- Linux-stable-mirror - lists.linaro.org

[PATCH v2] ipc/shm.c add ->pagesize function to shm_vm_ops

by Jane Chu

05ea88608d4e13 ("mm, hugetlbfs: introduce ->pagesize() to vm_operations_struct") adds a new ->pagesize() function to hugetlb_vm_ops, intended to cover all hugetlbfs backed files. With System V shared memory model, if "huge page" is specified, the "shared memory" is backed by hugetlbfs files, but the mappings initiated via shmget/shmat have their original vm_ops overwritten with shm_vm_ops, so we need to add a ->pagesize function to shm_vm_ops. Otherwise, vma_kernel_pagesize() returns PAGE_SIZE given a hugetlbfs backed vma, result in below BUG: fs/hugetlbfs/inode.c 443 if (unlikely(page_mapped(page))) { 444 BUG_ON(truncate_op); [ 242.268342] hugetlbfs: oracle (4592): Using mlock ulimits for SHM_HUGETLB is deprecated [ 282.653208] ------------[ cut here ]------------ [ 282.708447] kernel BUG at fs/hugetlbfs/inode.c:444! [ 282.818957] Modules linked in: nfsv3 rpcsec_gss_krb5 nfsv4 ... [ 284.025873] CPU: 35 PID: 5583 Comm: oracle_5583_sbt Not tainted 4.14.35-1829.el7uek.x86_64 #2 [ 284.246609] task: ffff9bf0507aaf80 task.stack: ffffa9e625628000 [ 284.317455] RIP: 0010:remove_inode_hugepages+0x3db/0x3e2 .... [ 285.292389] Call Trace: [ 285.321630] hugetlbfs_evict_inode+0x1e/0x3e [ 285.372707] evict+0xdb/0x1af [ 285.408185] iput+0x1a2/0x1f7 [ 285.443661] dentry_unlink_inode+0xc6/0xf0 [ 285.492661] __dentry_kill+0xd8/0x18d [ 285.536459] dput+0x1b5/0x1ed [ 285.571939] __fput+0x18b/0x216 [ 285.609495] ____fput+0xe/0x10 [ 285.646030] task_work_run+0x90/0xa7 [ 285.688788] exit_to_usermode_loop+0xdd/0x116 [ 285.740905] do_syscall_64+0x187/0x1ae [ 285.785740] entry_SYSCALL_64_after_hwframe+0x150/0x0 Link: http://lkml.kernel.org/r/20180727211727.5020-1-jane.chu@oracle.com Fixes: 05ea88608d4e13 ("mm, hugetlbfs: introduce ->pagesize() to vm_operations_struct") Signed-off-by: Jane Chu <jane.chu(a)oracle.com> Suggested-by: Mike Kravetz <mike.kravetz(a)oracle.com> Reviewed-by: Mike Kravetz <mike.kravetz(a)oracle.com> Acked-by: Michal Hocko <mhocko(a)suse.com> Acked-by: Davidlohr Bueso <dbueso(a)suse.de> Cc: Dan Williams <dan.j.williams(a)intel.com> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Jan Kara <jack(a)suse.cz> Cc: JÃ©rÃ´me Glisse <jglisse(a)redhat.com> Cc: Davidlohr Bueso <dave(a)stgolabs.net> Cc: Manfred Spraul <manfred(a)colorfullife.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- --- ipc/shm.c | 12 ++++++++++++ mm/hugetlb.c | 7 +++++++ 2 files changed, 19 insertions(+) diff --git a/ipc/shm.c b/ipc/shm.c index 051a3e1fb8df..fefa00d310fb 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -427,6 +427,17 @@ static int shm_split(struct vm_area_struct *vma, unsigned long addr) return 0; } +static unsigned long shm_pagesize(struct vm_area_struct *vma) +{ + struct file *file = vma->vm_file; + struct shm_file_data *sfd = shm_file_data(file); + + if (sfd->vm_ops->pagesize) + return sfd->vm_ops->pagesize(vma); + + return PAGE_SIZE; +} + #ifdef CONFIG_NUMA static int shm_set_policy(struct vm_area_struct *vma, struct mempolicy *new) { @@ -554,6 +565,7 @@ static const struct vm_operations_struct shm_vm_ops = { .close = shm_close, /* callback for when the vm-area is released */ .fault = shm_fault, .split = shm_split, + .pagesize = shm_pagesize, #if defined(CONFIG_NUMA) .set_policy = shm_set_policy, .get_policy = shm_get_policy, diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 039ddbc574e9..3103099f64fd 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -3167,6 +3167,13 @@ static vm_fault_t hugetlb_vm_op_fault(struct vm_fault *vmf) return 0; } +/* + * When a new function is introduced to vm_operations_struct and added + * to hugetlb_vm_ops, please consider adding the function to shm_vm_ops. + * This is because under System V memory model, mappings created via + * shmget/shmat with "huge page" specified are backed by hugetlbfs files, + * their original vm_ops are overwritten with shm_vm_ops. + */ const struct vm_operations_struct hugetlb_vm_ops = { .fault = hugetlb_vm_op_fault, .open = hugetlb_vm_op_open, -- 2.15.GIT

6 years, 11 months

1
0
0 0

[PATCH] arch/x86: Fix boot_cpu_data.microcode version output

by Prarit Bhargava

On systems where a runtime microcode update has occurred the microcode version output in a MCE log record is wrong because boot_cpu_data.microcode is not updated during runtime. Update boot_cpu_data.microcode when the BSP's microcode is updated. Fixes: fa94d0c6e0f3 ("x86/MCE: Save microcode revision in machine check records") Suggested-by: Borislav Petkov <bp(a)alien8.com> Signed-off-by: Prarit Bhargava <prarit(a)redhat.com> Cc: stable(a)vger.kernel.org Cc: sironi(a)amazon.de Cc: tony.luck(a)intel.com --- arch/x86/kernel/cpu/microcode/amd.c | 4 ++++ arch/x86/kernel/cpu/microcode/intel.c | 4 ++++ 2 files changed, 8 insertions(+) diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c index 0624957aa068..7f5b32535ac7 100644 --- a/arch/x86/kernel/cpu/microcode/amd.c +++ b/arch/x86/kernel/cpu/microcode/amd.c @@ -537,6 +537,10 @@ static enum ucode_state apply_microcode_amd(int cpu) uci->cpu_sig.rev = mc_amd->hdr.patch_id; c->microcode = mc_amd->hdr.patch_id; + /* Update boot_cpu_data's revision too, if we're on the BSP: */ + if (c->cpu_index == boot_cpu_data.cpu_index) + boot_cpu_data.microcode = rev; + return UCODE_UPDATED; } diff --git a/arch/x86/kernel/cpu/microcode/intel.c b/arch/x86/kernel/cpu/microcode/intel.c index 97ccf4c3b45b..256d336cbc04 100644 --- a/arch/x86/kernel/cpu/microcode/intel.c +++ b/arch/x86/kernel/cpu/microcode/intel.c @@ -851,6 +851,10 @@ static enum ucode_state apply_microcode_intel(int cpu) uci->cpu_sig.rev = rev; c->microcode = rev; + /* Update boot_cpu_data's revision too, if we're on the BSP: */ + if (c->cpu_index == boot_cpu_data.cpu_index) + boot_cpu_data.microcode = rev; + return UCODE_UPDATED; } -- 2.17.0

6 years, 11 months

2
2
0 0

[PATCH] qedi: Fix a potential buffer overflow

by Bart Van Assche

Tell snprintf() to store at most 255 characters in the output buffer instead of 256. This patch avoids that smatch reports the following warning: drivers/scsi/qedi/qedi_main.c:891: qedi_get_boot_tgt_info() error: snprintf() is printing too much 256 vs 255 Signed-off-by: Bart Van Assche <bart.vanassche(a)wdc.com> Cc: <QLogic-Storage-Upstream(a)cavium.com> Cc: <stable(a)vger.kernel.org> --- drivers/scsi/qedi/qedi_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/qedi/qedi_main.c b/drivers/scsi/qedi/qedi_main.c index 682f3ce31014..ea62180d9ec8 100644 --- a/drivers/scsi/qedi/qedi_main.c +++ b/drivers/scsi/qedi/qedi_main.c @@ -888,7 +888,7 @@ static void qedi_get_boot_tgt_info(struct nvm_iscsi_block *block, ipv6_en = !!(block->generic.ctrl_flags & NVM_ISCSI_CFG_GEN_IPV6_ENABLED); - snprintf(tgt->iscsi_name, NVM_ISCSI_CFG_ISCSI_NAME_MAX_LEN, "%s\n", + snprintf(tgt->iscsi_name, sizeof(tgt->iscsi_name), "%s\n", block->target[index].target_name.byte); tgt->ipv6_en = ipv6_en; -- 2.18.0

6 years, 11 months

3
2
0 0

Richard & Angela Maxwell Congratulates You..

by Richard & Angela Maxwell

My wife and I won the Euro Millions Lottery of 53 Million British Pounds and we have voluntarily decided to donate 1,000,000GBP(One Million British Pounds) to 5 individuals randomly as part of our own charity project. To verify our lottery winnings,please see our interview by visiting the web page below: http://www.telegraph.co.uk/news/newstopics/howaboutthat/11511467/Lincolnshi… Lincolnshire couple thought £53m EuroMillions win was ... A retired couple who have become one of the country's biggest ever National Lottery winners after bagging more than £53 million at first thought it was an April Fool's joke. Richard and Angela Maxwell, from Boston in Lincolnshire, scooped £53,193,914 after winning the EuroMillions jackpot draw on ... www.telegraph.co.uk After a computer spinball,your email address was among the emails which were submitted to us by the Google, Inc as a web user; ifyou have received our email, kindly send us the below details so that we can transfer your 1,000,000.00 GBP(One Million Pounds) in your name or direct our paying bank to effect the transfer of the funds to your designated bank account in your own country. Full Names: Mobile No: Age: Address: Send your response to: richangela203(a)hotmail.com or richangela22<3057927356(a)qq.com> Best Regards, Richard & Angela Maxwell

6 years, 11 months

1
0
0 0

[PATCH v3 8/8] drm/nouveau: Call pm_runtime_get_noresume() from hpd handlers

by Lyude Paul

We can't and don't need to try resuming the device from our hotplug handlers, but hotplug events are generally something we'd like to keep the device awake for whenever possible. So, grab a PM ref safely in our hotplug handlers using pm_runtime_get_noresume() and mark the device as busy once we're finished. Signed-off-by: Lyude Paul <lyude(a)redhat.com> Cc: stable(a)vger.kernel.org Cc: Lukas Wunner <lukas(a)wunner.de> Cc: Karol Herbst <karolherbst(a)gmail.com> --- drivers/gpu/drm/nouveau/nouveau_connector.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 8409c3f2c3a1..5a8e8c1ad647 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -1152,6 +1152,11 @@ nouveau_connector_hotplug(struct nvif_notify *notify) const char *name = connector->name; struct nouveau_encoder *nv_encoder; + /* Resuming the device here isn't possible; but the suspend PM ops + * will wait for us to finish our work before disabling us so this + * should be enough + */ + pm_runtime_get_noresume(drm->dev->dev); nv_connector->hpd_task = current; if (rep->mask & NVIF_NOTIFY_CONN_V0_IRQ) { @@ -1171,6 +1176,9 @@ nouveau_connector_hotplug(struct nvif_notify *notify) } nv_connector->hpd_task = NULL; + + pm_runtime_mark_last_busy(drm->dev->dev); + pm_runtime_put_autosuspend(drm->dev->dev); return NVIF_NOTIFY_KEEP; } -- 2.17.1

6 years, 11 months

1
0
0 0

[PATCH v3 7/8] drm/nouveau: Fix deadlocks in nouveau_connector_detect()

by Lyude Paul

When we disable hotplugging on the GPU, we need to be able to synchronize with each connector's hotplug interrupt handler before the interrupt is finally disabled. This can be a problem however, since nouveau_connector_detect() currently grabs a runtime power reference when handling connector probing. This will deadlock the runtime suspend handler like so: [ 861.480896] INFO: task kworker/0:2:61 blocked for more than 120 seconds. [ 861.483290] Tainted: G O 4.18.0-rc6Lyude-Test+ #1 [ 861.485158] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [ 861.486332] kworker/0:2 D 0 61 2 0x80000000 [ 861.487044] Workqueue: events nouveau_display_hpd_work [nouveau] [ 861.487737] Call Trace: [ 861.488394] __schedule+0x322/0xaf0 [ 861.489070] schedule+0x33/0x90 [ 861.489744] rpm_resume+0x19c/0x850 [ 861.490392] ? finish_wait+0x90/0x90 [ 861.491068] __pm_runtime_resume+0x4e/0x90 [ 861.491753] nouveau_display_hpd_work+0x22/0x60 [nouveau] [ 861.492416] process_one_work+0x231/0x620 [ 861.493068] worker_thread+0x44/0x3a0 [ 861.493722] kthread+0x12b/0x150 [ 861.494342] ? wq_pool_ids_show+0x140/0x140 [ 861.494991] ? kthread_create_worker_on_cpu+0x70/0x70 [ 861.495648] ret_from_fork+0x3a/0x50 [ 861.496304] INFO: task kworker/6:2:320 blocked for more than 120 seconds. [ 861.496968] Tainted: G O 4.18.0-rc6Lyude-Test+ #1 [ 861.497654] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [ 861.498341] kworker/6:2 D 0 320 2 0x80000080 [ 861.499045] Workqueue: pm pm_runtime_work [ 861.499739] Call Trace: [ 861.500428] __schedule+0x322/0xaf0 [ 861.501134] ? wait_for_completion+0x104/0x190 [ 861.501851] schedule+0x33/0x90 [ 861.502564] schedule_timeout+0x3a5/0x590 [ 861.503284] ? mark_held_locks+0x58/0x80 [ 861.503988] ? _raw_spin_unlock_irq+0x2c/0x40 [ 861.504710] ? wait_for_completion+0x104/0x190 [ 861.505417] ? trace_hardirqs_on_caller+0xf4/0x190 [ 861.506136] ? wait_for_completion+0x104/0x190 [ 861.506845] wait_for_completion+0x12c/0x190 [ 861.507555] ? wake_up_q+0x80/0x80 [ 861.508268] flush_work+0x1c9/0x280 [ 861.508990] ? flush_workqueue_prep_pwqs+0x1b0/0x1b0 [ 861.509735] nvif_notify_put+0xb1/0xc0 [nouveau] [ 861.510482] nouveau_display_fini+0xbd/0x170 [nouveau] [ 861.511241] nouveau_display_suspend+0x67/0x120 [nouveau] [ 861.511969] nouveau_do_suspend+0x5e/0x2d0 [nouveau] [ 861.512715] nouveau_pmops_runtime_suspend+0x47/0xb0 [nouveau] [ 861.513435] pci_pm_runtime_suspend+0x6b/0x180 [ 861.514165] ? pci_has_legacy_pm_support+0x70/0x70 [ 861.514897] __rpm_callback+0x7a/0x1d0 [ 861.515618] ? pci_has_legacy_pm_support+0x70/0x70 [ 861.516313] rpm_callback+0x24/0x80 [ 861.517027] ? pci_has_legacy_pm_support+0x70/0x70 [ 861.517741] rpm_suspend+0x142/0x6b0 [ 861.518449] pm_runtime_work+0x97/0xc0 [ 861.519144] process_one_work+0x231/0x620 [ 861.519831] worker_thread+0x44/0x3a0 [ 861.520522] kthread+0x12b/0x150 [ 861.521220] ? wq_pool_ids_show+0x140/0x140 [ 861.521925] ? kthread_create_worker_on_cpu+0x70/0x70 [ 861.522622] ret_from_fork+0x3a/0x50 [ 861.523299] INFO: task kworker/6:0:1329 blocked for more than 120 seconds. [ 861.523977] Tainted: G O 4.18.0-rc6Lyude-Test+ #1 [ 861.524644] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [ 861.525349] kworker/6:0 D 0 1329 2 0x80000000 [ 861.526073] Workqueue: events nvif_notify_work [nouveau] [ 861.526751] Call Trace: [ 861.527411] __schedule+0x322/0xaf0 [ 861.528089] schedule+0x33/0x90 [ 861.528758] rpm_resume+0x19c/0x850 [ 861.529399] ? finish_wait+0x90/0x90 [ 861.530073] __pm_runtime_resume+0x4e/0x90 [ 861.530798] nouveau_connector_detect+0x7e/0x510 [nouveau] [ 861.531459] ? ww_mutex_lock+0x47/0x80 [ 861.532097] ? ww_mutex_lock+0x47/0x80 [ 861.532819] ? drm_modeset_lock+0x88/0x130 [drm] [ 861.533481] drm_helper_probe_detect_ctx+0xa0/0x100 [drm_kms_helper] [ 861.534127] drm_helper_hpd_irq_event+0xa4/0x120 [drm_kms_helper] [ 861.534940] nouveau_connector_hotplug+0x98/0x120 [nouveau] [ 861.535556] nvif_notify_work+0x2d/0xb0 [nouveau] [ 861.536221] process_one_work+0x231/0x620 [ 861.536994] worker_thread+0x44/0x3a0 [ 861.537757] kthread+0x12b/0x150 [ 861.538463] ? wq_pool_ids_show+0x140/0x140 [ 861.539102] ? kthread_create_worker_on_cpu+0x70/0x70 [ 861.539815] ret_from_fork+0x3a/0x50 [ 861.540521] Showing all locks held in the system: [ 861.541696] 2 locks held by kworker/0:2/61: [ 861.542406] #0: 000000002dbf8af5 ((wq_completion)"events"){+.+.}, at: process_one_work+0x1b3/0x620 [ 861.543071] #1: 0000000076868126 ((work_completion)(&drm->hpd_work)){+.+.}, at: process_one_work+0x1b3/0x620 [ 861.543814] 1 lock held by khungtaskd/64: [ 861.544535] #0: 0000000059db4b53 (rcu_read_lock){....}, at: debug_show_all_locks+0x23/0x185 [ 861.545160] 3 locks held by kworker/6:2/320: [ 861.545896] #0: 00000000d9e1bc59 ((wq_completion)"pm"){+.+.}, at: process_one_work+0x1b3/0x620 [ 861.546702] #1: 00000000c9f92d84 ((work_completion)(&dev->power.work)){+.+.}, at: process_one_work+0x1b3/0x620 [ 861.547443] #2: 000000004afc5de1 (drm_connector_list_iter){.+.+}, at: nouveau_display_fini+0x96/0x170 [nouveau] [ 861.548146] 1 lock held by dmesg/983: [ 861.548889] 2 locks held by zsh/1250: [ 861.549605] #0: 00000000348e3cf6 (&tty->ldisc_sem){++++}, at: ldsem_down_read+0x37/0x40 [ 861.550393] #1: 000000007009a7a8 (&ldata->atomic_read_lock){+.+.}, at: n_tty_read+0xc1/0x870 [ 861.551122] 6 locks held by kworker/6:0/1329: [ 861.551957] #0: 000000002dbf8af5 ((wq_completion)"events"){+.+.}, at: process_one_work+0x1b3/0x620 [ 861.552765] #1: 00000000ddb499ad ((work_completion)(&notify->work)#2){+.+.}, at: process_one_work+0x1b3/0x620 [ 861.553582] #2: 000000006e013cbe (&dev->mode_config.mutex){+.+.}, at: drm_helper_hpd_irq_event+0x6c/0x120 [drm_kms_helper] [ 861.554357] #3: 000000004afc5de1 (drm_connector_list_iter){.+.+}, at: drm_helper_hpd_irq_event+0x78/0x120 [drm_kms_helper] [ 861.555227] #4: 0000000044f294d9 (crtc_ww_class_acquire){+.+.}, at: drm_helper_probe_detect_ctx+0x3d/0x100 [drm_kms_helper] [ 861.556133] #5: 00000000db193642 (crtc_ww_class_mutex){+.+.}, at: drm_modeset_lock+0x4b/0x130 [drm] [ 861.557864] ============================================= [ 861.559507] NMI backtrace for cpu 2 [ 861.560363] CPU: 2 PID: 64 Comm: khungtaskd Tainted: G O 4.18.0-rc6Lyude-Test+ #1 [ 861.561197] Hardware name: LENOVO 20EQS64N0B/20EQS64N0B, BIOS N1EET78W (1.51 ) 05/18/2018 [ 861.561948] Call Trace: [ 861.562757] dump_stack+0x8e/0xd3 [ 861.563516] nmi_cpu_backtrace.cold.3+0x14/0x5a [ 861.564269] ? lapic_can_unplug_cpu.cold.27+0x42/0x42 [ 861.565029] nmi_trigger_cpumask_backtrace+0xa1/0xae [ 861.565789] arch_trigger_cpumask_backtrace+0x19/0x20 [ 861.566558] watchdog+0x316/0x580 [ 861.567355] kthread+0x12b/0x150 [ 861.568114] ? reset_hung_task_detector+0x20/0x20 [ 861.568863] ? kthread_create_worker_on_cpu+0x70/0x70 [ 861.569598] ret_from_fork+0x3a/0x50 [ 861.570370] Sending NMI from CPU 2 to CPUs 0-1,3-7: [ 861.571426] NMI backtrace for cpu 6 skipped: idling at intel_idle+0x7f/0x120 [ 861.571429] NMI backtrace for cpu 7 skipped: idling at intel_idle+0x7f/0x120 [ 861.571432] NMI backtrace for cpu 3 skipped: idling at intel_idle+0x7f/0x120 [ 861.571464] NMI backtrace for cpu 5 skipped: idling at intel_idle+0x7f/0x120 [ 861.571467] NMI backtrace for cpu 0 skipped: idling at intel_idle+0x7f/0x120 [ 861.571469] NMI backtrace for cpu 4 skipped: idling at intel_idle+0x7f/0x120 [ 861.571472] NMI backtrace for cpu 1 skipped: idling at intel_idle+0x7f/0x120 [ 861.572428] Kernel panic - not syncing: hung_task: blocked tasks So: fix this with a new trick; store the current task_struct that's executing in the nouveau_connector structure, then avoid attempting to runtime resume the device when we know that we're just running from the context of our hotplug interrupt handler. Since hpd interrupts are only enabled while the device is runtime active, this should be totally safe. Signed-off-by: Lyude Paul <lyude(a)redhat.com> Cc: stable(a)vger.kernel.org Cc: Lukas Wunner <lukas(a)wunner.de> Cc: Karol Herbst <karolherbst(a)gmail.com> --- drivers/gpu/drm/nouveau/nouveau_connector.c | 16 ++++++++++------ drivers/gpu/drm/nouveau/nouveau_connector.h | 1 + 2 files changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 9714e09f17db..8409c3f2c3a1 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -572,13 +572,14 @@ nouveau_connector_detect(struct drm_connector *connector, bool force) nv_connector->edid = NULL; } - /* Outputs are only polled while runtime active, so resuming the - * device here is unnecessary (and would deadlock upon runtime suspend - * because it waits for polling to finish). We do however, want to - * prevent the autosuspend timer from elapsing during this operation - * if possible. + /* Output polling and HPD only happens while we're runtime active, so + * resuming the device here is unnecessary (and would deadlock upon + * runtime suspend because it waits for polling to finish). We do + * however, want to prevent the autosuspend timer from elapsing during + * this operation if possible. */ - if (drm_kms_helper_is_poll_worker()) { + if (drm_kms_helper_is_poll_worker() || + nv_connector->hpd_task == current) { pm_runtime_get_noresume(dev->dev); } else { ret = pm_runtime_get_sync(dev->dev); @@ -1151,6 +1152,8 @@ nouveau_connector_hotplug(struct nvif_notify *notify) const char *name = connector->name; struct nouveau_encoder *nv_encoder; + nv_connector->hpd_task = current; + if (rep->mask & NVIF_NOTIFY_CONN_V0_IRQ) { NV_DEBUG(drm, "service %s\n", name); if ((nv_encoder = find_encoder(connector, DCB_OUTPUT_DP))) @@ -1167,6 +1170,7 @@ nouveau_connector_hotplug(struct nvif_notify *notify) nouveau_connector_hotplug_probe(nv_connector); } + nv_connector->hpd_task = NULL; return NVIF_NOTIFY_KEEP; } diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h index 2d9d35a146a4..1964e682ba13 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.h +++ b/drivers/gpu/drm/nouveau/nouveau_connector.h @@ -45,6 +45,7 @@ struct nouveau_connector { u8 *dcb; struct nvif_notify hpd; + struct task_struct *hpd_task; struct drm_dp_aux aux; -- 2.17.1

6 years, 11 months

1
0
0 0

[PATCH v3 6/8] drm/nouveau: Respond to HPDs by probing one conn at a time

by Lyude Paul

There isn't actually any reason we need to call drm_hpd_irq_event() from our hotplug handler, as we already know which connector the hotplug event was fired for. We're also going to need to avoid probing all connectors needlessly from hotplug handlers anyway so that we can track when nouveau_connector_detect() is being called from the context of it's connector's hotplug handler in order to fix the next deadlocking issue. This is (slightly) faster anyway! Signed-off-by: Lyude Paul <lyude(a)redhat.com> Cc: stable(a)vger.kernel.org Cc: Lukas Wunner <lukas(a)wunner.de> Cc: Karol Herbst <karolherbst(a)gmail.com> --- drivers/gpu/drm/nouveau/nouveau_connector.c | 28 ++++++++++++++++++++- 1 file changed, 27 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 010d6db14cba..9714e09f17db 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -1114,6 +1114,32 @@ nouveau_connector_funcs_lvds = { .atomic_get_property = nouveau_conn_atomic_get_property, }; +static void +nouveau_connector_hotplug_probe(struct nouveau_connector *nv_conn) +{ + struct drm_modeset_acquire_ctx ctx; + struct drm_connector *conn = &nv_conn->base; + enum drm_connector_status old_status; + struct drm_device *dev = conn->dev; + bool changed; + + mutex_lock(&dev->mode_config.mutex); + + drm_modeset_acquire_init(&ctx, 0); + drm_modeset_lock(&dev->mode_config.connection_mutex, &ctx); + + old_status = conn->status; + conn->status = drm_helper_probe_detect(conn, &ctx, true); + changed = old_status != conn->status; + + drm_modeset_drop_locks(&ctx); + drm_modeset_acquire_fini(&ctx); + mutex_unlock(&dev->mode_config.mutex); + + if (changed) + drm_kms_helper_hotplug_event(dev); +} + static int nouveau_connector_hotplug(struct nvif_notify *notify) { @@ -1138,7 +1164,7 @@ nouveau_connector_hotplug(struct nvif_notify *notify) nv50_mstm_remove(nv_encoder->dp.mstm); } - drm_helper_hpd_irq_event(connector->dev); + nouveau_connector_hotplug_probe(nv_connector); } return NVIF_NOTIFY_KEEP; -- 2.17.1

6 years, 11 months

1
0
0 0

[PATCH v3 5/8] drm/nouveau: Use pm_runtime_get_noresume() in connector_detect()

by Lyude Paul

It's true we can't resume the device from poll workers in nouveau_connector_detect(). We can however, prevent the autosuspend timer from elapsing immediately if it hasn't already without risking any sort of deadlock with the runtime suspend/resume operations. So do that instead of entirely avoiding grabbing a power reference. Signed-off-by: Lyude Paul <lyude(a)redhat.com> Cc: stable(a)vger.kernel.org Cc: Lukas Wunner <lukas(a)wunner.de> Cc: Karol Herbst <karolherbst(a)gmail.com> --- drivers/gpu/drm/nouveau/nouveau_connector.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 2a45b4c2ceb0..010d6db14cba 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -572,12 +572,16 @@ nouveau_connector_detect(struct drm_connector *connector, bool force) nv_connector->edid = NULL; } - /* Outputs are only polled while runtime active, so acquiring a - * runtime PM ref here is unnecessary (and would deadlock upon - * runtime suspend because it waits for polling to finish). + /* Outputs are only polled while runtime active, so resuming the + * device here is unnecessary (and would deadlock upon runtime suspend + * because it waits for polling to finish). We do however, want to + * prevent the autosuspend timer from elapsing during this operation + * if possible. */ - if (!drm_kms_helper_is_poll_worker()) { - ret = pm_runtime_get_sync(connector->dev->dev); + if (drm_kms_helper_is_poll_worker()) { + pm_runtime_get_noresume(dev->dev); + } else { + ret = pm_runtime_get_sync(dev->dev); if (ret < 0 && ret != -EACCES) return conn_status; } @@ -655,10 +659,8 @@ nouveau_connector_detect(struct drm_connector *connector, bool force) out: - if (!drm_kms_helper_is_poll_worker()) { - pm_runtime_mark_last_busy(connector->dev->dev); - pm_runtime_put_autosuspend(connector->dev->dev); - } + pm_runtime_mark_last_busy(dev->dev); + pm_runtime_put_autosuspend(dev->dev); return conn_status; } -- 2.17.1

6 years, 11 months

1
0
0 0

[PATCH v3 4/8] drm/nouveau: Fix deadlock with fb_helper using new helpers

by Lyude Paul

This removes the potential of deadlocking with fb_helper entirely by preventing it from handling hotplugs during the runtime suspend process as early as possible in the suspend process. If it turns out this is not possible, due to some fb_helper action having been queued up before we got a time to disable hotplugging, we simply return -EBUSY so that the runtime PM core attempts autosuspending the device again once fb_helper isn't doing anything. This fixes one of the issues causing deadlocks on runtime suspend/resume with nouveau on my P50. Signed-off-by: Lyude Paul <lyude(a)redhat.com> Cc: stable(a)vger.kernel.org Cc: Lukas Wunner <lukas(a)wunner.de> Cc: Karol Herbst <karolherbst(a)gmail.com> --- drivers/gpu/drm/nouveau/nouveau_drm.c | 8 ++++++++ drivers/gpu/drm/nouveau/nouveau_fbcon.c | 1 + 2 files changed, 9 insertions(+) diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index ee2546db09c9..d47cb5b2af98 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -836,6 +836,14 @@ nouveau_pmops_runtime_suspend(struct device *dev) return -EBUSY; } + /* There's no way for us to stop fb_helper work in reaction to + * hotplugs later in the RPM process. First off: we don't want to, + * fb_helper should be able to keep the GPU awake. Second off: it is + * capable of grabbing basically any lock in existence. + */ + if (!drm_fb_helper_suspend_hotplug(drm_dev->fb_helper)) + return -EBUSY; + nouveau_switcheroo_optimus_dsm(); ret = nouveau_do_suspend(drm_dev, true); pci_save_state(pdev); diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c index 85c1f10bc2b6..963ba630fd04 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c +++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c @@ -466,6 +466,7 @@ nouveau_fbcon_set_suspend_work(struct work_struct *work) console_unlock(); if (state == FBINFO_STATE_RUNNING) { + drm_fb_helper_resume_hotplug(drm->dev->fb_helper); pm_runtime_mark_last_busy(drm->dev->dev); pm_runtime_put_sync(drm->dev->dev); } -- 2.17.1

6 years, 11 months

1
0
0 0

[PATCH v3 2/8] drm/nouveau: Enable polling even if we have runtime PM

by Lyude Paul

Having runtime PM makes no difference on whether or not we want polling, and it's now safe to just enable polling unconditionally in drm_load() thanks to d61a5c106351 ("drm/nouveau: Fix deadlock on runtime suspend") Signed-off-by: Lyude Paul <lyude(a)redhat.com> Cc: Lukas Wunner <lukas(a)wunner.de> Cc: Peter Ujfalusi <peter.ujfalusi(a)ti.com> Cc: stable(a)vger.kernel.org --- drivers/gpu/drm/nouveau/nouveau_drm.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 5fdc1fbe2ee5..ee2546db09c9 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -592,10 +592,11 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags) pm_runtime_allow(dev->dev); pm_runtime_mark_last_busy(dev->dev); pm_runtime_put(dev->dev); - } else { - /* enable polling for external displays */ - drm_kms_helper_poll_enable(dev); } + + /* enable polling for connectors without hpd */ + drm_kms_helper_poll_enable(dev); + return 0; fail_dispinit: -- 2.17.1

6 years, 11 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror