[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 2/8] rdma: core rdma logic
From: |
Paolo Bonzini |
Subject: |
Re: [Qemu-devel] [PATCH 2/8] rdma: core rdma logic |
Date: |
Fri, 12 Apr 2013 13:10:45 +0200 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130311 Thunderbird/17.0.4 |
Il 12/04/2013 07:52, address@hidden ha scritto:
>
> + *
> + * If we're using dynamic registration on the server-side, we have to
> + * send a registration command first.
> + */
> +static int __qemu_rdma_write(QEMUFile *f, RDMAContext *rdma,
> + int current_index,
> + uint64_t offset, uint64_t length,
> + uint64_t wr_id, enum ibv_send_flags flag)
> +{
No __ prefix, perhaps call this function qemu_rdma_flush_one?
> + /*
> + * Don't pin zero pages on the destination. Just return.
> + */
> + if (rdma->chunk_register_destination &&
> + (buffer_find_nonzero_offset(va, size) == size)) {
> + return size;
> + }
Is this the right place to test? Is it correct if a page first is
non-zero and then becomes zero? Perhaps you have to test where you add
a page to a chunk, instead.
Paolo
- [Qemu-devel] [PATCH 3/8] rdma: new QEMUFileOps hooks, (continued)
- [Qemu-devel] [PATCH 3/8] rdma: new QEMUFileOps hooks, mrhines, 2013/04/12
- [Qemu-devel] [PATCH 4/8] rdma: implement new QEMUFileOps hooks, mrhines, 2013/04/12
- [Qemu-devel] [PATCH 5/8] rdma: introduce capability for chunk registration, mrhines, 2013/04/12
- [Qemu-devel] [PATCH 6/8] rdma: send pc.ram, mrhines, 2013/04/12
- [Qemu-devel] [PATCH 7/8] rdma: print out throughput while debugging, mrhines, 2013/04/12
- [Qemu-devel] [PATCH 8/8] rdma: add documentation, mrhines, 2013/04/12
- Re: [Qemu-devel] [PATCH 2/8] rdma: core rdma logic,
Paolo Bonzini <=