1 What: /sys/devices/system/machineche 1 What: /sys/devices/system/machinecheck/machinecheckX/tolerant
2 Contact: Borislav Petkov <bp@suse.de> 2 Contact: Borislav Petkov <bp@suse.de>
3 Date: Dec, 2021 3 Date: Dec, 2021
4 Description: 4 Description:
5 Unused and obsolete after the 5 Unused and obsolete after the advent of recoverable machine
6 checks (see last sentence belo 6 checks (see last sentence below) and those are present since
7 2010 (Nehalem). 7 2010 (Nehalem).
8 8
9 Original description: 9 Original description:
10 10
11 The entries appear for each CP 11 The entries appear for each CPU, but they are truly shared
12 between all CPUs. 12 between all CPUs.
13 13
14 Tolerance level. When a machin 14 Tolerance level. When a machine check exception occurs for a
15 non corrected machine check th 15 non corrected machine check the kernel can take different
16 actions. 16 actions.
17 17
18 Since machine check exceptions 18 Since machine check exceptions can happen any time it is
19 sometimes risky for the kernel 19 sometimes risky for the kernel to kill a process because it
20 defies normal kernel locking r 20 defies normal kernel locking rules. The tolerance level
21 configures how hard the kernel 21 configures how hard the kernel tries to recover even at some
22 risk of deadlock. Higher toler 22 risk of deadlock. Higher tolerant values trade potentially
23 better uptime with the risk of 23 better uptime with the risk of a crash or even corruption
24 (for tolerant >= 3). 24 (for tolerant >= 3).
25 25
26 == ========================== 26 == ===========================================================
27 0 always panic on uncorrecte 27 0 always panic on uncorrected errors, log corrected errors
28 1 panic or SIGBUS on uncorre 28 1 panic or SIGBUS on uncorrected errors, log corrected errors
29 2 SIGBUS or log uncorrected 29 2 SIGBUS or log uncorrected errors, log corrected errors
30 3 never panic or SIGBUS, log 30 3 never panic or SIGBUS, log all errors (for testing only)
31 == ========================== 31 == ===========================================================
32 32
33 Default: 1 33 Default: 1
34 34
35 Note this only makes a differe 35 Note this only makes a difference if the CPU allows recovery
36 from a machine check exception 36 from a machine check exception. Current x86 CPUs generally
37 do not. 37 do not.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.