xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Julien Grall <julien.grall@linaro.org>
To: Arianna Avanzini <avanzini.arianna@gmail.com>
Cc: julien.grall@citrix.com, paolo.valente@unimore.it, keir@xen.org,
	stefano.stabellini@eu.citrix.com, tim@xen.org,
	dario.faggioli@citrix.com, Ian.Jackson@eu.citrix.com,
	xen-devel@lists.xen.org, Ian.Campbell@eu.citrix.com,
	etrudeau@broadcom.com, JBeulich@suse.com,
	viktor.kleinik@globallogic.com
Subject: Re: [PATCH v4 5/7] tools, libxl: parse optional start gfn from the iomem config option
Date: Tue, 25 Mar 2014 15:39:48 +0000	[thread overview]
Message-ID: <5331A344.9000801@linaro.org> (raw)
In-Reply-To: <1395712976-19454-6-git-send-email-avanzini.arianna@gmail.com>

Hi Arianna,

Thank you for the patch.

On 03/25/2014 02:02 AM, Arianna Avanzini wrote:
> diff --git a/tools/libxl/libxl.h b/tools/libxl/libxl.h
> index 06bbca6..13f5fe7 100644
> --- a/tools/libxl/libxl.h
> +++ b/tools/libxl/libxl.h
> @@ -95,6 +95,16 @@
>  #define LIBXL_HAVE_BUILDINFO_EVENT_CHANNELS 1
>  
>  /*
> + * LIBXL_HAVE_BUILDINFO_IOMEM_START_GFN indicated that it is possible
> + * to specify the start guest frame number used to map a range of I/O
> + * memory machine frame numbers via the 'gfn' field (of type uint64)
> + * of the 'iomem' structure. An array of iomem structures is embedded
> + * in libxl_domain_build_info and used to map the indicated memory
> + * ranges during domain build.
> + */
> +#define LIBXL_HAVE_BUILDINFO_IOMEM_START_GFN 1
> +
> +/*
>   * libxl ABI compatibility
>   *
>   * The only guarantee which libxl makes regarding ABI compatibility
> diff --git a/tools/libxl/libxl_types.idl b/tools/libxl/libxl_types.idl
> index 649ce50..4462586 100644
> --- a/tools/libxl/libxl_types.idl
> +++ b/tools/libxl/libxl_types.idl
> @@ -169,8 +169,9 @@ libxl_ioport_range = Struct("ioport_range", [
>      ])
>  
>  libxl_iomem_range = Struct("iomem_range", [
> -    ("start", uint64),
> -    ("number", uint64),
> +    ("start", uint64),  # start host frame number to be mapped to the guest
> +    ("number", uint64), # number of frames to be mapped
> +    ("gfn", uint64),    # guest frame number used as a start for the mapping
>      ])

When this structure will be used by another toolstack (e.g not xl), the
default value for gfn will be 0. This is wrong because we won't be able
to differentiate a user that will want to map to the GFN 0 and a 1:1
mapping. -1 seems the best default value for now.

You can use init_val in the IDL to set this value.

>  libxl_vga_interface_info = Struct("vga_interface_info", [
> diff --git a/tools/libxl/xl_cmdimpl.c b/tools/libxl/xl_cmdimpl.c
> index 4fc46eb..99a0958 100644
> --- a/tools/libxl/xl_cmdimpl.c
> +++ b/tools/libxl/xl_cmdimpl.c
> @@ -1208,13 +1208,20 @@ static void parse_config_data(const char *config_source,
>                          "xl: Unable to get element %d in iomem list\n", i);
>                  exit(1);
>              }

It's a bug on the current code. We have to init correctly iomem before
with libxl_iomem_range_init(&b_info->iomem);

> -            if(sscanf(buf, "%" SCNx64",%" SCNx64,
> -                     &b_info->iomem[i].start,
> -                     &b_info->iomem[i].number)
> -                  != 2) {
> -               fprintf(stderr,
> -                       "xl: Invalid argument parsing iomem: %s\n", buf);
> -               exit(1);
> +            switch(sscanf(buf, "%" SCNx64",%" SCNx64"@%" SCNx64,
> +                          &b_info->iomem[i].start,
> +                          &b_info->iomem[i].number,
> +                          &b_info->iomem[i].gfn)) {

I don't like the switch(sscanf(... it's hard to read.

I would prefer:

ret = sscanf(...);
switch (ret) {
}

> +                case 3: break;
> +                case 2:
> +                        /* default to 1:1 mapping */
> +                        b_info->iomem[i].gfn = b_info->iomem[i].start;
> +                        break;

If the iomem_range is initialized with the default value. You can defer
this initialization in libxl.

The result code here will be nicer:

ret = sscanf(...)

if ( ret < 2 )
  fprintf(stderr, ...);

Regards,

-- 
Julien Grall

  reply	other threads:[~2014-03-25 15:39 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-03-25  2:02 [PATCH v4 0/7] Implement the XEN_DOMCTL_memory_mapping hypercall for ARM Arianna Avanzini
2014-03-25  2:02 ` [PATCH v4 1/7] arch, arm: domain build: let dom0 access I/O memory of mapped devices Arianna Avanzini
2014-03-25 12:37   ` Julien Grall
2014-03-25  2:02 ` [PATCH v4 2/7] arch, arm: add consistency checks to REMOVE p2m changes Arianna Avanzini
2014-03-25 12:18   ` Stefano Stabellini
2014-03-25 12:51   ` Julien Grall
2014-03-25 13:10     ` Julien Grall
2014-03-25 17:41   ` Ian Campbell
2014-03-25  2:02 ` [PATCH v4 3/7] arch, arm: let map_mmio_regions() take pfn as parameters Arianna Avanzini
2014-03-25 12:22   ` Stefano Stabellini
2014-03-25 12:54     ` Julien Grall
2014-03-28 12:51       ` Arianna Avanzini
2014-03-28 13:31         ` Julien Grall
2014-03-25 13:00   ` Julien Grall
2014-03-25  2:02 ` [PATCH v4 4/7] xen, common: add the XEN_DOMCTL_memory_mapping hypercall Arianna Avanzini
2014-03-25  9:33   ` Jan Beulich
2014-03-28 13:24     ` Arianna Avanzini
2014-03-28 13:30       ` Jan Beulich
2014-03-25 12:35   ` Stefano Stabellini
2014-03-25 14:10     ` Jan Beulich
2014-03-25 15:10       ` Stefano Stabellini
2014-03-25 15:36         ` Jan Beulich
2014-03-25 15:42           ` Stefano Stabellini
2014-04-01 15:01             ` Ian Campbell
2014-04-01 15:18               ` Jan Beulich
2014-04-01 15:37                 ` Ian Campbell
2014-03-25 13:17   ` Julien Grall
2014-04-01 14:52     ` Ian Campbell
2014-04-01 15:16       ` Julien Grall
2014-04-01 15:39         ` Ian Campbell
2014-04-01 16:00           ` Julien Grall
2014-04-02  9:43             ` Ian Campbell
2014-04-02 10:06               ` Jan Beulich
2014-04-02 10:19                 ` Ian Campbell
2014-04-02 10:53                   ` Jan Beulich
2014-04-05 12:08                     ` Arianna Avanzini
2014-04-06 16:23                       ` Stefano Stabellini
2014-04-07  7:01                       ` Jan Beulich
2014-03-25  2:02 ` [PATCH v4 5/7] tools, libxl: parse optional start gfn from the iomem config option Arianna Avanzini
2014-03-25 15:39   ` Julien Grall [this message]
2014-03-25 15:45     ` Julien Grall
2014-03-25 16:27     ` Ian Campbell
2014-03-25  2:02 ` [PATCH v4 6/7] tools, libxl: add helpers to establish if guest is auto-translated Arianna Avanzini
2014-03-25  2:02 ` [PATCH v4 7/7] tools, libxl: handle the iomem parameter with the memory_mapping hcall Arianna Avanzini
2014-04-01 15:13   ` Ian Campbell
2014-04-01 15:26     ` Julien Grall
2014-04-01 15:34       ` Ian Campbell
2014-04-01 20:52       ` Daniel De Graaf
2014-04-02  9:45         ` Ian Campbell
2014-04-02 14:14           ` Daniel De Graaf

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=5331A344.9000801@linaro.org \
    --to=julien.grall@linaro.org \
    --cc=Ian.Campbell@eu.citrix.com \
    --cc=Ian.Jackson@eu.citrix.com \
    --cc=JBeulich@suse.com \
    --cc=avanzini.arianna@gmail.com \
    --cc=dario.faggioli@citrix.com \
    --cc=etrudeau@broadcom.com \
    --cc=julien.grall@citrix.com \
    --cc=keir@xen.org \
    --cc=paolo.valente@unimore.it \
    --cc=stefano.stabellini@eu.citrix.com \
    --cc=tim@xen.org \
    --cc=viktor.kleinik@globallogic.com \
    --cc=xen-devel@lists.xen.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).