From: Trent Piepho <tpiepho@kymetacorp.com>
To: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Barebox List <barebox@lists.infradead.org>
Subject: Re: [PATCH] of_path: Make partition names work again
Date: Thu, 7 Jan 2016 18:31:45 +0000 [thread overview]
Message-ID: <1452191516.4474.81.camel@rtred1test09.kymeta.local> (raw)
In-Reply-To: <1452166031-9973-1-git-send-email-s.hauer@pengutronix.de>
There's already a patch in barebox-next that fixes this bug, 985e773.
It doesn't delete the patname: parsing code, but does add a number of
comments that this patch doesn't.
On Thu, 2016-01-07 at 12:27 +0100, Sascha Hauer wrote:
> The device-path "partname:<name>" mechanism is broken by:
>
> 75b6827 of_path: of_find_path() factor out device detection logic into separate function
>
> This is because since said commit the device-path property is no longer
> searched in the original device node anymore, but in the device node
> of the device itself.
>
> Fix this by passing the partition name directly to __of_find_path.
>
> Originally it was intended to further extend the multi string property
> device-path further with more elements, like for example a filename. It
> turned out though that this is too complex and instead of further
> extending the property we should instead create additional properties,
> so this mechanism is removed with this patch.
>
> Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
> Cc: Marc Kleine-Budde <mkl@pengutronix.de>
> ---
> drivers/of/of_path.c | 125 ++++++++++++---------------------------------------
> 1 file changed, 29 insertions(+), 96 deletions(-)
>
> diff --git a/drivers/of/of_path.c b/drivers/of/of_path.c
> index 6903905..35d3576 100644
> --- a/drivers/of/of_path.c
> +++ b/drivers/of/of_path.c
> @@ -23,16 +23,6 @@
>
> #include <linux/mtd/mtd.h>
>
> -struct of_path {
> - struct cdev *cdev;
> - struct device_d *dev;
> -};
> -
> -struct of_path_type {
> - const char *name;
> - int (*parse)(struct of_path *op, const char *str);
> -};
> -
> struct device_d *of_find_device_by_node_path(const char *path)
> {
> struct device_d *dev;
> @@ -47,103 +37,34 @@ struct device_d *of_find_device_by_node_path(const char *path)
> return NULL;
> }
>
> -/**
> - * of_path_type_partname - find a partition based on physical device and
> - * partition name
> - * @op: of_path context
> - * @name: the partition name to find
> - */
> -static int of_path_type_partname(struct of_path *op, const char *name)
> -{
> - if (!op->dev)
> - return -EINVAL;
> -
> - op->cdev = device_find_partition(op->dev, name);
> - if (op->cdev) {
> - pr_debug("%s: found part '%s'\n", __func__, name);
> - return 0;
> - } else {
> - pr_debug("%s: cannot find part '%s'\n", __func__, name);
> - return -ENODEV;
> - }
> -}
> -
> -static struct of_path_type of_path_types[] = {
> - {
> - .name = "partname",
> - .parse = of_path_type_partname,
> - },
> -};
> -
> -static int of_path_parse_one(struct of_path *op, const char *str)
> -{
> - int i, ret;
> - char *name, *desc;
> -
> - pr_debug("parsing: %s\n", str);
> -
> - name = xstrdup(str);
> - desc = strchr(name, ':');
> - if (!desc) {
> - free(name);
> - return -EINVAL;
> - }
> -
> - *desc = 0;
> - desc++;
> -
> - for (i = 0; i < ARRAY_SIZE(of_path_types); i++) {
> - if (!strcmp(of_path_types[i].name, name)) {
> - ret = of_path_types[i].parse(op, desc);
> - goto out;
> - }
> - }
> -
> - ret = -EINVAL;
> -out:
> - free(name);
> -
> - return ret;
> -}
> -
> -static int __of_find_path(struct device_node *node, const char *propname, char **outpath, unsigned flags)
> +static int __of_find_path(struct device_node *node, const char *partname, char **outpath, unsigned flags)
> {
> - struct of_path op = {};
> - const char *str;
> + struct device_d *dev;
> + struct cdev *cdev;
> bool add_bb = false;
> - int i, ret;
>
> - op.dev = of_find_device_by_node_path(node->full_name);
> - if (!op.dev) {
> - op.dev = of_find_device_by_node_path(node->parent->full_name);
> - if (!op.dev)
> + dev = of_find_device_by_node_path(node->full_name);
> + if (!dev) {
> + dev = of_find_device_by_node_path(node->parent->full_name);
> + if (!dev)
> return -ENODEV;
> }
>
> - device_detect(op.dev);
> -
> - op.cdev = cdev_by_device_node(node);
> + device_detect(dev);
>
> - i = 1;
> + if (partname)
> + cdev = device_find_partition(dev, partname);
> + else
> + cdev = cdev_by_device_node(node);
>
> - while (propname) {
> - ret = of_property_read_string_index(node, propname, i++, &str);
> - if (ret)
> - break;
> -
> - ret = of_path_parse_one(&op, str);
> - if (ret)
> - return ret;
> - }
> -
> - if (!op.cdev)
> + if (!cdev)
> return -ENOENT;
>
> - if ((flags & OF_FIND_PATH_FLAGS_BB) && op.cdev->mtd &&
> - mtd_can_have_bb(op.cdev->mtd))
> + if ((flags & OF_FIND_PATH_FLAGS_BB) && cdev->mtd &&
> + mtd_can_have_bb(cdev->mtd))
> add_bb = true;
>
> - *outpath = asprintf("/dev/%s%s", op.cdev->name, add_bb ? ".bb" : "");
> + *outpath = asprintf("/dev/%s%s", cdev->name, add_bb ? ".bb" : "");
>
> return 0;
> }
> @@ -193,6 +114,8 @@ int of_find_path(struct device_node *node, const char *propname, char **outpath,
> {
> struct device_node *rnode;
> const char *path;
> + const char *partname = NULL;
> + char partnamestr[] = "partname:";
>
> path = of_get_property(node, propname, NULL);
> if (!path)
> @@ -202,5 +125,15 @@ int of_find_path(struct device_node *node, const char *propname, char **outpath,
> if (!rnode)
> return -ENODEV;
>
> - return __of_find_path(rnode, propname, outpath, flags);
> + of_property_read_string_index(node, propname, 1, &partname);
> + if (partname) {
> + if (!strncmp(partname, partnamestr, sizeof(partnamestr) - 1)) {
> + partname += sizeof(partnamestr) - 1;
> + } else {
> + pr_err("Invalid device-path: %s\n", partname);
> + return -EINVAL;
> + }
> + }
> +
> + return __of_find_path(rnode, partname, outpath, flags);
> }
_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox
next prev parent reply other threads:[~2016-01-07 18:32 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-01-07 11:27 Sascha Hauer
2016-01-07 18:31 ` Trent Piepho [this message]
2016-01-08 7:34 ` Sascha Hauer
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=1452191516.4474.81.camel@rtred1test09.kymeta.local \
--to=tpiepho@kymetacorp.com \
--cc=barebox@lists.infradead.org \
--cc=s.hauer@pengutronix.de \
/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