From: Simon Horman <horms@verge.net.au>
To: Andreas Fenkart <afenkart@gmail.com>
Cc: Andreas Fenkart <andreas.fenkart@dev.digitalstrom.org>,
kexec@lists.infradead.org
Subject: Re: [PATCH] kexec: determine size of block device
Date: Tue, 20 Oct 2015 10:27:14 +0900 [thread overview]
Message-ID: <20151020012714.GE11217@verge.net.au> (raw)
In-Reply-To: <CALtMJEDut5=vr6pDtEyuAK0yW09CX60KSfN8Ug7ZiPk5sEx0Sg@mail.gmail.com>
On Mon, Oct 19, 2015 at 11:32:58AM +0200, Andreas Fenkart wrote:
> Hi Simon,
>
> 2015-10-16 3:30 GMT+02:00 Simon Horman <horms@verge.net.au>:
> > On Fri, Oct 09, 2015 at 03:27:18PM +0200, Andreas Fenkart wrote:
> >> starting 'kexec -l /dev/mmcblk0p1' fails since the size of
> >> a block device can not be determined with stat
> >
> > Could you provide some motivation for kexec reading a kernel image
> > from a block device as per the example above?
>
> We plan to use it for our embedded system
>
> We have an interim initramfs, that selects the kernel we want to boot
> our real system. Something like a n+1 stage boot loader,
> where n is the last stage of u-boot. The motivation is to do a safe
> kernel upgrade on the production rootfs through package management.
> The new kernel is dropped into a folder where the interim system will
> selects the kernel with the highest priority. No reconfiguration of
> u-boot/rescue system needed. In case no kernel has yet been installed,
> we fall back to the factory default, which is installed in a raw flash
> partition on eMMC without filesystem.
>
> https://git.digitalstrom.org/bsp/dss-rescue-tools/blob/master/src/kernel-fallback-dss11-1gb-t1.config
Nice :)
> >> Signed-off-by: Andreas Fenkart <andreas.fenkart@dev.digitalstrom.org>
> >> ---
> >> kexec/kexec.c | 10 ++++++++++
> >> 1 file changed, 10 insertions(+)
> >>
> >> diff --git a/kexec/kexec.c b/kexec/kexec.c
> >> index ff024f3..8b8846e 100644
> >> --- a/kexec/kexec.c
> >> +++ b/kexec/kexec.c
> >> @@ -26,7 +26,9 @@
> >> #include <stdlib.h>
> >> #include <errno.h>
> >> #include <limits.h>
> >> +#include <sys/ioctl.h>
> >> #include <sys/mman.h>
> >> +#include <sys/mount.h>
> >> #include <sys/types.h>
> >> #include <sys/stat.h>
> >> #include <sys/reboot.h>
> >> @@ -554,6 +556,14 @@ char *slurp_file(const char *filename, off_t *r_size)
> >> die("Can not seek to the begin of file %s: %s\n",
> >> filename, strerror(errno));
> >> buf = slurp_fd(fd, filename, size, &nread);
> >> + } else if (S_ISBLK(stats.st_mode)) {
> >> + /* taken from blockdev */
> >
> > Is "blockdev" Linux kernel code?
>
> It is a util-linux tool, which is GPL of course. I added the comment
> as a reference, since I couldn't find a man page for BLKGETSIZE64.
> Since it's a straight forward ioctl call, there is probably no
> copyright violation at all, and I could well remove it.
Understood. Could you repost with the blockdev comment removed?
> >> + unsigned long long llu = -1;
> >> + err = ioctl(fd, BLKGETSIZE64, &llu);
> >> + if (err < 0)
> >> + die("Can't retrieve size of block device %s: %s\n",
> >> + filename, strerror(errno));
> >> + size = llu;
> >> } else {
> >> size = nread = stats.st_size;
> >> buf = mmap(NULL, size,
> >> --
> >> 2.5.1
> >>
> >>
> >> _______________________________________________
> >> kexec mailing list
> >> kexec@lists.infradead.org
> >> http://lists.infradead.org/mailman/listinfo/kexec
> >>
>
_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec
prev parent reply other threads:[~2015-10-20 1:27 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-09 13:27 [PATCH] kexec: determine size of block device Andreas Fenkart
2015-10-16 1:30 ` Simon Horman
2015-10-19 9:32 ` Andreas Fenkart
2015-10-20 1:27 ` Simon Horman [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=20151020012714.GE11217@verge.net.au \
--to=horms@verge.net.au \
--cc=afenkart@gmail.com \
--cc=andreas.fenkart@dev.digitalstrom.org \
--cc=kexec@lists.infradead.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