qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Yuval Shaia <yuval.shaia@oracle.com>
To: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: qemu-devel@nongnu.org, yuval.shaia@oracle.com
Subject: Re: [Qemu-devel] [PATCH 07/13] hw/rdma: Do not allocate memory for non-dma MR
Date: Sun, 5 Aug 2018 17:56:34 +0300	[thread overview]
Message-ID: <20180805145634.GA22209@lap1> (raw)
In-Reply-To: <42b81675-ea89-e9e9-f84c-ad723372fc1c@gmail.com>

On Tue, Jul 24, 2018 at 03:19:52PM +0300, Marcel Apfelbaum wrote:
> 
> Hi Yuval,
> 
> On 07/16/2018 10:40 AM, Yuval Shaia wrote:
> > There is no use in the memory allocated for non-dma MR (one with
> > host_virt equals to NULL).
> 
> No need for the (one with...)

Will remove.

> 
> > Delete the code that allocates it.
> > 
> > Signed-off-by: Yuval Shaia <yuval.shaia@oracle.com>
> > ---
> >   hw/rdma/rdma_rm.c | 52 +++++++++++++++++++----------------------------
> >   1 file changed, 21 insertions(+), 31 deletions(-)
> > 
> > diff --git a/hw/rdma/rdma_rm.c b/hw/rdma/rdma_rm.c
> > index 7403d24674..bf4a5c71b4 100644
> > --- a/hw/rdma/rdma_rm.c
> > +++ b/hw/rdma/rdma_rm.c
> > @@ -144,8 +144,6 @@ int rdma_rm_alloc_mr(RdmaDeviceResources *dev_res, uint32_t pd_handle,
> >       RdmaRmMR *mr;
> >       int ret = 0;
> >       RdmaRmPD *pd;
> > -    void *addr;
> > -    size_t length;
> >       pd = rdma_rm_get_pd(dev_res, pd_handle);
> >       if (!pd) {
> > @@ -158,40 +156,29 @@ int rdma_rm_alloc_mr(RdmaDeviceResources *dev_res, uint32_t pd_handle,
> >           pr_dbg("Failed to allocate obj in table\n");
> >           return -ENOMEM;
> >       }
> > +    pr_dbg("mr_handle=%d\n", *mr_handle);
> > -    if (!host_virt) {
> > -        /* TODO: This is my guess but not so sure that this needs to be
> > -         * done */
> > -        length = TARGET_PAGE_SIZE;
> > -        addr = g_malloc(length);
> > -    } else {
> > +    pr_dbg("host_virt=0x%p\n", host_virt);
> > +    pr_dbg("guest_start=0x%" PRIx64 "\n", guest_start);
> > +    pr_dbg("length=%zu\n", guest_length);
> > +
> > +    if (host_virt) {
> >           mr->virt = host_virt;
> > -        pr_dbg("host_virt=0x%p\n", mr->virt);
> > -        mr->length = guest_length;
> > -        pr_dbg("length=%zu\n", guest_length);
> >           mr->start = guest_start;

[1]

> > -        pr_dbg("guest_start=0x%" PRIx64 "\n", mr->start);
> > -
> > -        length = mr->length;
> > -        addr = mr->virt;
> > -    }
> > +        mr->length = guest_length;
> > -    ret = rdma_backend_create_mr(&mr->backend_mr, &pd->backend_pd, addr, length,
> > -                                 access_flags);
> > -    if (ret) {
> > -        pr_dbg("Fail in rdma_backend_create_mr, err=%d\n", ret);
> > -        ret = -EIO;
> > -        goto out_dealloc_mr;
> > +        ret = rdma_backend_create_mr(&mr->backend_mr, &pd->backend_pd, mr->virt,
> > +                                     mr->length, access_flags);
> > +        if (ret) {
> > +            pr_dbg("Fail in rdma_backend_create_mr, err=%d\n", ret);
> > +            ret = -EIO;
> > +            goto out_dealloc_mr;
> > +        }
> >       }
> > -    if (!host_virt) {
> > -        *lkey = mr->lkey = rdma_backend_mr_lkey(&mr->backend_mr);
> > -        *rkey = mr->rkey = rdma_backend_mr_rkey(&mr->backend_mr);
> > -    } else {
> > -        /* We keep mr_handle in lkey so send and recv get get mr ptr */
> > -        *lkey = *mr_handle;
> > -        *rkey = -1;
> > -    }
> > +    /* We keep mr_handle in lkey so send and recv get get mr ptr */
> > +    *lkey = *mr_handle;
> > +    *rkey = -1;
> 
> Before this change rkey whould get a value when !host_virt.
> But I suppose is OK since Remote DMA operations are not implemented yet.

The entire code that handled the case where host_virt is NULL was wrong
thus removed.
And yes, RDMA verb is not yet implemented so rkey is anyway not needed.

> 
> >       mr->pd_handle = pd_handle;
> > @@ -214,7 +201,10 @@ void rdma_rm_dealloc_mr(RdmaDeviceResources *dev_res, uint32_t mr_handle)
> >       if (mr) {
> >           rdma_backend_destroy_mr(&mr->backend_mr);
> > -        munmap(mr->virt, mr->length);
> > +        pr_dbg("start=0x%" PRIx64 "\n", mr->start);
> > +        if (mr->start) {
> 
> When is the mr->start inited?

res_tbl_alloc cleans the MR before giving it to caller so we expect
mr->start to be NULL.
Then if host_virt is given then mr->start is set to guest virtual address
[1].

> 
> Thanks,
> Marcel
> 
> > +            munmap(mr->virt, mr->length);
> > +        }
> >           res_tbl_dealloc(&dev_res->mr_tbl, mr_handle);
> >       }
> >   }
> 

  reply	other threads:[~2018-08-05 14:56 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-07-16  7:40 [Qemu-devel] [PATCH 00/13] Misc fixes for pvrdma device Yuval Shaia
2018-07-16  7:40 ` [Qemu-devel] [PATCH 01/13] hw/rdma: Make distinction between device init and start modes Yuval Shaia
2018-07-24 12:08   ` Marcel Apfelbaum
2018-07-24 19:29     ` Yuval Shaia
2018-07-24 19:53       ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 02/13] hw/pvrdma: Bugfix - provide the correct attr_mask to query_qp Yuval Shaia
2018-07-16 10:38   ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 03/13] hw/rdma: Modify debug macros Yuval Shaia
2018-07-24 12:10   ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 04/13] hw/pvrdma: Clean CQE before use Yuval Shaia
2018-07-16 10:41   ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 05/13] hw/pvrdma: Make default pkey 0xFFFF Yuval Shaia
2018-07-16 10:42   ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 06/13] hw/rdma: Get rid of unneeded structure Yuval Shaia
2018-07-16 10:44   ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 07/13] hw/rdma: Do not allocate memory for non-dma MR Yuval Shaia
2018-07-24 12:19   ` Marcel Apfelbaum
2018-08-05 14:56     ` Yuval Shaia [this message]
2018-07-16  7:40 ` [Qemu-devel] [PATCH 08/13] hw/rdma: Reorder resource cleanup Yuval Shaia
2018-07-16 10:44   ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 09/13] hw/pvrdma: Cosmetic change - indent right Yuval Shaia
2018-07-16 10:45   ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 10/13] hw/rdma: Cosmetic change - move to generic function Yuval Shaia
2018-07-16 10:46   ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 11/13] hw/rdma: Print backend QP number in hex format Yuval Shaia
2018-07-16 10:46   ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 12/13] hw/rdma: Bugfix: Support non-aligned buffers Yuval Shaia
2018-07-24 12:22   ` Marcel Apfelbaum
2018-07-16  7:40 ` [Qemu-devel] [PATCH 13/13] hw/rdma: Save pci dev in backend_dev Yuval Shaia
2018-07-16 10:49   ` Marcel Apfelbaum

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180805145634.GA22209@lap1 \
    --to=yuval.shaia@oracle.com \
    --cc=marcel.apfelbaum@gmail.com \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).