* next: Rpi4: Unexpected kernel BRK exception at EL1
@ 2023-06-21 12:36 Naresh Kamboju
2023-06-21 12:57 ` Mark Rutland
0 siblings, 1 reply; 11+ messages in thread
From: Naresh Kamboju @ 2023-06-21 12:36 UTC (permalink / raw)
To: Linux ARM, open list, linux-rpi-kernel, Netdev, lkft-triage
Cc: Arnd Bergmann, Dan Carpenter, Linus Walleij, Catalin Marinas,
Will Deacon, Mark Rutland, Anshuman Khandual
Following boot warnings and crashes noticed on arm64 Rpi4 device running
Linux next-20230621 kernel.
Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
boot log:
[ 22.331748] Kernel text patching generated an invalid instruction
at 0xffff8000835d6580!
[ 22.340579] Unexpected kernel BRK exception at EL1
[ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
.
[ 22.353814] Modules linked in: hci_uart(+) brcmfmac(+) btqca
brcmutil btbcm vc4(+) cfg80211 bluetooth reset_raspberrypi
clk_raspberrypi crct10dif_ce raspberrypi_hwmon snd_soc_hdmi_codec cec
v3d bcm2711_thermal rfkill drm_display_helper drm_shmem_helper
pcie_brcmstb drm_dma_helper pwm_bcm2835 i2c_bcm2835 gpu_sched
drm_kms_helper fuse drm
[ 22.376947] vc4-drm gpu: bound fe004000.txp (ops vc4_txp_ops [vc4])
[ 22.384754] CPU: 3 PID: 159 Comm: systemd-udevd Not tainted
6.4.0-rc7-next-20230621 #1
[ 22.384769] Hardware name: Raspberry Pi 4 Model B (DT)
[ 22.384776] pstate: 40000005 (nZcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 22.384789] pc : 0xffff8000835d6580
[ 22.384798] lr : sk_filter_trim_cap+0x80/0x2a0
[ 22.384825] sp : ffff800083cdb9d0
[ 22.384831] x29: ffff800083cdb9d0 x28: 0000000000000000 x27: 0000000000000001
[ 22.384853] x26: ffff000041ece000 x25: ffff0000423ac800 x24: ffff800083468e00
[ 22.384872] x23: 0000000000000000 x22: ffff000044c41300 x21: 0000000000000001
[ 22.384891] x20: ffff800083bed000 x19: ffff000044c41300 x18: 0000000000000000
[ 22.384909] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000015798650
[ 22.384928] x14: 414d003633373832 x13: 3332323d44455a49 x12: 4c414954494e495f
[ 22.384946] x11: 4345535500343331 x10: 343d4d554e514553 x9 : ffff80008124f500
[ 22.384965] x8 : ffff800083cdb7d8 x7 : 0000000000000000 x6 : 0000000000000001
[ 22.384983] x5 : ffff800082def000 x4 : ffff800082def2e8 x3 : 0000000000000000
[ 22.385001] x2 : ffff8000835d657c x1 : ffff800083bed048 x0 : ffff000044c41300
[ 22.385020] Call trace:
[ 22.385025] 0xffff8000835d6580
[ 22.385033] netlink_broadcast+0x1f0/0x4e8
[ 22.385047] netlink_sendmsg+0x318/0x420
[ 22.385056] ____sys_sendmsg+0x1cc/0x2c8
[ 22.385075] ___sys_sendmsg+0x88/0xf0
[ 22.385084] __sys_sendmsg+0x70/0xd8
[ 22.385093] __arm64_sys_sendmsg+0x2c/0x40
[ 22.385102] invoke_syscall+0x50/0x128
[ 22.385120] el0_svc_common.constprop.0+0xf4/0x120
[ 22.385136] do_el0_svc+0x44/0xb8
[ 22.385152] el0_svc+0x30/0x98
[ 22.385163] el0t_64_sync_handler+0x13c/0x158
[ 22.385174] el0t_64_sync+0x190/0x198
[ 22.385190] Code: d4202000 d4202000 d4202000 910003c9 (d503201f)
[ 22.385199] ---[ end trace 0000000000000000 ]---
[ 22.385206] note: systemd-udevd[159] exited with irqs disabled
[ 22.385378] note: systemd-udevd[159] exited with preempt_count 1
[ 22.395083] uart-pl011 fe201000.serial: no DMA platform data
[ 22.427073] vc4-drm gpu: bound fe206000.pixelvalve (ops vc4_crtc_ops [vc4])
[ 22.434295] ------------[ cut here ]------------
[ 22.474787] Bluetooth: HCI UART protocol Marvell registered
[ 22.475105] Voluntary context switch within RCU read-side critical section!
[ 22.475129] WARNING: CPU: 3 PID: 159 at
kernel/rcu/tree_plugin.h:320 rcu_note_context_switch+0x458/0x530
[ 22.553049] vc4-drm gpu: bound fe207000.pixelvalve (ops vc4_crtc_ops [vc4])
[ 22.555403] Modules linked in: hci_uart(+) brcmfmac(+) btqca
brcmutil btbcm vc4(+) cfg80211 bluetooth reset_raspberrypi
clk_raspberrypi crct10dif_ce
[ 22.583877] vc4-drm gpu: bound fe20a000.pixelvalve (ops vc4_crtc_ops [vc4])
[ 22.584920] raspberrypi_hwmon snd_soc_hdmi_codec
[ 22.614889] vc4-drm gpu: bound fe216000.pixelvalve (ops vc4_crtc_ops [vc4])
[ 22.618972] cec v3d bcm2711_thermal rfkill drm_display_helper
drm_shmem_helper pcie_brcmstb drm_dma_helper pwm_bcm2835 i2c_bcm2835
gpu_sched drm_kms_helper fuse drm
[ 22.666409] CPU: 3 PID: 159 Comm: systemd-udevd Tainted: G D
6.4.0-rc7-next-20230621 #1
[ 22.666424] Hardware name: Raspberry Pi 4 Model B (DT)
[ 22.666431] pstate: 600000c5 (nZCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 22.666443] pc : rcu_note_context_switch+0x458/0x530
[ 22.666463] lr : rcu_note_context_switch+0x458/0x530
[ 22.666478] sp : ffff800083cdb3a0
[ 22.666483] x29: ffff800083cdb3a0 x28: ffff0000423acc08 x27: 0000000000000000
[ 22.666506] x26: ffff00004205a080 x25: ffff80008146d7c8 x24: 0000000000000000
[ 22.666525] x23: 0000000000000000 x22: ffff00004205a080 x21: ffff80008335dd18
[ 22.666544] x20: ffff80008237d2c0 x19: ffff0000fb5f1140 x18: ffffffffffffffff
[ 22.666562] x17: 3035663432313830 x16: 3030386666666620 x15: 3a20397820333535
[ 22.666581] x14: 3431356534353564 x13: 216e6f6974636573 x12: 206c616369746972
[ 22.666600] x11: 6320656469732d64 x10: 6165722055435220 x9 : ffff80008012b740
[ 22.752425] x8 : 6863746977732074 x7 : 7865746e6f632079 x6 : 7261746e756c6f56
[ 22.759681] x5 : 0000000000000000 x4 : 0000000000000000 x3 : 0000000000000027
[ 22.766935] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff00004205a080
[ 22.774190] Call trace:
[ 22.776668] rcu_note_context_switch+0x458/0x530
[ 22.781364] __schedule+0xb8/0xd78
[ 22.784824] schedule+0x60/0x100
[ 22.788105] netlink_table_grab.part.0+0x8c/0xf8
[ 22.792801] netlink_release+0x5dc/0x6d8
[ 22.796782] __sock_release+0x4c/0xc8
[ 22.800505] sock_close+0x20/0x38
[ 22.803872] __fput+0xbc/0x280
[ 22.806979] ____fput+0x18/0x30
[ 22.810172] task_work_run+0x78/0xd8
[ 22.813802] do_exit+0x2f8/0x9a8
[ 22.817085] make_task_dead+0xa4/0x1a8
[ 22.820896] die+0x254/0x260
[ 22.823819] arm64_notify_die+0xbc/0xe0
[ 22.827712] do_debug_exception+0xe0/0x118
[ 22.831876] el1_dbg+0x70/0x90
[ 22.834975] el1h_64_sync_handler+0xc8/0xe8
[ 22.839222] el1h_64_sync+0x64/0x68
[ 22.842761] 0xffff8000835d6580
[ 22.845949] netlink_broadcast+0x1f0/0x4e8
[ 22.850105] netlink_sendmsg+0x318/0x420
[ 22.854084] ____sys_sendmsg+0x1cc/0x2c8
[ 22.858072] ___sys_sendmsg+0x88/0xf0
[ 22.861788] __sys_sendmsg+0x70/0xd8
[ 22.865414] __arm64_sys_sendmsg+0x2c/0x40
[ 22.869571] invoke_syscall+0x50/0x128
[ 22.873383] el0_svc_common.constprop.0+0xf4/0x120
[ 22.878251] do_el0_svc+0x44/0xb8
[ 22.881621] el0_svc+0x30/0x98
[ 22.884721] el0t_64_sync_handler+0x13c/0x158
[ 22.889144] el0t_64_sync+0x190/0x198
[ 22.892860] ---[ end trace 0000000000000000 ]---
Links:
- https://lkft.validation.linaro.org/scheduler/job/6531518#L886
- https://qa-reports.linaro.org/lkft/linux-next-master-sanity/build/next-20230621/testrun/17697990/suite/log-parser-test/tests/
- https://qa-reports.linaro.org/lkft/linux-next-master-sanity/build/next-20230621/testrun/17697990/suite/log-parser-test/test/check-kernel-warning/log
metadata:
git_ref: master
git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next
git_sha: 15e71592dbae49a674429c618a10401d7f992ac3
git_describe: next-20230621
kernel_version: 6.4.0-rc7
kernel-config:
https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVA7srTtdxlVq1QVEgrzc0AAKp/config
artifact-location:
https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVA7srTtdxlVq1QVEgrzc0AAKp/
toolchain: gcc-11
--
Linaro LKFT
https://lkft.linaro.org
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: next: Rpi4: Unexpected kernel BRK exception at EL1
2023-06-21 12:36 next: Rpi4: Unexpected kernel BRK exception at EL1 Naresh Kamboju
@ 2023-06-21 12:57 ` Mark Rutland
2023-06-21 13:38 ` Naresh Kamboju
2023-06-22 9:42 ` Mark Rutland
0 siblings, 2 replies; 11+ messages in thread
From: Mark Rutland @ 2023-06-21 12:57 UTC (permalink / raw)
To: Naresh Kamboju
Cc: Linux ARM, open list, linux-rpi-kernel, Netdev, lkft-triage,
Arnd Bergmann, Dan Carpenter, Linus Walleij, Catalin Marinas,
Will Deacon, Anshuman Khandual, Puranjay Mohan, Song Liu,
Alexei Starovoitov
On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> Following boot warnings and crashes noticed on arm64 Rpi4 device running
> Linux next-20230621 kernel.
>
> Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
>
> boot log:
>
> [ 22.331748] Kernel text patching generated an invalid instruction
> at 0xffff8000835d6580!
> [ 22.340579] Unexpected kernel BRK exception at EL1
> [ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
This indicates execution of AARCH64_BREAK_FAULT.
That could be from dodgy arguments to aarch64_insn_gen_*(), or elsewhere, and
given this is in the networking code I suspect this'll be related to BPF.
Looking at next-20230621 I see commit:
49703aa2adfaff28 ("bpf, arm64: use bpf_jit_binary_pack_alloc")
... which changed the way BPF allocates memory, and has code that pads memory
with a bunch of AARCH64_BREAK_FAULT, so it looks like that *might* be related.
Are you able to bisect this?
In the mean time, I've Cc'd the relevant BPF people to give them a heads-up.
Thanks,
Mark.
> .
> [ 22.353814] Modules linked in: hci_uart(+) brcmfmac(+) btqca
> brcmutil btbcm vc4(+) cfg80211 bluetooth reset_raspberrypi
> clk_raspberrypi crct10dif_ce raspberrypi_hwmon snd_soc_hdmi_codec cec
> v3d bcm2711_thermal rfkill drm_display_helper drm_shmem_helper
> pcie_brcmstb drm_dma_helper pwm_bcm2835 i2c_bcm2835 gpu_sched
> drm_kms_helper fuse drm
> [ 22.376947] vc4-drm gpu: bound fe004000.txp (ops vc4_txp_ops [vc4])
> [ 22.384754] CPU: 3 PID: 159 Comm: systemd-udevd Not tainted
> 6.4.0-rc7-next-20230621 #1
> [ 22.384769] Hardware name: Raspberry Pi 4 Model B (DT)
> [ 22.384776] pstate: 40000005 (nZcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [ 22.384789] pc : 0xffff8000835d6580
> [ 22.384798] lr : sk_filter_trim_cap+0x80/0x2a0
> [ 22.384825] sp : ffff800083cdb9d0
> [ 22.384831] x29: ffff800083cdb9d0 x28: 0000000000000000 x27: 0000000000000001
> [ 22.384853] x26: ffff000041ece000 x25: ffff0000423ac800 x24: ffff800083468e00
> [ 22.384872] x23: 0000000000000000 x22: ffff000044c41300 x21: 0000000000000001
> [ 22.384891] x20: ffff800083bed000 x19: ffff000044c41300 x18: 0000000000000000
> [ 22.384909] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000015798650
> [ 22.384928] x14: 414d003633373832 x13: 3332323d44455a49 x12: 4c414954494e495f
> [ 22.384946] x11: 4345535500343331 x10: 343d4d554e514553 x9 : ffff80008124f500
> [ 22.384965] x8 : ffff800083cdb7d8 x7 : 0000000000000000 x6 : 0000000000000001
> [ 22.384983] x5 : ffff800082def000 x4 : ffff800082def2e8 x3 : 0000000000000000
> [ 22.385001] x2 : ffff8000835d657c x1 : ffff800083bed048 x0 : ffff000044c41300
> [ 22.385020] Call trace:
> [ 22.385025] 0xffff8000835d6580
> [ 22.385033] netlink_broadcast+0x1f0/0x4e8
> [ 22.385047] netlink_sendmsg+0x318/0x420
> [ 22.385056] ____sys_sendmsg+0x1cc/0x2c8
> [ 22.385075] ___sys_sendmsg+0x88/0xf0
> [ 22.385084] __sys_sendmsg+0x70/0xd8
> [ 22.385093] __arm64_sys_sendmsg+0x2c/0x40
> [ 22.385102] invoke_syscall+0x50/0x128
> [ 22.385120] el0_svc_common.constprop.0+0xf4/0x120
> [ 22.385136] do_el0_svc+0x44/0xb8
> [ 22.385152] el0_svc+0x30/0x98
> [ 22.385163] el0t_64_sync_handler+0x13c/0x158
> [ 22.385174] el0t_64_sync+0x190/0x198
> [ 22.385190] Code: d4202000 d4202000 d4202000 910003c9 (d503201f)
> [ 22.385199] ---[ end trace 0000000000000000 ]---
> [ 22.385206] note: systemd-udevd[159] exited with irqs disabled
> [ 22.385378] note: systemd-udevd[159] exited with preempt_count 1
> [ 22.395083] uart-pl011 fe201000.serial: no DMA platform data
> [ 22.427073] vc4-drm gpu: bound fe206000.pixelvalve (ops vc4_crtc_ops [vc4])
> [ 22.434295] ------------[ cut here ]------------
> [ 22.474787] Bluetooth: HCI UART protocol Marvell registered
> [ 22.475105] Voluntary context switch within RCU read-side critical section!
> [ 22.475129] WARNING: CPU: 3 PID: 159 at
> kernel/rcu/tree_plugin.h:320 rcu_note_context_switch+0x458/0x530
> [ 22.553049] vc4-drm gpu: bound fe207000.pixelvalve (ops vc4_crtc_ops [vc4])
> [ 22.555403] Modules linked in: hci_uart(+) brcmfmac(+) btqca
> brcmutil btbcm vc4(+) cfg80211 bluetooth reset_raspberrypi
> clk_raspberrypi crct10dif_ce
> [ 22.583877] vc4-drm gpu: bound fe20a000.pixelvalve (ops vc4_crtc_ops [vc4])
> [ 22.584920] raspberrypi_hwmon snd_soc_hdmi_codec
> [ 22.614889] vc4-drm gpu: bound fe216000.pixelvalve (ops vc4_crtc_ops [vc4])
> [ 22.618972] cec v3d bcm2711_thermal rfkill drm_display_helper
> drm_shmem_helper pcie_brcmstb drm_dma_helper pwm_bcm2835 i2c_bcm2835
> gpu_sched drm_kms_helper fuse drm
> [ 22.666409] CPU: 3 PID: 159 Comm: systemd-udevd Tainted: G D
> 6.4.0-rc7-next-20230621 #1
> [ 22.666424] Hardware name: Raspberry Pi 4 Model B (DT)
> [ 22.666431] pstate: 600000c5 (nZCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [ 22.666443] pc : rcu_note_context_switch+0x458/0x530
> [ 22.666463] lr : rcu_note_context_switch+0x458/0x530
> [ 22.666478] sp : ffff800083cdb3a0
> [ 22.666483] x29: ffff800083cdb3a0 x28: ffff0000423acc08 x27: 0000000000000000
> [ 22.666506] x26: ffff00004205a080 x25: ffff80008146d7c8 x24: 0000000000000000
> [ 22.666525] x23: 0000000000000000 x22: ffff00004205a080 x21: ffff80008335dd18
> [ 22.666544] x20: ffff80008237d2c0 x19: ffff0000fb5f1140 x18: ffffffffffffffff
> [ 22.666562] x17: 3035663432313830 x16: 3030386666666620 x15: 3a20397820333535
> [ 22.666581] x14: 3431356534353564 x13: 216e6f6974636573 x12: 206c616369746972
> [ 22.666600] x11: 6320656469732d64 x10: 6165722055435220 x9 : ffff80008012b740
> [ 22.752425] x8 : 6863746977732074 x7 : 7865746e6f632079 x6 : 7261746e756c6f56
> [ 22.759681] x5 : 0000000000000000 x4 : 0000000000000000 x3 : 0000000000000027
> [ 22.766935] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff00004205a080
> [ 22.774190] Call trace:
> [ 22.776668] rcu_note_context_switch+0x458/0x530
> [ 22.781364] __schedule+0xb8/0xd78
> [ 22.784824] schedule+0x60/0x100
> [ 22.788105] netlink_table_grab.part.0+0x8c/0xf8
> [ 22.792801] netlink_release+0x5dc/0x6d8
> [ 22.796782] __sock_release+0x4c/0xc8
> [ 22.800505] sock_close+0x20/0x38
> [ 22.803872] __fput+0xbc/0x280
> [ 22.806979] ____fput+0x18/0x30
> [ 22.810172] task_work_run+0x78/0xd8
> [ 22.813802] do_exit+0x2f8/0x9a8
> [ 22.817085] make_task_dead+0xa4/0x1a8
> [ 22.820896] die+0x254/0x260
> [ 22.823819] arm64_notify_die+0xbc/0xe0
> [ 22.827712] do_debug_exception+0xe0/0x118
> [ 22.831876] el1_dbg+0x70/0x90
> [ 22.834975] el1h_64_sync_handler+0xc8/0xe8
> [ 22.839222] el1h_64_sync+0x64/0x68
> [ 22.842761] 0xffff8000835d6580
> [ 22.845949] netlink_broadcast+0x1f0/0x4e8
> [ 22.850105] netlink_sendmsg+0x318/0x420
> [ 22.854084] ____sys_sendmsg+0x1cc/0x2c8
> [ 22.858072] ___sys_sendmsg+0x88/0xf0
> [ 22.861788] __sys_sendmsg+0x70/0xd8
> [ 22.865414] __arm64_sys_sendmsg+0x2c/0x40
> [ 22.869571] invoke_syscall+0x50/0x128
> [ 22.873383] el0_svc_common.constprop.0+0xf4/0x120
> [ 22.878251] do_el0_svc+0x44/0xb8
> [ 22.881621] el0_svc+0x30/0x98
> [ 22.884721] el0t_64_sync_handler+0x13c/0x158
> [ 22.889144] el0t_64_sync+0x190/0x198
> [ 22.892860] ---[ end trace 0000000000000000 ]---
>
>
> Links:
> - https://lkft.validation.linaro.org/scheduler/job/6531518#L886
> - https://qa-reports.linaro.org/lkft/linux-next-master-sanity/build/next-20230621/testrun/17697990/suite/log-parser-test/tests/
> - https://qa-reports.linaro.org/lkft/linux-next-master-sanity/build/next-20230621/testrun/17697990/suite/log-parser-test/test/check-kernel-warning/log
>
> metadata:
> git_ref: master
> git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next
> git_sha: 15e71592dbae49a674429c618a10401d7f992ac3
> git_describe: next-20230621
> kernel_version: 6.4.0-rc7
> kernel-config:
> https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVA7srTtdxlVq1QVEgrzc0AAKp/config
> artifact-location:
> https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVA7srTtdxlVq1QVEgrzc0AAKp/
> toolchain: gcc-11
>
>
> --
> Linaro LKFT
> https://lkft.linaro.org
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: next: Rpi4: Unexpected kernel BRK exception at EL1
2023-06-21 12:57 ` Mark Rutland
@ 2023-06-21 13:38 ` Naresh Kamboju
2023-06-21 14:16 ` Puranjay Mohan
2023-06-22 9:42 ` Mark Rutland
1 sibling, 1 reply; 11+ messages in thread
From: Naresh Kamboju @ 2023-06-21 13:38 UTC (permalink / raw)
To: Mark Rutland, Puranjay Mohan, Alexei Starovoitov, Song Liu
Cc: Linux ARM, open list, linux-rpi-kernel, Netdev, lkft-triage,
Arnd Bergmann, Dan Carpenter, Linus Walleij, Catalin Marinas,
Will Deacon, Anshuman Khandual, bpf
On Wed, 21 Jun 2023 at 18:27, Mark Rutland <mark.rutland@arm.com> wrote:
>
> On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> > Following boot warnings and crashes noticed on arm64 Rpi4 device running
> > Linux next-20230621 kernel.
> >
> > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> >
> > boot log:
> >
> > [ 22.331748] Kernel text patching generated an invalid instruction
> > at 0xffff8000835d6580!
> > [ 22.340579] Unexpected kernel BRK exception at EL1
> > [ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
>
> This indicates execution of AARCH64_BREAK_FAULT.
I see kernel panic with kselftest merge configs on Juno-r2 and Rpi4.
>
> That could be from dodgy arguments to aarch64_insn_gen_*(), or elsewhere, and
> given this is in the networking code I suspect this'll be related to BPF.
>
> Looking at next-20230621 I see commit:
>
> 49703aa2adfaff28 ("bpf, arm64: use bpf_jit_binary_pack_alloc")
>
> ... which changed the way BPF allocates memory, and has code that pads memory
> with a bunch of AARCH64_BREAK_FAULT, so it looks like that *might* be related.
>
> Are you able to bisect this?
I have not started bisection on this issue yet.
Let me give it a try.
>
> In the mean time, I've Cc'd the relevant BPF people to give them a heads-up.
Thanks.
Extra information from boot failures.
This is always reproducible on Juno-r2 and Rpi4 devices.
Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
Boot crash log:
[ 3.605232] Kernel text patching generated an invalid instruction
at bpf_prog_99a0cd861b84ee07___loader.prog+0x0/0x728!
[ 3.616052] Unexpected kernel BRK exception at EL1
[ 3.620849] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
[ 3.627736] Modules linked in:
[ 3.630796] CPU: 1 PID: 1 Comm: swapper/0 Not tainted
6.4.0-rc7-next-20230621 #1
[ 3.638140] hub 1-1:1.0: USB hub found
[ 3.638206] Hardware name: ARM Juno development board (r2) (DT)
[ 3.638210] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 3.642431] hub 1-1:1.0: 4 ports detected
[ 3.647879] pc : bpf_prog_99a0cd861b84ee07___loader.prog+0x0/0x728
[ 3.647891] lr : kern_sys_bpf+0x130/0x218
[ 3.669061] sp : ffff80008391bc10
[ 3.672376] x29: ffff80008391bc10 x28: ffff8000826e70d8 x27: ffff800082450110
[ 3.679533] x26: ffff8000820ed948 x25: ffff800082427b10 x24: 0000000000000289
[ 3.686687] x23: ffff000800acfa00 x22: ffff8000837f8000 x21: ffff000823dbc240
[ 3.693841] x20: ffff8000839b1000 x19: ffff80008391bca8 x18: 000000001d03406d
[ 3.700995] x17: ffff800080464204 x16: ffff8000804640b4 x15: ffff8000803f8af0
[ 3.708149] x14: ffff8000803f88f8 x13: ffff800081717720 x12: ffff8000824514b4
[ 3.715302] x11: ffff800080015788 x10: ffff800082470304 x9 : ffff8000800f3338
[ 3.722456] x8 : ffff80008391bcf8 x7 : 0000000000000000 x6 : 0000000000000001
[ 3.729609] x5 : 0000000000000001 x4 : ffff8000831f0000 x3 : ffff8008fc63d000
[ 3.736763] x2 : ffff800083b6d88c x1 : ffff8000839b1048 x0 : ffff000823dbc240
[ 3.743917] Call trace:
[ 3.746362] bpf_prog_99a0cd861b84ee07___loader.prog+0x0/0x728
[ 3.752210] bpf_load_and_run.constprop.0+0x120/0x1d8
[ 3.757270] load+0xf4/0x278
[ 3.760159] do_one_initcall+0x50/0x2f0
[ 3.764001] kernel_init_freeable+0x224/0x438
[ 3.768368] kernel_init+0x30/0x200
[ 3.771862] ret_from_fork+0x10/0x20
[ 3.775447] Code: d4202000 00000780 d4202000 d4202000 (910003c9)
[ 3.781550] ---[ end trace 0000000000000000 ]---
[ 3.786172] note: swapper/0[1] exited with irqs disabled
[ 3.791526] note: swapper/0[1] exited with preempt_count 1
[ 3.797043] Kernel panic - not syncing: Attempted to kill init!
exitcode=0x0000000b
[ 3.804711] SMP: stopping secondary CPUs
[ 3.808843] Kernel Offset: disabled
[ 3.812331] CPU features: 0x40000106,1e010000,0000421b
[ 3.817476] Memory Limit: none
[ 3.820536] ---[ end Kernel panic - not syncing: Attempted to kill
init! exitcode=0x0000000b ]---
Links:
https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20230621/testrun/17701148/suite/log-parser-test/tests/
https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20230621/testrun/17701148/suite/log-parser-test/test/check-kernel-panic/log
https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20230621/testrun/17701148/suite/log-parser-test/test/check-kernel-panic/details/
metadata:
git_ref: master
git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next
git_sha: 15e71592dbae49a674429c618a10401d7f992ac3
git_describe: next-20230621
kernel_version: 6.4.0-rc7
kernel-config:
https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVAA4lj35ia3YDkqaoV6ztyqdW/config
artifact-location:
https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVAA4lj35ia3YDkqaoV6ztyqdW/
toolchain: gcc-11
build_name: gcc-11-lkftconfig-kselftest
--
Linaro LKFT
https://lkft.linaro.org
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: next: Rpi4: Unexpected kernel BRK exception at EL1
2023-06-21 13:38 ` Naresh Kamboju
@ 2023-06-21 14:16 ` Puranjay Mohan
2023-06-21 14:40 ` Naresh Kamboju
0 siblings, 1 reply; 11+ messages in thread
From: Puranjay Mohan @ 2023-06-21 14:16 UTC (permalink / raw)
To: Naresh Kamboju
Cc: Mark Rutland, Alexei Starovoitov, Song Liu, Linux ARM, open list,
linux-rpi-kernel, Netdev, lkft-triage, Arnd Bergmann,
Dan Carpenter, Linus Walleij, Catalin Marinas, Will Deacon,
Anshuman Khandual, bpf
Hi,
On Wed, Jun 21, 2023 at 3:39 PM Naresh Kamboju
<naresh.kamboju@linaro.org> wrote:
>
> On Wed, 21 Jun 2023 at 18:27, Mark Rutland <mark.rutland@arm.com> wrote:
> >
> > On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> > > Following boot warnings and crashes noticed on arm64 Rpi4 device running
> > > Linux next-20230621 kernel.
> > >
> > > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> > >
> > > boot log:
> > >
> > > [ 22.331748] Kernel text patching generated an invalid instruction
> > > at 0xffff8000835d6580!
> > > [ 22.340579] Unexpected kernel BRK exception at EL1
> > > [ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
> >
> > This indicates execution of AARCH64_BREAK_FAULT.
>
> I see kernel panic with kselftest merge configs on Juno-r2 and Rpi4.
Is there a way to reproduce this setup on Qemu?
I am able to build the linux-next kernel with the config given below.
But the bug doesn't reproduce in Qemu with debian rootfs.
I guess I would need the Rootfs that is being used here to reproduce it.
Can you point me to the rootfs for this?
Thanks,
Puranjay
> metadata:
> git_ref: master
> git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next
> git_sha: 15e71592dbae49a674429c618a10401d7f992ac3
> git_describe: next-20230621
> kernel_version: 6.4.0-rc7
> kernel-config:
> https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVAA4lj35ia3YDkqaoV6ztyqdW/config
> artifact-location:
> https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVAA4lj35ia3YDkqaoV6ztyqdW/
> toolchain: gcc-11
> build_name: gcc-11-lkftconfig-kselftest
>
>
> --
> Linaro LKFT
> https://lkft.linaro.org
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: next: Rpi4: Unexpected kernel BRK exception at EL1
2023-06-21 14:16 ` Puranjay Mohan
@ 2023-06-21 14:40 ` Naresh Kamboju
2023-06-21 15:19 ` Puranjay Mohan
0 siblings, 1 reply; 11+ messages in thread
From: Naresh Kamboju @ 2023-06-21 14:40 UTC (permalink / raw)
To: Puranjay Mohan
Cc: Mark Rutland, Alexei Starovoitov, Song Liu, Linux ARM, open list,
linux-rpi-kernel, Netdev, lkft-triage, Arnd Bergmann,
Dan Carpenter, Linus Walleij, Catalin Marinas, Will Deacon,
Anshuman Khandual, bpf
On Wed, 21 Jun 2023 at 19:46, Puranjay Mohan <puranjay12@gmail.com> wrote:
>
> Hi,
>
> On Wed, Jun 21, 2023 at 3:39 PM Naresh Kamboju
> <naresh.kamboju@linaro.org> wrote:
> >
> > On Wed, 21 Jun 2023 at 18:27, Mark Rutland <mark.rutland@arm.com> wrote:
> > >
> > > On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> > > > Following boot warnings and crashes noticed on arm64 Rpi4 device running
> > > > Linux next-20230621 kernel.
> > > >
> > > > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> > > >
> > > > boot log:
> > > >
> > > > [ 22.331748] Kernel text patching generated an invalid instruction
> > > > at 0xffff8000835d6580!
> > > > [ 22.340579] Unexpected kernel BRK exception at EL1
> > > > [ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
> > >
> > > This indicates execution of AARCH64_BREAK_FAULT.
> >
> > I see kernel panic with kselftest merge configs on Juno-r2 and Rpi4.
>
> Is there a way to reproduce this setup on Qemu?
Not reproducible on Qemu-arm64.
I see only on arm64 devices Juno-r2 and Rpi4.
>
> I am able to build the linux-next kernel with the config given below.
> But the bug doesn't reproduce in Qemu with debian rootfs.
>
> I guess I would need the Rootfs that is being used here to reproduce it.
> Can you point me to the rootfs for this?
Here is the link for rootfs - OE one.
https://storage.tuxsuite.com/public/linaro/lkft/oebuilds/2RVA7dHPf73agY0gDJD6XEdBQBI/images/juno/
>
> Thanks,
> Puranjay
>
> > metadata:
> > git_ref: master
> > git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next
> > git_sha: 15e71592dbae49a674429c618a10401d7f992ac3
> > git_describe: next-20230621
> > kernel_version: 6.4.0-rc7
> > kernel-config:
> > https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVAA4lj35ia3YDkqaoV6ztyqdW/config
> > artifact-location:
> > https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVAA4lj35ia3YDkqaoV6ztyqdW/
> > toolchain: gcc-11
> > build_name: gcc-11-lkftconfig-kselftest
> >
> >
> > --
> > Linaro LKFT
> > https://lkft.linaro.org
- Naresh
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: next: Rpi4: Unexpected kernel BRK exception at EL1
2023-06-21 14:40 ` Naresh Kamboju
@ 2023-06-21 15:19 ` Puranjay Mohan
0 siblings, 0 replies; 11+ messages in thread
From: Puranjay Mohan @ 2023-06-21 15:19 UTC (permalink / raw)
To: Naresh Kamboju
Cc: Mark Rutland, Alexei Starovoitov, Song Liu, Linux ARM, open list,
linux-rpi-kernel, Netdev, lkft-triage, Arnd Bergmann,
Dan Carpenter, Linus Walleij, Catalin Marinas, Will Deacon,
Anshuman Khandual, bpf
Hi,
On Wed, Jun 21, 2023 at 4:41 PM Naresh Kamboju
<naresh.kamboju@linaro.org> wrote:
>
> On Wed, 21 Jun 2023 at 19:46, Puranjay Mohan <puranjay12@gmail.com> wrote:
> >
> > Hi,
> >
> > On Wed, Jun 21, 2023 at 3:39 PM Naresh Kamboju
> > <naresh.kamboju@linaro.org> wrote:
> > >
> > > On Wed, 21 Jun 2023 at 18:27, Mark Rutland <mark.rutland@arm.com> wrote:
> > > >
> > > > On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> > > > > Following boot warnings and crashes noticed on arm64 Rpi4 device running
> > > > > Linux next-20230621 kernel.
> > > > >
> > > > > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> > > > >
> > > > > boot log:
> > > > >
> > > > > [ 22.331748] Kernel text patching generated an invalid instruction
> > > > > at 0xffff8000835d6580!
> > > > > [ 22.340579] Unexpected kernel BRK exception at EL1
> > > > > [ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
> > > >
> > > > This indicates execution of AARCH64_BREAK_FAULT.
> > >
> > > I see kernel panic with kselftest merge configs on Juno-r2 and Rpi4.
> >
> > Is there a way to reproduce this setup on Qemu?
>
> Not reproducible on Qemu-arm64.
> I see only on arm64 devices Juno-r2 and Rpi4.
>
> >
> > I am able to build the linux-next kernel with the config given below.
> > But the bug doesn't reproduce in Qemu with debian rootfs.
> >
> > I guess I would need the Rootfs that is being used here to reproduce it.
> > Can you point me to the rootfs for this?
>
> Here is the link for rootfs - OE one.
> https://storage.tuxsuite.com/public/linaro/lkft/oebuilds/2RVA7dHPf73agY0gDJD6XEdBQBI/images/juno/
I tested this rootfs and couldn't reproduce on Qemu. Now, I will try
to use my raspberry pi and try to reproduce this.
Thanks.
>
> >
> > Thanks,
> > Puranjay
> >
> > > metadata:
> > > git_ref: master
> > > git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next
> > > git_sha: 15e71592dbae49a674429c618a10401d7f992ac3
> > > git_describe: next-20230621
> > > kernel_version: 6.4.0-rc7
> > > kernel-config:
> > > https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVAA4lj35ia3YDkqaoV6ztyqdW/config
> > > artifact-location:
> > > https://storage.tuxsuite.com/public/linaro/lkft/builds/2RVAA4lj35ia3YDkqaoV6ztyqdW/
> > > toolchain: gcc-11
> > > build_name: gcc-11-lkftconfig-kselftest
> > >
> > >
> > > --
> > > Linaro LKFT
> > > https://lkft.linaro.org
>
> - Naresh
--
Thanks and Regards
Yours Truly,
Puranjay Mohan
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: next: Rpi4: Unexpected kernel BRK exception at EL1
2023-06-21 12:57 ` Mark Rutland
2023-06-21 13:38 ` Naresh Kamboju
@ 2023-06-22 9:42 ` Mark Rutland
2023-06-22 12:35 ` Naresh Kamboju
1 sibling, 1 reply; 11+ messages in thread
From: Mark Rutland @ 2023-06-22 9:42 UTC (permalink / raw)
To: Naresh Kamboju
Cc: Linux ARM, open list, linux-rpi-kernel, Netdev, lkft-triage,
Arnd Bergmann, Dan Carpenter, Linus Walleij, Catalin Marinas,
Will Deacon, Anshuman Khandual, Puranjay Mohan, Song Liu,
Alexei Starovoitov
On Wed, Jun 21, 2023 at 01:57:21PM +0100, Mark Rutland wrote:
> On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> > Following boot warnings and crashes noticed on arm64 Rpi4 device running
> > Linux next-20230621 kernel.
> >
> > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> >
> > boot log:
> >
> > [ 22.331748] Kernel text patching generated an invalid instruction
> > at 0xffff8000835d6580!
> > [ 22.340579] Unexpected kernel BRK exception at EL1
> > [ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
>
> This indicates execution of AARCH64_BREAK_FAULT.
>
> That could be from dodgy arguments to aarch64_insn_gen_*(), or elsewhere, and
> given this is in the networking code I suspect this'll be related to BPF.
>
> Looking at next-20230621 I see commit:
>
> 49703aa2adfaff28 ("bpf, arm64: use bpf_jit_binary_pack_alloc")
>
> ... which changed the way BPF allocates memory, and has code that pads memory
> with a bunch of AARCH64_BREAK_FAULT, so it looks like that *might* be related.
For the benefit of those just looknig at this thread, there has been some
discussion in the original thread for this commit. Summary and links below.
We identified a potential issue with missing cache maintenance:
https://lore.kernel.org/linux-arm-kernel/ZJMXqTffB22LSOkd@FVFF77S0Q05N/
Puranjay verified that was causing the problem seen here:
https://lore.kernel.org/linux-arm-kernel/CANk7y0h5ucxmMz4K8sGx7qogFyx6PRxYxmFtwTRO7=0Y=B4ugw@mail.gmail.com/
Alexei has dropped this commit for now:
https://lore.kernel.org/linux-arm-kernel/CAADnVQJqDOMABEx8JuU6r_Dehyf=SkDfRNChx1oNfqPoo7pSrw@mail.gmail.com/
Thanks,
Mark.
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: next: Rpi4: Unexpected kernel BRK exception at EL1
2023-06-22 9:42 ` Mark Rutland
@ 2023-06-22 12:35 ` Naresh Kamboju
2023-06-26 9:04 ` Puranjay Mohan
0 siblings, 1 reply; 11+ messages in thread
From: Naresh Kamboju @ 2023-06-22 12:35 UTC (permalink / raw)
To: Mark Rutland
Cc: Linux ARM, open list, linux-rpi-kernel, Netdev, lkft-triage,
Arnd Bergmann, Dan Carpenter, Linus Walleij, Catalin Marinas,
Will Deacon, Anshuman Khandual, Puranjay Mohan, Song Liu,
Alexei Starovoitov
Hi Mark,
On Thu, 22 Jun 2023 at 15:12, Mark Rutland <mark.rutland@arm.com> wrote:
>
> On Wed, Jun 21, 2023 at 01:57:21PM +0100, Mark Rutland wrote:
> > On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> > > Following boot warnings and crashes noticed on arm64 Rpi4 device running
> > > Linux next-20230621 kernel.
> > >
> > > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> > >
> > > boot log:
> > >
> > > [ 22.331748] Kernel text patching generated an invalid instruction
> > > at 0xffff8000835d6580!
> > > [ 22.340579] Unexpected kernel BRK exception at EL1
> > > [ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
> >
> > This indicates execution of AARCH64_BREAK_FAULT.
> >
> > That could be from dodgy arguments to aarch64_insn_gen_*(), or elsewhere, and
> > given this is in the networking code I suspect this'll be related to BPF.
> >
> > Looking at next-20230621 I see commit:
> >
> > 49703aa2adfaff28 ("bpf, arm64: use bpf_jit_binary_pack_alloc")
> >
> > ... which changed the way BPF allocates memory, and has code that pads memory
> > with a bunch of AARCH64_BREAK_FAULT, so it looks like that *might* be related.
>
> For the benefit of those just looknig at this thread, there has been some
> discussion in the original thread for this commit. Summary and links below.
>
> We identified a potential issue with missing cache maintenance:
>
> https://lore.kernel.org/linux-arm-kernel/ZJMXqTffB22LSOkd@FVFF77S0Q05N/
>
> Puranjay verified that was causing the problem seen here:
>
> https://lore.kernel.org/linux-arm-kernel/CANk7y0h5ucxmMz4K8sGx7qogFyx6PRxYxmFtwTRO7=0Y=B4ugw@mail.gmail.com/
>
> Alexei has dropped this commit for now:
>
> https://lore.kernel.org/linux-arm-kernel/CAADnVQJqDOMABEx8JuU6r_Dehyf=SkDfRNChx1oNfqPoo7pSrw@mail.gmail.com/
Thanks for the detailed information.
I am happy to test any proposed fix patches.
>
> Thanks,
> Mark.
- Naresh
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: next: Rpi4: Unexpected kernel BRK exception at EL1
2023-06-22 12:35 ` Naresh Kamboju
@ 2023-06-26 9:04 ` Puranjay Mohan
2023-07-03 9:36 ` Puranjay Mohan
0 siblings, 1 reply; 11+ messages in thread
From: Puranjay Mohan @ 2023-06-26 9:04 UTC (permalink / raw)
To: Naresh Kamboju
Cc: Mark Rutland, Linux ARM, open list, linux-rpi-kernel, Netdev,
lkft-triage, Arnd Bergmann, Dan Carpenter, Linus Walleij,
Catalin Marinas, Will Deacon, Anshuman Khandual, Song Liu,
Alexei Starovoitov
Hi Naresh,
On Thu, Jun 22, 2023 at 2:35 PM Naresh Kamboju
<naresh.kamboju@linaro.org> wrote:
>
> Hi Mark,
>
> On Thu, 22 Jun 2023 at 15:12, Mark Rutland <mark.rutland@arm.com> wrote:
> >
> > On Wed, Jun 21, 2023 at 01:57:21PM +0100, Mark Rutland wrote:
> > > On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> > > > Following boot warnings and crashes noticed on arm64 Rpi4 device running
> > > > Linux next-20230621 kernel.
> > > >
> > > > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> > > >
> > > > boot log:
> > > >
> > > > [ 22.331748] Kernel text patching generated an invalid instruction
> > > > at 0xffff8000835d6580!
> > > > [ 22.340579] Unexpected kernel BRK exception at EL1
> > > > [ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
> > >
> > > This indicates execution of AARCH64_BREAK_FAULT.
> > >
> > > That could be from dodgy arguments to aarch64_insn_gen_*(), or elsewhere, and
> > > given this is in the networking code I suspect this'll be related to BPF.
> > >
> > > Looking at next-20230621 I see commit:
> > >
> > > 49703aa2adfaff28 ("bpf, arm64: use bpf_jit_binary_pack_alloc")
> > >
> > > ... which changed the way BPF allocates memory, and has code that pads memory
> > > with a bunch of AARCH64_BREAK_FAULT, so it looks like that *might* be related.
> >
> > For the benefit of those just looknig at this thread, there has been some
> > discussion in the original thread for this commit. Summary and links below.
> >
> > We identified a potential issue with missing cache maintenance:
> >
> > https://lore.kernel.org/linux-arm-kernel/ZJMXqTffB22LSOkd@FVFF77S0Q05N/
> >
> > Puranjay verified that was causing the problem seen here:
> >
> > https://lore.kernel.org/linux-arm-kernel/CANk7y0h5ucxmMz4K8sGx7qogFyx6PRxYxmFtwTRO7=0Y=B4ugw@mail.gmail.com/
> >
> > Alexei has dropped this commit for now:
> >
> > https://lore.kernel.org/linux-arm-kernel/CAADnVQJqDOMABEx8JuU6r_Dehyf=SkDfRNChx1oNfqPoo7pSrw@mail.gmail.com/
>
> Thanks for the detailed information.
> I am happy to test any proposed fix patches.
I have sent the v4 of the patch series:
https://lore.kernel.org/bpf/20230626085811.3192402-1-puranjay12@gmail.com/T/#t
This works on my raspberry pi 4 setup. If possible can you test this
on the similar setup where it was failing earlier?
>
>
> >
> > Thanks,
> > Mark.
>
> - Naresh
Thanks,
Puranjay
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: next: Rpi4: Unexpected kernel BRK exception at EL1
2023-06-26 9:04 ` Puranjay Mohan
@ 2023-07-03 9:36 ` Puranjay Mohan
2023-07-25 6:33 ` Naresh Kamboju
0 siblings, 1 reply; 11+ messages in thread
From: Puranjay Mohan @ 2023-07-03 9:36 UTC (permalink / raw)
To: Naresh Kamboju
Cc: Mark Rutland, Linux ARM, open list, linux-rpi-kernel, Netdev,
lkft-triage, Arnd Bergmann, Dan Carpenter, Linus Walleij,
Catalin Marinas, Will Deacon, Anshuman Khandual, Song Liu,
Alexei Starovoitov
Hi Naresh,
On Mon, Jun 26, 2023 at 11:04 AM Puranjay Mohan <puranjay12@gmail.com> wrote:
>
> Hi Naresh,
>
> On Thu, Jun 22, 2023 at 2:35 PM Naresh Kamboju
> <naresh.kamboju@linaro.org> wrote:
> >
> > Hi Mark,
> >
> > On Thu, 22 Jun 2023 at 15:12, Mark Rutland <mark.rutland@arm.com> wrote:
> > >
> > > On Wed, Jun 21, 2023 at 01:57:21PM +0100, Mark Rutland wrote:
> > > > On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> > > > > Following boot warnings and crashes noticed on arm64 Rpi4 device running
> > > > > Linux next-20230621 kernel.
> > > > >
> > > > > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> > > > >
> > > > > boot log:
> > > > >
> > > > > [ 22.331748] Kernel text patching generated an invalid instruction
> > > > > at 0xffff8000835d6580!
> > > > > [ 22.340579] Unexpected kernel BRK exception at EL1
> > > > > [ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
> > > >
> > > > This indicates execution of AARCH64_BREAK_FAULT.
> > > >
> > > > That could be from dodgy arguments to aarch64_insn_gen_*(), or elsewhere, and
> > > > given this is in the networking code I suspect this'll be related to BPF.
> > > >
> > > > Looking at next-20230621 I see commit:
> > > >
> > > > 49703aa2adfaff28 ("bpf, arm64: use bpf_jit_binary_pack_alloc")
> > > >
> > > > ... which changed the way BPF allocates memory, and has code that pads memory
> > > > with a bunch of AARCH64_BREAK_FAULT, so it looks like that *might* be related.
> > >
> > > For the benefit of those just looknig at this thread, there has been some
> > > discussion in the original thread for this commit. Summary and links below.
> > >
> > > We identified a potential issue with missing cache maintenance:
> > >
> > > https://lore.kernel.org/linux-arm-kernel/ZJMXqTffB22LSOkd@FVFF77S0Q05N/
> > >
> > > Puranjay verified that was causing the problem seen here:
> > >
> > > https://lore.kernel.org/linux-arm-kernel/CANk7y0h5ucxmMz4K8sGx7qogFyx6PRxYxmFtwTRO7=0Y=B4ugw@mail.gmail.com/
> > >
> > > Alexei has dropped this commit for now:
> > >
> > > https://lore.kernel.org/linux-arm-kernel/CAADnVQJqDOMABEx8JuU6r_Dehyf=SkDfRNChx1oNfqPoo7pSrw@mail.gmail.com/
> >
> > Thanks for the detailed information.
> > I am happy to test any proposed fix patches.
>
> I have sent the v4 of the patch series:
> https://lore.kernel.org/bpf/20230626085811.3192402-1-puranjay12@gmail.com/T/#t
> This works on my raspberry pi 4 setup. If possible can you test this
> on the similar setup where it was failing earlier?
I think my previous email was missed.
Can you test the V4 series in the same setup?
This is still not applied to the bpf-next tree.
Thanks,
Puranjay.
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: next: Rpi4: Unexpected kernel BRK exception at EL1
2023-07-03 9:36 ` Puranjay Mohan
@ 2023-07-25 6:33 ` Naresh Kamboju
0 siblings, 0 replies; 11+ messages in thread
From: Naresh Kamboju @ 2023-07-25 6:33 UTC (permalink / raw)
To: Puranjay Mohan
Cc: Mark Rutland, Linux ARM, open list, linux-rpi-kernel, Netdev,
lkft-triage, Arnd Bergmann, Dan Carpenter, Linus Walleij,
Catalin Marinas, Will Deacon, Anshuman Khandual, Song Liu,
Alexei Starovoitov
Hi Puranjay,
On Mon, 3 Jul 2023 at 15:07, Puranjay Mohan <puranjay12@gmail.com> wrote:
>
> Hi Naresh,
>
> On Mon, Jun 26, 2023 at 11:04 AM Puranjay Mohan <puranjay12@gmail.com> wrote:
> >
> > Hi Naresh,
> >
> > On Thu, Jun 22, 2023 at 2:35 PM Naresh Kamboju
> > <naresh.kamboju@linaro.org> wrote:
> > >
> > > Hi Mark,
> > >
> > > On Thu, 22 Jun 2023 at 15:12, Mark Rutland <mark.rutland@arm.com> wrote:
> > > >
> > > > On Wed, Jun 21, 2023 at 01:57:21PM +0100, Mark Rutland wrote:
> > > > > On Wed, Jun 21, 2023 at 06:06:51PM +0530, Naresh Kamboju wrote:
> > > > > > Following boot warnings and crashes noticed on arm64 Rpi4 device running
> > > > > > Linux next-20230621 kernel.
> > > > > >
> > > > > > Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> > > > > >
> > > > > > boot log:
> > > > > >
> > > > > > [ 22.331748] Kernel text patching generated an invalid instruction
> > > > > > at 0xffff8000835d6580!
> > > > > > [ 22.340579] Unexpected kernel BRK exception at EL1
> > > > > > [ 22.346141] Internal error: BRK handler: 00000000f2000100 [#1] PREEMPT SMP
> > > > >
> > > > > This indicates execution of AARCH64_BREAK_FAULT.
> > > > >
> > > > > That could be from dodgy arguments to aarch64_insn_gen_*(), or elsewhere, and
> > > > > given this is in the networking code I suspect this'll be related to BPF.
> > > > >
> > > > > Looking at next-20230621 I see commit:
> > > > >
> > > > > 49703aa2adfaff28 ("bpf, arm64: use bpf_jit_binary_pack_alloc")
> > > > >
> > > > > ... which changed the way BPF allocates memory, and has code that pads memory
> > > > > with a bunch of AARCH64_BREAK_FAULT, so it looks like that *might* be related.
> > > >
> > > > For the benefit of those just looknig at this thread, there has been some
> > > > discussion in the original thread for this commit. Summary and links below.
> > > >
> > > > We identified a potential issue with missing cache maintenance:
> > > >
> > > > https://lore.kernel.org/linux-arm-kernel/ZJMXqTffB22LSOkd@FVFF77S0Q05N/
> > > >
> > > > Puranjay verified that was causing the problem seen here:
> > > >
> > > > https://lore.kernel.org/linux-arm-kernel/CANk7y0h5ucxmMz4K8sGx7qogFyx6PRxYxmFtwTRO7=0Y=B4ugw@mail.gmail.com/
> > > >
> > > > Alexei has dropped this commit for now:
> > > >
> > > > https://lore.kernel.org/linux-arm-kernel/CAADnVQJqDOMABEx8JuU6r_Dehyf=SkDfRNChx1oNfqPoo7pSrw@mail.gmail.com/
> > >
> > > Thanks for the detailed information.
> > > I am happy to test any proposed fix patches.
> >
> > I have sent the v4 of the patch series:
> > https://lore.kernel.org/bpf/20230626085811.3192402-1-puranjay12@gmail.com/T/#t
> > This works on my raspberry pi 4 setup. If possible can you test this
> > on the similar setup where it was failing earlier?
>
> I think my previous email was missed.
> Can you test the V4 series in the same setup?
I have tested V4 series and reported issues got fixed.
Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Thank you !
> This is still not applied to the bpf-next tree.
>
> Thanks,
> Puranjay.
- Naresh
^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2023-07-25 6:34 UTC | newest]
Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-06-21 12:36 next: Rpi4: Unexpected kernel BRK exception at EL1 Naresh Kamboju
2023-06-21 12:57 ` Mark Rutland
2023-06-21 13:38 ` Naresh Kamboju
2023-06-21 14:16 ` Puranjay Mohan
2023-06-21 14:40 ` Naresh Kamboju
2023-06-21 15:19 ` Puranjay Mohan
2023-06-22 9:42 ` Mark Rutland
2023-06-22 12:35 ` Naresh Kamboju
2023-06-26 9:04 ` Puranjay Mohan
2023-07-03 9:36 ` Puranjay Mohan
2023-07-25 6:33 ` Naresh Kamboju
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).