From: anis chali <chalianis1@gmail.com>
To: Marco Felsch <m.felsch@pengutronix.de>
Cc: BAREBOX <barebox@lists.infradead.org>,
Ahmad Fatoum <a.fatoum@pengutronix.de>
Subject: Re: [PATCH] blspec: derive root prefix from blspec config path instead of mountpoint
Date: Fri, 14 Nov 2025 11:49:55 -0500 [thread overview]
Message-ID: <CAL+1fyD1DN+6TLmJKKNaMHeLccTXH_VENRQvEm9LMLrDhLzw0A@mail.gmail.com> (raw)
In-Reply-To: <CAL+1fyCeZdxMhpDNYjpvbteGre+mgDi0pRyUn_24wUkKJa_+4w@mail.gmail.com>
Hi Marco,
As Ahmad mentioned this will serve to boot blspec files directly, a
practical use case is when using ostree without separating the /boot
from the rest of rootfs in one partition, in that case the loader is
under /boot. And inside the same entry directory they put two configs.
This force the user to use the full config path to boot specific
system. Without this patch the boot fails because the path to the
kernel is broken because it puts only the prefix with the mount point
and not mount point/boot/....
Thanks.
Anis C.
Le ven. 14 nov. 2025 à 10:58, anis chali <chalianis1@gmail.com> a écrit :
>
> Hi Marco,
>
> As Ahmad mentioned this will serve to boot blspec files directly, a practical use case is when using ostree without separating the /boot from the rest of rootfs in one partition, in that case the loader is under /boot. And inside the same entry directory they put two configs. This force the user to use the full config path to boot specific system. Without this patch the boot fails because the path to the kernel is broken because it puts only the prefix with the mount point and not mount point/boot/....
>
>
> Thanks.
>
> Anis C.
>
> Le ven. 14 nov. 2025, 08:51, Ahmad Fatoum <a.fatoum@pengutronix.de> a écrit :
>>
>> Hello Marco,
>>
>> On 11/14/25 11:11 AM, Marco Felsch wrote:
>> > Hi,
>> >
>> > thanks for your patch.
>> >
>> > On 25-11-14, chalianis1@gmail.com wrote:
>> >> From: Chali Anis <chalianis1@gmail.com>
>> >>
>> >> Add get_prefix_path() to compute the filesystem prefix of a BLSpec
>> >> entry by locating the deepest /loader/entries/ component in the
>> >
>> > What exactly do you mean by "deepest"? The specification says that the
>> > entries should be under /loader/entries of $BOOT or $EFI partition [1].
>> >
>> > There is also a note:
>> > """""""""
>> > In all cases the /loader/entries/ directory should be located directly
>> > in the root of the ESP or XBOOTLDR filesystem. Specifically, the
>> > /loader/entries/ directory should not be located under the /EFI/
>> > subdirectory on the ESP.
>> > """""""""
>> >
>> > So searching in subdirs seems not to be spec compatible. May I ask why
>> > your loader entries are within some subdir in the first place?
>>
>> I was also confused when Anis first mentioned this on IRC, but the
>> missing piece of information was that get_mounted_path() (and now
>> get_blspec_prefix_path()) is only used when booting a specific
>> bootloader spec config by name, e.g.,
>>
>> bootm /mnt/.../loader/entries/my.conf
>>
>> as opposed to:
>>
>> bootm /mnt/.../
>>
>>
>> In that case, it makes sense to not go up to the first mount, but just
>> take the loader/entries the file is inside.
>>
>> I believe this can also be useful for testing: include a loader/
>> directory in the barebox environment and then run tests over it directly
>> without having to copy /env/loader to /loader first, because that's the
>> file system root.
>>
>> Cheers,
>> Ahmad
>>
>> >
>> > Regards,
>> > Marco
>> >
>> > [1] https://uapi-group.org/specifications/specs/boot_loader_specification/#type-1-boot-loader-specification-entries
>> >
>> >> configuration path. This ensures correct rootpath extraction for
>> >> nested loader directories, mixed path layouts, duplicate segments,
>> >> and paths without valid loader entries. If no prefix is found,
>> >> fall back to get_mounted_path().
>> >>
>> >> Signed-off-by: Chali Anis <chalianis1@gmail.com>
>> >> ---
>> >> common/blspec.c | 25 ++++++++++++++++++++++++-
>> >> 1 file changed, 24 insertions(+), 1 deletion(-)
>> >>
>> >> diff --git a/common/blspec.c b/common/blspec.c
>> >> index 80ebe256ceaf..ad94d4d6be64 100644
>> >> --- a/common/blspec.c
>> >> +++ b/common/blspec.c
>> >> @@ -391,6 +391,29 @@ static bool entry_is_match_machine_id(struct blspec_entry *entry)
>> >> return ret;
>> >> }
>> >>
>> >> +/*
>> >> + * Return the path prefix before the deepest "/loader/entries/" component
>> >> + * in @configname. Falls back to the mounted path if none is found.
>> >> + */
>> >> +static char *get_blspec_prefix_path(char *configname)
>> >> +{
>> >> + char *p;
>> >> + char *scfg = configname;
>> >> + char *save = NULL;
>> >> +
>> >> + while ((p = strstr(scfg, "/loader/entries/")) != NULL) {
>> >> + if (p - configname > 0)
>> >> + save = p;
>> >> +
>> >> + scfg = p + 1;
>> >> + }
>> >> +
>> >> + if (save && save - configname > 0)
>> >> + return xstrndup(configname, save - configname);
>> >> +
>> >> + return get_mounted_path(configname);
>> >> +}
>> >> +
>> >> static int __blspec_scan_file(struct bootentries *bootentries, const char *root,
>> >> const char *configname)
>> >> {
>> >> @@ -404,7 +427,7 @@ static int __blspec_scan_file(struct bootentries *bootentries, const char *root,
>> >> if (IS_ERR(entry))
>> >> return PTR_ERR(entry);
>> >>
>> >> - root = root ?: get_mounted_path(configname);
>> >> + root = root ?: get_blspec_prefix_path(configname);
>> >> entry->rootpath = xstrdup_const(root);
>> >> entry->configpath = xstrdup_const(configname);
>> >> entry->cdev = get_cdev_by_mountpath(root);
>> >> --
>> >> 2.49.0
>> >>
>> >>
>> >>
>> >
>>
>> --
>> Pengutronix e.K. | |
>> Steuerwalder Str. 21 | http://www.pengutronix.de/ |
>> 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
>> Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
>>
prev parent reply other threads:[~2025-11-14 16:50 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-14 5:00 chalianis1
2025-11-14 10:11 ` Marco Felsch
2025-11-14 13:51 ` Ahmad Fatoum
[not found] ` <CAL+1fyCeZdxMhpDNYjpvbteGre+mgDi0pRyUn_24wUkKJa_+4w@mail.gmail.com>
2025-11-14 16:49 ` anis chali [this message]
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=CAL+1fyD1DN+6TLmJKKNaMHeLccTXH_VENRQvEm9LMLrDhLzw0A@mail.gmail.com \
--to=chalianis1@gmail.com \
--cc=a.fatoum@pengutronix.de \
--cc=barebox@lists.infradead.org \
--cc=m.felsch@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