On Mon, 13 Apr 2020 at 13:56, Greg KH gregkh@linuxfoundation.org wrote:
On Mon, Apr 13, 2020 at 05:03:26AM +0000, nobuhiro1.iwamatsu@toshiba.co.jp wrote:
Hi,
-----Original Message----- From: stable-owner@vger.kernel.org [mailto:stable-owner@vger.kernel.org] On Behalf Of Greg Kroah-Hartman Sent: Saturday, April 11, 2020 9:10 PM To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman gregkh@linuxfoundation.org; stable@vger.kernel.org; Stephen Boyd sboyd@kernel.org; Stephen Boyd swboyd@chromium.org; Jordan Crouse jcrouse@codeaurora.org; Rob Clark robdclark@chromium.org; Sean Paul seanpaul@chromium.org; Lee Jones lee.jones@linaro.org Subject: [PATCH 4.19 50/54] drm/msm: stop abusing dma_map/unmap for cache
From: Rob Clark robdclark@chromium.org
commit 0036bc73ccbe7e600a3468bf8e8879b122252274 upstream.
Recently splats like this started showing up:
WARNING: CPU: 4 PID: 251 at drivers/iommu/dma-iommu.c:451 __iommu_dma_unmap+0xb8/0xc0 Modules linked in: ath10k_snoc ath10k_core fuse msm ath mac80211 uvcvideo cfg80211 videobuf2_vmalloc videobuf2_memops vide CPU: 4 PID: 251 Comm: kworker/u16:4 Tainted: G W 5.2.0-rc5-next-20190619+ #2317 Hardware name: LENOVO 81JL/LNVNB161216, BIOS 9UCN23WW(V1.06) 10/25/2018 Workqueue: msm msm_gem_free_work [msm] pstate: 80c00005 (Nzcv daif +PAN +UAO) pc : __iommu_dma_unmap+0xb8/0xc0 lr : __iommu_dma_unmap+0x54/0xc0 sp : ffff0000119abce0 x29: ffff0000119abce0 x28: 0000000000000000 x27: ffff8001f9946648 x26: ffff8001ec271068 x25: 0000000000000000 x24: ffff8001ea3580a8 x23: ffff8001f95ba010 x22: ffff80018e83ba88 x21: ffff8001e548f000 x20: fffffffffffff000 x19: 0000000000001000 x18: 00000000c00001fe x17: 0000000000000000 x16: 0000000000000000 x15: ffff000015b70068 x14: 0000000000000005 x13: 0003142cc1be1768 x12: 0000000000000001 x11: ffff8001f6de9100 x10: 0000000000000009 x9 : ffff000015b78000 x8 : 0000000000000000 x7 : 0000000000000001 x6 : fffffffffffff000 x5 : 0000000000000fff x4 : ffff00001065dbc8 x3 : 000000000000000d x2 : 0000000000001000 x1 : fffffffffffff000 x0 : 0000000000000000 Call trace: __iommu_dma_unmap+0xb8/0xc0 iommu_dma_unmap_sg+0x98/0xb8 put_pages+0x5c/0xf0 [msm] msm_gem_free_work+0x10c/0x150 [msm] process_one_work+0x1e0/0x330 worker_thread+0x40/0x438 kthread+0x12c/0x130 ret_from_fork+0x10/0x18 ---[ end trace afc0dc5ab81a06bf ]---
Not quite sure what triggered that, but we really shouldn't be abusing dma_{map,unmap}_sg() for cache maint.
Cc: Stephen Boyd sboyd@kernel.org Tested-by: Stephen Boyd swboyd@chromium.org Reviewed-by: Jordan Crouse jcrouse@codeaurora.org Signed-off-by: Rob Clark robdclark@chromium.org Signed-off-by: Sean Paul seanpaul@chromium.org Link: https://patchwork.freedesktop.org/patch/msgid/20190630124735.27786-1-robdcla... Signed-off-by: Lee Jones lee.jones@linaro.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
This commit also requires the following commits:
commit 3de433c5b38af49a5fc7602721e2ab5d39f1e69c Author: Rob Clark robdclark@chromium.org Date: Tue Jul 30 14:46:28 2019 -0700
drm/msm: Use the correct dma_sync calls in msm_gem [subject was: drm/msm: shake fist angrily at dma-mapping] So, using dma_sync_* for our cache needs works out w/ dma iommu ops, but it falls appart with dma direct ops. The problem is that, depending on display generation, we can have either set of dma ops (mdp4 and dpu have iommu wired to mdss node, which maps to toplevel drm device, but mdp5 has iommu wired up to the mdp sub-node within mdss). Fixes this splat on mdp5 devices: Unable to handle kernel paging request at virtual address ffffffff80000000 Mem abort info: ESR = 0x96000144 Exception class = DABT (current EL), IL = 32 bits SET = 0, FnV = 0 EA = 0, S1PTW = 0 Data abort info: ISV = 0, ISS = 0x00000144 CM = 1, WnR = 1 swapper pgtable: 4k pages, 48-bit VAs, pgdp=00000000810e4000 [ffffffff80000000] pgd=0000000000000000 Internal error: Oops: 96000144 [#1] SMP Modules linked in: btqcomsmd btqca bluetooth cfg80211 ecdh_generic ecc rfkill libarc4 panel_simple msm wcnss_ctrl qrtr_smd drm_kms_helper venus_enc venus_dec videobuf2_dma_sg videobuf2_memops drm venus_core ipv6 qrtr qcom_wcnss_pil v4l2_mem2mem qcom_sysmon videobuf2_v4l2 qmi_helpers videobuf2_common crct10dif_ce mdt_loader qcom_common videodev qcom_glink_smem remoteproc bmc150_accel_i2c bmc150_magn_i2c bmc150_accel_core bmc150_magn snd_soc_lpass_apq8016 snd_soc_msm8916_analog mms114 mc nf_defrag_ipv6 snd_soc_lpass_cpu snd_soc_apq8016_sbc industrialio_triggered_buffer kfifo_buf snd_soc_lpass_platform snd_soc_msm8916_digital drm_panel_orientation_quirks CPU: 2 PID: 33 Comm: kworker/2:1 Not tainted 5.3.0-rc2 #1 Hardware name: Samsung Galaxy A5U (EUR) (DT) Workqueue: events deferred_probe_work_func pstate: 80000005 (Nzcv daif -PAN -UAO) pc : __clean_dcache_area_poc+0x20/0x38 lr : arch_sync_dma_for_device+0x28/0x30
We have noticed this problem on stable-rc 4.19.118-rc1 running on arm64 qualcomm dragonboard-410c device while booting.
[ 5.474942] msm 1a00000.mdss: A306: using IOMMU [ 5.483399] Unable to handle kernel paging request at virtual address ffffffff80000000 [ 5.487564] Mem abort info: [ 5.498182] ESR = 0x96000144 [ 5.507101] SET = 0, FnV = 0 [ 5.507114] EA = 0, S1PTW = 0 [ 5.509154] Data abort info: [ 5.512253] ISV = 0, ISS = 0x00000144 [ 5.515376] CM = 1, WnR = 1 [ 5.518877] swapper pgtable: 4k pages, 48-bit VAs, pgdp = (____ptrval____) [ 5.522072] [ffffffff80000000] pgd=0000000000000000 [ 5.528819] Internal error: Oops: 96000144 [#1] PREEMPT SMP [ 5.533491] Modules linked in: msm(+) crc32_ce adv7511 cec mdt_loader drm_kms_helper drm drm_panel_orientation_quirks fuse [ 5.539057] Process systemd-udevd (pid: 2807, stack limit = 0x(____ptrval____)) [ 5.550162] CPU: 0 PID: 2807 Comm: systemd-udevd Not tainted 4.19.118-rc1-00065-gb5f03cd61ab6 #1 [ 5.557366] Hardware name: Qualcomm Technologies, Inc. APQ 8016 SBC (DT) [ 5.566392] pstate: 80000005 (Nzcv daif -PAN -UAO) [ 5.573079] pc : __clean_dcache_area_poc+0x20/0x38 [ 5.577676] lr : __swiotlb_sync_sg_for_device+0x74/0xa0 [ 5.582447] sp : ffff00000dcab490 [ 5.587567] x29: ffff00000dcab490 x28: 0000000000000001 [ 5.591043] x27: ffff000000d97e40 x26: ffff80003bb9e000 [ 5.596423] x25: ffff80003c14c410 x24: ffff000009066178 [ 5.601717] x23: ffff80003c14c810 x22: 0000000000000000 [ 5.607013] x21: 0000000000000001 x20: 0000000000000001 [ 5.612308] x19: ffff80003634cf80 x18: 0000000000000400 [ 5.617603] x17: 0000000000000000 x16: 0000000000000000 [ 5.622899] x15: 0000000000000400 x14: 0000000000000400 [ 5.628194] x13: 0000000000000001 x12: 0000000000000000 [ 5.633489] x11: 0000800036c50000 x10: ffff80003639fba8 [ 5.638783] x9 : 0000000000001000 x8 : ffff7e0000e7c080 [ 5.644079] x7 : 0000000000000001 x6 : 0000000000000000 [ 5.649375] x5 : 0000000000000000 x4 : ffffffff80000000 [ 5.654669] x3 : 000000000000003f x2 : 0000000000000040 [ 5.659964] x1 : ffffffff80001000 x0 : ffffffff80000000 [ 5.665260] Call trace: [ 5.670556] __clean_dcache_area_poc+0x20/0x38 [ 5.672850] get_pages+0x1cc/0x240 [msm] [ 5.677355] msm_gem_get_iova+0x94/0x138 [msm] [ 5.681428] _msm_gem_kernel_new+0x40/0xb0 [msm] [ 5.685679] msm_gem_kernel_new+0x10/0x18 [msm] [ 5.690452] msm_gpu_init+0x300/0x568 [msm] [ 5.694698] adreno_gpu_init+0x14c/0x268 [msm] [ 5.698861] a3xx_gpu_init+0x7c/0x108 [msm] [ 5.703375] adreno_bind+0x144/0x238 [msm] [ 5.707365] component_bind_all+0x110/0x270 [ 5.711627] msm_drm_bind+0x104/0x760 [msm] [ 5.715609] try_to_bring_up_master+0x14c/0x1a8 [ 5.719775] component_master_add_with_match+0xc0/0x100 [ 5.724388] msm_pdev_probe+0x280/0x320 [msm] [ 5.729499] platform_drv_probe+0x50/0xa0 [ 5.734010] really_probe+0x1f4/0x290 [ 5.738003] driver_probe_device+0x54/0xe8 [ 5.741649] __driver_attach+0xe0/0xe8 [ 5.745644] bus_for_each_dev+0x70/0xb8 [ 5.749374] driver_attach+0x20/0x28 [ 5.753108] bus_add_driver+0x1a0/0x210 [ 5.756927] driver_register+0x60/0x110 [ 5.760485] __platform_driver_register+0x44/0x50 [ 5.764407] msm_drm_register+0x54/0x68 [msm] [ 5.769169] do_one_initcall+0x54/0x154 [ 5.773509] do_init_module+0x54/0x1c8 [ 5.777152] load_module+0x1bf4/0x2190 [ 5.780972] __se_sys_finit_module+0xb8/0xc8 [ 5.784706] __arm64_sys_finit_module+0x18/0x20 [ 5.789135] el0_svc_common+0x70/0x168 [ 5.793385] el0_svc_handler+0x2c/0x80 [ 5.797204] el0_svc+0x8/0xc [ 5.800939] Code: 9ac32042 8b010001 d1000443 8a230000 (d50b7e20) [ 5.803980] ---[ end trace 004276cd8aee46e8 ]---
Ref: Full test logs. https://qa-reports.linaro.org/lkft/linux-stable-rc-4.19-oe/build/v4.19.117-6... https://lkft.validation.linaro.org/scheduler/job/1387568#L3575
kernel configs link, https://builds.tuxbuild.com/TcvobwCBir3uhOd2MA-ndw/kernel.config
-- Linaro LKFT https://lkft.linaro.org
Hi,
Thanks for your report.
-----Original Message----- From: Naresh Kamboju [mailto:naresh.kamboju@linaro.org] Sent: Thursday, April 23, 2020 5:24 AM To: Greg KH gregkh@linuxfoundation.org Cc: iwamatsu nobuhiro(岩松 信洋 □ＳＷＣ◯ＡＣＴ) nobuhiro1.iwamatsu@toshiba.co.jp; open list linux-kernel@vger.kernel.org; linux- stable stable@vger.kernel.org; Stephen Boyd sboyd@kernel.org; swboyd@chromium.org; jcrouse@codeaurora.org; Rob Clark robdclark@chromium.org; seanpaul@chromium.org; Lee Jones lee.jones@linaro.org; lkft-triage@lists.linaro.org Subject: Re: [PATCH 4.19 50/54] drm/msm: stop abusing dma_map/unmap for cache
On Mon, 13 Apr 2020 at 13:56, Greg KH gregkh@linuxfoundation.org wrote:
On Mon, Apr 13, 2020 at 05:03:26AM +0000, nobuhiro1.iwamatsu@toshiba.co.jp wrote:
Hi,
-----Original Message----- From: stable-owner@vger.kernel.org [mailto:stable-owner@vger.kernel.org] On Behalf Of Greg Kroah-Hartman Sent: Saturday, April 11, 2020 9:10 PM To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman gregkh@linuxfoundation.org; stable@vger.kernel.org; Stephen Boyd sboyd@kernel.org;
Stephen
Boyd swboyd@chromium.org; Jordan Crouse jcrouse@codeaurora.org; Rob Clark robdclark@chromium.org; Sean
Paul
seanpaul@chromium.org; Lee Jones lee.jones@linaro.org Subject: [PATCH 4.19 50/54] drm/msm: stop abusing dma_map/unmap for cache
From: Rob Clark robdclark@chromium.org
commit 0036bc73ccbe7e600a3468bf8e8879b122252274 upstream.
Recently splats like this started showing up:
WARNING: CPU: 4 PID: 251 at drivers/iommu/dma-iommu.c:451 __iommu_dma_unmap+0xb8/0xc0 Modules linked in: ath10k_snoc ath10k_core fuse msm ath mac80211 uvcvideo cfg80211 videobuf2_vmalloc
videobuf2_memops
vide CPU: 4 PID: 251 Comm: kworker/u16:4 Tainted: G W 5.2.0-rc5-next-20190619+ #2317 Hardware name: LENOVO 81JL/LNVNB161216, BIOS 9UCN23WW(V1.06) 10/25/2018 Workqueue: msm msm_gem_free_work [msm] pstate: 80c00005 (Nzcv daif +PAN +UAO) pc : __iommu_dma_unmap+0xb8/0xc0 lr : __iommu_dma_unmap+0x54/0xc0 sp : ffff0000119abce0 x29: ffff0000119abce0 x28: 0000000000000000 x27: ffff8001f9946648 x26: ffff8001ec271068 x25: 0000000000000000 x24: ffff8001ea3580a8 x23: ffff8001f95ba010 x22: ffff80018e83ba88 x21: ffff8001e548f000 x20: fffffffffffff000 x19: 0000000000001000 x18: 00000000c00001fe x17: 0000000000000000 x16: 0000000000000000 x15: ffff000015b70068 x14: 0000000000000005 x13: 0003142cc1be1768 x12: 0000000000000001 x11: ffff8001f6de9100 x10: 0000000000000009 x9 : ffff000015b78000 x8 : 0000000000000000 x7 : 0000000000000001 x6 : fffffffffffff000 x5 : 0000000000000fff x4 : ffff00001065dbc8 x3 : 000000000000000d x2 : 0000000000001000 x1 : fffffffffffff000 x0 : 0000000000000000 Call trace: __iommu_dma_unmap+0xb8/0xc0 iommu_dma_unmap_sg+0x98/0xb8 put_pages+0x5c/0xf0 [msm] msm_gem_free_work+0x10c/0x150 [msm] process_one_work+0x1e0/0x330 worker_thread+0x40/0x438 kthread+0x12c/0x130 ret_from_fork+0x10/0x18 ---[ end trace afc0dc5ab81a06bf ]---
Not quite sure what triggered that, but we really shouldn't be abusing dma_{map,unmap}_sg() for cache maint.
Cc: Stephen Boyd sboyd@kernel.org Tested-by: Stephen Boyd swboyd@chromium.org Reviewed-by: Jordan Crouse jcrouse@codeaurora.org Signed-off-by: Rob Clark robdclark@chromium.org Signed-off-by: Sean Paul seanpaul@chromium.org Link: https://patchwork.freedesktop.org/patch/msgid/20190630124735.27786-1-robdcla... Signed-off-by: Lee Jones lee.jones@linaro.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
This commit also requires the following commits:
commit 3de433c5b38af49a5fc7602721e2ab5d39f1e69c Author: Rob Clark robdclark@chromium.org Date: Tue Jul 30 14:46:28 2019 -0700
drm/msm: Use the correct dma_sync calls in msm_gem [subject was: drm/msm: shake fist angrily at dma-mapping] So, using dma_sync_* for our cache needs works out w/ dma iommu ops, but it falls appart with dma direct ops. The problem is that, depending on display generation, we can have either set of dma ops (mdp4 and dpu have iommu wired to mdss node, which maps to toplevel drm device, but mdp5 has iommu wired up to the mdp sub-node within mdss). Fixes this splat on mdp5 devices: Unable to handle kernel paging request at virtual address ffffffff80000000 Mem abort info: ESR = 0x96000144 Exception class = DABT (current EL), IL = 32 bits SET = 0, FnV = 0 EA = 0, S1PTW = 0 Data abort info: ISV = 0, ISS = 0x00000144 CM = 1, WnR = 1 swapper pgtable: 4k pages, 48-bit VAs, pgdp=00000000810e4000 [ffffffff80000000] pgd=0000000000000000 Internal error: Oops: 96000144 [#1] SMP Modules linked in: btqcomsmd btqca bluetooth cfg80211 ecdh_generic ecc rfkill libarc4 panel_simple msmwcnss_ctrl qrtr_smd drm_kms_helper venus_enc venus_dec videobuf2_dma_sg videobuf2_memops drm venus_core ipv6 qrtr qcom_wcnss_pil v4l2_mem2mem qcom_sysmon videobuf2_v4l2 qmi_helpers videobuf2_common crct10dif_ce mdt_loader qcom_common videodev qcom_glink_smem remoteproc bmc150_accel_i2c bmc150_magn_i2c bmc150_accel_core bmc150_magn snd_soc_lpass_apq8016 snd_soc_msm8916_analog mms114 mc nf_defrag_ipv6 snd_soc_lpass_cpu snd_soc_apq8016_sbc industrialio_triggered_buffer kfifo_buf snd_soc_lpass_platform snd_soc_msm8916_digital drm_panel_orientation_quirks
CPU: 2 PID: 33 Comm: kworker/2:1 Not tainted 5.3.0-rc2 #1 Hardware name: Samsung Galaxy A5U (EUR) (DT) Workqueue: events deferred_probe_work_func pstate: 80000005 (Nzcv daif -PAN -UAO) pc : __clean_dcache_area_poc+0x20/0x38 lr : arch_sync_dma_for_device+0x28/0x30We have noticed this problem on stable-rc 4.19.118-rc1 running on arm64 qualcomm dragonboard-410c device while booting.
[ 5.474942] msm 1a00000.mdss: A306: using IOMMU [ 5.483399] Unable to handle kernel paging request at virtual address ffffffff80000000 [ 5.487564] Mem abort info: [ 5.498182] ESR = 0x96000144 [ 5.507101] SET = 0, FnV = 0 [ 5.507114] EA = 0, S1PTW = 0 [ 5.509154] Data abort info: [ 5.512253] ISV = 0, ISS = 0x00000144 [ 5.515376] CM = 1, WnR = 1 [ 5.518877] swapper pgtable: 4k pages, 48-bit VAs, pgdp = (____ptrval____) [ 5.522072] [ffffffff80000000] pgd=0000000000000000 [ 5.528819] Internal error: Oops: 96000144 [#1] PREEMPT SMP [ 5.533491] Modules linked in: msm(+) crc32_ce adv7511 cec mdt_loader drm_kms_helper drm drm_panel_orientation_quirks fuse [ 5.539057] Process systemd-udevd (pid: 2807, stack limit = 0x(____ptrval____)) [ 5.550162] CPU: 0 PID: 2807 Comm: systemd-udevd Not tainted 4.19.118-rc1-00065-gb5f03cd61ab6 #1 [ 5.557366] Hardware name: Qualcomm Technologies, Inc. APQ 8016 SBC (DT) [ 5.566392] pstate: 80000005 (Nzcv daif -PAN -UAO) [ 5.573079] pc : __clean_dcache_area_poc+0x20/0x38 [ 5.577676] lr : __swiotlb_sync_sg_for_device+0x74/0xa0 [ 5.582447] sp : ffff00000dcab490 [ 5.587567] x29: ffff00000dcab490 x28: 0000000000000001 [ 5.591043] x27: ffff000000d97e40 x26: ffff80003bb9e000 [ 5.596423] x25: ffff80003c14c410 x24: ffff000009066178 [ 5.601717] x23: ffff80003c14c810 x22: 0000000000000000 [ 5.607013] x21: 0000000000000001 x20: 0000000000000001 [ 5.612308] x19: ffff80003634cf80 x18: 0000000000000400 [ 5.617603] x17: 0000000000000000 x16: 0000000000000000 [ 5.622899] x15: 0000000000000400 x14: 0000000000000400 [ 5.628194] x13: 0000000000000001 x12: 0000000000000000 [ 5.633489] x11: 0000800036c50000 x10: ffff80003639fba8 [ 5.638783] x9 : 0000000000001000 x8 : ffff7e0000e7c080 [ 5.644079] x7 : 0000000000000001 x6 : 0000000000000000 [ 5.649375] x5 : 0000000000000000 x4 : ffffffff80000000 [ 5.654669] x3 : 000000000000003f x2 : 0000000000000040 [ 5.659964] x1 : ffffffff80001000 x0 : ffffffff80000000 [ 5.665260] Call trace: [ 5.670556] __clean_dcache_area_poc+0x20/0x38 [ 5.672850] get_pages+0x1cc/0x240 [msm] [ 5.677355] msm_gem_get_iova+0x94/0x138 [msm] [ 5.681428] _msm_gem_kernel_new+0x40/0xb0 [msm] [ 5.685679] msm_gem_kernel_new+0x10/0x18 [msm] [ 5.690452] msm_gpu_init+0x300/0x568 [msm] [ 5.694698] adreno_gpu_init+0x14c/0x268 [msm] [ 5.698861] a3xx_gpu_init+0x7c/0x108 [msm] [ 5.703375] adreno_bind+0x144/0x238 [msm] [ 5.707365] component_bind_all+0x110/0x270 [ 5.711627] msm_drm_bind+0x104/0x760 [msm] [ 5.715609] try_to_bring_up_master+0x14c/0x1a8 [ 5.719775] component_master_add_with_match+0xc0/0x100 [ 5.724388] msm_pdev_probe+0x280/0x320 [msm] [ 5.729499] platform_drv_probe+0x50/0xa0 [ 5.734010] really_probe+0x1f4/0x290 [ 5.738003] driver_probe_device+0x54/0xe8 [ 5.741649] __driver_attach+0xe0/0xe8 [ 5.745644] bus_for_each_dev+0x70/0xb8 [ 5.749374] driver_attach+0x20/0x28 [ 5.753108] bus_add_driver+0x1a0/0x210 [ 5.756927] driver_register+0x60/0x110 [ 5.760485] __platform_driver_register+0x44/0x50 [ 5.764407] msm_drm_register+0x54/0x68 [msm] [ 5.769169] do_one_initcall+0x54/0x154 [ 5.773509] do_init_module+0x54/0x1c8 [ 5.777152] load_module+0x1bf4/0x2190 [ 5.780972] __se_sys_finit_module+0xb8/0xc8 [ 5.784706] __arm64_sys_finit_module+0x18/0x20 [ 5.789135] el0_svc_common+0x70/0x168 [ 5.793385] el0_svc_handler+0x2c/0x80 [ 5.797204] el0_svc+0x8/0xc [ 5.800939] Code: 9ac32042 8b010001 d1000443 8a230000 (d50b7e20) [ 5.803980] ---[ end trace 004276cd8aee46e8 ]---
Ref: Full test logs. https://qa-reports.linaro.org/lkft/linux-stable-rc-4.19-oe/build/v4.19.117-6... https://lkft.validation.linaro.org/scheduler/job/1387568#L3575
kernel configs link, https://builds.tuxbuild.com/TcvobwCBir3uhOd2MA-ndw/kernel.config
I think the following patch is needed for this.
9f614197c744002f9968e82c649fdf7fe778e1e7 3de433c5b38af49a5fc7602721e2ab5d39f1e69c
But I have no environment to check this now.
Best regards, Nobuhiro
-- Linaro LKFT https://lkft.linaro.org
On Thu, 23 Apr 2020 at 05:03, nobuhiro1.iwamatsu@toshiba.co.jp wrote: <trim>
I think the following patch is needed for this.
9f614197c744002f9968e82c649fdf7fe778e1e7 3de433c5b38af49a5fc7602721e2ab5d39f1e69c
But I have no environment to check this now.
The above suggested two patches already in stable-rc 4.19 branch but i still notice Internal error: Oops: 96000144 [#1] PREEMPT SMP on arm64 qualcomm dragonboard-410c device while booting.
[ 7.906343] msm 1a00000.mdss: bound 1a98000.dsi (ops dsi_ops [msm]) [ 7.912697] adreno 1c00000.gpu: 1c00000.gpu supply vdd not found, using dummy regulator [ 7.918567] adreno 1c00000.gpu: Linked as a consumer to regulator.0 [ 7.926521] adreno 1c00000.gpu: 1c00000.gpu supply vddcx not found, using dummy regulator [ 7.932759] msm 1a00000.mdss: A306: using IOMMU [ 7.941207] Unable to handle kernel paging request at virtual address ffffffff80000000 [ 7.945375] Mem abort info: [ 7.953353] ESR = 0x96000144 [ 7.956034] Exception class = DABT (current EL), IL = 32 bits [ 7.970501] EA = 0, S1PTW = 0 [ 7.970516] Data abort info: [ 7.972485] ISV = 0, ISS = 0x00000144 [ 7.975577] CM = 1, WnR = 1 [ 7.979169] swapper pgtable: 4k pages, 48-bit VAs, pgdp = (____ptrval____) [ 7.982295] [ffffffff80000000] pgd=0000000000000000 [ 7.989099] Internal error: Oops: 96000144 [#1] PREEMPT SMP [ 7.993802] Modules linked in: msm(+) crc32_ce adv7511 cec mdt_loader drm_kms_helper drm drm_panel_orientation_quirks fuse [ 7.999367] Process systemd-udevd (pid: 2849, stack limit = 0x(____ptrval____)) [ 8.010474] CPU: 1 PID: 2849 Comm: systemd-udevd Not tainted 4.19.125-rc1-00077-g0708fb235b9c #1 [ 8.017677] Hardware name: Qualcomm Technologies, Inc. APQ 8016 SBC (DT) [ 8.026703] pstate: 80000005 (Nzcv daif -PAN -UAO) [ 8.033390] pc : __clean_dcache_area_poc+0x20/0x38 [ 8.037998] lr : __swiotlb_sync_sg_for_device+0x74/0xa0 [ 8.038001] sp : ffff00000e313490 [ 8.038004] x29: ffff00000e313490 x28: 0000000000000001 [ 8.038009] x27: ffff000000d56e40 x26: ffff800036301000 [ 8.038013] x25: ffff80003c108410 x24: ffff000009069998 [ 8.038018] x23: ffff80003c108810 x22: 0000000000000000 [ 8.038023] x21: 0000000000000001 x20: 0000000000000001 [ 8.038027] x19: ffff80003acd4480 x18: ffff0000092298c8 [ 8.038032] x17: 0000000000000000 x16: 0000000000000000 [ 8.038036] x15: 0000000000000010 x14: ffffffffffffffff [ 8.038042] x13: ffff0000893987d7 x12: 0000000000000000 [ 8.038046] x11: 0000800036c56000 x10: ffff8000362ef368 [ 8.038050] x9 : 0000000000001000 x8 : ffff7e0000e71640 [ 8.038055] x7 : 0000000000000001 x6 : 0000000000000000 [ 8.038059] x5 : 0000000000000000 x4 : ffffffff80000000 [ 8.038064] x3 : 000000000000003f x2 : 0000000000000040 [ 8.038068] x1 : ffffffff80001000 x0 : ffffffff80000000 [ 8.038072] Call trace: [ 8.038078] __clean_dcache_area_poc+0x20/0x38 [ 8.038204] get_pages+0x1cc/0x240 [msm] [ 8.038327] msm_gem_get_iova+0x94/0x138 [msm] [ 8.141741] _msm_gem_kernel_new+0x40/0xb0 [msm] [ 8.145989] msm_gem_kernel_new+0x10/0x18 [msm] [ 8.150759] msm_gpu_init+0x300/0x568 [msm] [ 8.155004] adreno_gpu_init+0x14c/0x268 [msm] [ 8.159171] a3xx_gpu_init+0x7c/0x108 [msm] [ 8.163682] adreno_bind+0x144/0x238 [msm] [ 8.167676] component_bind_all+0x110/0x278 [ 8.171939] msm_drm_bind+0x104/0x760 [msm] [ 8.175921] try_to_bring_up_master+0x14c/0x1b0 [ 8.180086] component_master_add_with_match+0xc0/0x100 [ 8.184697] msm_pdev_probe+0x280/0x320 [msm] [ 8.189810] platform_drv_probe+0x50/0xa0 [ 8.194322] really_probe+0x1f4/0x290 [ 8.198314] driver_probe_device+0x54/0xe8 [ 8.201960] __driver_attach+0xe0/0xe8 [ 8.205952] bus_for_each_dev+0x70/0xb8 [ 8.209685] driver_attach+0x20/0x28 [ 8.213417] bus_add_driver+0x1a0/0x210 [ 8.217237] driver_register+0x60/0x110 [ 8.220797] __platform_driver_register+0x44/0x50 [ 8.224717] msm_drm_register+0x54/0x68 [msm] [ 8.229479] do_one_initcall+0x54/0x154 [ 8.233819] do_init_module+0x54/0x1c8 [ 8.237463] load_module+0x1bf4/0x2190 [ 8.241282] __se_sys_finit_module+0xb8/0xc8 [ 8.245016] __arm64_sys_finit_module+0x18/0x20 [ 8.249445] el0_svc_common+0x70/0x168 [ 8.253695] el0_svc_handler+0x2c/0x80 [ 8.257514] el0_svc+0x8/0xc [ 8.261250] Code: 9ac32042 8b010001 d1000443 8a230000 (d50b7e20) [ 8.264290] ---[ end trace 2effae58ca65f06b ]---
on stable-rc 4.19 branch git log show this info, $ git log --oneline drivers/gpu/drm/msm/msm_gem.c | head 05fe33cad985 drm/msm: Use the correct dma_sync calls harder 39718d086d9b drm/msm: Use the correct dma_sync calls in msm_gem 9c23e00804f8 drm/msm: stop abusing dma_map/unmap for cache a5f74ec7d3cb gpu: drm: msm: Change return type to vm_fault_t 3976626ea3d2 drm/msm: Fix possible null dereference on failure of get_pages() d71b6bd80d96 drm/msm/dsi: fix direct caller of msm_gem_free_object() dc9a9b32053e drm/msm: Replace gem_object deprecated functions 62e3a3e342af drm/msm: fix leak in failed get_pages 7a88cbd8d65d Backmerge tag 'v4.14-rc7' into drm-next fad33f4b1073 drm/msm: add special _get_vaddr_active() for cmdstream dumps
link to full boot log, https://qa-reports.linaro.org/lkft/linux-stable-rc-4.19-oe/build/v4.19.124-7... https://qa-reports.linaro.org/lkft/linux-stable-rc-4.19-oe/build/v4.19.124-7...
Kernel config: https://builds.tuxbuild.com/Xp5Fh9e52QxohQeW6nazPA/kernel.config
Best regards, Nobuhiro
-- Linaro LKFT https://lkft.linaro.org
On Tue, May 26, 2020 at 7:33 AM Naresh Kamboju naresh.kamboju@linaro.org wrote:
On Thu, 23 Apr 2020 at 05:03, nobuhiro1.iwamatsu@toshiba.co.jp wrote:
<trim> > > I think the following patch is needed for this. > > 9f614197c744002f9968e82c649fdf7fe778e1e7 > 3de433c5b38af49a5fc7602721e2ab5d39f1e69c > > But I have no environment to check this now.
The above suggested two patches already in stable-rc 4.19 branch but i still notice Internal error: Oops: 96000144 [#1] PREEMPT SMP on arm64 qualcomm dragonboard-410c device while booting.
[ 7.906343] msm 1a00000.mdss: bound 1a98000.dsi (ops dsi_ops [msm]) [ 7.912697] adreno 1c00000.gpu: 1c00000.gpu supply vdd not found, using dummy regulator [ 7.918567] adreno 1c00000.gpu: Linked as a consumer to regulator.0 [ 7.926521] adreno 1c00000.gpu: 1c00000.gpu supply vddcx not found, using dummy regulator [ 7.932759] msm 1a00000.mdss: A306: using IOMMU [ 7.941207] Unable to handle kernel paging request at virtual address ffffffff80000000 [ 7.945375] Mem abort info: [ 7.953353] ESR = 0x96000144 [ 7.956034] Exception class = DABT (current EL), IL = 32 bits [ 7.970501] EA = 0, S1PTW = 0 [ 7.970516] Data abort info: [ 7.972485] ISV = 0, ISS = 0x00000144 [ 7.975577] CM = 1, WnR = 1 [ 7.979169] swapper pgtable: 4k pages, 48-bit VAs, pgdp = (____ptrval____) [ 7.982295] [ffffffff80000000] pgd=0000000000000000 [ 7.989099] Internal error: Oops: 96000144 [#1] PREEMPT SMP [ 7.993802] Modules linked in: msm(+) crc32_ce adv7511 cec mdt_loader drm_kms_helper drm drm_panel_orientation_quirks fuse [ 7.999367] Process systemd-udevd (pid: 2849, stack limit = 0x(____ptrval____)) [ 8.010474] CPU: 1 PID: 2849 Comm: systemd-udevd Not tainted 4.19.125-rc1-00077-g0708fb235b9c #1 [ 8.017677] Hardware name: Qualcomm Technologies, Inc. APQ 8016 SBC (DT) [ 8.026703] pstate: 80000005 (Nzcv daif -PAN -UAO) [ 8.033390] pc : __clean_dcache_area_poc+0x20/0x38 [ 8.037998] lr : __swiotlb_sync_sg_for_device+0x74/0xa0 [ 8.038001] sp : ffff00000e313490 [ 8.038004] x29: ffff00000e313490 x28: 0000000000000001 [ 8.038009] x27: ffff000000d56e40 x26: ffff800036301000 [ 8.038013] x25: ffff80003c108410 x24: ffff000009069998 [ 8.038018] x23: ffff80003c108810 x22: 0000000000000000 [ 8.038023] x21: 0000000000000001 x20: 0000000000000001 [ 8.038027] x19: ffff80003acd4480 x18: ffff0000092298c8 [ 8.038032] x17: 0000000000000000 x16: 0000000000000000 [ 8.038036] x15: 0000000000000010 x14: ffffffffffffffff [ 8.038042] x13: ffff0000893987d7 x12: 0000000000000000 [ 8.038046] x11: 0000800036c56000 x10: ffff8000362ef368 [ 8.038050] x9 : 0000000000001000 x8 : ffff7e0000e71640 [ 8.038055] x7 : 0000000000000001 x6 : 0000000000000000 [ 8.038059] x5 : 0000000000000000 x4 : ffffffff80000000 [ 8.038064] x3 : 000000000000003f x2 : 0000000000000040 [ 8.038068] x1 : ffffffff80001000 x0 : ffffffff80000000 [ 8.038072] Call trace: [ 8.038078] __clean_dcache_area_poc+0x20/0x38 [ 8.038204] get_pages+0x1cc/0x240 [msm] [ 8.038327] msm_gem_get_iova+0x94/0x138 [msm] [ 8.141741] _msm_gem_kernel_new+0x40/0xb0 [msm] [ 8.145989] msm_gem_kernel_new+0x10/0x18 [msm] [ 8.150759] msm_gpu_init+0x300/0x568 [msm] [ 8.155004] adreno_gpu_init+0x14c/0x268 [msm] [ 8.159171] a3xx_gpu_init+0x7c/0x108 [msm] [ 8.163682] adreno_bind+0x144/0x238 [msm] [ 8.167676] component_bind_all+0x110/0x278 [ 8.171939] msm_drm_bind+0x104/0x760 [msm] [ 8.175921] try_to_bring_up_master+0x14c/0x1b0 [ 8.180086] component_master_add_with_match+0xc0/0x100 [ 8.184697] msm_pdev_probe+0x280/0x320 [msm] [ 8.189810] platform_drv_probe+0x50/0xa0 [ 8.194322] really_probe+0x1f4/0x290 [ 8.198314] driver_probe_device+0x54/0xe8 [ 8.201960] __driver_attach+0xe0/0xe8 [ 8.205952] bus_for_each_dev+0x70/0xb8 [ 8.209685] driver_attach+0x20/0x28 [ 8.213417] bus_add_driver+0x1a0/0x210 [ 8.217237] driver_register+0x60/0x110 [ 8.220797] __platform_driver_register+0x44/0x50 [ 8.224717] msm_drm_register+0x54/0x68 [msm] [ 8.229479] do_one_initcall+0x54/0x154 [ 8.233819] do_init_module+0x54/0x1c8 [ 8.237463] load_module+0x1bf4/0x2190 [ 8.241282] __se_sys_finit_module+0xb8/0xc8 [ 8.245016] __arm64_sys_finit_module+0x18/0x20 [ 8.249445] el0_svc_common+0x70/0x168 [ 8.253695] el0_svc_handler+0x2c/0x80 [ 8.257514] el0_svc+0x8/0xc [ 8.261250] Code: 9ac32042 8b010001 d1000443 8a230000 (d50b7e20) [ 8.264290] ---[ end trace 2effae58ca65f06b ]---
on stable-rc 4.19 branch git log show this info, $ git log --oneline drivers/gpu/drm/msm/msm_gem.c | head 05fe33cad985 drm/msm: Use the correct dma_sync calls harder 39718d086d9b drm/msm: Use the correct dma_sync calls in msm_gem 9c23e00804f8 drm/msm: stop abusing dma_map/unmap for cache a5f74ec7d3cb gpu: drm: msm: Change return type to vm_fault_t 3976626ea3d2 drm/msm: Fix possible null dereference on failure of get_pages() d71b6bd80d96 drm/msm/dsi: fix direct caller of msm_gem_free_object() dc9a9b32053e drm/msm: Replace gem_object deprecated functions 62e3a3e342af drm/msm: fix leak in failed get_pages 7a88cbd8d65d Backmerge tag 'v4.14-rc7' into drm-next fad33f4b1073 drm/msm: add special _get_vaddr_active() for cmdstream dumps
link to full boot log, https://qa-reports.linaro.org/lkft/linux-stable-rc-4.19-oe/build/v4.19.124-7... https://qa-reports.linaro.org/lkft/linux-stable-rc-4.19-oe/build/v4.19.124-7...
Kernel config: https://builds.tuxbuild.com/Xp5Fh9e52QxohQeW6nazPA/kernel.config
I suspect there is some difference between qcom_iommu and arm-smmu? Maybe qcom_iommu (esp. back in v4.19) is missing something for automatic hookup of iommu-dma-ops?
All the ugliness around dma-sync is related to trying to avoid using dma_map_*/dma_unmap_* when arm-smmu is in use (and the iommu dma-ops are installed, since they end up fighting w/ drm/msm's direct use of it's iommus). It looks like we are going to finally get a way to solve this uglyness a bit better once some arm-smmu patches[1] land in v5.8. I suppose that doesn't really help v4.19.
But taking a step back, I'm not entirely sure whether this was a problem yet on v4.19. Presumably dropping these would make db410c (qcom_iommu) work:
05fe33cad985 drm/msm: Use the correct dma_sync calls harder 39718d086d9b drm/msm: Use the correct dma_sync calls in msm_gem 9c23e00804f8 drm/msm: stop abusing dma_map/unmap for cache
What I'm not entirely sure of (v4.19 seems so long ago now) is whether that would cause problems for db820c (arm-smmu)? Looks like the first of those patches landed around v5.4, so presumably the problem it was meant to workaround started happening sometime after v4.19?
BR, -R
[1] https://lkml.org/lkml/2020/4/20/1142
Best regards, Nobuhiro
-- Linaro LKFT https://lkft.linaro.org
-
Naresh Kamboju -
nobuhiro1.iwamatsu＠toshiba.co.jp -
Rob Clark