> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Wednesday, February 11, 2015 10:16 PM
> To: Stephen Hemminger
> Cc: dev at dpdk.org; Vincent JARDIN; Ouyang, Changchun
> Subject: Re: [dpdk-dev] [PATCH v4 17/26] virtio: Use
2015-02-11 05:50, Stephen Hemminger:
> But driver needs to tell application via driver flags
I think it's done:
> + pci_dev->intr_handle.type = RTE_INTR_HANDLE_UIO;
> + pci_drv->drv_flags |= RTE_PCI_DRV_INTR_LSC;
In my understanding, Changchun chose to use ioports as a fallback if uio
is
On 11/02/2015 05:50, Ouyang, Changchun wrote:
> As we know proc/ioports hasn't interrupt, 6wind implementation also don't
> introduce interrupt in it.
correct, it is running without interrupt, they are not mandatory.
But driver needs to tell application via driver flags
> -Original Message-
> From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> Sent: Wednesday, February 11, 2015 12:33 PM
> To: Ouyang, Changchun
> Cc: dev at dpdk.org; Xie, Huawei; Cao, Waterman; Xu, Qian Q
> Subject: Re: [PATCH v4 17/26] virtio: Use port IO to get PCI resource.
On Mon, 9 Feb 2015 09:14:06 +0800
Ouyang Changchun wrote:
> Make virtio not require UIO for some security reasons, this is to match
> 6Wind's virtio-net-pmd.
>
> Signed-off-by: Changchun Ouyang
This breaks applications that use link state interrupt.
If the non-UIO mode is used link state in
Make virtio not require UIO for some security reasons, this is to match 6Wind's
virtio-net-pmd.
Signed-off-by: Changchun Ouyang
---
changes in v3:
Remove macro RTE_EAL_PORT_IO;
virtio pmd could support uio and ioports method to get the address;
lib/librte_pmd_virtio/Makefile| 2
7 matches
Mail list logo
