forked from mchehab/rasdaemon
-
Notifications
You must be signed in to change notification settings - Fork 0
/
mce-intel-ivb.c
181 lines (159 loc) · 4.81 KB
/
mce-intel-ivb.c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
/*
* The code below came from Tony Luck mcelog code,
* released under GNU Public General License, v.2
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include <string.h>
#include <stdio.h>
#include "ras-mce-handler.h"
#include "bitfield.h"
/* See IA32 SDM Vol3B Table 16-17 */
static char *pcu_1[] = {
[0] = "No error",
[1] = "Non_IMem_Sel",
[2] = "I_Parity_Error",
[3] = "Bad_OpCode",
[4] = "I_Stack_Underflow",
[5] = "I_Stack_Overflow",
[6] = "D_Stack_Underflow",
[7] = "D_Stack_Overflow",
[8] = "Non-DMem_Sel",
[9] = "D_Parity_Error"
};
static char *pcu_2[] = {
[0x00] = "No Error",
[0x0D] = "MC_IMC_FORCE_SR_S3_TIMEOUT",
[0x0E] = "MC_MC_CPD_UNCPD_ST_TIMEOUT",
[0x0F] = "MC_PKGS_SAFE_WP_TIMEOUT",
[0x43] = "MC_PECI_MAILBOX_QUIESCE_TIMEOUT",
[0x44] = "MC_CRITICAL_VR_FAILED",
[0x45] = "MC_ICC_MAX-NOTSUPPORTED",
[0x5C] = "MC_MORE_THAN_ONE_LT_AGENT",
[0x60] = "MC_INVALID_PKGS_REQ_PCH",
[0x61] = "MC_INVALID_PKGS_REQ_QPI",
[0x62] = "MC_INVALID_PKGS_RES_QPI",
[0x63] = "MC_INVALID_PKGC_RES_PCH",
[0x64] = "MC_INVALID_PKG_STATE_CONFIG",
[0x70] = "MC_WATCHDG_TIMEOUT_PKGC_SLAVE",
[0x71] = "MC_WATCHDG_TIMEOUT_PKGC_MASTER",
[0x72] = "MC_WATCHDG_TIMEOUT_PKGS_MASTER",
[0x7A] = "MC_HA_FAILSTS_CHANGE_DETECTED",
[0x7B] = "MC_PCIE_R2PCIE-RW_BLOCK_ACK_TIMEOUT",
[0x81] = "MC_RECOVERABLE_DIE_THERMAL_TOO_HOT",
};
static struct field pcu_mc4[] = {
FIELD(16, pcu_1),
FIELD(24, pcu_2),
{}
};
/* See IA32 SDM Vol3B Table 16-18 */
static char *memctrl_1[] = {
[0x001] = "Address parity error",
[0x002] = "HA Wrt buffer Data parity error",
[0x004] = "HA Wrt byte enable parity error",
[0x008] = "Corrected patrol scrub error",
[0x010] = "Uncorrected patrol scrub error",
[0x020] = "Corrected spare error",
[0x040] = "Uncorrected spare error",
[0x080] = "Corrected memory read error",
[0x100] = "iMC, WDB, parity errors",
};
static struct field memctrl_mc9[] = {
FIELD(16, memctrl_1),
{}
};
void ivb_decode_model(struct ras_events *ras, struct mce_event *e)
{
struct mce_priv *mce = ras->mce_priv;
uint64_t status = e->status;
uint32_t mca = status & 0xffff;
unsigned rank0 = -1, rank1 = -1, chan;
switch (e->bank) {
case 4:
// Wprintf("PCU: ");
decode_bitfield(e, e->status, pcu_mc4);
// Wprintf("\n");
break;
case 5:
if (mce->cputype == CPU_IVY_BRIDGE_EPEX) {
/* MCACOD already decoded */
mce_snprintf(e->bank_name, "QPI");
}
break;
case 9: case 10: case 11: case 12:
case 13: case 14: case 15: case 16:
// Wprintf("MemCtrl: ");
decode_bitfield(e, e->status, memctrl_mc9);
break;
}
/*
* Memory error specific code. Returns if the error is not a MC one
*/
/* Check if the error is at the memory controller */
if ((mca >> 7) != 1)
return;
/* Ignore unless this is an corrected extended error from an iMC bank */
if (e->bank < 9 || e->bank > 16 || (status & MCI_STATUS_UC) ||
!test_prefix(7, status & 0xefff))
return;
/*
* Parse the reported channel and ranks
*/
chan = EXTRACT(status, 0, 3);
if (chan == 0xf)
return;
mce_snprintf(e->mc_location, "memory_channel=%d", chan);
if (EXTRACT(e->misc, 62, 62))
rank0 = EXTRACT(e->misc, 46, 50);
if (EXTRACT(e->misc, 63, 63))
rank1 = EXTRACT(e->misc, 51, 55);
/*
* FIXME: The conversion from rank to dimm requires to parse the
* DMI tables and call failrank2dimm().
*/
if (rank0 >= 0 && rank1 >= 0)
mce_snprintf(e->mc_location, "ranks=%d and %d",
rank0, rank1);
else if (rank0 >= 0)
mce_snprintf(e->mc_location, "rank=%d", rank0);
else
mce_snprintf(e->mc_location, "rank=%d", rank1);
}
/*
* Ivy Bridge EP and EX processors (family 6, model 62) support additional
* logging for corrected errors in the integrated memory controller (IMC)
* banks. The mode is off by default, but can be enabled by setting the
* "MemError Log Enable" * bit in MSR_ERROR_CONTROL (MSR 0x17f).
* The SDM leaves it as an exercise for the reader to convert the
* faling rank to a DIMM slot.
*/
#if 0
static int failrank2dimm(unsigned failrank, int socket, int channel)
{
switch (failrank) {
case 0: case 1: case 2: case 3:
return 0;
case 4: case 5:
return 1;
case 6: case 7:
if (get_memdimm(socket, channel, 2, 0))
return 2;
else
return 1;
}
return -1;
}
#endif