ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Subsystems (3)
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
From: Kevin Hilman <[email protected]>
The option CONFIG_EXYNOS5420_MCPM is causing imprecise external aborts during boot testing, causing various userspace startup failures.
Disable until it has gotten more testing.
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.
[1] http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/288344.html
arch/arm/configs/exynos_defconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/configs/exynos_defconfig +1 -1
@@ -10,7 +10,7 @@ CONFIG_MODULE_UNLOAD=y CONFIG_PARTITION_ADVANCED=y CONFIG_ARCH_EXYNOS=y CONFIG_ARCH_EXYNOS3=y -CONFIG_EXYNOS5420_MCPM=y +CONFIG_EXYNOS5420_MCPM=n CONFIG_SMP=y CONFIG_BIG_LITTLE=y CONFIG_BL_SWITCHER=y
Signature · 1 line
2.1.0
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>
Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.
Hi Kevin,
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend. Anyway let me apply this into -fixes and then let's enable after test its functionality in -next in a couple of days.
Thanks,
Kukjin
Quoted 20 lines
[1] http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/288344.html
arch/arm/configs/exynos_defconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)arch/arm/configs/exynos_defconfig +1
-2@@ -10,7 +10,7 @@ CONFIG_MODULE_UNLOAD=y CONFIG_PARTITION_ADVANCED=y CONFIG_ARCH_EXYNOS=y CONFIG_ARCH_EXYNOS3=y -CONFIG_EXYNOS5420_MCPM=y +CONFIG_EXYNOS5420_MCPM=n CONFIG_SMP=y CONFIG_BIG_LITTLE=y CONFIG_BL_SWITCHER=y --2.1.0
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 22 lines · Kukjin Kim
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.
Quoted 6 lines
Hi Kevin,
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend.
Which of the recently queued patches are expected to address the imprecise abort issue? I'd be happy to test them out.
Quoted 2 lines
Anyway let me apply this into -fixes and
then let's enable after test its functionality in -next in a couple of days.
Yes, I think this needs to be applied until these aborts are understood and fixed.
Thanks,
Kevin
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Kukjin,
Quoted 26 lines · Kevin Hilman
Quoted 24 lines · Kukjin Kim
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.Hi Kevin,
Quoted 7 lines
Quoted 4 lines
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend.
Which of the recently queued patches are expected to address the imprecise abort issue? I'd be happy to test them out.
Exynos5 MCPM is still broken in linux-next and still causing an imprecise abort.
What is the status of $SUBJECT patch?
Quoted 5 lines
Quoted 2 lines
Anyway let me apply this into -fixes and
then let's enable after test its functionality in -next in a couple of days.Yes, I think this needs to be applied until these aborts are understood and fixed.
Is anyone at Samsung actually looking into these MCPM issues?
Kevin
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 41 lines · Kevin Hilman
Kukjin,
Quoted 26 lines · Kevin Hilman
Quoted 24 lines · Kukjin Kim
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.Hi Kevin,
Quoted 7 lines
Quoted 4 lines
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend.
Which of the recently queued patches are expected to address the imprecise abort issue? I'd be happy to test them out.
Exynos5 MCPM is still broken in linux-next and still causing an imprecise abort.
What is the status of $SUBJECT patch?
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines
Anyway let me apply this into -fixes and
then let's enable after test its functionality in -next in a couple of days.Yes, I think this needs to be applied until these aborts are understood and fixed.
Is anyone at Samsung actually looking into these MCPM issues?
Hi Kevin,
What hardware are you having problems with? 5420 or 5422/5800?
-Olof
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 45 lines · Olof Johansson
Quoted 44 lines · Kevin Hilman
Kukjin,
Quoted 26 lines · Kevin Hilman
Quoted 24 lines · Kukjin Kim
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.Hi Kevin,
Quoted 7 lines
Quoted 4 lines
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend.
Which of the recently queued patches are expected to address the imprecise abort issue? I'd be happy to test them out.
Exynos5 MCPM is still broken in linux-next and still causing an imprecise abort.
What is the status of $SUBJECT patch?
Quoted 3 lines
Quoted 2 lines
Anyway let me apply this into -fixes and
then let's enable after test its functionality in -next in a couple of days.
Quoted 8 lines
Quoted 4 lines
Quoted 2 lines
Yes, I think this needs to be applied until these aborts are understood and fixed.
Is anyone at Samsung actually looking into these MCPM issues?
Hi Kevin,
What hardware are you having problems with? 5420 or 5422/5800?
Yes. :)
exynos5420-arndale-octa:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5420-arndale-octa.html
exynos5422-odroid-xu3:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5422-odroid-xu3.html
My boot tests seem to pass fine because I have such a minimal userspace, but Tyler Baker reported that with a "real" userspace, he can't boot to a shell:
http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/286203.html
Kevin
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 62 lines · Kevin Hilman
Quoted 45 lines · Olof Johansson
Quoted 44 lines · Kevin Hilman
Kukjin,
Quoted 26 lines · Kevin Hilman
Quoted 24 lines · Kukjin Kim
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.Hi Kevin,
Quoted 7 lines
Quoted 4 lines
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend.
Which of the recently queued patches are expected to address the imprecise abort issue? I'd be happy to test them out.
Exynos5 MCPM is still broken in linux-next and still causing an imprecise abort.
What is the status of $SUBJECT patch?
Quoted 3 lines
Quoted 2 lines
Anyway let me apply this into -fixes and
then let's enable after test its functionality in -next in a couple of days.Quoted 8 lines
Quoted 4 lines
Quoted 2 lines
Yes, I think this needs to be applied until these aborts are understood and fixed.
Is anyone at Samsung actually looking into these MCPM issues?
Hi Kevin,
What hardware are you having problems with? 5420 or 5422/5800?
Yes. :)
exynos5420-arndale-octa:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5420-arndale-octa.html
exynos5422-odroid-xu3:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5422-odroid-xu3.html
Quoted 5 lines
My boot tests seem to pass fine because I have such a minimal userspace, but Tyler Baker reported that with a "real" userspace, he can't boot to a shell:
http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/286203.html
I'm not surprised that 5420 has issues, but I have not seen any external aborts on neither Chromebook that I have in my farm.
Sounds like the secondary cpus should be disabled on those device trees instead, doesn't it?
-Olof
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 67 lines · Olof Johansson
Quoted 66 lines · Kevin Hilman
Quoted 45 lines · Olof Johansson
Quoted 44 lines · Kevin Hilman
Kukjin,
Quoted 26 lines · Kevin Hilman
Quoted 24 lines · Kukjin Kim
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.Hi Kevin,
Quoted 7 lines
Quoted 4 lines
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend.
Which of the recently queued patches are expected to address the imprecise abort issue? I'd be happy to test them out.
Exynos5 MCPM is still broken in linux-next and still causing an imprecise abort.
What is the status of $SUBJECT patch?
Quoted 3 lines
Quoted 2 lines
Anyway let me apply this into -fixes and
then let's enable after test its functionality in -next in a couple of days.Quoted 8 lines
Quoted 4 lines
Quoted 2 lines
Yes, I think this needs to be applied until these aborts are understood and fixed.
Is anyone at Samsung actually looking into these MCPM issues?
Hi Kevin,
What hardware are you having problems with? 5420 or 5422/5800?
Yes. :)
exynos5420-arndale-octa:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5420-arndale-octa.html
exynos5422-odroid-xu3:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5422-odroid-xu3.htmlMy boot tests seem to pass fine because I have such a minimal userspace, but Tyler Baker reported that with a "real" userspace, he can't boot to a shell:
Quoted 7 lines
I'm not surprised that 5420 has issues, but I have not seen any external aborts on neither Chromebook that I have in my farm.
Sounds like the secondary cpus should be disabled on those device trees instead, doesn't it?
Er, cluster, not cpus.
-Olof
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 58 lines · Olof Johansson
Quoted 56 lines · Olof Johansson
Quoted 55 lines · Kevin Hilman
Quoted 53 lines · Olof Johansson
Quoted 41 lines · Kevin Hilman
Kukjin,
Quoted 26 lines · Kevin Hilman
Quoted 24 lines · Kukjin Kim
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.Hi Kevin,
Quoted 7 lines
Quoted 4 lines
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend.
Which of the recently queued patches are expected to address the imprecise abort issue? I'd be happy to test them out.
Exynos5 MCPM is still broken in linux-next and still causing an imprecise abort.
What is the status of $SUBJECT patch?
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines
Anyway let me apply this into -fixes and
then let's enable after test its functionality in -next in a couple of days.Yes, I think this needs to be applied until these aborts are understood and fixed.
Is anyone at Samsung actually looking into these MCPM issues?
Hi Kevin,
What hardware are you having problems with? 5420 or 5422/5800?
Quoted 15 lines
Quoted 4 lines
Quoted 4 lines
Yes. :)
exynos5420-arndale-octa:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5420-arndale-octa.html
Quoted 2 lines
Quoted 2 lines
exynos5422-odroid-xu3:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5422-odroid-xu3.html
Quoted 7 lines
Quoted 6 lines
My boot tests seem to pass fine because I have such a minimal userspace, but Tyler Baker reported that with a "real" userspace, he can't boot to a shell:
http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/286203.html
Hmm...his report was in Sep...I think it should be fine with current -next?
To be honest, since I don't have the exynos5420 arndale, chromebook...but smdk which has different bootloader, I couldn't test it...I'll try to make a test farm like you guys...
Quoted 1 line
Quoted 1 line
I'm not surprised that 5420 has issues, but I have not seen any
Sorry.
Quoted 6 lines
Quoted 4 lines
external aborts on neither Chromebook that I have in my farm.
Sounds like the secondary cpus should be disabled on those device trees instead, doesn't it?
Er, cluster, not cpus.
- Kukjin
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 68 lines · Kukjin Kim
Quoted 67 lines · Olof Johansson
Quoted 52 lines · Olof Johansson
Quoted 51 lines · Kevin Hilman
Quoted 50 lines · Olof Johansson
Quoted 48 lines · Kevin Hilman
Kukjin,
Quoted 26 lines · Kevin Hilman
Quoted 24 lines · Kukjin Kim
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.Hi Kevin,
Quoted 7 lines
Quoted 4 lines
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend.
Which of the recently queued patches are expected to address the imprecise abort issue? I'd be happy to test them out.
Exynos5 MCPM is still broken in linux-next and still causing an imprecise abort.
What is the status of $SUBJECT patch?
Quoted 5 lines
Quoted 2 lines
Anyway let me apply this into -fixes and
then let's enable after test its functionality in -next in a couple of days.Yes, I think this needs to be applied until these aborts are understood and fixed.
Is anyone at Samsung actually looking into these MCPM issues?
Quoted 8 lines
Quoted 8 lines
Quoted 3 lines
Hi Kevin,
What hardware are you having problems with? 5420 or 5422/5800?
Yes. :)
exynos5420-arndale-octa:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5420-arndale-octa.html
Quoted 2 lines
Quoted 2 lines
exynos5422-odroid-xu3:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5422-odroid-xu3.html
Quoted 1 line
Quoted 1 line
Quoted 7 lines
Quoted 6 lines
Quoted 6 lines
Quoted 5 lines
My boot tests seem to pass fine because I have such a minimal userspace, but Tyler Baker reported that with a "real" userspace, he can't boot to a shell:
http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/286203.html
Hmm...his report was in Sep...I think it should be fine with current -next?
No, it is still broken in linux-next (as I stated above.)
Moreover, earlier in this thread you mentioned you were merging some MCPM patches that should address this, but did not respond when I asked which patches you thing should address this issue
Quoted 3 lines
To be honest, since I don't have the exynos5420 arndale, chromebook...but smdk which has different bootloader, I couldn't test it...I'll try to make a test farm like you guys...
Do you have some colleagues with any other 542x hardware? I had assumed that linux-next was being better tested on the publicaly available, and widely available boards like odroid-xu3 and Chromebook2, but I've come to realize the hard way that that is not the case. You mention your board has a different bootloader. Do you suspect there's a bootloader issue on these other platforms? If so, could you elaborate on possible fixes? I'm more than willing to test any proposed fixes, but I'm not familiar enough yet with these SoCs to figure out the underlying issues alone.
Until you have a working board farm, you could start having a closer look at the boot logs we're already producing. Admittedly linux-next broken in many ways besides this one for exynos currently, but it has been having these imprecise aborts well before the other recent issues.
Also, It's very possible that this issue is not even MCPM related at all, and MCPM is just uncovering a previously hidden bug. It would be very helpful if people more familiar with this hardware and SoC would investigate bug reports like these.
Kevin
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Hello Kevin,
Quoted 83 lines · Kevin Hilman
Quoted 68 lines · Kukjin Kim
Quoted 67 lines · Olof Johansson
Quoted 52 lines · Olof Johansson
Quoted 51 lines · Kevin Hilman
Quoted 50 lines · Olof Johansson
Quoted 48 lines · Kevin Hilman
Kukjin,
Quoted 26 lines · Kevin Hilman
Quoted 24 lines · Kukjin Kim
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.Hi Kevin,
Quoted 7 lines
Quoted 4 lines
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend.
Which of the recently queued patches are expected to address the imprecise abort issue? I'd be happy to test them out.
Exynos5 MCPM is still broken in linux-next and still causing an imprecise abort.
What is the status of $SUBJECT patch?
Quoted 5 lines
Quoted 2 lines
Anyway let me apply this into -fixes and
then let's enable after test its functionality in -next in a couple of days.Yes, I think this needs to be applied until these aborts are understood and fixed.
Is anyone at Samsung actually looking into these MCPM issues?
Quoted 8 lines
Quoted 8 lines
Quoted 3 lines
Hi Kevin,
What hardware are you having problems with? 5420 or 5422/5800?
Yes. :)
exynos5420-arndale-octa:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5420-arndale-octa.html
Quoted 2 lines
Quoted 2 lines
exynos5422-odroid-xu3:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5422-odroid-xu3.html
Quoted 1 line
Quoted 1 line
Quoted 7 lines
Quoted 6 lines
Quoted 6 lines
Quoted 5 lines
My boot tests seem to pass fine because I have such a minimal userspace, but Tyler Baker reported that with a "real" userspace, he can't boot to a shell:
http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/286203.html
Hmm...his report was in Sep...I think it should be fine with current -next?
No, it is still broken in linux-next (as I stated above.)
Moreover, earlier in this thread you mentioned you were merging some MCPM patches that should address this, but did not respond when I asked which patches you thing should address this issue
Quoted 8 lines
Quoted 3 lines
To be honest, since I don't have the exynos5420 arndale, chromebook...but smdk which has different bootloader, I couldn't test it...I'll try to make a test farm like you guys...
Do you have some colleagues with any other 542x hardware? I had assumed that linux-next was being better tested on the publicaly available, and widely available boards like odroid-xu3 and Chromebook2, but I've come to realize the hard way that that is not
Are you seeing this on Chromebook2 (Peach-Pi 5800) too ?
Quoted 12 lines
the case. You mention your board has a different bootloader. Do you suspect there's a bootloader issue on these other platforms? If so, could you elaborate on possible fixes? I'm more than willing to test any proposed fixes, but I'm not familiar enough yet with these SoCs to figure out the underlying issues alone.
Until you have a working board farm, you could start having a closer look at the boot logs we're already producing. Admittedly linux-next broken in many ways besides this one for exynos currently, but it has been having these imprecise aborts well before the other recent issues.
Quoted 4 lines
Also, It's very possible that this issue is not even MCPM related at all, and MCPM is just uncovering a previously hidden bug. It would be very helpful if people more familiar with this hardware and SoC would investigate bug reports like these.
The 3 boards I have access to (SMDK5420, Chromebook Peach-Pi and Chromebook Peach-Pit) work fine with MCPM enabled. I am not sure why it is failing only on the above mentioned boards as there is nothing specific to them in the MCPM back-end.
I assume that when you default to platsmp (on disabling MCPM), the non-working boards boot all cores upto userspace without any issues ?
Based on the timeline (problems started about 2.5 months back), there have only been a couple of changes in the 5420 MCPM back-end. Could you revert the following commits and check if things improve.
20fe6f9 ARM: EXYNOS: Support cluster power off on exynos5420/5800 fbb0499 ARM: 8083/1: exynos: activate the CCI on boot CPU/cluster using the MCPM loopback
These might not revert cleanly, so instead of the above you could also comment the following 2 lines:
(quoted diff) +0 -0
b/arch/arm/mach-exynos/mcpm-exynos.c
index dc9a764..9a07188 100644
arch/arm/mach-exynos/mcpm-exynos.c +3 -3
@@ -152,7 +152,7 @@ static void exynos_power_down(void)exynos_cpu_power_down (cpunr);if (exynos_cluster_unused (cluster)) { -exynos_cluster_power_down (cluster); +//exynos_cluster_power_down(cluster); last_man = true; } }else if (cpu_use_count[cpu][cluster] ==1 ) {
@@ -356,8 +356,8 @@ static int __init exynos_mcpm_init(void) ret =mcpm_platform_register (&exynos_power_ops);if (!ret) ret =mcpm_sync_init (exynos_pm_power_up_setup); -if (!ret) - ret =mcpm_loopback (exynos_cache_off);/* turn on the CCI */ +//if (!ret) +//ret = mcpm_loopback(exynos_cache_off); /* turn on the CCI */ if (ret) {iounmap (ns_sram_base_addr);return ret;
If you still get aborts then I suspect that the problem is with the bootloader configuration but am not sure. I am OK with disabling 5420_MCPM in the default configuration in such a case. This would however mean that S2R also stops working by default on 5420.
Regards,
Abhilash
Quoted 7 lines
Kevin
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Hi Abhilash,
Abhilash Kesavan <[email protected]> writes:
[...]
Quoted 10 lines
Quoted 8 lines
Quoted 3 lines
To be honest, since I don't have the exynos5420 arndale, chromebook...but smdk which has different bootloader, I couldn't test it...I'll try to make a test farm like you guys...
Do you have some colleagues with any other 542x hardware? I had assumed that linux-next was being better tested on the publicaly available, and widely available boards like odroid-xu3 and Chromebook2, but I've come to realize the hard way that that is not
Are you seeing this on Chromebook2 (Peach-Pi 5800) too ?
No, it seems that my exynos5800-peach-pi is not having this problem, which suggests it's a bootloader setup issue.
Quoted 12 lines
Quoted 12 lines
the case. You mention your board has a different bootloader. Do you suspect there's a bootloader issue on these other platforms? If so, could you elaborate on possible fixes? I'm more than willing to test any proposed fixes, but I'm not familiar enough yet with these SoCs to figure out the underlying issues alone.
Until you have a working board farm, you could start having a closer look at the boot logs we're already producing. Admittedly linux-next broken in many ways besides this one for exynos currently, but it has been having these imprecise aborts well before the other recent issues.
Quoted 7 lines
Quoted 4 lines
Also, It's very possible that this issue is not even MCPM related at all, and MCPM is just uncovering a previously hidden bug. It would be very helpful if people more familiar with this hardware and SoC would investigate bug reports like these.
The 3 boards I have access to (SMDK5420, Chromebook Peach-Pi and Chromebook Peach-Pit) work fine with MCPM enabled.
Thanks for helping look into this.
Quoted 6 lines
I am not sure why
it is failing only on the above mentioned boards as there is nothing specific to them in the MCPM back-end.I assume that when you default to platsmp (on disabling MCPM), the non-working boards boot all cores upto userspace without any issues ?
Nope. With MCPM disabled:
- 5420/arndale-octa: CPU0-3 come up (A15s)
- 5422/odroid-xu3: only CPU0 (A7)
- 5800/peach-pi: only CPU0 (A15)
Note that with MCPM enabled, the arndale-octa gets the same result. Peach-pi on the other hand gets all 8 CPUs, and the odroid-xu3 only gets 6/8 CPUs (see other thread on that topic.)
Quoted 18 lines
Based on the timeline (problems started about 2.5 months back), there have only been a couple of changes in the 5420 MCPM back-end. Could you revert the following commits and check if things improve.
20fe6f9 ARM: EXYNOS: Support cluster power off on exynos5420/5800 fbb0499 ARM: 8083/1: exynos: activate the CCI on boot CPU/cluster using the MCPM loopback
These might not revert cleanly, so instead of the above you could also comment the following 2 lines:
arch/arm/mach-exynos/mcpm-exynos.c +0
-0
arch/arm/mach-exynos/mcpm-exynos.c static void exynos_power_down(void) · line 157
Quoted 7 lines
arch/arm/mach-exynos/mcpm-exynos.c +1
-1@@ -152,7 +152,7 @@ static void exynos_power_down(void)exynos_cpu_power_down (cpunr);if (exynos_cluster_unused (cluster)) { -exynos_cluster_power_down (cluster); +//exynos_cluster_power_down(cluster); last_man = true; }
2> } else if (cpu_use_count[cpu][cluster] == 1) {
Quoted 14 lines
arch/arm/mach-exynos/mcpm-exynos.c +2
-2@@ -356,8 +356,8 @@ static int __init exynos_mcpm_init(void) ret =mcpm_platform_register (&exynos_power_ops);if (!ret) ret =mcpm_sync_init (exynos_pm_power_up_setup); -if (!ret) - ret =mcpm_loopback (exynos_cache_off);/* turn on the CCI */ +//if (!ret) +//ret = mcpm_loopback(exynos_cache_off); /* turn on the CCI */ if (ret) {iounmap (ns_sram_base_addr);return ret;
Quoted 2 lines
If you still get aborts then I suspect that the problem is with the bootloader configuration but am not sure.
Nice. With those lines commented out, the arndale-octa is not geting imprecise aborts anymore, and this is the platform where those aborts seem to prevent booting into a full userspace (as originally reported by Tyler.)
More specifically, with only the loopback call to turn off CCI commented out, the imprecise aborts go away.
The odroid-xu3 is still getting them, but these seem to happen whether or not MCPM is enabled, so must a different issue related to the bootloader setup.
Quoted 3 lines
I am OK with disabling
5420_MCPM in the default configuration in such a case. This would however mean that S2R also stops working by default on 5420.
Disabling the option isn't my first choice either, I would rather see this issue debugged and fixed by folks that are more familiar with MCPM on Exynos.
Kevin
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Hi Kevin,
Quoted 107 lines · Kevin Hilman
Hi Abhilash,
Abhilash Kesavan <[email protected]> writes:
[...]
Quoted 10 lines
Quoted 8 lines
Quoted 3 lines
To be honest, since I don't have the exynos5420 arndale, chromebook...but smdk which has different bootloader, I couldn't test it...I'll try to make a test farm like you guys...
Do you have some colleagues with any other 542x hardware? I had assumed that linux-next was being better tested on the publicaly available, and widely available boards like odroid-xu3 and Chromebook2, but I've come to realize the hard way that that is not
Are you seeing this on Chromebook2 (Peach-Pi 5800) too ?
No, it seems that my exynos5800-peach-pi is not having this problem, which suggests it's a bootloader setup issue.
Quoted 12 lines
Quoted 12 lines
the case. You mention your board has a different bootloader. Do you suspect there's a bootloader issue on these other platforms? If so, could you elaborate on possible fixes? I'm more than willing to test any proposed fixes, but I'm not familiar enough yet with these SoCs to figure out the underlying issues alone.
Until you have a working board farm, you could start having a closer look at the boot logs we're already producing. Admittedly linux-next broken in many ways besides this one for exynos currently, but it has been having these imprecise aborts well before the other recent issues.
Quoted 7 lines
Quoted 4 lines
Also, It's very possible that this issue is not even MCPM related at all, and MCPM is just uncovering a previously hidden bug. It would be very helpful if people more familiar with this hardware and SoC would investigate bug reports like these.
The 3 boards I have access to (SMDK5420, Chromebook Peach-Pi and Chromebook Peach-Pit) work fine with MCPM enabled.
Thanks for helping look into this.
Quoted 6 lines
I am not sure why
it is failing only on the above mentioned boards as there is nothing specific to them in the MCPM back-end.I assume that when you default to platsmp (on disabling MCPM), the non-working boards boot all cores upto userspace without any issues ?
Nope. With MCPM disabled:
- 5420/arndale-octa: CPU0-3 come up (A15s)
- 5422/odroid-xu3: only CPU0 (A7)
- 5800/peach-pi: only CPU0 (A15)Note that with MCPM enabled, the arndale-octa gets the same result. Peach-pi on the other hand gets all 8 CPUs, and the odroid-xu3 only gets 6/8 CPUs (see other thread on that topic.)
Quoted 18 lines
Based on the timeline (problems started about 2.5 months back), there have only been a couple of changes in the 5420 MCPM back-end. Could you revert the following commits and check if things improve.
20fe6f9 ARM: EXYNOS: Support cluster power off on exynos5420/5800 fbb0499 ARM: 8083/1: exynos: activate the CCI on boot CPU/cluster using the MCPM loopback
These might not revert cleanly, so instead of the above you could also comment the following 2 lines:
arch/arm/mach-exynos/mcpm-exynos.c +0
-0arch/arm/mach-exynos/mcpm-exynos.c static void exynos_power_down(void) · line 157
Quoted 7 lines
arch/arm/mach-exynos/mcpm-exynos.c +1
-1@@ -152,7 +152,7 @@ static void exynos_power_down(void)exynos_cpu_power_down (cpunr);if (exynos_cluster_unused (cluster)) { -exynos_cluster_power_down (cluster); +//exynos_cluster_power_down(cluster); last_man = true; }2> } else if (cpu_use_count[cpu][cluster] == 1) {
Quoted 14 lines
arch/arm/mach-exynos/mcpm-exynos.c +2
-2@@ -356,8 +356,8 @@ static int __init exynos_mcpm_init(void) ret =mcpm_platform_register (&exynos_power_ops);if (!ret) ret =mcpm_sync_init (exynos_pm_power_up_setup); -if (!ret) - ret =mcpm_loopback (exynos_cache_off);/* turn on the CCI */ +//if (!ret) +//ret = mcpm_loopback(exynos_cache_off); /* turn on the CCI */ if (ret) {iounmap (ns_sram_base_addr);return ret;Quoted 2 lines
If you still get aborts then I suspect that the problem is with the bootloader configuration but am not sure.
Nice. With those lines commented out, the arndale-octa is not geting imprecise aborts anymore, and this is the platform where those aborts seem to prevent booting into a full userspace (as originally reported by Tyler.)
Quoted 2 lines
More specifically, with only the loopback call to turn off CCI commented out, the imprecise aborts go away.
I can't see how enabling snoops for the boot cluster is causing these aborts. Perhaps as Krzysztof commented it has something to do with the secure firmware/tz software on these boards ? Other than there does not appear to be any difference between the working/non-working setups.
Abhilash
Quoted 14 lines
The odroid-xu3 is still getting them, but these seem to happen whether or not MCPM is enabled, so must a different issue related to the bootloader setup.
Quoted 3 lines
I am OK with disabling
5420_MCPM in the default configuration in such a case. This would however mean that S2R also stops working by default on 5420.Disabling the option isn't my first choice either, I would rather see this issue debugged and fixed by folks that are more familiar with MCPM on Exynos.
Kevin
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 110 lines · Abhilash Kesavan
Hi Kevin,
Quoted 107 lines · Kevin Hilman
Hi Abhilash,
Abhilash Kesavan <[email protected]> writes:
[...]
Quoted 10 lines
Quoted 8 lines
Quoted 3 lines
To be honest, since I don't have the exynos5420 arndale, chromebook...but smdk which has different bootloader, I couldn't test it...I'll try to make a test farm like you guys...
Do you have some colleagues with any other 542x hardware? I had assumed that linux-next was being better tested on the publicaly available, and widely available boards like odroid-xu3 and Chromebook2, but I've come to realize the hard way that that is not
Are you seeing this on Chromebook2 (Peach-Pi 5800) too ?
No, it seems that my exynos5800-peach-pi is not having this problem, which suggests it's a bootloader setup issue.
Quoted 12 lines
Quoted 12 lines
the case. You mention your board has a different bootloader. Do you suspect there's a bootloader issue on these other platforms? If so, could you elaborate on possible fixes? I'm more than willing to test any proposed fixes, but I'm not familiar enough yet with these SoCs to figure out the underlying issues alone.
Until you have a working board farm, you could start having a closer look at the boot logs we're already producing. Admittedly linux-next broken in many ways besides this one for exynos currently, but it has been having these imprecise aborts well before the other recent issues.
Quoted 7 lines
Quoted 4 lines
Also, It's very possible that this issue is not even MCPM related at all, and MCPM is just uncovering a previously hidden bug. It would be very helpful if people more familiar with this hardware and SoC would investigate bug reports like these.
The 3 boards I have access to (SMDK5420, Chromebook Peach-Pi and Chromebook Peach-Pit) work fine with MCPM enabled.
Thanks for helping look into this.
Quoted 6 lines
I am not sure why
it is failing only on the above mentioned boards as there is nothing specific to them in the MCPM back-end.I assume that when you default to platsmp (on disabling MCPM), the non-working boards boot all cores upto userspace without any issues ?
Nope. With MCPM disabled:
- 5420/arndale-octa: CPU0-3 come up (A15s)
- 5422/odroid-xu3: only CPU0 (A7)
- 5800/peach-pi: only CPU0 (A15)Note that with MCPM enabled, the arndale-octa gets the same result. Peach-pi on the other hand gets all 8 CPUs, and the odroid-xu3 only gets 6/8 CPUs (see other thread on that topic.)
Quoted 18 lines
Based on the timeline (problems started about 2.5 months back), there have only been a couple of changes in the 5420 MCPM back-end. Could you revert the following commits and check if things improve.
20fe6f9 ARM: EXYNOS: Support cluster power off on exynos5420/5800 fbb0499 ARM: 8083/1: exynos: activate the CCI on boot CPU/cluster using the MCPM loopback
These might not revert cleanly, so instead of the above you could also comment the following 2 lines:
arch/arm/mach-exynos/mcpm-exynos.c +0
-0arch/arm/mach-exynos/mcpm-exynos.c static void exynos_power_down(void) · line 157
Quoted 7 lines
arch/arm/mach-exynos/mcpm-exynos.c +1
-1@@ -152,7 +152,7 @@ static void exynos_power_down(void)exynos_cpu_power_down (cpunr);if (exynos_cluster_unused (cluster)) { -exynos_cluster_power_down (cluster); +//exynos_cluster_power_down(cluster); last_man = true; }2> } else if (cpu_use_count[cpu][cluster] == 1) {
Quoted 14 lines
arch/arm/mach-exynos/mcpm-exynos.c +2
-2@@ -356,8 +356,8 @@ static int __init exynos_mcpm_init(void) ret =mcpm_platform_register (&exynos_power_ops);if (!ret) ret =mcpm_sync_init (exynos_pm_power_up_setup); -if (!ret) - ret =mcpm_loopback (exynos_cache_off);/* turn on the CCI */ +//if (!ret) +//ret = mcpm_loopback(exynos_cache_off); /* turn on the CCI */ if (ret) {iounmap (ns_sram_base_addr);return ret;Quoted 2 lines
If you still get aborts then I suspect that the problem is with the bootloader configuration but am not sure.
Nice. With those lines commented out, the arndale-octa is not geting imprecise aborts anymore, and this is the platform where those aborts seem to prevent booting into a full userspace (as originally reported by Tyler.)
Quoted 8 lines
Quoted 2 lines
More specifically, with only the loopback call to turn off CCI commented out, the imprecise aborts go away.
I can't see how enabling snoops for the boot cluster is causing these aborts. Perhaps as Krzysztof commented it has something to do with the secure
firmware/tzsoftware on these boards ? Other than there does not appear to be any difference between the working/non-working setups.
Perhaps the secure firmware is preventing the CCI to be enabled by the kernel, and that is causing the imprecise abort?
Is there a way to update/replace the BL1/BL2/TZ firmware blobs with something that is known to be working better?
Kevin
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 121 lines · Kevin Hilman
Quoted 110 lines · Abhilash Kesavan
Hi Kevin,
Quoted 107 lines · Kevin Hilman
Hi Abhilash,
Abhilash Kesavan <[email protected]> writes:
[...]
Quoted 10 lines
Quoted 8 lines
Quoted 3 lines
To be honest, since I don't have the exynos5420 arndale, chromebook...but smdk which has different bootloader, I couldn't test it...I'll try to make a test farm like you guys...
Do you have some colleagues with any other 542x hardware? I had assumed that linux-next was being better tested on the publicaly available, and widely available boards like odroid-xu3 and Chromebook2, but I've come to realize the hard way that that is not
Are you seeing this on Chromebook2 (Peach-Pi 5800) too ?
No, it seems that my exynos5800-peach-pi is not having this problem, which suggests it's a bootloader setup issue.
Quoted 12 lines
Quoted 12 lines
the case. You mention your board has a different bootloader. Do you suspect there's a bootloader issue on these other platforms? If so, could you elaborate on possible fixes? I'm more than willing to test any proposed fixes, but I'm not familiar enough yet with these SoCs to figure out the underlying issues alone.
Until you have a working board farm, you could start having a closer look at the boot logs we're already producing. Admittedly linux-next broken in many ways besides this one for exynos currently, but it has been having these imprecise aborts well before the other recent issues.
Quoted 7 lines
Quoted 4 lines
Also, It's very possible that this issue is not even MCPM related at all, and MCPM is just uncovering a previously hidden bug. It would be very helpful if people more familiar with this hardware and SoC would investigate bug reports like these.
The 3 boards I have access to (SMDK5420, Chromebook Peach-Pi and Chromebook Peach-Pit) work fine with MCPM enabled.
Thanks for helping look into this.
Quoted 6 lines
I am not sure why
it is failing only on the above mentioned boards as there is nothing specific to them in the MCPM back-end.I assume that when you default to platsmp (on disabling MCPM), the non-working boards boot all cores upto userspace without any issues ?
Nope. With MCPM disabled:
- 5420/arndale-octa: CPU0-3 come up (A15s)
- 5422/odroid-xu3: only CPU0 (A7)
- 5800/peach-pi: only CPU0 (A15)Note that with MCPM enabled, the arndale-octa gets the same result. Peach-pi on the other hand gets all 8 CPUs, and the odroid-xu3 only gets 6/8 CPUs (see other thread on that topic.)
Quoted 18 lines
Based on the timeline (problems started about 2.5 months back), there have only been a couple of changes in the 5420 MCPM back-end. Could you revert the following commits and check if things improve.
20fe6f9 ARM: EXYNOS: Support cluster power off on exynos5420/5800 fbb0499 ARM: 8083/1: exynos: activate the CCI on boot CPU/cluster using the MCPM loopback
These might not revert cleanly, so instead of the above you could also comment the following 2 lines:
arch/arm/mach-exynos/mcpm-exynos.c +0
-0arch/arm/mach-exynos/mcpm-exynos.c static void exynos_power_down(void) · line 157
Quoted 7 lines
arch/arm/mach-exynos/mcpm-exynos.c +1
-1@@ -152,7 +152,7 @@ static void exynos_power_down(void)exynos_cpu_power_down (cpunr);if (exynos_cluster_unused (cluster)) { -exynos_cluster_power_down (cluster); +//exynos_cluster_power_down(cluster); last_man = true; }2> } else if (cpu_use_count[cpu][cluster] == 1) {
Quoted 14 lines
arch/arm/mach-exynos/mcpm-exynos.c +2
-2@@ -356,8 +356,8 @@ static int __init exynos_mcpm_init(void) ret =mcpm_platform_register (&exynos_power_ops);if (!ret) ret =mcpm_sync_init (exynos_pm_power_up_setup); -if (!ret) - ret =mcpm_loopback (exynos_cache_off);/* turn on the CCI */ +//if (!ret) +//ret = mcpm_loopback(exynos_cache_off); /* turn on the CCI */ if (ret) {iounmap (ns_sram_base_addr);return ret;Quoted 2 lines
If you still get aborts then I suspect that the problem is with the bootloader configuration but am not sure.
Nice. With those lines commented out, the arndale-octa is not geting imprecise aborts anymore, and this is the platform where those aborts seem to prevent booting into a full userspace (as originally reported by Tyler.)
Quoted 8 lines
Quoted 2 lines
More specifically, with only the loopback call to turn off CCI commented out, the imprecise aborts go away.
I can't see how enabling snoops for the boot cluster is causing these aborts. Perhaps as Krzysztof commented it has something to do with the secure
firmware/tzsoftware on these boards ? Other than there does not appear to be any difference between the working/non-working setups.
Quoted 6 lines
Perhaps the secure firmware is preventing the CCI to be enabled by the kernel, and that is causing the imprecise abort?
Is there a way to update/replace the BL1/BL2/TZ firmware blobs with something that is known to be working better?
Seems current problem you mentioned is due to different bootloader as I commented before, but to release bootloader images (bl1, bl2 and so on) should be handled by board manufacture not SoC vendor I think...even though the images are provided by vendor for manufacture. To be honest I'm not sure what procedure should be passed in Samsung side for now because we including Abhilash are belong to just development team. Need some time but I can't confirm that...sorry. Let us try.
BTW, Kevin do you know current version for bootloader images on the boards?
Thanks,
Kukjin
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
In reply to Kevin HilmanQuoted 10 lines · Kevin Hilman
Quoted 8 lines · Abhilash Kesavan
Hi Kevin,
Quoted 4 lines · Kevin Hilman
[...]
More specifically, with only the loopback call to turn off CCI commented out, the imprecise aborts go away.
Quoted 8 lines
Quoted 5 lines
I can't see how enabling snoops for the boot cluster is causing these aborts. Perhaps as Krzysztof commented it has something to do with the secure
firmware/tzsoftware on these boards ? Other than there does not appear to be any difference between the working/non-working setups.Perhaps the secure firmware is preventing the CCI to be enabled by the kernel, and that is causing the imprecise abort?
That is well possible.
Now...... if the bootloader/firmware does not let Linux deal with both the CCI and caches then MCPM simply has no more purpose for this board. The whole point of MCPM is actually to handle the CCI properly and the most efficient way despite all the possible races and opportunities for memory corruptions. And yes, this is a complex task.
So there is actually two choices: the firmware let Linux take care of it via the MCPM layer (easy), or the firmware has to implement it all _properly_ (hard) behind an interface such as PSCI, at which point MCPM should be configured out.
If the firmware does not let Linux interact with the CCI _and_ does not implement full MCPM-like services then the platform is broken and only a firmware upgrade could fix that. It might still be possible to boot all CPUs through other means, but power management would then be severely limited.
Nicolas
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Hi Kevin,
Quoted 34 lines · Nicolas Pitre
Quoted 10 lines · Kevin Hilman
Quoted 8 lines · Abhilash Kesavan
Hi Kevin,
Quoted 4 lines · Kevin Hilman
[...]
More specifically, with only the loopback call to turn off CCI commented out, the imprecise aborts go away.
Quoted 8 lines
Quoted 5 lines
I can't see how enabling snoops for the boot cluster is causing these aborts. Perhaps as Krzysztof commented it has something to do with the secure
firmware/tzsoftware on these boards ? Other than there does not appear to be any difference between the working/non-working setups.Perhaps the secure firmware is preventing the CCI to be enabled by the kernel, and that is causing the imprecise abort?
That is well possible.
Now...... if the bootloader/firmware does not let Linux deal with both the CCI and caches then MCPM simply has no more purpose for this board. The whole point of MCPM is actually to handle the CCI properly and the most efficient way despite all the possible races and opportunities for memory corruptions. And yes, this is a complex task.
So there is actually two choices: the firmware let Linux take care of it via the MCPM layer (easy), or the firmware has to implement it all _properly_ (hard) behind an interface such as PSCI, at which point MCPM should be configured out.
Quoted 5 lines
If the firmware does not let Linux interact with the CCI _and_ does not implement full MCPM-like services then the platform is broken and only a firmware upgrade could fix that. It might still be possible to boot all CPUs through other means, but power management would then be severely limited.
How about restricting the mcpm initialization to only known working boards like chromebooks and smdk. This would be better than disabling the config altogether from exynos_defconfig. The non-working boards would then default to platsmp. Assuming that the firmware handles all CCI/cache activities then platsmp may work for secondary core boot-up ?
Can you please apply the below diff and test the non-working boards with CONFIG_EXYNOS5420_MCPM enabled.
(quoted diff) +0 -0
b/arch/arm/mach-exynos/mcpm-exynos.c
index b0d3c2e..34d77bb 100644
arch/arm/mach-exynos/mcpm-exynos.c +3 -2
@@ -316,8 +316,9 @@ static void __init exynos_cache_off(void) }static const struct of_device_id exynos_dt_mcpm_match[] = { - { .compatible ="samsung,exynos5420" }, - { .compatible ="samsung,exynos5800" }, + { .compatible ="samsung,smdk5420" }, + { .compatible ="google,pi" }, + { .compatible ="google,pit" }, {}, };
On a different note, I did not see any mainline support for Odroid Xu3, are you testing this board with a non-mainline kernel ?
Regards,
Abhilash
Quoted 4 lines
Nicolas
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 51 lines · Abhilash Kesavan
Hi Kevin,
Quoted 34 lines · Nicolas Pitre
Quoted 10 lines · Kevin Hilman
Quoted 8 lines · Abhilash Kesavan
Hi Kevin,
Quoted 4 lines · Kevin Hilman
[...]
More specifically, with only the loopback call to turn off CCI commented out, the imprecise aborts go away.
Quoted 8 lines
Quoted 5 lines
I can't see how enabling snoops for the boot cluster is causing these aborts. Perhaps as Krzysztof commented it has something to do with the secure
firmware/tzsoftware on these boards ? Other than there does not appear to be any difference between the working/non-working setups.Perhaps the secure firmware is preventing the CCI to be enabled by the kernel, and that is causing the imprecise abort?
That is well possible.
Now...... if the bootloader/firmware does not let Linux deal with both the CCI and caches then MCPM simply has no more purpose for this board. The whole point of MCPM is actually to handle the CCI properly and the most efficient way despite all the possible races and opportunities for memory corruptions. And yes, this is a complex task.
So there is actually two choices: the firmware let Linux take care of it via the MCPM layer (easy), or the firmware has to implement it all _properly_ (hard) behind an interface such as PSCI, at which point MCPM should be configured out.
Quoted 5 lines
If the firmware does not let Linux interact with the CCI _and_ does not implement full MCPM-like services then the platform is broken and only a firmware upgrade could fix that. It might still be possible to boot all CPUs through other means, but power management would then be severely limited.
How about restricting the mcpm initialization to only known working boards like chromebooks and smdk. This would be better than disabling the config altogether from
exynos_defconfig. The non-working boards would then default to platsmp. Assuming that the firmware handles all CCI/cache activities then platsmp may work for secondary core boot-up ?
Quoted 2 lines
Can you please apply the below diff and test the non-working boards with
CONFIG_EXYNOS5420_MCPMenabled.
I'd much prefer if the CCI is non accessible on some board that the device tree file for that board be modified instead by marking the CCI as unavailable.
Quoted 29 lines
arch/arm/mach-exynos/mcpm-exynos.c +3
-2@@ -316,8 +316,9 @@ static void __init exynos_cache_off(void) }static const struct of_device_id exynos_dt_mcpm_match[] = { - { .compatible ="samsung,exynos5420" }, - { .compatible ="samsung,exynos5800" }, + { .compatible ="samsung,smdk5420" }, + { .compatible ="google,pi" }, + { .compatible ="google,pit" }, {}, };On a different note, I did not see any mainline support for Odroid Xu3, are you testing this board with a non-mainline kernel ?
Regards,
AbhilashQuoted 4 lines
Nicolas
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Hi Nicolas,
Quoted 53 lines · Nicolas Pitre
Quoted 51 lines · Abhilash Kesavan
Hi Kevin,
Quoted 34 lines · Nicolas Pitre
Quoted 10 lines · Kevin Hilman
Quoted 8 lines · Abhilash Kesavan
Hi Kevin,
Quoted 4 lines · Kevin Hilman
[...]
More specifically, with only the loopback call to turn off CCI commented out, the imprecise aborts go away.
Quoted 8 lines
Quoted 5 lines
I can't see how enabling snoops for the boot cluster is causing these aborts. Perhaps as Krzysztof commented it has something to do with the secure
firmware/tzsoftware on these boards ? Other than there does not appear to be any difference between the working/non-working setups.Perhaps the secure firmware is preventing the CCI to be enabled by the kernel, and that is causing the imprecise abort?
That is well possible.
Now...... if the bootloader/firmware does not let Linux deal with both the CCI and caches then MCPM simply has no more purpose for this board. The whole point of MCPM is actually to handle the CCI properly and the most efficient way despite all the possible races and opportunities for memory corruptions. And yes, this is a complex task.
So there is actually two choices: the firmware let Linux take care of it via the MCPM layer (easy), or the firmware has to implement it all _properly_ (hard) behind an interface such as PSCI, at which point MCPM should be configured out.
Quoted 5 lines
If the firmware does not let Linux interact with the CCI _and_ does not implement full MCPM-like services then the platform is broken and only a firmware upgrade could fix that. It might still be possible to boot all CPUs through other means, but power management would then be severely limited.
How about restricting the mcpm initialization to only known working boards like chromebooks and smdk. This would be better than disabling the config altogether from
exynos_defconfig. The non-working boards would then default to platsmp. Assuming that the firmware handles all CCI/cache activities then platsmp may work for secondary core boot-up ?
Quoted 6 lines
Quoted 2 lines
Can you please apply the below diff and test the non-working boards with
CONFIG_EXYNOS5420_MCPMenabled.I'd much prefer if the CCI is non accessible on some board that the device tree file for that board be modified instead by marking the CCI as unavailable.
I will post a patch disabling CCI for Arndale-Octa.
Abhilasj
Quoted 31 lines
Quoted 31 lines
arch/arm/mach-exynos/mcpm-exynos.c +3
-2@@ -316,8 +316,9 @@ static void __init exynos_cache_off(void) }static const struct of_device_id exynos_dt_mcpm_match[] = { - { .compatible ="samsung,exynos5420" }, - { .compatible ="samsung,exynos5800" }, + { .compatible ="samsung,smdk5420" }, + { .compatible ="google,pi" }, + { .compatible ="google,pit" }, {}, };On a different note, I did not see any mainline support for Odroid Xu3, are you testing this board with a non-mainline kernel ?
Regards,
AbhilashQuoted 4 lines
Nicolas
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
In reply to Nicolas PitreQuoted 34 lines · Nicolas Pitre
Quoted 10 lines · Kevin Hilman
Quoted 8 lines · Abhilash Kesavan
Hi Kevin,
Quoted 4 lines · Kevin Hilman
[...]
More specifically, with only the loopback call to turn off CCI commented out, the imprecise aborts go away.
Quoted 8 lines
Quoted 5 lines
I can't see how enabling snoops for the boot cluster is causing these aborts. Perhaps as Krzysztof commented it has something to do with the secure
firmware/tzsoftware on these boards ? Other than there does not appear to be any difference between the working/non-working setups.Perhaps the secure firmware is preventing the CCI to be enabled by the kernel, and that is causing the imprecise abort?
That is well possible.
Now...... if the bootloader/firmware does not let Linux deal with both the CCI and caches then MCPM simply has no more purpose for this board. The whole point of MCPM is actually to handle the CCI properly and the most efficient way despite all the possible races and opportunities for memory corruptions. And yes, this is a complex task.
So there is actually two choices: the firmware let Linux take care of it via the MCPM layer (easy), or the firmware has to implement it all _properly_ (hard) behind an interface such as PSCI, at which point MCPM should be configured out.
Quoted 6 lines
If the firmware does not let Linux interact with the CCI _and_ does not implement full MCPM-like services then the platform is broken and only a firmware upgrade could fix that. It might still be possible to boot all CPUs through other means, but power management would then be severely limited.
Thanks Nico for the detailed description on the requirements for using MCPM. This is the kind of issue I was worried in the other thread on Fijitsu platform. That's the reason I was asking the information about their secure firmware and what exactly it configures so that we won't end up with similar situation on there too and definitely not to push PSCI. I completely agree with you that making a some change in firmware to give control of CCI to kernel is easy.
Probably if the vendors disagree to apply this small fix to the firmware we should provide them with *only choice* of PSCI implementation which is quite complex and easy to get it wrong. That might trigger them to provide a small fix to use MCPM.
Regards,
Sudeep
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
In reply to Kevin HilmanQuoted 57 lines · Kevin Hilman
Quoted 22 lines · Kukjin Kim
Quoted 21 lines · Olof Johansson
Quoted 6 lines · Olof Johansson
Quoted 5 lines · Kevin Hilman
Quoted 4 lines · Olof Johansson
Quoted 2 lines · Kevin Hilman
Is anyone at Samsung actually looking into these MCPM issues?
Quoted 8 lines
Quoted 8 lines
Quoted 3 lines
Hi Kevin,
What hardware are you having problems with? 5420 or 5422/5800?
Yes. :)
exynos5420-arndale-octa:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5420-arndale-octa.html
Quoted 2 lines
Quoted 2 lines
exynos5422-odroid-xu3:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5422-odroid-xu3.html
Quoted 1 line
Quoted 1 line
Quoted 7 lines
Quoted 6 lines
Quoted 6 lines
Quoted 5 lines
My boot tests seem to pass fine because I have such a minimal userspace, but Tyler Baker reported that with a "real" userspace, he can't boot to a shell:
http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/286203.html
Hmm...his report was in Sep...I think it should be fine with current -next?
No, it is still broken in linux-next (as I stated above.)
Moreover, earlier in this thread you mentioned you were merging some MCPM patches that should address this, but did not respond when I asked which patches you thing should address this issue
Quoted 3 lines
To be honest, since I don't have the exynos5420 arndale, chromebook...but smdk which has different bootloader, I couldn't test it...I'll try to make a test farm like you guys...
Do you have some colleagues with any other 542x hardware? I had assumed that linux-next was being better tested on the publicaly available, and widely available boards like odroid-xu3 and Chromebook2, but I've come to realize the hard way that that is not the case. You mention your board has a different bootloader. Do you suspect there's a bootloader issue on these other platforms? If so, could you elaborate on possible fixes? I'm more than willing to test any proposed fixes, but I'm not familiar enough yet with these SoCs to figure out the underlying issues alone.
Until you have a working board farm, you could start having a closer look at the boot logs we're already producing. Admittedly linux-next broken in many ways besides this one for exynos currently, but it has been having these imprecise aborts well before the other recent issues.
Quoted 4 lines
Also, It's very possible that this issue is not even MCPM related at all, and MCPM is just uncovering a previously hidden bug. It would be very helpful if people more familiar with this hardware and SoC would investigate bug reports like these.
Interesting thing can be found in exynos5420.dtsi:
mdma1: mdma at 11C10000 {
...
/*
* MDMA1 can support both secure and non-secure
* AXI transactions. When this is enabled in the kernel
* for boards that run in secure mode, we are getting
* imprecise external aborts causing the kernel to oops.
*/
status = "disabled";
};I am booting Arndale Octa on some other config and exynos. However with or without MCPM the imprecise aborts are still present (but not fatal, shell comes up).
My board boots also under secure firmware (I am using Linaro's ubuntu image). Maybe that is the cause?
Best regards,
Krzysztof
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
Quoted 75 lines · Krzysztof Kozlowski
Quoted 57 lines · Kevin Hilman
Quoted 22 lines · Kukjin Kim
Quoted 21 lines · Olof Johansson
Quoted 6 lines · Olof Johansson
Quoted 5 lines · Kevin Hilman
Quoted 4 lines · Olof Johansson
Quoted 2 lines · Kevin Hilman
Is anyone at Samsung actually looking into these MCPM issues?
Quoted 8 lines
Quoted 8 lines
Quoted 3 lines
Hi Kevin,
What hardware are you having problems with? 5420 or 5422/5800?
Yes. :)
exynos5420-arndale-octa:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5420-arndale-octa.html
Quoted 2 lines
Quoted 2 lines
exynos5422-odroid-xu3:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5422-odroid-xu3.html
Quoted 1 line
Quoted 1 line
Quoted 7 lines
Quoted 6 lines
Quoted 6 lines
Quoted 5 lines
My boot tests seem to pass fine because I have such a minimal userspace, but Tyler Baker reported that with a "real" userspace, he can't boot to a shell:
http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/286203.html
Hmm...his report was in Sep...I think it should be fine with current -next?
No, it is still broken in linux-next (as I stated above.)
Moreover, earlier in this thread you mentioned you were merging some MCPM patches that should address this, but did not respond when I asked which patches you thing should address this issue
Quoted 3 lines
To be honest, since I don't have the exynos5420 arndale, chromebook...but smdk which has different bootloader, I couldn't test it...I'll try to make a test farm like you guys...
Do you have some colleagues with any other 542x hardware? I had assumed that linux-next was being better tested on the publicaly available, and widely available boards like odroid-xu3 and Chromebook2, but I've come to realize the hard way that that is not the case. You mention your board has a different bootloader. Do you suspect there's a bootloader issue on these other platforms? If so, could you elaborate on possible fixes? I'm more than willing to test any proposed fixes, but I'm not familiar enough yet with these SoCs to figure out the underlying issues alone.
Until you have a working board farm, you could start having a closer look at the boot logs we're already producing. Admittedly linux-next broken in many ways besides this one for exynos currently, but it has been having these imprecise aborts well before the other recent issues.
Quoted 4 lines
Also, It's very possible that this issue is not even MCPM related at all, and MCPM is just uncovering a previously hidden bug. It would be very helpful if people more familiar with this hardware and SoC would investigate bug reports like these.
Interesting thing can be found in
exynos5420.dtsi:mdma1: mdma at 11C10000 { ... /* * MDMA1 can support both secure and non-secure * AXI transactions. When this is enabled in the kernel * for boards that run in secure mode, we are getting * imprecise external aborts causing the kernel to oops. */ status = "disabled"; };
Quoted 6 lines
I am booting Arndale Octa on some other config and exynos. However with or without MCPM the imprecise aborts are still present (but not fatal, shell comes up).
My board boots also under secure firmware (I am using Linaro's ubuntu image). Maybe that is the cause?
One update: some fatal imprecise aborts (hanging boot) happen also on exynos_defconfig without MCPM. It looks random... one boot fails, next succeeds (however also with "imprecise external abort" message but shell comes up).
Best regards,
Krzysztof
PATCH 1/1 ARM: exynos_defconfig: disable CONFIG_EXYNOS5420_MCPM; not stable
In reply to Olof JohanssonQuoted 67 lines · Olof Johansson
Quoted 66 lines · Kevin Hilman
Quoted 45 lines · Olof Johansson
Quoted 44 lines · Kevin Hilman
Kukjin,
Quoted 26 lines · Kevin Hilman
Quoted 24 lines · Kukjin Kim
Quoted 17 lines · Kevin Hilman
From: Kevin Hilman <[email protected]>
The option
CONFIG_EXYNOS5420_MCPMis causing imprecise external aborts during boot testing, causing various userspace startup failures.Disable until it has gotten more testing.
Cc Kukjin Kim <[email protected]>,Cc Javier Martinez Canillas <[email protected]>,Cc Sachin Kamat <[email protected]>,Cc Doug Anderson <[email protected]>,Cc Bartlomiej Zolnierkiewicz <[email protected]>,Cc Krzysztof Kozlowski <[email protected]>,Cc Tushar Behera <[email protected]>,Cc stable at vger.kernel.org # v3.17+Signed-off-by Kevin Hilman <[email protected]>Quoted 4 lines
---
This has been reported by a few people[1], but not investigated or fixed, so it's time to disable this feature until it can be fixed.Hi Kevin,
Quoted 7 lines
Quoted 4 lines
Yeah I agree with your opinion.
But as you can see my tree, I've queued regarding mcpm patches for 3.19 will be shown in -next in this weekend.
Which of the recently queued patches are expected to address the imprecise abort issue? I'd be happy to test them out.
Exynos5 MCPM is still broken in linux-next and still causing an imprecise abort.
What is the status of $SUBJECT patch?
Quoted 3 lines
Quoted 2 lines
Anyway let me apply this into -fixes and
then let's enable after test its functionality in -next in a couple of days.Quoted 8 lines
Quoted 4 lines
Quoted 2 lines
Yes, I think this needs to be applied until these aborts are understood and fixed.
Is anyone at Samsung actually looking into these MCPM issues?
Hi Kevin,
What hardware are you having problems with? 5420 or 5422/5800?
Yes. :)
exynos5420-arndale-octa:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5420-arndale-octa.html
exynos5422-odroid-xu3:
http://storage.armcloud.us/kernel-ci/mainline/v3.18-rc6/arm-exynos_defconfig/boot-exynos5422-odroid-xu3.htmlMy boot tests seem to pass fine because I have such a minimal userspace, but Tyler Baker reported that with a "real" userspace, he can't boot to a shell:
Quoted 7 lines
I'm not surprised that 5420 has issues, but I have not seen any external aborts on neither Chromebook that I have in my farm.
Sounds like the secondary cpus should be disabled on those device trees instead, doesn't it?
That's possible.
Unfortunately, I've gotten very little support from Samsung on this and it was originally reported 2.5 months ago[2], so I think that the 5420 MCPM should be disabled until they can propose the right fix, and actually test it.
Also, I tried disabling some CPUs at boot time, but the exynos5422-odroid-xu3 wont' even boot with nr_cpus=1 or 4 (or anything less than 8)
Kevin
[1] http://lists.infradead.org/pipermail/linux-arm-kernel/2014-September/286203.html