linux-kernel - Re: [PATCH v11 10/20] cxl/pci: Introduce config option PCIEAER

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <da5bc564-76c4-4078-97ab-1d543933071e@amd.com>
Date:   Mon, 16 Oct 2023 08:40:23 -0500
From:   Terry Bowman <Terry.Bowman@....com>
To:     Jonathan Cameron <Jonathan.Cameron@...wei.com>,
        Robert Richter <rrichter@....com>
Cc:     Alison Schofield <alison.schofield@...el.com>,
        Vishal Verma <vishal.l.verma@...el.com>,
        Ira Weiny <ira.weiny@...el.com>,
        Ben Widawsky <bwidawsk@...nel.org>,
        Dan Williams <dan.j.williams@...el.com>,
        Davidlohr Bueso <dave@...olabs.net>,
        Dave Jiang <dave.jiang@...el.com>, linux-cxl@...r.kernel.org,
        linux-kernel@...r.kernel.org, Bjorn Helgaas <bhelgaas@...gle.com>
Subject: Re: [PATCH v11 10/20] cxl/pci: Introduce config option PCIEAER_CXL

Hi Jonathan,

I added a response below.

On 10/9/23 09:44, Terry Bowman wrote:
> Hi Jonathan,
> 
> I added responses inline below.
> 
> On 10/2/23 09:46, Jonathan Cameron wrote:
>> On Wed, 27 Sep 2023 17:43:29 +0200
>> Robert Richter <rrichter@....com> wrote:
>>
>>> CXL error handling depends on AER.
>>>
>>> Introduce config option PCIEAER_CXL in preparation of the AER dport
>>> error handling. Also, introduce the stub function
>>> devm_cxl_setup_parent_dport() to setup dports.
>>>
>>> This is in preparation of follow on patches.
>>>
>>> Note the Kconfg part of the option is added in a later patch to enable
>>> it once coding of the feature is complete.
>>>
>>> Signed-off-by: Robert Richter <rrichter@....com>
>>
>> Feels like it should just be combined with a later patch that fills
>> some of this in as on it's own it's just a weird snippet of code :)
>>
> 
> We will look to merge with the following patch.
> 
>> Still, one comment inline anyway.
>>
>>
>>> ---
>>>  drivers/cxl/core/pci.c | 9 +++++++++
>>>  drivers/cxl/cxl.h      | 7 +++++++
>>>  drivers/cxl/mem.c      | 2 ++
>>>  3 files changed, 18 insertions(+)
>>>
>>> diff --git a/drivers/cxl/core/pci.c b/drivers/cxl/core/pci.c
>>> index c7a7887ebdcf..6ba3b7370816 100644
>>> --- a/drivers/cxl/core/pci.c
>>> +++ b/drivers/cxl/core/pci.c
>>> @@ -718,6 +718,15 @@ static bool cxl_report_and_clear(struct cxl_dev_state *cxlds)
>>>  	return true;
>>>  }
>>>  
>>> +#ifdef CONFIG_PCIEAER_CXL
>>> +
>>> +void devm_cxl_setup_parent_dport(struct device *host, struct cxl_dport *dport)
>>> +{
>>> +}
>>> +EXPORT_SYMBOL_NS_GPL(devm_cxl_setup_parent_dport, CXL);
>>> +
>>> +#endif
>>> +
>>>  pci_ers_result_t cxl_error_detected(struct pci_dev *pdev,
>>>  				    pci_channel_state_t state)
>>>  {
>>> diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
>>> index c07064e0c136..cfa2f6bede41 100644
>>> --- a/drivers/cxl/cxl.h
>>> +++ b/drivers/cxl/cxl.h
>>> @@ -704,6 +704,13 @@ struct cxl_dport *devm_cxl_add_rch_dport(struct cxl_port *port,
>>>  					 struct device *dport_dev, int port_id,
>>>  					 resource_size_t rcrb);
>>>  
>>> +#ifdef CONFIG_PCIEAER_CXL
>>> +void devm_cxl_setup_parent_dport(struct device *host, struct cxl_dport *dport);
>>> +#else
>>> +static inline void devm_cxl_setup_parent_dport(struct device *host,
>>> +					       struct cxl_dport *dport) { }
>>> +#endif
>>> +
>>>  struct cxl_decoder *to_cxl_decoder(struct device *dev);
>>>  struct cxl_root_decoder *to_cxl_root_decoder(struct device *dev);
>>>  struct cxl_switch_decoder *to_cxl_switch_decoder(struct device *dev);
>>> diff --git a/drivers/cxl/mem.c b/drivers/cxl/mem.c
>>> index 04107058739b..61ca21c020fa 100644
>>> --- a/drivers/cxl/mem.c
>>> +++ b/drivers/cxl/mem.c
>>> @@ -157,6 +157,8 @@ static int cxl_mem_probe(struct device *dev)
>>>  	else
>>>  		endpoint_parent = &parent_port->dev;
>>>  
>>> +	devm_cxl_setup_parent_dport(dev, dport);
>>
>> devm calls can always fail (because if nothing else you have to register
>> some cleanup and that involves an allocation.  If you want to ignore
>> that I'd expect a comment here.
>>
> 
> We will add error handling here.
> 
> Regards,
> Terry
>

Found devm_cxl_setup_parent_dport() is a NULL function without return value.
 
Regards,
Terry