RE: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs

2019-07-04 Thread Idan Burstein
org; Yamin Friedman ; Max Gurtovoy Subject: Re: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs Hey Idan, > " Please don't. This is a bad choice to opt it in by default." > > I disagree here. I'd prefer Linux to have good out of the box experience &

Re: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs

2019-07-04 Thread Leon Romanovsky
On Wed, Jul 03, 2019 at 11:56:04AM -0700, Sagi Grimberg wrote: > > > Hi Sagi, > > > > I'm not sharing your worries about bad out-of-the-box experience for a > > number of reasons. > > > > First of all, this code is part of upstream kernel and will take time > > till users actually start to use it a

Re: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs

2019-07-03 Thread Sagi Grimberg
Hi Sagi, I'm not sharing your worries about bad out-of-the-box experience for a number of reasons. First of all, this code is part of upstream kernel and will take time till users actually start to use it as is and not as part of some distro backports or MOFED packages. True, but I am still

Re: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs

2019-07-01 Thread Leon Romanovsky
On Mon, Jul 01, 2019 at 10:36:41PM -0700, Sagi Grimberg wrote: > Hey Idan, > > > " Please don't. This is a bad choice to opt it in by default." > > > > I disagree here. I'd prefer Linux to have good out of the box experience > > (e.g. reach 100G in 4K NVMeOF on Intel servers) with the default > >

Re: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs

2019-07-01 Thread Sagi Grimberg
Hey Idan, " Please don't. This is a bad choice to opt it in by default." I disagree here. I'd prefer Linux to have good out of the box experience (e.g. reach 100G in 4K NVMeOF on Intel servers) with the default parameters. Especially since Yamin have shown it is beneficial / not hurting in te

Re: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs

2019-06-26 Thread Yamin Friedman
On 6/26/2019 12:14 AM, Sagi Grimberg wrote: > > >> +static int ib_poll_dim_handler(struct irq_poll *iop, int budget) >> +{ >> +    struct ib_cq *cq = container_of(iop, struct ib_cq, iop); >> +    struct dim *dim = cq->dim; >> +    int completed; >> + >> +    completed = __ib_process_cq(cq, budget,

RE: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs

2019-06-26 Thread Idan Burstein
t: Re: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs > +static int ib_poll_dim_handler(struct irq_poll *iop, int budget) { > + struct ib_cq *cq = container_of(iop, struct ib_cq, iop); > + struct dim *dim = cq->dim; > + int completed; > + > +

Re: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs

2019-06-25 Thread Sagi Grimberg
+static int ib_poll_dim_handler(struct irq_poll *iop, int budget) +{ + struct ib_cq *cq = container_of(iop, struct ib_cq, iop); + struct dim *dim = cq->dim; + int completed; + + completed = __ib_process_cq(cq, budget, cq->wc, IB_POLL_BATCH); + if (completed < budg

[for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs

2019-06-25 Thread Saeed Mahameed
From: Yamin Friedman Added the interface in the infiniband driver that applies the rdma_dim adaptive moderation. There is now a special function for allocating an ib_cq that uses rdma_dim. Performance improvement (ConnectX-5 100GbE, x86) running FIO benchmark over NVMf between two equal end-host