From: Boris Ostrovsky <boris.ostrovsky@oracle.com>
To: Wei Liu <wei.liu2@citrix.com>,
Xen-devel <xen-devel@lists.xenproject.org>
Cc: Dario Faggioli <dario.faggioli@citrix.com>,
Ian Jackson <Ian.Jackson@eu.citrix.com>,
Ian Campbell <ian.campbell@citrix.com>
Subject: Re: [PATCH v3 3/4] libxc: rework vnuma bits in setup_guest
Date: Mon, 01 Jun 2015 16:18:28 -0400 [thread overview]
Message-ID: <556CBE14.3010509@oracle.com> (raw)
In-Reply-To: <1433153954-27994-4-git-send-email-wei.liu2@citrix.com>
On 06/01/2015 06:19 AM, Wei Liu wrote:
> Make the setup process similar to PV counterpart. That is, to allocate a
> P2M array that covers the whole memory range and start from there. This
> is clearer than using an array with no holes in it.
>
> Also the dummy layout should take MMIO hole into consideration. We might
> end up having two vmemranges in the dummy layout.
>
> Signed-off-by: Wei Liu <wei.liu2@citrix.com>
> Cc: Ian Campbell <ian.campbell@citrix.com>
> Cc: Ian Jackson <ian.jackson@eu.citrix.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
with a couple of nits below that you might consider.
> @@ -325,16 +352,23 @@ static int setup_guest(xc_interface *xch,
> DPRINTF(" TOTAL: %016"PRIx64"->%016"PRIx64"\n", v_start, v_end);
> DPRINTF(" ENTRY: %016"PRIx64"\n", elf_uval(&elf, elf.ehdr, e_entry));
>
> - if ( (page_array = malloc(nr_pages * sizeof(xen_pfn_t))) == NULL )
> + if ( (page_array = malloc(p2m_size * sizeof(xen_pfn_t))) == NULL )
> {
> PERROR("Could not allocate memory.");
> goto error_out;
> }
>
> - for ( i = 0; i < nr_pages; i++ )
> - page_array[i] = i;
> - for ( i = args->mmio_start >> PAGE_SHIFT; i < nr_pages; i++ )
> - page_array[i] += args->mmio_size >> PAGE_SHIFT;
> + for ( i = 0; i < p2m_size; i++ )
> + page_array[i] = ((xen_pfn_t)-1);
For large guests memset-ting page_array to 0xf should be much faster
(but obviously not type-safe).
> + for ( vmemid = 0; vmemid < args->nr_vmemranges; vmemid++ )
> + {
> + uint64_t pfn;
> +
> + for ( pfn = args->vmemranges[vmemid].start >> PAGE_SHIFT;
> + pfn < args->vmemranges[vmemid].end >> PAGE_SHIFT;
> + pfn++ )
> + page_array[pfn] = pfn;
> + }
>
> /*
> * Try to claim pages for early warning of insufficient memory available.
> @@ -645,6 +679,12 @@ static int setup_guest(xc_interface *xch,
> error_out:
> rc = -1;
> out:
> + if ( use_dummy )
Or 'if (args->vmemranges == dummy_vmemrange)' and drop use_dummy variable.
-boris
> + {
> + args->nr_vnodes = 0;
> + args->vmemranges = NULL;
> + args->vnode_to_pnode = NULL;
> + }
> if ( elf_check_broken(&elf) )
> ERROR("HVM ELF broken: %s", elf_check_broken(&elf));
>
next prev parent reply other threads:[~2015-06-01 20:18 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-01 10:19 [PATCH v3 0/4] Fix HVM vNUMA Wei Liu
2015-06-01 10:19 ` [PATCH v3 1/4] libxc/libxl: fill xc_hvm_build_args in libxl Wei Liu
2015-06-01 10:19 ` [PATCH v3 2/4] libxc: print more error messages when failed Wei Liu
2015-06-01 10:19 ` [PATCH v3 3/4] libxc: rework vnuma bits in setup_guest Wei Liu
2015-06-01 20:18 ` Boris Ostrovsky [this message]
2015-06-03 10:34 ` Ian Campbell
2015-06-03 10:39 ` Wei Liu
2015-06-03 10:57 ` Ian Campbell
2015-06-03 11:04 ` Wei Liu
2015-06-01 10:19 ` [PATCH v3 4/4] libxl: fix HVM vNUMA Wei Liu
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=556CBE14.3010509@oracle.com \
--to=boris.ostrovsky@oracle.com \
--cc=Ian.Jackson@eu.citrix.com \
--cc=dario.faggioli@citrix.com \
--cc=ian.campbell@citrix.com \
--cc=wei.liu2@citrix.com \
--cc=xen-devel@lists.xenproject.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.