Re: [linux-yocto] [PATCH] kmemleak: Turn kmemleak_lock to raw spinlock on RT
On 2018-11-14 2:27 p.m., Bruce Ashfield wrote: On 11/13/18 4:11 AM, zhe...@windriver.com wrote: From: He Zhe See https://lore.kernel.org/patchwork/patch/1011368/ for upstream status. kmemleak_lock, as a rwlock on RT, can possibly be held in atomic context and causes the follow BUG. Thanks. I'm waiting to see if there's any upstream feedback, and will wait a couple more days before deciding to merge. The alternative to making this a raw lock would be to simply declare it incompatible with -rt (Which is what we've done with different debug mechanisms in the past). Since there really should be relatively few (none??) -rt specific kmemleaks, and if you are really concerned about -rt, you likely wouldn't be running with it enabled. I still see no comments on this upstream, so I've gone ahead and merged this to the 4.18 branches. Hopefully there won't be any side effects. Bruce Bruce BUG: scheduling while atomic: migration/15/132/0x0002 Modules linked in: iTCO_wdt iTCO_vendor_support intel_rapl pcc_cpufreq pnd2_edac intel_powerclamp coretemp crct10dif_pclmul crct10dif_common aesni_intel matroxfb_base aes_x86_64 matroxfb_g450 matroxfb_accel crypto_simd matroxfb_DAC1064 cryptd glue_helper g450_pll matroxfb_misc i2c_ismt i2c_i801 acpi_cpufreq Preemption disabled at: [] cpu_stopper_thread+0x71/0x100 CPU: 15 PID: 132 Comm: migration/15 Not tainted 4.19.0-rt1-preempt-rt #1 Hardware name: Intel Corp. Harcuvar/Server, BIOS HAVLCRB1.X64.0015.D62.1708310404 08/31/2017 Call Trace: dump_stack+0x4f/0x6a ? cpu_stopper_thread+0x71/0x100 __schedule_bug.cold.16+0x38/0x55 __schedule+0x484/0x6c0 schedule+0x3d/0xe0 rt_spin_lock_slowlock_locked+0x118/0x2a0 rt_spin_lock_slowlock+0x57/0x90 __rt_spin_lock+0x26/0x30 __write_rt_lock+0x23/0x1a0 ? intel_pmu_cpu_dying+0x67/0x70 rt_write_lock+0x2a/0x30 find_and_remove_object+0x1e/0x80 delete_object_full+0x10/0x20 kmemleak_free+0x32/0x50 kfree+0x104/0x1f0 ? x86_pmu_starting_cpu+0x30/0x30 intel_pmu_cpu_dying+0x67/0x70 x86_pmu_dying_cpu+0x1a/0x30 cpuhp_invoke_callback+0x92/0x700 take_cpu_down+0x70/0xa0 multi_cpu_stop+0x62/0xc0 ? cpu_stop_queue_work+0x130/0x130 cpu_stopper_thread+0x79/0x100 smpboot_thread_fn+0x20f/0x2d0 kthread+0x121/0x140 ? sort_range+0x30/0x30 ? kthread_park+0x90/0x90 ret_from_fork+0x35/0x40 The following call trace, caused by grabbing kmemleak_lock twice, is also observed. kernel BUG at kernel/locking/rtmutex.c:1048! invalid opcode: [#1] PREEMPT SMP PTI CPU: 5 PID: 689 Comm: mkfs.ext4 Not tainted 4.18.16-rt9-preempt-rt #1 Hardware name: Intel Corp. Harcuvar/Server, BIOS HAVLCRB1.X64.0015.D62.1708310404 08/31/2017 RIP: 0010:rt_spin_lock_slowlock_locked+0x277/0x2a0 Code: e8 5e 64 61 ff e9 bc fe ff ff e8 54 64 61 ff e9 b7 fe ff ff 0f 0b e8 98 57 53 ff e9 43 fe ff ff e8 8e 57 53 ff e9 74 ff ff ff <0f> 0b 0f 0b 0f 0b 48 8b 43 10 48 85 c0 74 06 48 3b 58 38 75 0b 49 RSP: 0018:936846d4f3b0 EFLAGS: 00010046 RAX: 8e3680361e00 RBX: 83a8b240 RCX: 0001 RDX: RSI: 8e3680361e00 RDI: 83a8b258 RBP: 936846d4f3e8 R08: 8e3680361e01 R09: 82adfdf0 R10: 827ede18 R11: R12: 936846d4f3f8 R13: 8e3680361e00 R14: 936846d4f3f8 R15: 0246 FS: 7fc8b6bfd780() GS:8e369f34() knlGS: CS: 0010 DS: ES: CR0: 80050033 CR2: 55fb5659e000 CR3: 0007fdd14000 CR4: 003406e0 Call Trace: ? preempt_count_add+0x74/0xc0 rt_spin_lock_slowlock+0x57/0x90 ? __kernel_text_address+0x12/0x40 ? __save_stack_trace+0x75/0x100 __rt_spin_lock+0x26/0x30 __write_rt_lock+0x23/0x1a0 rt_write_lock+0x2a/0x30 create_object+0x17d/0x2b0 kmemleak_alloc+0x34/0x50 kmem_cache_alloc+0x146/0x220 ? mempool_alloc_slab+0x15/0x20 mempool_alloc_slab+0x15/0x20 mempool_alloc+0x65/0x170 sg_pool_alloc+0x21/0x60 __sg_alloc_table+0x101/0x160 ? sg_free_table_chained+0x30/0x30 sg_alloc_table_chained+0x8b/0xb0 scsi_init_sgtable+0x31/0x90 scsi_init_io+0x44/0x130 sd_setup_write_same16_cmnd+0xef/0x150 sd_init_command+0x6bf/0xaa0 ? cgroup_base_stat_cputime_account_end.isra.0+0x26/0x60 ? elv_rb_del+0x2a/0x40 scsi_setup_cmnd+0x8e/0x140 scsi_prep_fn+0x5d/0x140 blk_peek_request+0xda/0x2f0 scsi_request_fn+0x33/0x550 ? cfq_rb_erase+0x23/0x40 __blk_run_queue+0x43/0x60 cfq_insert_request+0x2f3/0x5d0 __elv_add_request+0x160/0x290 blk_flush_plug_list+0x204/0x230 schedule+0x87/0xe0 __write_rt_lock+0x18b/0x1a0 rt_write_lock+0x2a/0x30 create_object+0x17d/0x2b0 kmemleak_alloc+0x34/0x50 __kmalloc_node+0x1cd/0x340 alloc_request_size+0x30/0x70 mempool_alloc+0x65/0x170 ? ioc_lookup_icq+0x54/0x70 get_request+0x4e3/0x8d0 ? wait_woken+0x80/0x80 blk_queue_bio+0x153/0x470 generic_make_request+0x1dc/0x3f0 submit_bio+0x49/0x140 ? next_bio+0x38/0x40 submit_bio_wait+0x59/0x90 blkdev_issue_discard+0x7a/0xd0 ?
Re: [linux-yocto] [PATCH] kmemleak: Turn kmemleak_lock to raw spinlock on RT
On 11/13/18 4:11 AM, zhe...@windriver.com wrote: From: He Zhe See https://lore.kernel.org/patchwork/patch/1011368/ for upstream status. kmemleak_lock, as a rwlock on RT, can possibly be held in atomic context and causes the follow BUG. Thanks. I'm waiting to see if there's any upstream feedback, and will wait a couple more days before deciding to merge. The alternative to making this a raw lock would be to simply declare it incompatible with -rt (Which is what we've done with different debug mechanisms in the past). Since there really should be relatively few (none??) -rt specific kmemleaks, and if you are really concerned about -rt, you likely wouldn't be running with it enabled. Bruce BUG: scheduling while atomic: migration/15/132/0x0002 Modules linked in: iTCO_wdt iTCO_vendor_support intel_rapl pcc_cpufreq pnd2_edac intel_powerclamp coretemp crct10dif_pclmul crct10dif_common aesni_intel matroxfb_base aes_x86_64 matroxfb_g450 matroxfb_accel crypto_simd matroxfb_DAC1064 cryptd glue_helper g450_pll matroxfb_misc i2c_ismt i2c_i801 acpi_cpufreq Preemption disabled at: [] cpu_stopper_thread+0x71/0x100 CPU: 15 PID: 132 Comm: migration/15 Not tainted 4.19.0-rt1-preempt-rt #1 Hardware name: Intel Corp. Harcuvar/Server, BIOS HAVLCRB1.X64.0015.D62.1708310404 08/31/2017 Call Trace: dump_stack+0x4f/0x6a ? cpu_stopper_thread+0x71/0x100 __schedule_bug.cold.16+0x38/0x55 __schedule+0x484/0x6c0 schedule+0x3d/0xe0 rt_spin_lock_slowlock_locked+0x118/0x2a0 rt_spin_lock_slowlock+0x57/0x90 __rt_spin_lock+0x26/0x30 __write_rt_lock+0x23/0x1a0 ? intel_pmu_cpu_dying+0x67/0x70 rt_write_lock+0x2a/0x30 find_and_remove_object+0x1e/0x80 delete_object_full+0x10/0x20 kmemleak_free+0x32/0x50 kfree+0x104/0x1f0 ? x86_pmu_starting_cpu+0x30/0x30 intel_pmu_cpu_dying+0x67/0x70 x86_pmu_dying_cpu+0x1a/0x30 cpuhp_invoke_callback+0x92/0x700 take_cpu_down+0x70/0xa0 multi_cpu_stop+0x62/0xc0 ? cpu_stop_queue_work+0x130/0x130 cpu_stopper_thread+0x79/0x100 smpboot_thread_fn+0x20f/0x2d0 kthread+0x121/0x140 ? sort_range+0x30/0x30 ? kthread_park+0x90/0x90 ret_from_fork+0x35/0x40 The following call trace, caused by grabbing kmemleak_lock twice, is also observed. kernel BUG at kernel/locking/rtmutex.c:1048! invalid opcode: [#1] PREEMPT SMP PTI CPU: 5 PID: 689 Comm: mkfs.ext4 Not tainted 4.18.16-rt9-preempt-rt #1 Hardware name: Intel Corp. Harcuvar/Server, BIOS HAVLCRB1.X64.0015.D62.1708310404 08/31/2017 RIP: 0010:rt_spin_lock_slowlock_locked+0x277/0x2a0 Code: e8 5e 64 61 ff e9 bc fe ff ff e8 54 64 61 ff e9 b7 fe ff ff 0f 0b e8 98 57 53 ff e9 43 fe ff ff e8 8e 57 53 ff e9 74 ff ff ff <0f> 0b 0f 0b 0f 0b 48 8b 43 10 48 85 c0 74 06 48 3b 58 38 75 0b 49 RSP: 0018:936846d4f3b0 EFLAGS: 00010046 RAX: 8e3680361e00 RBX: 83a8b240 RCX: 0001 RDX: RSI: 8e3680361e00 RDI: 83a8b258 RBP: 936846d4f3e8 R08: 8e3680361e01 R09: 82adfdf0 R10: 827ede18 R11: R12: 936846d4f3f8 R13: 8e3680361e00 R14: 936846d4f3f8 R15: 0246 FS: 7fc8b6bfd780() GS:8e369f34() knlGS: CS: 0010 DS: ES: CR0: 80050033 CR2: 55fb5659e000 CR3: 0007fdd14000 CR4: 003406e0 Call Trace: ? preempt_count_add+0x74/0xc0 rt_spin_lock_slowlock+0x57/0x90 ? __kernel_text_address+0x12/0x40 ? __save_stack_trace+0x75/0x100 __rt_spin_lock+0x26/0x30 __write_rt_lock+0x23/0x1a0 rt_write_lock+0x2a/0x30 create_object+0x17d/0x2b0 kmemleak_alloc+0x34/0x50 kmem_cache_alloc+0x146/0x220 ? mempool_alloc_slab+0x15/0x20 mempool_alloc_slab+0x15/0x20 mempool_alloc+0x65/0x170 sg_pool_alloc+0x21/0x60 __sg_alloc_table+0x101/0x160 ? sg_free_table_chained+0x30/0x30 sg_alloc_table_chained+0x8b/0xb0 scsi_init_sgtable+0x31/0x90 scsi_init_io+0x44/0x130 sd_setup_write_same16_cmnd+0xef/0x150 sd_init_command+0x6bf/0xaa0 ? cgroup_base_stat_cputime_account_end.isra.0+0x26/0x60 ? elv_rb_del+0x2a/0x40 scsi_setup_cmnd+0x8e/0x140 scsi_prep_fn+0x5d/0x140 blk_peek_request+0xda/0x2f0 scsi_request_fn+0x33/0x550 ? cfq_rb_erase+0x23/0x40 __blk_run_queue+0x43/0x60 cfq_insert_request+0x2f3/0x5d0 __elv_add_request+0x160/0x290 blk_flush_plug_list+0x204/0x230 schedule+0x87/0xe0 __write_rt_lock+0x18b/0x1a0 rt_write_lock+0x2a/0x30 create_object+0x17d/0x2b0 kmemleak_alloc+0x34/0x50 __kmalloc_node+0x1cd/0x340 alloc_request_size+0x30/0x70 mempool_alloc+0x65/0x170 ? ioc_lookup_icq+0x54/0x70 get_request+0x4e3/0x8d0 ? wait_woken+0x80/0x80 blk_queue_bio+0x153/0x470 generic_make_request+0x1dc/0x3f0 submit_bio+0x49/0x140 ? next_bio+0x38/0x40 submit_bio_wait+0x59/0x90 blkdev_issue_discard+0x7a/0xd0 ? _raw_spin_unlock_irqrestore+0x18/0x50 blk_ioctl_discard+0xc7/0x110 blkdev_ioctl+0x57e/0x960 ? __wake_up+0x13/0x20 block_ioctl+0x3d/0x50 do_vfs_ioctl+0xa8/0x610 ? vfs_write+0x166/0x1b0 ksys_ioctl+0x67/0x90