Skip to contentSkip to conversation

fs/cpio: make initramfs init script survive 'console=' kernel argument

PATCH New Timo Ketola 3 participants 10 messages Sep 23, 2019 – Aug 20, 2020 1 message not in the archive

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 fs/cpio: make initramfs init script survive 'console=' kernel argument

When booting with 'console=<empty>' in the kernel command line (as e.g. U-Boot does with silent flags in effect), opening /dev/console fails. That is fatal in the /init script and kernel will panic. It is also needless, because the kernel tries to open it anyway (well, as long as we have console node in initramfs /dev; cpio.mk creates that alright).

---
 fs/cpio/init | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
fs/cpio/init +1 -3
index dbe09ac..72b2401 100755
@@ -1,7 +1,5 @@
 #!/bin/sh
 # devtmpfs does not get automounted for initramfs
 /bin/mount -t devtmpfs devtmpfs /dev
-exec 0</dev/console
-exec 1>/dev/console
-exec 2>/dev/console
+
 exec /sbin/init "$@"
Signature · 1 line
2.7.4
#1 lore raw

PATCH 1/1 fs/cpio: make initramfs init script survive 'console=' kernel argument

Quoted 7 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Timo" == Timo Ketola <[email protected]> writes:

When booting with 'console=<empty>' in the kernel command line (as e.g. U-Boot does with silent flags in effect), opening /dev/console fails. That is fatal in the /init script and kernel will panic. It is also needless, because the kernel tries to open it anyway (well, as long as we have console node in initramfs /dev; cpio.mk creates that alright).

Also when devtmpfs is used? We manually mount devtmpfs in the initramfs, so the kernel presumably cannot have opened /dev/console before executing /init, and we do not add a static device node for /dev/console.

Quoted 24 lines
---
 fs/cpio/init | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
fs/cpio/init +1 -3
index dbe09ac..72b2401 100755
@@ -1,7 +1,5 @@
 #!/bin/sh
 # devtmpfs does not get automounted for initramfs
 /bin/mount -t devtmpfs devtmpfs /dev
-exec 0</dev/console
-exec 1>/dev/console
-exec 2>/dev/console
+
 exec /sbin/init "$@"
Signature · 5 lines
2.7.4
_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot
Signature · 1 line
Bye, Peter Korsgaard
#2 lore raw

PATCH 1/1 fs/cpio: make initramfs init script survive 'console=' kernel argument

Quoted 8 lines · Peter Korsgaard
Quoted 7 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Timo" == Timo Ketola <[email protected]> writes:

When booting with 'console=<empty>' in the kernel command line (as e.g. U-Boot does with silent flags in effect), opening /dev/console fails. That is fatal in the /init script and kernel will panic. It is also needless, because the kernel tries to open it anyway (well, as long as we have console node in initramfs /dev; cpio.mk creates that alright).

Quoted 4 lines

Also when devtmpfs is used? We manually mount devtmpfs in the initramfs, so the kernel presumably cannot have opened /dev/console before executing /init, and we do not add a static device node for /dev/console.

But I thought BR does just that here:

https://git.busybox.net/buildroot/tree/fs/cpio/cpio.mk

line 24, 25

Signature · 1 line
Timo

A message that is not in the archive (sent off-list or not yet received): [email protected]

#4 lore raw

PATCH 1/1 fs/cpio: make initramfs init script survive 'console=' kernel argument

Quoted 16 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Timo" == Timo Ketola <[email protected]> writes:

Quoted 8 lines · Peter Korsgaard
Quoted 7 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Timo" == Timo Ketola <[email protected]> writes:

When booting with 'console=<empty>' in the kernel command line (as e.g. U-Boot does with silent flags in effect), opening /dev/console fails. That is fatal in the /init script and kernel will panic. It is also needless, because the kernel tries to open it anyway (well, as long as we have console node in initramfs /dev; cpio.mk creates that alright).

Quoted 4 lines

Also when devtmpfs is used? We manually mount devtmpfs in the initramfs, so the kernel presumably cannot have opened /dev/console before executing /init, and we do not add a static device node for /dev/console.

Quoted 5 lines

But I thought BR does just that here:

https://git.busybox.net/buildroot/tree/fs/cpio/cpio.mk

line 24, 25

Heh, indeed - I forgot about that logic. Then I guess it is good, unless something else in the boot sequence doesn't like a missing stdin/stdout/stderr.

Signature · 1 line
Bye, Peter Korsgaard
#5 lore raw

PATCH 1/1 fs/cpio: make initramfs init script survive 'console=' kernel argument

Peter,

Quoted 10 lines · Peter Korsgaard
Quoted 6 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Timo" == Timo Ketola <[email protected]> writes:

But I thought BR does just that here:

https://git.busybox.net/buildroot/tree/fs/cpio/cpio.mk

line 24, 25

Heh, indeed - I forgot about that logic. Then I guess it is good, unless something else in the boot sequence doesn't like a missing stdin/stdout/stderr.

I think BusyBox's init takes care about that. See my conversation with Arnout:

http://lists.busybox.net/pipermail/buildroot/2019-September/259889.html

And thanks for looking at this.

--

Timo

#6 lore raw

PATCH 1/1 fs/cpio: make initramfs init script survive 'console=' kernel argument

In reply to Timo Ketola

Hi there,

I tested this as per Arnout's suggestion by booting with 'console=' / 'console=ttymxc0,115200' and with / without initramfs. The result depends only on console setting:

# cat /proc/cmdline; ls -l /proc/1/fd
console=
total 0
lrwx------    1 root     root            64 Sep 24 05:01 0 -> /dev/null
lrwx------    1 root     root            64 Sep 24 05:01 1 -> /dev/null
lrwx------    1 root     root            64 Sep 24 05:01 2 -> /dev/null
# cat /proc/cmdline; ls -l /proc/1/fd
console=ttymxc0,115200
total 0
lrwx------    1 root     root            64 Sep 24 05:20 0 -> /dev/console
lrwx------    1 root     root            64 Sep 24 05:20 1 -> /dev/console
lrwx------    1 root     root            64 Sep 24 05:20 2 -> /dev/console

--

Timo

#8 lore raw

PATCH 1/1 fs/cpio: make initramfs init script survive 'console=' kernel argument

In reply to Timo Ketola

Timo, All,

On 2019-09-23 10:58 +0000, Timo Ketola spake thusly:

Quoted 7 lines

When booting with 'console=<empty>' in the kernel command line (as e.g. U-Boot does with silent flags in effect), opening /dev/console fails. That is fatal in the /init script and kernel will panic. It is also needless, because the kernel tries to open it anyway (well, as long as we have console node in initramfs /dev; cpio.mk creates that alright).

I've applied to master, after extending the commit log with all the gory details explaininng why we can indeed safely remove those redisrections now.

Thank you for the patience on this patch.

Thanks also to Peter for helping on IRC about this research.

Regards,
Yann E. MORIN.

Quoted 23 lines
---
 fs/cpio/init | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
fs/cpio/init +1 -3
index dbe09ac..72b2401 100755
@@ -1,7 +1,5 @@
 #!/bin/sh
 # devtmpfs does not get automounted for initramfs
 /bin/mount -t devtmpfs devtmpfs /dev
-exec 0</dev/console
-exec 1>/dev/console
-exec 2>/dev/console
+
 exec /sbin/init "$@"
Signature · 5 lines
2.7.4
_______________________________________________
buildroot mailing list
buildroot at busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot
Signature · 6 lines
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 561 099 427 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'
#9 lore raw

PATCH 1/1 fs/cpio: make initramfs init script survive 'console=' kernel argument

Quoted 12 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Yann" == Yann E MORIN <[email protected]> writes:

Timo, All,
On 2019-09-23 10:58 +0000, Timo Ketola spake thusly:

Quoted 7 lines

When booting with 'console=<empty>' in the kernel command line (as e.g. U-Boot does with silent flags in effect), opening /dev/console fails. That is fatal in the /init script and kernel will panic. It is also needless, because the kernel tries to open it anyway (well, as long as we have console node in initramfs /dev; cpio.mk creates that alright).

Quoted 7 lines

I've applied to master, after extending the commit log with all the gory details explaininng why we can indeed safely remove those redisrections now.

Thank you for the patience on this patch.

Thanks also to Peter for helping on IRC about this research.

Committed to 2020.02.x, thanks.

Signature · 1 line
Bye, Peter Korsgaard
#10 lore raw

PATCH 1/1 fs/cpio: make initramfs init script survive 'console=' kernel argument

In reply to Yann E. MORIN
Quoted 12 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Yann" == Yann E MORIN <[email protected]> writes:

Timo, All,
On 2019-09-23 10:58 +0000, Timo Ketola spake thusly:

Quoted 7 lines

When booting with 'console=<empty>' in the kernel command line (as e.g. U-Boot does with silent flags in effect), opening /dev/console fails. That is fatal in the /init script and kernel will panic. It is also needless, because the kernel tries to open it anyway (well, as long as we have console node in initramfs /dev; cpio.mk creates that alright).

Quoted 7 lines

I've applied to master, after extending the commit log with all the gory details explaininng why we can indeed safely remove those redisrections now.

Thank you for the patience on this patch.

Thanks also to Peter for helping on IRC about this research.

It unfortunately causes a regression :/

I have a setup booting from an initramfs, using devtmpfs, busybox init and glibc. One of the initscripts used ttyname_r(3) (through the busybox tty applet), and with this change ttyname_3() fails rather than returning /dev/console.

Looking at the glibc code, it does a fstat on fd(0), and then compares st_ino / st_dev (and optionally st_rdev) against all the entries under /dev.

As the kernel does not mount devtmpfs for us, fd 0 refers to the /dev/console entry in the rootfs (cpio), but after devtmpfs is mounted, /dev/console is the entry in the devtmpfs so st_dev / st_ino will no longer match :/

This logic seems to be added to support containers. See the following glibc commit:

commit 15e9a4f378c8607c2ae1aa465436af4321db0e23

Author: Christian Brauner <[email protected]>
Date:   Fri Jan 27 15:59:59 2017 +0100

linux ttyname and ttyname_r: do not return wrong results

If a link (say /proc/self/fd/0) pointing to a device, say /dev/pts/2, in a parent mount namespace is passed to ttyname, and a /dev/pts/2 exists (in a different devpts) in the current namespace, then it returns /dev/pts/2. But /dev/pts/2 is NOT the current tty, it is a different file and device.

Detect this case and return ENODEV. Userspace can choose to take this as a hint that the fd points to a tty device but to act on the fd rather than the link.

Notice: Once logged in, tty does work correctly. Presumably login reopens stdin/stdout/stderr.

Any great ideas about how we can fix this? I would prefer to not add a lot of logic in /init, so preferably not having to mount /proc and so on. If we don't have a good solution, what then? Do we revert or keep it like this?

Signature · 1 line
Bye, Peter Korsgaard