In the Linux kernel, the following vulnerability has been resolved:[EOL][EOL]drm/amdgpu: csa unmap use uninterruptible lock[EOL][EOL]After process exit to unmap csa and free GPU vm, if signal is accepted[EOL]and then waiting to take vm lock is interrupted and return, it causes[EOL]memory leaking and below warning backtrace.[EOL][EOL]Change to use uninterruptible wait lock fix the issue.[EOL][EOL]WARNING: CPU: 69 PID: 167800 at amd/amdgpu/amdgpu_kms.c:1525[EOL] amdgpu_driver_postclose_kms+0x294/0x2a0 [amdgpu][EOL] Call Trace:[EOL] <TASK>[EOL] drm_file_free.part.0+0x1da/0x230 [drm][EOL] drm_close_helper.isra.0+0x65/0x70 [drm][EOL] drm_release+0x6a/0x120 [drm][EOL] amdgpu_drm_release+0x51/0x60 [amdgpu][EOL] __fput+0x9f/0x280[EOL] ____fput+0xe/0x20[EOL] task_work_run+0x67/0xa0[EOL] do_exit+0x217/0x3c0[EOL] do_group_exit+0x3b/0xb0[EOL] get_signal+0x14a/0x8d0[EOL] arch_do_signal_or_restart+0xde/0x100[EOL] exit_to_user_mode_loop+0xc1/0x1a0[EOL] exit_to_user_mode_prepare+0xf4/0x100[EOL] syscall_exit_to_user_mode+0x17/0x40[EOL] do_syscall_64+0x69/0xc0[EOL][EOL](cherry picked from commit 7dbbfb3c171a6f63b01165958629c9c26abf38ab)
