[dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line

2014-05-19 Thread Anatoly Burakov
Unlike igb_uio, VFIO interrupt type is not set by kernel module parameters but is set up via ioctl() calls at runtime. This warrants a new EAL command-line parameter. It will have no effect if VFIO is not compiled, but will set VFIO interrupt type to either "legacy" or "msix" if VFIO support is com

[dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line

2014-05-20 Thread Stephen Hemminger
I really wish the code did automatic fall back based on PCI config. It is possible to know the right mode, and do the right thing. Rather than punting the problem out to command line which is totally unusable in hot plug and generic application. On Mon, May 19, 2014 at 8:51 AM, Anatoly Burakov wr

[dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line

2014-05-20 Thread Burakov, Anatoly
Hi Stephen, > I really wish the code did automatic fall back based on PCI config. It is > possible to know the right mode, and do the right thing. > Rather than punting the problem out to command line which is totally unusable > in hot plug and generic application. You mean we should use whatev

[dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line

2014-05-20 Thread Stephen Hemminger
On Tue, 20 May 2014 08:33:43 + "Burakov, Anatoly" wrote: > Hi Stephen, > > > I really wish the code did automatic fall back based on PCI config. It is > > possible to know the right mode, and do the right thing. > > Rather than punting the problem out to command line which is totally > > u

[dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line

2014-05-20 Thread Burakov, Anatoly
Hi Stephen, > I am not sure that MSI-X has any advantage with only one IRQ, so MSI would > do. Igb_uio doesn't support MSI, so I never included MSI support. It can be added though, but I don't see much point. > Then have the code look at the PCI capability of device and fallback to INTX > if >

[dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line

2014-05-21 Thread Stephen Hemminger
Originally igb_uio had code for MSI, but it was broken. See my patches which fixed that and several other bugs. On Tue, May 20, 2014 at 8:26 PM, Burakov, Anatoly wrote: > Hi Stephen, > > > I am not sure that MSI-X has any advantage with only one IRQ, so MSI > would > > do. > > Igb_uio doesn't s

[dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line

2014-05-22 Thread Thomas Monjalon
2014-05-19 16:51, Anatoly Burakov: > Unlike igb_uio, VFIO interrupt type is not set by kernel module > parameters but is set up via ioctl() calls at runtime. This warrants > a new EAL command-line parameter. It will have no effect if VFIO is > not compiled, but will set VFIO interrupt type to eithe

[dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line

2014-05-27 Thread Xu, HuilongX
Anatoly Burakov anatoly.burakov at intel.com Mon May 19 17:51:43 CEST 2014 Previous message: [dpdk-dev] [PATCH v2 09/16] Enable VFIO device binding Next message: [dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line Messages sorted by: [ date ] [ thread

[dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line

2014-05-28 Thread Burakov, Anatoly
Hi Thomas, > > + } > > else if (!strcmp(lgopts[option_index].name, > OPT_CREATE_UIO_DEV)) > > another code style issue reported by checkpatch.pl ;) > > But it should be fixed by removing this code as Stephen suggests. I'm not sure this could should be remov

[dpdk-dev] [PATCH v2 10/16] Added support for selecting VFIO interrupt type from EAL command-line

2014-05-28 Thread Thomas Monjalon
2014-05-28 10:35, Burakov, Anatoly: > Hi Thomas, > > > > + } > > > > > > else if (!strcmp(lgopts[option_index].name, > > > > OPT_CREATE_UIO_DEV)) > > > > another code style issue reported by checkpatch.pl ;) > > > > But it should be fixed by removing this code