https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6863/fi-apl-guc/igt@gem_exec_gttfill@basic.html

<6> [76.877592] [IGT] gem_exec_gttfill: starting subtest basic
<6> [76.882725] gem_exec_gttfil (1094): drop_caches: 4
<4> [76.884330] 
<4> [76.884342] ======================================================
<4> [76.884347] WARNING: possible circular locking dependency detected
<4> [76.884354] 5.3.0-rc8-CI-CI_DRM_6863+ #1 Tainted: G     U           
<4> [76.884359] ------------------------------------------------------
<4> [76.884364] gem_exec_gttfil/1094 is trying to acquire lock:
<4> [76.884371] 00000000d7b6d922 (cpu_hotplug_lock.rw_sem){++++}, at: stop_machine+0x12/0x30
<4> [76.884384] 
but task is already holding lock:
<4> [76.884389] 00000000b2ffc63a (fs_reclaim){+.+.}, at: fs_reclaim_acquire.part.23+0x0/0x30
<4> [76.884399] 
which lock already depends on the new lock.

<4> [76.884405] 
the existing dependency chain (in reverse order) is:
<4> [76.884411] 
-> #1 (fs_reclaim){+.+.}:
<4> [76.884418]        fs_reclaim_acquire.part.23+0x24/0x30
<4> [76.884425]        kmem_cache_alloc_trace+0x2a/0x300
<4> [76.884431]        intel_cpuc_prepare+0x3b/0x1a0
<4> [76.884438]        cpuhp_invoke_callback+0x9b/0xa20
<4> [76.884444]        _cpu_up+0xa2/0x140
<4> [76.884449]        do_cpu_up+0x6d/0xb0
<4> [76.884455]        smp_init+0x5d/0xa4
<4> [76.884462]        kernel_init_freeable+0x16f/0x354
<4> [76.884469]        kernel_init+0x5/0x100
<4> [76.884474]        ret_from_fork+0x3a/0x50
<4> [76.884478] 
-> #0 (cpu_hotplug_lock.rw_sem){++++}:
<4> [76.884487]        __lock_acquire+0x15d8/0x1e90
<4> [76.884493]        lock_acquire+0xa6/0x1c0
<4> [76.884498]        cpus_read_lock+0x34/0xe0
<4> [76.884503]        stop_machine+0x12/0x30
<4> [76.884607]        bxt_vtd_ggtt_clear_range__BKL+0x33/0x50 [i915]
<4> [76.884695]        ggtt_unbind_vma+0x49/0x60 [i915]
<4> [76.884778]        i915_vma_unbind+0x1a2/0x490 [i915]
<4> [76.884860]        i915_gem_object_unbind+0x153/0x1c0 [i915]
<4> [76.884939]        i915_gem_shrink+0x2b3/0x640 [i915]
<4> [76.885018]        i915_gem_shrink_all+0x38/0x60 [i915]
<4> [76.885089]        i915_drop_caches_set+0x206/0x2f0 [i915]
<4> [76.885102]        simple_attr_write+0xb0/0xd0
<4> [76.885108]        full_proxy_write+0x51/0x80
<4> [76.885115]        vfs_write+0xbd/0x1d0
<4> [76.885120]        ksys_write+0x8f/0xe0
<4> [76.885126]        do_syscall_64+0x55/0x1c0
<4> [76.885132]        entry_SYSCALL_64_after_hwframe+0x49/0xbe
<4> [76.885137] 
other info that might help us debug this:

<4> [76.885144]  Possible unsafe locking scenario:

<4> [76.885149]        CPU0                    CPU1
<4> [76.885153]        ----                    ----
<4> [76.885157]   lock(fs_reclaim);
<4> [76.885162]                                lock(cpu_hotplug_lock.rw_sem);
<4> [76.885168]                                lock(fs_reclaim);
<4> [76.885173]   lock(cpu_hotplug_lock.rw_sem);
<4> [76.885177] 
 *** DEADLOCK ***