From mboxrd@z Thu Jan  1 00:00:00 1970
From: Guillaume Gaudonville <guillaume.gaudonville-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>
Subject: Re: [PATCH 1/2] igb_uio: fix compability on old kernel
Date: Mon, 01 Sep 2014 17:07:04 +0200
Message-ID: <54048B98.708@6wind.com>
References: <20140725103627.4ca989b7@haswell.linuxnetplumber.net>
 <9317787.qKVt3oyQbN@xps13> <D01CF3B4.28B5%rsanford@akamai.com>
 <CA+cr1coJs9yokaDT1ya=5ChSWuSYWiF4TqmAhpcP0qLZKC4odA@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
To: Robert Sanford <rsanford2-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
 Thomas Monjalon <thomas.monjalon-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>, "dev-VfR2kkLFssw@public.gmane.org" <dev-VfR2kkLFssw@public.gmane.org>
Return-path: <dev-bounces-VfR2kkLFssw@public.gmane.org>
In-Reply-To: <CA+cr1coJs9yokaDT1ya=5ChSWuSYWiF4TqmAhpcP0qLZKC4odA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request-VfR2kkLFssw@public.gmane.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev-VfR2kkLFssw@public.gmane.org>
List-Help: <mailto:dev-request-VfR2kkLFssw@public.gmane.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request-VfR2kkLFssw@public.gmane.org?subject=subscribe>
Errors-To: dev-bounces-VfR2kkLFssw@public.gmane.org
Sender: "dev" <dev-bounces-VfR2kkLFssw@public.gmane.org>

On 08/22/2014 08:09 PM, Robert Sanford wrote:
> This is what we came up with. It works for us. In our kernel headers'
> linux/pci.h, pci_num_vf is enclosed within "#ifdef CONFIG_PCI_IOV/#endif";
> pci_intx_mask_supported and pci_check_and_mask_intx are enclosed within
> "#ifdef HAVE_PCI_SET_MWI/#endif".
Testing on HAVE_PCI_SET_MWI seems not correct. It is defined since linux 
2.6.12.
There is no define associated with the commit that adds 
pci_intx_mask_supported and
pci_check_and_mask_intx.

So I think we'll have to check the distribution, something like:

-#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 3, 0)
+#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 3, 0) && \
+       (!(RHEL_RELEASE_CODE && RHEL_RELEASE_CODE == 
RHEL_RELEASE_VERSION(6,5)))

What do you think?

--
Best regards,
Guillaume
>
> What do you think?
>
> --
> Thanks,
> Robert
>
>
> ---
>   lib/librte_eal/linuxapp/igb_uio/compat.h |    4 ++--
>   1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/lib/librte_eal/linuxapp/igb_uio/compat.h
> b/lib/librte_eal/linuxapp/igb_uio/compat.h
> index 2a16540..f7404d8 100644
> --- a/lib/librte_eal/linuxapp/igb_uio/compat.h
> +++ b/lib/librte_eal/linuxapp/igb_uio/compat.h
> @@ -17,7 +17,7 @@
>   #define   PCI_MSIX_ENTRY_CTRL_MASKBIT   1
>   #endif
>
> -#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 34)
> +#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 34) &&
> !defined(CONFIG_PCI_IOV)
>   static int pci_num_vf(struct pci_dev *dev)
>   {
>      struct iov {
> @@ -38,7 +38,7 @@ static int pci_num_vf(struct pci_dev *dev)
>   #endif
>
>
> -#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 3, 0)
> +#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 3, 0) &&
> !defined(HAVE_PCI_SET_MWI)
>
>   /* Check if INTX works to control irq's.
>    * Set's INTX_DISABLE flag and reads it back