linux-kernel - Re: [PATCH] PCI: endpoint: Return error code when callback is not implemented

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] [day] [month] [year] [list]

Message-ID: <a6ce7776-68c7-3fd2-32ed-3981da4bf33b@kernel.org>
Date:   Fri, 18 Aug 2023 11:51:46 +0900
From:   Damien Le Moal <dlemoal@...nel.org>
To:     Rick Wertenbroek <rick.wertenbroek@...il.com>,
        rick.wertenbroek@...g-vd.ch
Cc:     Lorenzo Pieralisi <lpieralisi@...nel.org>,
        Krzysztof Wilczyński <kw@...ux.com>,
        Manivannan Sadhasivam <mani@...nel.org>,
        Kishon Vijay Abraham I <kishon@...nel.org>,
        Bjorn Helgaas <bhelgaas@...gle.com>, linux-pci@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] PCI: endpoint: Return error code when callback is not
 implemented

On 2023/08/17 22:33, Rick Wertenbroek wrote:
> Return an error code when the callback for an endpoint controller
> function is not implemented (is null). Among all the functions, only
> map_msi_irq() had an error returned if the callback was null. Extend
> this error handling to other functions as well.
> 
> Signed-off-by: Rick Wertenbroek <rick.wertenbroek@...il.com>

We should check for the mandatory ops definition in pci_epf_register_driver()
for function drivers and in __pci_epc_create() for controller drivers.

> ---
>  drivers/pci/endpoint/pci-epc-core.c | 14 +++++++-------
>  1 file changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/pci/endpoint/pci-epc-core.c b/drivers/pci/endpoint/pci-epc-core.c
> index 5a4a8b0be626..14f4256b9410 100644
> --- a/drivers/pci/endpoint/pci-epc-core.c
> +++ b/drivers/pci/endpoint/pci-epc-core.c
> @@ -196,7 +196,7 @@ int pci_epc_start(struct pci_epc *epc)
>  		return -EINVAL;
>  
>  	if (!epc->ops->start)
> -		return 0;
> +		return -EINVAL;
>  
>  	mutex_lock(&epc->lock);
>  	ret = epc->ops->start(epc);
> @@ -228,7 +228,7 @@ int pci_epc_raise_irq(struct pci_epc *epc, u8 func_no, u8 vfunc_no,
>  		return -EINVAL;
>  
>  	if (!epc->ops->raise_irq)
> -		return 0;
> +		return -EINVAL;

ENOTSUPP ?

>  
>  	mutex_lock(&epc->lock);
>  	ret = epc->ops->raise_irq(epc, func_no, vfunc_no, type, interrupt_num);
> @@ -340,7 +340,7 @@ int pci_epc_set_msi(struct pci_epc *epc, u8 func_no, u8 vfunc_no, u8 interrupts)
>  		return -EINVAL;
>  
>  	if (!epc->ops->set_msi)
> -		return 0;
> +		return -EINVAL;

ENOTSUPP ?

>  
>  	encode_int = order_base_2(interrupts);
>  
> @@ -408,7 +408,7 @@ int pci_epc_set_msix(struct pci_epc *epc, u8 func_no, u8 vfunc_no,
>  		return -EINVAL;
>  
>  	if (!epc->ops->set_msix)
> -		return 0;
> +		return -EINVAL;

ENOTSUPP ?

>  
>  	mutex_lock(&epc->lock);
>  	ret = epc->ops->set_msix(epc, func_no, vfunc_no, interrupts - 1, bir,
> @@ -469,7 +469,7 @@ int pci_epc_map_addr(struct pci_epc *epc, u8 func_no, u8 vfunc_no,
>  		return -EINVAL;
>  
>  	if (!epc->ops->map_addr)
> -		return 0;
> +		return -EINVAL;

This one is definitely mandatory and should always be defined. So check in
__pci_epc_create() ?

>  
>  	mutex_lock(&epc->lock);
>  	ret = epc->ops->map_addr(epc, func_no, vfunc_no, phys_addr, pci_addr,
> @@ -537,7 +537,7 @@ int pci_epc_set_bar(struct pci_epc *epc, u8 func_no, u8 vfunc_no,
>  		return -EINVAL;
>  
>  	if (!epc->ops->set_bar)
> -		return 0;
> +		return -EINVAL;

Same.

>  
>  	mutex_lock(&epc->lock);
>  	ret = epc->ops->set_bar(epc, func_no, vfunc_no, epf_bar);
> @@ -575,7 +575,7 @@ int pci_epc_write_header(struct pci_epc *epc, u8 func_no, u8 vfunc_no,
>  		return -EINVAL;
>  
>  	if (!epc->ops->write_header)
> -		return 0;
> +		return -EINVAL;

Same.

>  
>  	mutex_lock(&epc->lock);
>  	ret = epc->ops->write_header(epc, func_no, vfunc_no, header);

-- 
Damien Le Moal
Western Digital Research