netdev - Re: [PATCH net] net: fealnx: fix missing pci_disable

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]

Date:   Tue, 25 Oct 2022 19:25:48 +0800
From:   Yang Yingliang <yangyingliang@...wei.com>
To:     Leon Romanovsky <leon@...nel.org>
CC:     <netdev@...r.kernel.org>, <davem@...emloft.net>, <kuba@...nel.org>
Subject: Re: [PATCH net] net: fealnx: fix missing pci_disable_device()


On 2022/10/25 18:27, Leon Romanovsky wrote:
> On Tue, Oct 25, 2022 at 05:01:58PM +0800, Yang Yingliang wrote:
>> On 2022/10/25 15:47, Leon Romanovsky wrote:
>>> On Mon, Oct 24, 2022 at 09:57:28PM +0800, Yang Yingliang wrote:
>>>> pci_disable_device() need be called while module exiting, switch
>>>> to use pcim_enable(), pci_disable_device() and pci_release_regions()
>>>> will be called in pcim_release() while unbinding device.
>>> I didn't understand the description at all, maybe people in CC will.
>>> Most likely, you wanted something like this:
>> After using pcim_enable_device(), pcim_release() will be called while
>> unbinding
>> device, pcim_release() calls both pci_release_regions() and
>> pci_disable_device().
> I'm not sure that you can mix managed with unmanaged PCI APIs.
>
> Thanks
After pcim_enable_device() being called, the region_mask is set in
__pci_request_region(). pcim_release() will call pci_release_region()
because region_mask is set.

Without device managed, the pci_release_region() and pci_disable_device()
is called at end of remove() function which is called in 
device_remove(), and
with managed, they are called in device_unbind_cleanup() which is called
after device_remove(). So I think it's OK to use this management.

Thanks,
Yang
>
>> Thanks,
>> Yang
>>> diff --git a/drivers/net/ethernet/fealnx.c b/drivers/net/ethernet/fealnx.c
>>> index ed18450fd2cc..78107dd4aa57 100644
>>> --- a/drivers/net/ethernet/fealnx.c
>>> +++ b/drivers/net/ethernet/fealnx.c
>>> @@ -690,6 +690,7 @@ static void fealnx_remove_one(struct pci_dev *pdev)
>>>                   pci_iounmap(pdev, np->mem);
>>>                   free_netdev(dev);
>>>                   pci_release_regions(pdev);
>>> +               pci_disable_device(pdev);
>>>           } else
>>>                   printk(KERN_ERR "fealnx: remove for unknown device\n");
>>>
>>>
>>>> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
>>>> Signed-off-by: Yang Yingliang <yangyingliang@...wei.com>
>>>> ---
>>>>    drivers/net/ethernet/fealnx.c | 4 +---
>>>>    1 file changed, 1 insertion(+), 3 deletions(-)
>>>>
>>>> diff --git a/drivers/net/ethernet/fealnx.c b/drivers/net/ethernet/fealnx.c
>>>> index ed18450fd2cc..fb139f295b67 100644
>>>> --- a/drivers/net/ethernet/fealnx.c
>>>> +++ b/drivers/net/ethernet/fealnx.c
>>>> @@ -494,7 +494,7 @@ static int fealnx_init_one(struct pci_dev *pdev,
>>>>    	option = card_idx < MAX_UNITS ? options[card_idx] : 0;
>>>> -	i = pci_enable_device(pdev);
>>>> +	i = pcim_enable_device(pdev);
>>>>    	if (i) return i;
>>>>    	pci_set_master(pdev);
>>>> @@ -670,7 +670,6 @@ static int fealnx_init_one(struct pci_dev *pdev,
>>>>    err_out_unmap:
>>>>    	pci_iounmap(pdev, ioaddr);
>>>>    err_out_res:
>>>> -	pci_release_regions(pdev);
>>>>    	return err;
>>>>    }
>>>> @@ -689,7 +688,6 @@ static void fealnx_remove_one(struct pci_dev *pdev)
>>>>    		unregister_netdev(dev);
>>>>    		pci_iounmap(pdev, np->mem);
>>>>    		free_netdev(dev);
>>>> -		pci_release_regions(pdev);
>>>>    	} else
>>>>    		printk(KERN_ERR "fealnx: remove for unknown device\n");
>>>>    }
>>>> -- 
>>>> 2.25.1
>>>>
>>> .
> .