From: Tim Deegan <tim@xen.org>
To: Ian Campbell <ian.campbell@citrix.com>
Cc: xen-devel@lists.xen.org
Subject: Re: [PATCH 04/12] arm: parse modules from DT during early boot.
Date: Thu, 29 Nov 2012 17:05:01 +0000 [thread overview]
Message-ID: <20121129170501.GF80627@ocelot.phlegethon.org> (raw)
In-Reply-To: <1352823804-28482-4-git-send-email-ian.campbell@citrix.com>
At 16:23 +0000 on 13 Nov (1352823796), Ian Campbell wrote:
> The bootloader should populate /chosen/module@<N>/ for each module it
> wishes to pass to the hypervisor. The content of these nodes is
> described in docs/misc/arm/device-tree/booting.txt
>
> The hypervisor allows for 2 modules (@1==kernel and @2==initrd).
> Currently we don't do anything with them.
>
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
> ---
> v2: Reserve the zeroeth module for Xen itself (not used yet)
> Use a more idiomatic DT layout
> Document said layout
> ---
> docs/misc/arm/device-tree/booting.txt | 27 ++++++++++++
> xen/common/device_tree.c | 75 +++++++++++++++++++++++++++++++++
> xen/include/xen/device_tree.h | 14 ++++++
> 3 files changed, 116 insertions(+), 0 deletions(-)
> create mode 100644 docs/misc/arm/device-tree/booting.txt
>
> diff --git a/docs/misc/arm/device-tree/booting.txt b/docs/misc/arm/device-tree/booting.txt
> new file mode 100644
> index 0000000..2609450
> --- /dev/null
> +++ b/docs/misc/arm/device-tree/booting.txt
> @@ -0,0 +1,27 @@
> +Xen is passed the dom0 kernel and initrd via a reference in the /chosen
> +node of the device tree.
> +
> +Each node has the form /chosen/module@<N> and contains the following
> +properties:
> +
> +- compatible
> +
> + Must be "xen,multiboot-module"
> +
> +- start
> +
> + Physical address of the start of this module
> +
> +- end
> +
> + Physical address of the end of this module
> +
> +- bootargs (optional)
> +
> + Command line associated with this module
> +
> +The following modules are understood
> +
> +- 1 -- the domain 0 kernel
> +- 2 -- the domain 0 ramdisk
> +
> diff --git a/xen/common/device_tree.c b/xen/common/device_tree.c
> index 3d1f0f4..efd1663 100644
> --- a/xen/common/device_tree.c
> +++ b/xen/common/device_tree.c
> @@ -229,6 +229,79 @@ static void __init process_memory_node(const void *fdt, int node,
> }
> }
>
> +static void __init process_chosen_node(const void *fdt, int node,
> + const char *name,
> + u32 address_cells, u32 size_cells)
> +{
> + const struct fdt_property *prop;
> + const u32 *cell;
> + paddr_t size;
> + int nr, depth, nr_modules = 0;
> + struct dt_mb_module *mod;
> + int len;
> +
> + for ( depth = 0;
> + depth >= 0;
> + node = fdt_next_node(fdt, node, &depth) )
> + {
> + name = fdt_get_name(fdt, node, NULL);
> + if ( strncmp(name, "module@", strlen("module@")) == 0 ) {
> +
> + if ( fdt_node_check_compatible(fdt, node,
> + "xen,multiboot-module" ) != 0 )
> + early_panic("%s not a compatible module node\n", name);
> +
> + nr = simple_strtol(name + strlen("module@"), NULL, 10);
> + if ( nr <= 0 )
> + early_panic("Invalid module number %d\n", nr);
> +
> + if ( nr > NR_MODULES )
> + early_panic("too many modules %d > %d\n", nr, NR_MODULES);
> + if ( nr > nr_modules )
> + nr_modules = nr;
> +
> + mod = &early_info.modules.module[nr];
> +
> + prop = fdt_get_property(fdt, node, "start", NULL);
> + if ( !prop )
> + early_panic("no start for module %d\n", nr);
> +
> + cell = (const u32 *)prop->data;
> + device_tree_get_reg(&cell, address_cells, size_cells,
> + &mod->start, &size);
This get_reg returns a start + size -- can/should we encode the module
as one of these rather than encdong start + end separately and
discarding the 'size' fields?
> +
> + prop = fdt_get_property(fdt, node, "end", NULL);
> + if ( !prop )
> + early_panic("no end for module %d\n", nr);
> +
> + cell = (const u32 *)prop->data;
> + device_tree_get_reg(&cell, address_cells, size_cells,
> + &mod->size, &size);
> + mod->size -= mod->start;
> +
> + prop = fdt_get_property(fdt, node, "bootargs", &len);
> + if ( prop )
> + {
> + if ( len > sizeof(mod->cmdline) )
> + early_panic("module %d command line too long\n", nr);
> +
> + safe_strcpy(mod->cmdline, prop->data);
> + }
> + else
> + mod->cmdline[0] = 0;
> + }
> + }
> +
> + for ( nr = 1 ; nr < nr_modules ; nr++ )
> + {
> + mod = &early_info.modules.module[nr];
> + if ( !mod->start || !mod->size )
> + early_panic("module %d missing / invalid\n", nr);
> + }
> +
> + early_info.modules.nr_mods = nr_modules;
> +}
> +
> static int __init early_scan_node(const void *fdt,
> int node, const char *name, int depth,
> u32 address_cells, u32 size_cells,
> @@ -236,6 +309,8 @@ static int __init early_scan_node(const void *fdt,
> {
> if ( device_tree_node_matches(fdt, node, "memory") )
> process_memory_node(fdt, node, name, address_cells, size_cells);
> + else if ( device_tree_node_matches(fdt, node, "chosen") )
> + process_chosen_node(fdt, node, name, address_cells, size_cells);
>
> return 0;
> }
> diff --git a/xen/include/xen/device_tree.h b/xen/include/xen/device_tree.h
> index 4d010c0..c383677 100644
> --- a/xen/include/xen/device_tree.h
> +++ b/xen/include/xen/device_tree.h
> @@ -15,6 +15,7 @@
> #define DEVICE_TREE_MAX_DEPTH 16
>
> #define NR_MEM_BANKS 8
> +#define NR_MODULES 2
>
> struct membank {
> paddr_t start;
> @@ -26,8 +27,21 @@ struct dt_mem_info {
> struct membank bank[NR_MEM_BANKS];
> };
>
> +struct dt_mb_module {
> + paddr_t start;
> + paddr_t size;
> + char cmdline[1024];
> +};
> +
> +struct dt_module_info {
> + int nr_mods;
> + /* Module 0 is Xen itself, followed by the provided modules-proper */
> + struct dt_mb_module module[NR_MODULES + 1];
> +};
> +
> struct dt_early_info {
> struct dt_mem_info mem;
> + struct dt_module_info modules;
> };
>
> typedef int (*device_tree_node_func)(const void *fdt,
> --
> 1.7.9.1
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel
next prev parent reply other threads:[~2012-11-29 17:05 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-13 16:22 [PATCH V2 00/12] arm: support for initial modules (e.g. dom0) and DTB supplied in RAM Ian Campbell
2012-11-13 16:23 ` [PATCH 01/12] arm: Enable build without CONFIG_DTB_FILE Ian Campbell
2012-11-13 16:23 ` [PATCH 02/12] arm: create a raw binary target Ian Campbell
2012-11-13 16:23 ` [PATCH 03/12] arm: handle xenheap which isn't at the start of RAM Ian Campbell
2012-11-13 16:23 ` [PATCH 04/12] arm: parse modules from DT during early boot Ian Campbell
2012-11-29 17:05 ` Tim Deegan [this message]
2012-11-29 17:13 ` Ian Campbell
2012-11-30 15:14 ` Stefano Stabellini
2012-11-30 15:11 ` Stefano Stabellini
2012-12-03 16:19 ` Ian Campbell
2012-12-04 12:42 ` Stefano Stabellini
2012-12-04 13:44 ` Ian Campbell
2012-11-13 16:23 ` [PATCH 05/12] arm: avoid placing Xen over any modules Ian Campbell
2012-11-13 16:23 ` [PATCH 06/12] arm: avoid allocating the heaps over modules or xen itself Ian Campbell
2012-11-29 17:06 ` Tim Deegan
2012-11-29 17:19 ` Ian Campbell
2012-11-29 17:45 ` Tim Deegan
2012-11-13 16:23 ` [PATCH 07/12] arm: const-correctness in virt_to_maddr Ian Campbell
2012-11-13 16:23 ` [PATCH 08/12] device-tree: get_val cannot cope with cells > 2, add a BUG Ian Campbell
2012-11-29 17:09 ` Tim Deegan
2012-11-29 17:14 ` Ian Campbell
2012-11-13 16:23 ` [PATCH 09/12] arm: load dom0 kernel from first boot module Ian Campbell
2012-11-29 17:15 ` Tim Deegan
2012-11-29 17:24 ` Ian Campbell
2012-11-29 17:55 ` Tim Deegan
2012-11-13 16:23 ` [PATCH 10/12] arm: discard boot modules after building domain 0 Ian Campbell
2012-11-13 16:23 ` [PATCH 11/12] arm: use /chosen/module@1/bootargs for domain 0 command line Ian Campbell
2012-11-13 16:23 ` [PATCH 12/12] xen: strip /chosen/module@<N>/* from dom0 device tree Ian Campbell
2012-11-13 16:38 ` [PATCH V2 00/12] arm: support for initial modules (e.g. dom0) and DTB supplied in RAM Ian Campbell
2012-11-29 17:59 ` Tim Deegan
2012-11-29 18:05 ` Ian Campbell
2012-11-30 12:20 ` Ian Campbell
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=20121129170501.GF80627@ocelot.phlegethon.org \
--to=tim@xen.org \
--cc=ian.campbell@citrix.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).