systemd: enable getty generator by default
RFC PATCH 1/3 systemd: enable getty generator by default
Until recently, even when the getty generator was disabled in the systemd recipe it was actually still active. This was because the old behaviour was to delete the serial-getty template unit if the generator was disabled, but the systemd-serialgetty package shipped then shipped the same files so the generator continued to run. This was a bug in the original commit[1] so this behaviour has been present since 2016.
My recent fixes[2] changed this: if the getty generator was disabled then the generator itself is deleted. This makes the actual behaviour match the intention, but the consequence was to demonstrate that some modern platforms were relying on this unexpected behaviour: specifically the genericarm64 BSP which intends to support a number of virtual and physical boards with a number of serial console ports that are not really suitable to be hardcoded into SERIAL_CONSOLES:
- ttyS0
- ttyAMA0 (AMBA PL011 uart)
- ttyS2 (BeagleBone Play, S0 and S1 are internal)
- hvc0 (KVM)
- ttyPS1 (AMD KV260)
- And most likely more
Restore the existing behaviour by explicitly enabling the serial getty
generator: this means that systemd will automatically bring up a getty on the first serial console it finds.
In the future we should extend some level of dynamic console-finding to sysvinit-based systems by searching for a console device in inittab, but for now this reverts the unintentional regression.
[1] oe-core 2a8d0df47c9 ("systemd: make systemd-serialgetty optional")
[2] oe-core 2beb3170af6 ("systemd: if getty generator is disabled remove the generator, not the units")
--- meta/recipes-core/systemd/systemd_257.4.bb | 1 + 1 file changed, 1 insertion(+)
meta/recipes-core/systemd/systemd_257.4.bb +1 -0
@@ -92,6 +92,7 @@ PACKAGECONFIG ??= " \ quotacheck \ randomseed \ resolved \ + serial-getty-generator \ set-time-epoch \ sysusers \ timedated \
Signature · 1 line
2.43.0
RFC PATCH 2/3 systemd: always depend on the explicit serial console units
The systemd-serialgetty recipe generates explicit units for consoles that are defined in SERIAL_CONSOLES, and if that variable is not defined then just produces an empty package.
Even when systemd has been configured to use the getty generator, if there are explicit consoles defined then we should respect them. Don't conditionalise the dependency on systemd-serialgetty so that we always pull in the explicit consoles.
--- meta/recipes-core/systemd/systemd_257.4.bb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
meta/recipes-core/systemd/systemd_257.4.bb +1 -1
@@ -754,7 +754,7 @@ FILES:${PN} = " ${base_bindir}/* \ FILES:${PN}-dev += "${base_libdir}/security/*.la ${datadir}/dbus-1/interfaces/ ${sysconfdir}/rpm/macros.systemd" RDEPENDS:${PN} += "kmod dbus util-linux-mount util-linux-umount udev (= ${EXTENDPKGV}) systemd-udev-rules util-linux-agetty util-linux-fsck util-linux-swaponoff util-linux-mkswap" -RDEPENDS:${PN} += "${@bb.utils.contains('PACKAGECONFIG', 'serial-getty-generator', '', 'systemd-serialgetty', d)}" +RDEPENDS:${PN} += "systemd-serialgetty" RDEPENDS:${PN} += "volatile-binds" RRECOMMENDS:${PN} += "${PN}-extra-utils \
Signature · 1 line
2.43.0
Re: [PATCH RFC walnascar 2/3] systemd: always depend on the explicit serial console units
Hi,
Quoted 8 lines · Ross Burton
The systemd-serialgetty recipe generates explicit units for consoles that are defined in
SERIAL_CONSOLES, and if that variable is not defined then just produces an empty package.Even when systemd has been configured to use the getty generator, if there are explicit consoles defined then we should respect them. Don't conditionalise the dependency on systemd-serialgetty so that we always pull in the explicit consoles.
So all boots of genericarm64 on all HW now wait for the configured SERIAL_CONSOLES timeouts.
How should that be fixed? Unsetting SERIAL_CONSOLES in genericarm64 machine config? So systemd-serialgetty is allowed to be machine specific but not systemd itself?
Cheers,
-Mikko
Re: [PATCH RFC walnascar 2/3] systemd: always depend on the explicit serial console units
Quoted 13 lines · Mikko Rapeli
Hi,
Quoted 8 lines · Ross Burton
The systemd-serialgetty recipe generates explicit units for consoles that are defined in
SERIAL_CONSOLES, and if that variable is not defined then just produces an empty package.Even when systemd has been configured to use the getty generator, if there are explicit consoles defined then we should respect them. Don't conditionalise the dependency on systemd-serialgetty so that we always pull in the explicit consoles.
Quoted 5 lines
So all boots of genericarm64 on all HW now wait for the configured
SERIAL_CONSOLEStimeouts.How should that be fixed? Unsetting
SERIAL_CONSOLESin genericarm64 machine config? So systemd-serialgetty is allowed to be machine specific but not systemd itself?
Yes, systemd-serialgetty is explicitly machine-specific because it uses SERIAL_CONSOLES. When we have support for sysvinit to probe a serial console, we can just set SERIALCONSOLES=“” in the machine and rely on probing.
Ross
A message that is not in the archive (sent off-list or not yet received): [email protected]
Re: [OE-core] [PATCH RFC walnascar 2/3] systemd: always depend on the explicit serial console units
Hi,
Quoted 10 lines · Mikko Rapeli via lists.openembedded.org
Quoted 8 lines · Ross Burton
The systemd-serialgetty recipe generates explicit units for consoles that are defined in
SERIAL_CONSOLES, and if that variable is not defined then just produces an empty package.Even when systemd has been configured to use the getty generator, if there are explicit consoles defined then we should respect them. Don't conditionalise the dependency on systemd-serialgetty so that we always pull in the explicit consoles.
Quoted 5 lines
So all boots of genericarm64 on all HW now wait for the configured
SERIAL_CONSOLEStimeouts.How should that be fixed? Unsetting
SERIAL_CONSOLESin genericarm64 machine config? So systemd-serialgetty is allowed to be machine specific but not systemd itself?
To fix the "poky-altcfg" distro or "poky" with systemd boot slowdown on genericarm64, systemd-serialgetty should either be removed or see an empty SERIAL_CONSOLES set currently in the machine config.
A genericarm64 machine specific bbappend could do this, or in genericarm64 machine config could have:
SERIAL_CONSOLES:poky-altcfg ?= ""
This sets for all which may break some other things so maybe it's better to do this only for systemd-serialgetty in machine config with:
SERIAL_CONSOLES:poky-altcfg:pn-systemd-serialgetty ?= ""
This would not work for plain "poky" distro if someone enabled systemd as init. For that, something like:
SERIAL_CONSOLES:pn-systemd-serialgetty ?= "
${@bb.utils.contains('DISTRO_FEATURES', 'systemd', \
'', '115200;ttyAMA0 115200;hvc0 115200;ttyS0 115200;ttyS1 115200;ttyS2', d)} \"
These either make SERIAL_CONSOLES depend on "systemd" in DISTRO_FEATURES inside machine config, so I'm not sure if this is allowed or not.
A genericarm64 machine specific bbappend to systemd-serialgetty to disable the package or clear out SERIAL_CONSOLES is "systemd" in DISTRO_FEATURES would be better then?
systemd would remain machine agnostic but systemd-serialgetty would be machine specific since it takes SERIAL_CONSOLES as input anyway.
I can't decide between these. :/
Cheers,
-Mikko
RFC PATCH 3/3 genericarm64: add ttyPS1 for KV260
In reply to Ross BurtonThe AMD KV260 has a serial console on ttyPS1, so until we can dynamically detect the console under sysvinit we can add it to SERIAL_CONSOLES so this platform has a working console when not running systemd.
--- meta-yocto-bsp/conf/machine/genericarm64.conf | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
meta-yocto-bsp/conf/machine/genericarm64.conf +1 -1
@@ -28,7 +28,7 @@ WKS_FILE ?= "genericarm64.wks.in" EFI_PROVIDER ?= "${@bb.utils.contains("DISTRO_FEATURES", "systemd", "systemd-boot", "grub-efi", d)}" # Try to bring up a selection of physical or virtual serial consoles -SERIAL_CONSOLES ?= "115200;ttyAMA0 115200;hvc0 115200;ttyS0 115200;ttyS1 115200;ttyS2" +SERIAL_CONSOLES ?= "115200;ttyAMA0 115200;hvc0 115200;ttyS0 115200;ttyS1 115200;ttyS2 115200;ttyPS1" # Allow u-boot to be built for use with qemu-system-aarch64. # This u-boot is _not_ suitable for use with real hardware, and the expectation
Signature · 1 line
2.43.0
Re: [PATCH RFC walnascar 3/3] genericarm64: add ttyPS1 for KV260
Hi,
Quoted 3 lines · Ross Burton
The AMD KV260 has a serial console on ttyPS1, so until we can dynamically detect the console under sysvinit we can add it to
SERIAL_CONSOLESso this platform has a working console when not running systemd.
So full circle back to
https://lists.yoctoproject.org/g/poky/topic/109273447#msg13461 then :)
To me this change is fine but it's also ok to not apply this and focus on "poky-altcfg" distro with systemd which is much better at detecting HW support at runtime.
NB, should this have been sent to poky list?
Cheers,
-Mikko
Quoted 21 lines
Signed-off-by Ross Burton <[email protected]>--- meta-yocto-bsp/conf/machine/genericarm64.conf | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)meta-yocto-bsp/conf/machine/genericarm64.conf +1
-1@@ -28,7 +28,7 @@ WKS_FILE ?= "genericarm64.wks.in" EFI_PROVIDER ?= "${@bb.utils.contains("DISTRO_FEATURES", "systemd", "systemd-boot", "grub-efi", d)}" # Try to bring up a selection of physical or virtual serial consoles -SERIAL_CONSOLES ?= "115200;ttyAMA0 115200;hvc0 115200;ttyS0 115200;ttyS1 115200;ttyS2" +SERIAL_CONSOLES ?= "115200;ttyAMA0 115200;hvc0 115200;ttyS0 115200;ttyS1 115200;ttyS2 115200;ttyPS1" # Allow u-boot to be built for use with qemu-system-aarch64. # This u-boot is _not_ suitable for use with real hardware, and the expectationSignature · 1 line
2.43.0
Re: [PATCH RFC walnascar 3/3] genericarm64: add ttyPS1 for KV260
Quoted 10 lines · Mikko Rapeli
Hi,
Quoted 3 lines · Ross Burton
The AMD KV260 has a serial console on ttyPS1, so until we can dynamically detect the console under sysvinit we can add it to
SERIAL_CONSOLESso this platform has a working console when not running systemd.So full circle back to
https://lists.yoctoproject.org/g/poky/topic/109273447#msg13461 then :)
Pretty much. It’s horrible but…
Quoted 3 lines
To me this change is fine but it's also ok to not apply this and focus on "poky-altcfg" distro with systemd which is much better at detecting HW support at runtime.
Agreed, but also sysv shouldn’t be inoperable out of the box where possible.
Quoted 1 line
NB, should this have been sent to poky list?
Yes…
Ross
With this patchset, the effect of serial-getty-generator PACKAGECONFIG is to control whether to remove systemd-getty-generator or not. But is there any case we want to remove this generator? If not, I think we should just remove this PACKAGECONFIG.
Regards,
Qi
Quoted 60 lines · Ross Burton via lists.openembedded.org
Until recently, even when the getty generator was disabled in the systemd recipe it was actually still active. This was because the old behaviour was to delete the serial-getty template unit if the generator was disabled, but the systemd-serialgetty package shipped then shipped the same files so the generator continued to run. This was a bug in the original commit[1] so this behaviour has been present since 2016.
My recent fixes[2] changed this: if the getty generator was disabled then the generator itself is deleted. This makes the actual behaviour match the intention, but the consequence was to demonstrate that some modern platforms were relying on this unexpected behaviour: specifically the genericarm64 BSP which intends to support a number of virtual and physical boards with a number of serial console ports that are not really suitable to be hardcoded into
SERIAL_CONSOLES:- ttyS0
- ttyAMA0 (AMBA PL011 uart)
- ttyS2 (BeagleBone Play, S0 and S1 are internal)
- hvc0 (KVM)
- ttyPS1 (AMD KV260)
- And most likely moreRestore the existing behaviour by explicitly enabling the serial getty
generator: this means that systemd will automatically bring up a getty on the first serial console it finds.In the future we should extend some level of dynamic console-finding to sysvinit-based systems by searching for a console device in inittab, but for now this reverts the unintentional regression.
[1] oe-core
2a8d0df47c9("systemd: make systemd-serialgetty optional")
[2] oe-core2beb3170af6("systemd: if getty generator is disabled remove the generator, not the units")Signed-off-by Ross Burton <[email protected]>--- meta/recipes-core/systemd/systemd_257.4.bb | 1 + 1 file changed, 1 insertion(+)meta/recipes-core/systemd/systemd_257.4.bb +1
-1@@ -92,6 +92,7 @@ PACKAGECONFIG ??= " \ quotacheck \ randomseed \ resolved \ + serial-getty-generator \ set-time-epoch \ sysusers \ timedated \ -=-=-=-=-=-=-=-=-=-=-=-Links: You receive all messages sent to this group.
View/Reply Online (#215242): https://lists.openembedded.org/g/openembedded-core/message/215242
Mute This Topic: https://lists.openembedded.org/mt/112401496/7304865
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-
Hi Ross,
Quoted 22 lines · Ross Burton
Until recently, even when the getty generator was disabled in the systemd recipe it was actually still active. This was because the old behaviour was to delete the serial-getty template unit if the generator was disabled, but the systemd-serialgetty package shipped then shipped the same files so the generator continued to run. This was a bug in the original commit[1] so this behaviour has been present since 2016.
My recent fixes[2] changed this: if the getty generator was disabled then the generator itself is deleted. This makes the actual behaviour match the intention, but the consequence was to demonstrate that some modern platforms were relying on this unexpected behaviour: specifically the genericarm64 BSP which intends to support a number of virtual and physical boards with a number of serial console ports that are not really suitable to be hardcoded into
SERIAL_CONSOLES:- ttyS0
- ttyAMA0 (AMBA PL011 uart)
- ttyS2 (BeagleBone Play, S0 and S1 are internal)
- hvc0 (KVM)
- ttyPS1 (AMD KV260)
- And most likely more
Quoted 3 lines
Restore the existing behaviour by explicitly enabling the serial getty
generator: this means that systemd will automatically bring up a getty on the first serial console it finds.
Is there a need to be backwards compatible this much? Many things may break or get refactored on master branch between releases and this is just one of them.
Then with these patches there still is the bug/issue seen in previous state too where systemd tries to start agetty on all SERIAL_CONSOLES defined at build time which slows down boot to "running" or "degraded" state by more than one minute. This happens also on qemu where core-image-base (I've added systemd-analyze with dependencies there to measure these):
root@genericarm64:~# journalctl -b -a|grep tty | tail -15 Apr 23 09:39:31 genericarm64 systemd[1]: dev-ttyS1.device: Job dev-ttyS1.device/start timed out. Apr 23 09:39:31 genericarm64 systemd[1]: Timed out waiting for device /dev/ttyS1. Apr 23 09:39:31 genericarm64 systemd[1]: Dependency failed for Serial Getty on ttyS1. Apr 23 09:39:31 genericarm64 systemd[1]: [email protected]: Job [email protected]/start failed with result 'dependency'. Apr 23 09:39:31 genericarm64 systemd[1]: dev-ttyS1.device: Job dev-ttyS1.device/start failed with result 'timeout'. Apr 23 09:39:31 genericarm64 systemd[1]: dev-ttyS0.device: Job dev-ttyS0.device/start timed out. Apr 23 09:39:31 genericarm64 systemd[1]: Timed out waiting for device /dev/ttyS0. Apr 23 09:39:31 genericarm64 systemd[1]: Dependency failed for Serial Getty on ttyS0. Apr 23 09:39:31 genericarm64 systemd[1]: [email protected]: Job [email protected]/start failed with result 'dependency'. Apr 23 09:39:31 genericarm64 systemd[1]: dev-ttyS0.device: Job dev-ttyS0.device/start failed with result 'timeout'. Apr 23 09:39:31 genericarm64 systemd[1]: dev-ttyS2.device: Job dev-ttyS2.device/start timed out. Apr 23 09:39:31 genericarm64 systemd[1]: Timed out waiting for device /dev/ttyS2. Apr 23 09:39:31 genericarm64 systemd[1]: Dependency failed for Serial Getty on ttyS2. Apr 23 09:39:31 genericarm64 systemd[1]: [email protected]: Job [email protected]/start failed with result 'dependency'. Apr 23 09:39:31 genericarm64 systemd[1]: dev-ttyS2.device: Job dev-ttyS2.device/start failed with result 'timeout'.
root@genericarm64:~# systemd-analyze Startup finished in 2.405s (firmware) + 5.109s (loader) + 10.315s (kernel) + 1min 34.041s (userspace) = 1min 51.872s multi-user.target reached after 1min 34.035s in userspace.
Thus I think this breaking change is fine for master branch and upcoming release and that anyone expecting different, more robust behavior should just use the systemd agetty generator.
For genericarm64 I think https://lists.yoctoproject.org/g/poky/message/13592
is the way to go to fix both completely missing agetty
https://bugzilla.yoctoproject.org/show_bug.cgi?id=15830
and the slow boot time. Both this series and
https://lists.yoctoproject.org/g/poky/message/13592 applied does not seem to work since SERIAL_CONSOLES is now used all the time, so delayed boot issue remains unfixed.
The boot time is important because testing on target HW should not start before system is a stable state which for systemd is "running" when all services succeed or "degraded" when something went wrong. Spending one minute or more, doubling the boot time, due to tty detection timeouts is not nice when udev has already scanned all buses and loaded needed drivers (multiple times, once in initrd and once in main rootfs).
For manually detecting this without systemd-analyze, login to target over serial console once prompt appears and see that "systemctl status" shows "starting" status for a long time after everything already calmed down, and "journalctl -b -a" eventually shows the tty timeout messages. The issue happens on qemu and on real target HW like AMD KV260.
Cheers,
-Mikko
Quoted 6 lines · Mikko Rapeli
Quoted 3 lines
Restore the existing behaviour by explicitly enabling the serial getty
generator: this means that systemd will automatically bring up a getty on the first serial console it finds.Is there a need to be backwards compatible this much? Many things may break or get refactored on master branch between releases and this is just one of them.
Yes, because this is the behaviour that both fixes the KV260 issue and also doesn’t break all of the selftests due to how testimage+qemu interact.
Quoted 5 lines
Then with these patches there still is the bug/issue seen in previous state too where systemd tries to start agetty on all
SERIAL_CONSOLESdefined at build time which slows down boot to "running" or "degraded" state by more than one minute. This happens also on qemu where core-image-base (I've added systemd-analyze with dependencies there to measure these):
Yes. As I said, the proper solution here is to actually probe the setup. Once we’ve got walnascar out of the way I’ll finish my branch that does the same system inspection for sysvinit so we don’t need to hardcode SERIAL_CONSOLES, instead it will bring up a console on the first tty it can find. That is likely to cause problems with testimage (as it hooks up a second tty for failures) but will be isolated to genericarm64 and not every qemu machine.
Quoted 4 lines
Thus I think this breaking change is fine for master branch and upcoming release and that anyone expecting different, more robust behavior should just use the systemd agetty generator.
Absolutely.
Quoted 7 lines
For genericarm64 I think https://lists.yoctoproject.org/g/poky/message/13592
is the way to go to fix both completely missing agetty
https://bugzilla.yoctoproject.org/show_bug.cgi?id=15830
and the slow boot time. Both this series and
https://lists.yoctoproject.org/g/poky/message/13592 applied does not seem to work sinceSERIAL_CONSOLESis now used all the time, so delayed boot issue remains unfixed.
Making systemd machine-specific would be disastrous for rebuilds as it is the provider of libudev and libsystemd, so would mean that _every_ recipe that uses those is also now machine-specific.
As was discussed in the Tuesday tech call last week, I suspect a better solution would be to have a machine-specific configuration recipe that can drop in configuration files, mask or enable units, etc. This way specific machines can configure systemd in specific ways, without actually touching the core recipe.
Alternatively, splitting systemd into ‘userspace’ and ’system’ recipes where the former is tune-specific and contains the userspace-facing libraries and the latter is machine specific and everything else could make sense.
Ross
Hi,
Quoted 10 lines · Ross Burton
Quoted 6 lines · Mikko Rapeli
Quoted 3 lines
Restore the existing behaviour by explicitly enabling the serial getty
generator: this means that systemd will automatically bring up a getty on the first serial console it finds.Is there a need to be backwards compatible this much? Many things may break or get refactored on master branch between releases and this is just one of them.
Yes, because this is the behaviour that both fixes the KV260 issue and also doesn’t break all of the selftests due to how testimage+qemu interact.
Quoted 7 lines
Quoted 5 lines
Then with these patches there still is the bug/issue seen in previous state too where systemd tries to start agetty on all
SERIAL_CONSOLESdefined at build time which slows down boot to "running" or "degraded" state by more than one minute. This happens also on qemu where core-image-base (I've added systemd-analyze with dependencies there to measure these):Yes. As I said, the proper solution here is to actually probe the setup. Once we’ve got walnascar out of the way I’ll finish my branch that does the same system inspection for sysvinit so we don’t need to hardcode
SERIAL_CONSOLES, instead it will bring up a console on the first tty it can find. That is likely to cause problems with testimage (as it hooks up a second tty for failures) but will be isolated to genericarm64 and not every qemu machine.
Is there some bugzilla ticket or other location for details about the qemu and testimage issue with serial consoles?
I've been running some selftests on genericarm64 with my proposed patches applied and have not run into any serial console related problems. I've only seen a lot of pseudo aborts on aarch64 build machine, e.g. wic touching images created by bitbake.
Quoted 14 lines
Quoted 4 lines
Thus I think this breaking change is fine for master branch and upcoming release and that anyone expecting different, more robust behavior should just use the systemd agetty generator.
Absolutely.
Quoted 7 lines
For genericarm64 I think https://lists.yoctoproject.org/g/poky/message/13592
is the way to go to fix both completely missing agetty
https://bugzilla.yoctoproject.org/show_bug.cgi?id=15830
and the slow boot time. Both this series and
https://lists.yoctoproject.org/g/poky/message/13592 applied does not seem to work sinceSERIAL_CONSOLESis now used all the time, so delayed boot issue remains unfixed.
Quoted 5 lines
Making systemd machine-specific would be disastrous for rebuilds as it is the provider of libudev and libsystemd, so would mean that _every_ recipe that uses those is also now machine-specific.
As was discussed in the Tuesday tech call last week, I suspect a better solution would be to have a machine-specific configuration recipe that can drop in configuration files, mask or enable units, etc. This way specific machines can configure systemd in specific ways, without actually touching the core recipe.
Alternatively, splitting systemd into ‘userspace’ and ’system’ recipes where the former is tune-specific and contains the userspace-facing libraries and the latter is machine specific and everything else could make sense.
Ok so machine/BSP config should not change systemd setup then, got it.
Since "poky-altcfg" distro uses systemd by default, could it not also use systemd to manage agetty startup and ignore SERIAL_CONSOLES from build time?
This sounds like the "efi" support issue I have with systemd. Patches out, decision pending...
Cheers,
-Mikko