>> +++ b/drivers/pci/controller/cadence/pcie-cadence.h
>> @@ -557,6 +557,9 @@ static inline int cdns_pcie_ep_setup(struct cdns_pcie_ep *ep)
>>   }
>>   #endif
>>   
>> +u8 cdns_pcie_find_capability(struct cdns_pcie *pcie, u8 cap);
>> +u16 cdns_pcie_find_ext_capability(struct cdns_pcie *pcie, u8 cap);
>> +
>>   void cdns_pcie_detect_quiet_min_delay_set(struct cdns_pcie *pcie);
>>   
>>   void cdns_pcie_set_outbound_region(struct cdns_pcie *pcie, u8 busnr, u8 fn,
>>
> 

Best regards,
Hans


  reply	other threads:[~2025-03-24 14:30 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-03-23 16:48 [v6 0/5] Introduce generic capability search functions Hans Zhang
2025-03-23 16:48 ` [v6 1/5] PCI: " Hans Zhang
2025-03-24 13:28   ` Ilpo Järvinen
2025-03-24 14:39     ` Hans Zhang
2025-03-24 14:52       ` Ilpo Järvinen
2025-03-25  2:58         ` Hans Zhang
2025-03-27 16:57   ` Manivannan Sadhasivam
2025-03-28  9:41     ` Hans Zhang
2025-03-27 16:58   ` Manivannan Sadhasivam
2025-03-28  9:42     ` Hans Zhang
2025-03-23 16:48 ` [v6 2/5] PCI: dwc: Use common PCI host bridge APIs for finding the capabilities Hans Zhang
2025-03-23 16:48 ` [v6 3/5] PCI: cadence: " Hans Zhang
2025-03-23 18:33   ` kernel test robot
2025-03-24  1:07     ` Hans Zhang
2025-03-23 19:26   ` kernel test robot
2025-03-24  1:08     ` Hans Zhang
2025-03-24 13:44   ` Ilpo Järvinen
2025-03-24 14:29     ` Hans Zhang [this message]
2025-03-24 15:02       ` Ilpo Järvinen
2025-03-25  2:59         ` Hans Zhang
2025-03-25 11:15           ` Ilpo Järvinen
2025-03-25 12:16             ` Hans Zhang
2025-03-25 14:47               ` Hans Zhang
2025-03-25 15:18                 ` Ilpo Järvinen
2025-03-25 15:37                   ` Hans Zhang
2025-03-28 10:33                     ` Hans Zhang
2025-03-28 11:42                       ` Ilpo Järvinen
2025-03-29 16:03                         ` Hans Zhang
2025-03-31 16:39                           ` Ilpo Järvinen
2025-04-01 13:20                             ` Hans Zhang
2025-03-23 16:48 ` [v6 4/5] PCI: cadence: Use cdns_pcie_find_*capability to avoid hardcode Hans Zhang
2025-03-23 16:48 ` [v6 5/5] MAINTAINERS: Add entry for PCI host controller helpers Hans Zhang
2025-03-27 17:01   ` Manivannan Sadhasivam
2025-03-28 10:36     ` Hans Zhang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=d370b69a-3b70-4e3b-94a3-43e0bc1305cd@163.com \
    --to=18255117159@163.com \
    --cc=bhelgaas@google.com \
    --cc=ilpo.jarvinen@linux.intel.com \
    --cc=jingoohan1@gmail.com \
    --cc=kw@linux.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=lpieralisi@kernel.org \
    --cc=manivannan.sadhasivam@linaro.org \
    --cc=robh@kernel.org \
    --cc=thomas.richard@bootlin.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox