Message-ID: <20161116143746.zoxdxrfqvmx35wln@treble>
Date:   Wed, 16 Nov 2016 08:37:46 -0600
From:   Josh Poimboeuf <jpoimboe@...hat.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Vince Weaver <vincent.weaver@...ne.edu>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Ingo Molnar <mingo@...hat.com>,
        Arnaldo Carvalho de Melo <acme@...nel.org>,
        "davej@...emonkey.org.uk" <davej@...emonkey.org.uk>,
        "dvyukov@...gle.com" <dvyukov@...gle.com>,
        Stephane Eranian <eranian@...il.com>
Subject: Re: perf: fuzzer KASAN unwind_get_return_address

On Wed, Nov 16, 2016 at 02:03:37PM +0100, Peter Zijlstra wrote:
> On Tue, Nov 15, 2016 at 02:57:48PM -0600, Josh Poimboeuf wrote:
> > Would you mind posting a disassembly of unwind_get_return_address()?
> > Any idea how recreatable it is?  (In particular I'd be interested in
> > seeing this dump with the latest unwinder improvements in the -tip tree,
> > which dump the pt_regs associated with an interrupt.)
> 
> Fairly reproducable it seems, doesn't seem to include pt_regs dumps
> though :/
> 
> tip/master as of this morning.

Thanks.  This is actually a different issue than the one reported by
Vince.  In this case FRAME_POINTER is disabled, so it uses the "guess"
unwinder which scans every address on the stack, looking for text
addresses.  So the kasan errors are expected.

(The missing pt_regs are also expected: the guess unwinder doesn't show
them.)

I'll work up a patch to fix this.  I still have no idea what's causing
Vince's bug in the frame pointer unwinder.

> 3==================================================================
> 3BUG: KASAN: stack-out-of-bounds in unwind_next_frame+0x1ba/0x1f0 at addr ffff88042fc87be0
> 3Read of size 8 by task swapper/28/0
> 0page:ffffea0010bf21c0 count:1 mapcount:0 mapping:          (null) index:0x0c
> 0flags: 0x2ffff8000000400(reserved)
> 1page dumped because: kasan: bad access detected
> dCPU: 28 PID: 0 Comm: swapper/28 Not tainted 4.9.0-rc5-00530-gd8866fc-dirty #2
> dHardware name: Intel Corporation S2600GZ/S2600GZ, BIOS SE5C600.86B.02.02.0002.122320131210 12/23/2013
> dCall Trace:
> d <NMI>
> d ? dump_stack+0x5e/0x89
> d ? kasan_report_error+0x4a5/0x4d0
> d ? __asan_report_load8_noabort+0x45/0x50
> d ? __kernel_text_address+0x20/0xa0
> d ? unwind_next_frame+0x1ba/0x1f0
> d ? unwind_next_frame+0x1ba/0x1f0
> d ? perf_callchain_kernel+0x33c/0x540
> d ? arch_perf_update_userpage+0x340/0x340
> d ? get_perf_callchain+0x24d/0x610
> d ? put_callchain_buffers+0x50/0x50
> d ? number+0x653/0x830
> d ? perf_callchain+0x126/0x190
> d ? perf_prepare_sample+0x720/0x1010
> d ? perf_event_output_forward+0x81/0xf0
> d ? perf_prepare_sample+0x1010/0x1010
> d ? pointer+0x880/0x880
> d ? perf_event_update_userpage+0x16/0x730
> d ? __perf_event_overflow+0x1a0/0x510
> d ? intel_pmu_handle_irq+0x34b/0xa90
> d ? intel_pmu_save_and_restart+0xd0/0xd0
> d ? acpi_os_read_memory+0x205/0x23c
> d ? format_decode+0xc5/0x7a0
> d ? vunmap_page_range+0x26a/0x400
> d ? ghes_copy_tofrom_phys+0x141/0x270
> d ? ghes_read_estatus+0x112/0x5a0
> d ? ghes_copy_tofrom_phys+0x270/0x270
> d ? early_printk+0xa4/0xd0
> d ? devkmsg_sysctl_set_loglvl+0x160/0x160
> d ? perf_event_nmi_handler+0x28/0x40
> d ? nmi_handle+0xa1/0x250
> d ? default_do_nmi+0x61/0x170
> d ? do_nmi+0x191/0x200
> d ? end_repeat_nmi+0x1a/0x1e
> d ? format_decode+0xc5/0x7a0
> d ? format_decode+0xc5/0x7a0
> d ? format_decode+0xc5/0x7a0
> d <EOE>
> d <IRQ>
> d ? vsnprintf+0xfc/0x15e0
> d ? pointer+0x880/0x880
> d ? x86_pmu_enable_all+0x1c0/0x1c0
> d ? vscnprintf+0x9/0x30
> d ? early_vprintk+0xb0/0x130
> d ? trace_raw_output_console+0x160/0x160
> d ? memcpy+0x34/0x50
> d ? x86_pmu_commit_txn+0x180/0x260
> d ? events_sysfs_show+0xb0/0xb0
> d ? save_stack+0x33/0xb0
> d ? hrtimer_init+0x120/0x120
> d ? timerqueue_del+0x62/0x140
> d ? perf_event_update_userpage+0x16/0x730
> d ? perf_event_update_userpage+0x16/0x730
> d ? x86_perf_event_set_period+0x239/0x450
> d ? perf_event_update_userpage+0x16/0x730
> d ? x86_pmu_enable+0x5f7/0xaa0
> d ? printk+0xb6/0xef
> d ? printk_emit+0xa0/0xa0
> d ? _raw_spin_unlock_irqrestore+0x42/0x70
> d ? ___ratelimit+0x1e4/0x3f0
> d ? irq_work_run_list+0xa1/0xf0
> d ? irq_work_run+0x14/0x40
> d ? smp_call_function_single_interrupt+0x60/0x80
> d ? call_function_single_interrupt+0x89/0x90
> d <EOI>
> d ? cpuidle_enter_state+0x113/0x780
> d ? cpuidle_enter_state+0x10e/0x780
> d ? cpu_load_update_nohz_stop+0x155/0x1b0
> d ? cpu_startup_entry+0x19a/0x2c0
> d ? start_cpu+0x5/0x14
> 3Memory state around the buggy address:
> 3 ffff88042fc87a80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 3 ffff88042fc87b00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 3>ffff88042fc87b80: 00 00 00 00 00 00 00 00 00 00 00 00 f1 f1 f1 f1
> 3                                                       ^
> 3 ffff88042fc87c00: 00 f4 f4 f4 f3 f3 f3 f3 00 00 00 00 00 00 00 00
> 3 ffff88042fc87c80: 00 f1 f1 f1 f1 00 00 00 00 00 00 00 00 00 00 00
> 3==================================================================
> 
> 
> 3==================================================================
> 3BUG: KASAN: stack-out-of-bounds in unwind_next_frame+0x1ba/0x1f0 at addr ffff880424a97878
> 3Read of size 8 by task perf_fuzzer/3451
> 0page:ffffea001092a5c0 count:0 mapcount:0 mapping:          (null) index:0x0c
> 0flags: 0x2ffff8000000000()
> 1page dumped because: kasan: bad access detected
> dCPU: 28 PID: 3451 Comm: perf_fuzzer Not tainted 4.9.0-rc5-00530-gd8866fc-dirty #2
> dHardware name: Intel Corporation S2600GZ/S2600GZ, BIOS SE5C600.86B.02.02.0002.122320131210 12/23/2013
> dCall Trace:
> d ? dump_stack+0x5e/0x89
> d ? kasan_report_error+0x4a5/0x4d0
> d ? __asan_report_load8_noabort+0x45/0x50
> d ? __kernel_text_address+0x20/0xa0
> d ? unwind_next_frame+0x1ba/0x1f0
> d ? unwind_next_frame+0x1ba/0x1f0
> d ? perf_callchain_kernel+0x33c/0x540
> d ? arch_perf_update_userpage+0x340/0x340
> d ? get_perf_callchain+0x24d/0x610
> d ? put_callchain_buffers+0x50/0x50
> d ? ipv6_flowlabel_opt+0x1111/0x17d0
> d ? perf_log_itrace_start+0x3a0/0x3a0
> d ? cpumask_next_and+0x5a/0xa0
> d ? ktime_get_raw_fast_ns+0xd3/0x1e0
> d ? perf_callchain+0x126/0x190
> d ? perf_prepare_sample+0x720/0x1010
> d ? perf_event_output_forward+0x81/0xf0
> d ? perf_prepare_sample+0x1010/0x1010
> d ? perf_event_update_userpage+0x16/0x730
> d ? kasan_unpoison_shadow+0x31/0x40
> d ? get_page_from_freelist+0x52e/0x2310
> d ? perf_output_begin+0x3a1/0x9b0
> d ? cpu_clock_event_add+0x17/0x20
> d ? __perf_event_overflow+0x1a0/0x510
> d ? perf_swevent_overflow+0x156/0x1f0
> d ? perf_tp_event+0x3e8/0x5c0
> d ? perf_output_begin_backward+0x960/0x960
> d ? perf_tp_event_match.isra.85.part.86+0x140/0x140
> d ? __mark_inode_dirty+0x459/0xa50
> d ? legitimize_path.isra.28+0x6b/0x150
> d ? unlazy_walk+0x456/0x790
> d ? memset+0x1f/0x40
> d ? perf_trace_writeback_dirty_inode_template+0x3af/0x610
> d ? save_stack+0x33/0xb0
> d ? inode_congested+0x450/0x450
> d ? dput+0x1de/0x530
> d ? walk_component+0x2cc/0xdc0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? pick_link+0xbe0/0xbe0
> d ? inode_congested+0x450/0x450
> d ? __mark_inode_dirty+0x459/0xa50
> d ? proc_sys_setattr+0x84/0xb0
> d ? notify_change+0x4d6/0xc40
> d ? security_inode_need_killpriv+0x58/0x80
> d ? do_truncate+0xd7/0x160
> d ? file_open_root+0x1a0/0x1a0
> d ? path_openat+0x97f/0x3b30
> d ? vfs_rename+0x14a0/0x14a0
> d ? getname_flags+0xba/0x500
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? do_filp_open+0x175/0x230
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? may_open_dev+0xc0/0xc0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? save_stack+0x33/0xb0
> d ? do_sys_open+0x16d/0x310
> d ? SyS_write+0xab/0x160
> d ? filp_open+0x50/0x50
> d ? task_stopped_code+0xf0/0xf0
> d ? trace_hardirqs_on_thunk+0x1a/0x1c
> d ? entry_SYSCALL_64_fastpath+0x18/0xa8
> 3Memory state around the buggy address:
> 3 ffff880424a97700: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> 3 ffff880424a97780: 00 00 00 00 00 00 00 00 f3 f3 f3 f3 f3 f3 f3 f3
> 3>ffff880424a97800: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 f1
> 3                                                                ^
> 3 ffff880424a97880: f1 f1 f1 04 f4 f4 f4 f2 f2 f2 f2 00 f4 f4 f4 f3
> 3 ffff880424a97900: f3 f3 f3 00 00 00 00 00 00 00 00 00 00 00 00 00
> 3==================================================================

-- 
Josh

Powered by blists - more mailing lists