linux-kernel - Re: [PATCH 9/21] MSI: Expand pci_msi

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-next>] [day] [month] [year] [list]

Message-ID: <m1slbqrn70.fsf@ebiederm.dsl.xmission.com>
Date:	Tue, 27 Mar 2007 22:45:07 -0600
From:	ebiederm@...ssion.com (Eric W. Biederman)
To:	Michael Ellerman <michael@...erman.id.au>
Cc:	linux-pci@...ey.karlin.mff.cuni.cz,
	Greg Kroah-Hartman <greg@...ah.com>,
	"David S. Miller" <davem@...emloft.net>,
	Benjamin Herrenschmidt <benh@...nel.crashing.org>,
	<linux-kernel@...r.kernel.org>, Andrew Morton <akpm@...l.org>,
	<daniel.e.wolstenholme@...el.com>
Subject: Re: [PATCH 9/21] MSI: Expand pci_msi_supported()

Michael Ellerman <michael@...erman.id.au> writes:

> pci_enable_msi() and pci_enable_msix() both search for the MSI/MSI-X
> capability, we can fold this into pci_msi_supported() by passing the
> type in.
>
> Update the code to match the comment for pci_msi_supported(). That is
> it returns 0 on success, and anything else indicates an error.

Ok.  Looking at this one I don't see a bug exactly but there
is one very confusing piece.

Currently we have a function pci_msi_supported that sounds like it
sounds like it should return a boolean value.

However instead it returns 0 for success and -EINVAL for failure.

Reading through the code before this patch it is clear it does this
weird thing because we check for < 0.

After this patch we are simply checking to see if there was a
return value at all.

Can we please change the return value here so it is actually boolean.
1 for supported 0 for not supported.

There aren't any useful return values anyway so this would just make
the code easier to read and maintain.

Eric


>
> Signed-off-by: Michael Ellerman <michael@...erman.id.au>
> ---
>
>  drivers/pci/msi.c |   19 ++++++++-----------
>  1 file changed, 8 insertions(+), 11 deletions(-)
>
> Index: msi-new/drivers/pci/msi.c
> ===================================================================
> --- msi-new.orig/drivers/pci/msi.c
> +++ msi-new/drivers/pci/msi.c
> @@ -433,12 +433,13 @@ static int msix_capability_init(struct p
>  /**
>   * pci_msi_supported - check whether MSI may be enabled on device
>   * @dev: pointer to the pci_dev data structure of MSI device function
> + * @type: are we checking for MSI or MSI-X ?
>   *
>   * Look at global flags, the device itself, and its parent busses
>   * to return 0 if MSI are supported for the device.
>   **/
>  static
> -int pci_msi_supported(struct pci_dev * dev)
> +int pci_msi_supported(struct pci_dev * dev, int type)
>  {
>  	struct pci_bus *bus;
>  
> @@ -456,6 +457,9 @@ int pci_msi_supported(struct pci_dev * d
>  		if (bus->bus_flags & PCI_BUS_FLAGS_NO_MSI)
>  			return -EINVAL;
>  
> +	if (!pci_find_capability(dev, type))
> +		return -EINVAL;
> +
>  	return 0;
>  }
>  
> @@ -471,13 +475,9 @@ int pci_msi_supported(struct pci_dev * d
>   **/
>  int pci_enable_msi(struct pci_dev* dev)
>  {
> -	int pos, status;
> -
> -	if (pci_msi_supported(dev) < 0)
> -		return -EINVAL;
> +	int status;
>  
> -	pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
> -	if (!pos)
> +	if (pci_msi_supported(dev, PCI_CAP_ID_MSI))
>  		return -EINVAL;
>  
>  	WARN_ON(!!dev->msi_enabled);
> @@ -580,13 +580,10 @@ int pci_enable_msix(struct pci_dev* dev,
>  	int i, j;
>  	u16 control;
>  
> -	if (!entries || pci_msi_supported(dev) < 0)
> +	if (!entries || pci_msi_supported(dev, PCI_CAP_ID_MSIX))
>   		return -EINVAL;
>  
>  	pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
> -	if (!pos)
> - 		return -EINVAL;
> -
>  	pci_read_config_word(dev, msi_control_reg(pos), &control);
>  	nr_entries = multi_msix_capable(control);
>  	if (nvec > nr_entries)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/