1 .. include:: <isonum.txt>
3 ============================================
4 Reliability, Availability and Serviceability
5 ============================================
10 Reliability, Availability and Serviceability (RAS) is a concept used on
11 servers meant to measure their robustness.
14 is the probability that a system will produce correct outputs.
16 * Generally measured as Mean Time Between Failures (MTBF)
17 * Enhanced by features that help to avoid, detect and repair hardware faults
20 is the probability that a system is operational at a given time
22 * Generally measured as a percentage of downtime per a period of time
23 * Often uses mechanisms to detect and correct hardware faults in
26 Serviceability (or maintainability)
27 is the simplicity and speed with which a system can be repaired or
30 * Generally measured on Mean Time Between Repair (MTBR)
35 In order to reduce systems downtime, a system should be capable of detecting
36 hardware errors, and, when possible correcting them in runtime. It should
37 also provide mechanisms to detect hardware degradation, in order to warn
38 the system administrator to take the action of replacing a component before
39 it causes data loss or system downtime.
41 Among the monitoring measures, the most usual ones include:
43 * CPU – detect errors at instruction execution and at L1/L2/L3 caches;
44 * Memory – add error correction logic (ECC) to detect and correct errors;
45 * I/O – add CRC checksums for transferred data;
46 * Storage – RAID, journal file systems, checksums,
47 Self-Monitoring, Analysis and Reporting Technology (SMART).
49 By monitoring the number of occurrences of error detections, it is possible
50 to identify if the probability of hardware errors is increasing, and, on such
51 case, do a preventive maintenance to replace a degraded component while
52 those errors are correctable.
57 Most mechanisms used on modern systems use technologies like Hamming
58 Codes that allow error correction when the number of errors on a bit packet
59 is below a threshold. If the number of errors is above, those mechanisms
60 can indicate with a high degree of confidence that an error happened, but
63 Also, sometimes an error occur on a component that it is not used. For
64 example, a part of the memory that it is not currently allocated.
66 That defines some categories of errors:
68 * **Correctable Error (CE)** - the error detection mechanism detected and
69 corrected the error. Such errors are usually not fatal, although some
70 Kernel mechanisms allow the system administrator to consider them as fatal.
72 * **Uncorrected Error (UE)** - the amount of errors happened above the error
73 correction threshold, and the system was unable to auto-correct.
75 * **Fatal Error** - when an UE error happens on a critical component of the
76 system (for example, a piece of the Kernel got corrupted by an UE), the
77 only reliable way to avoid data corruption is to hang or reboot the machine.
79 * **Non-fatal Error** - when an UE error happens on an unused component,
80 like a CPU in power down state or an unused memory bank, the system may
81 still run, eventually replacing the affected hardware by a hot spare,
84 Also, when an error happens on a userspace process, it is also possible to
85 kill such process and let userspace restart it.
87 The mechanism for handling non-fatal errors is usually complex and may
88 require the help of some userspace application, in order to apply the
89 policy desired by the system administrator.
91 Identifying a bad hardware component
92 ------------------------------------
94 Just detecting a hardware flaw is usually not enough, as the system needs
95 to pinpoint to the minimal replaceable unit (MRU) that should be exchanged
96 to make the hardware reliable again.
98 So, it requires not only error logging facilities, but also mechanisms that
99 will translate the error message to the silkscreen or component label for
102 Typically, it is very complex for memory, as modern CPUs interlace memory
103 from different memory modules, in order to provide a better performance. The
104 DMI BIOS usually have a list of memory module labels, with can be obtained
105 using the ``dmidecode`` tool. For example, on a desktop machine, it shows::
113 Locator: ChannelA-DIMM0
116 Type Detail: Synchronous
119 Configured Clock Speed: 2133 MHz
121 On the above example, a DDR4 SO-DIMM memory module is located at the
122 system's memory labeled as "BANK 0", as given by the *bank locator* field.
123 Please notice that, on such system, the *total width* is equal to the
124 *data width*. It means that such memory module doesn't have error
125 detection/correction mechanisms.
127 Unfortunately, not all systems use the same field to specify the memory
128 bank. On this example, from an older server, ``dmidecode`` shows::
132 Error Information Handle: Not Provided
139 Bank Locator: Not Specified
141 Type Detail: Synchronous Registered (Buffered)
144 Configured Clock Speed: 1600 MHz
146 There, the DDR3 RDIMM memory module is located at the system's memory labeled
147 as "DIMM_A1", as given by the *locator* field. Please notice that this
148 memory module has 64 bits of *data width* and 72 bits of *total width*. So,
149 it has 8 extra bits to be used by error detection and correction mechanisms.
150 Such kind of memory is called Error-correcting code memory (ECC memory).
152 To make things even worse, it is not uncommon that systems with different
153 labels on their system's board to use exactly the same BIOS, meaning that
154 the labels provided by the BIOS won't match the real ones.
159 As mentioned in the previous section, ECC memory has extra bits to be
160 used for error correction. In the above example, a memory module has
161 64 bits of *data width*, and 72 bits of *total width*. The extra 8
162 bits which are used for the error detection and correction mechanisms
163 are referred to as the *syndrome*\ [#f1]_\ [#f2]_.
165 So, when the cpu requests the memory controller to write a word with
166 *data width*, the memory controller calculates the *syndrome* in real time,
167 using Hamming code, or some other error correction code, like SECDED+,
168 producing a code with *total width* size. Such code is then written
169 on the memory modules.
171 At read, the *total width* bits code is converted back, using the same
172 ECC code used on write, producing a word with *data width* and a *syndrome*.
173 The word with *data width* is sent to the CPU, even when errors happen.
175 The memory controller also looks at the *syndrome* in order to check if
176 there was an error, and if the ECC code was able to fix such error.
177 If the error was corrected, a Corrected Error (CE) happened. If not, an
178 Uncorrected Error (UE) happened.
180 The information about the CE/UE errors is stored on some special registers
181 at the memory controller and can be accessed by reading such registers,
182 either by BIOS, by some special CPUs or by Linux EDAC driver. On x86 64
183 bit CPUs, such errors can also be retrieved via the Machine Check
184 Architecture (MCA)\ [#f3]_.
186 .. [#f1] Please notice that several memory controllers allow operation on a
187 mode called "Lock-Step", where it groups two memory modules together,
188 doing 128-bit reads/writes. That gives 16 bits for error correction, with
189 significantly improves the error correction mechanism, at the expense
190 that, when an error happens, there's no way to know what memory module is
191 to blame. So, it has to blame both memory modules.
193 .. [#f2] Some memory controllers also allow using memory in mirror mode.
194 On such mode, the same data is written to two memory modules. At read,
195 the system checks both memory modules, in order to check if both provide
196 identical data. On such configuration, when an error happens, there's no
197 way to know what memory module is to blame. So, it has to blame both
198 memory modules (or 4 memory modules, if the system is also on Lock-step
201 .. [#f3] For more details about the Machine Check Architecture (MCA),
202 please read Documentation/x86/x86_64/machinecheck.rst at the Kernel tree.
204 EDAC - Error Detection And Correction
205 *************************************
209 "bluesmoke" was the name for this device driver subsystem when it
210 was "out-of-tree" and maintained at http://bluesmoke.sourceforge.net.
211 That site is mostly archaic now and can be used only for historical
214 When the subsystem was pushed upstream for the first time, on
215 Kernel 2.6.16, it was renamed to ``EDAC``.
220 The ``edac`` kernel module's goal is to detect and report hardware errors
221 that occur within the computer system running under linux.
226 Memory Correctable Errors (CE) and Uncorrectable Errors (UE) are the
227 primary errors being harvested. These types of errors are harvested by
228 the ``edac_mc`` device.
230 Detecting CE events, then harvesting those events and reporting them,
231 **can** but must not necessarily be a predictor of future UE events. With
232 CE events only, the system can and will continue to operate as no data
233 has been damaged yet.
235 However, preventive maintenance and proactive part replacement of memory
236 modules exhibiting CEs can reduce the likelihood of the dreaded UE events
239 Other hardware elements
240 -----------------------
242 A new feature for EDAC, the ``edac_device`` class of device, was added in
243 the 2.6.23 version of the kernel.
245 This new device type allows for non-memory type of ECC hardware detectors
246 to have their states harvested and presented to userspace via the sysfs
249 Some architectures have ECC detectors for L1, L2 and L3 caches,
250 along with DMA engines, fabric switches, main data path switches,
251 interconnections, and various other hardware data paths. If the hardware
252 reports it, then a edac_device device probably can be constructed to
253 harvest and present that to userspace.
259 In addition, PCI devices are scanned for PCI Bus Parity and SERR Errors
260 in order to determine if errors are occurring during data transfers.
262 The presence of PCI Parity errors must be examined with a grain of salt.
263 There are several add-in adapters that do **not** follow the PCI specification
264 with regards to Parity generation and reporting. The specification says
265 the vendor should tie the parity status bits to 0 if they do not intend
266 to generate parity. Some vendors do not do this, and thus the parity bit
267 can "float" giving false positives.
269 There is a PCI device attribute located in sysfs that is checked by
270 the EDAC PCI scanning code. If that attribute is set, PCI parity/error
271 scanning is skipped for that device. The attribute is::
275 and is located in ``/sys/devices/pci<XXX>/0000:XX:YY.Z`` directories for
282 EDAC is composed of a "core" module (``edac_core.ko``) and several Memory
283 Controller (MC) driver modules. On a given system, the CORE is loaded
284 and one MC driver will be loaded. Both the CORE and the MC driver (or
285 ``edac_device`` driver) have individual versions that reflect current
286 release level of their respective modules.
288 Thus, to "report" on what version a system is running, one must report
289 both the CORE's and the MC driver's versions.
295 If ``edac`` was statically linked with the kernel then no loading
296 is necessary. If ``edac`` was built as modules then simply modprobe
297 the ``edac`` pieces that you need. You should be able to modprobe
298 hardware-specific modules and have the dependencies load the necessary
303 $ modprobe amd76x_edac
305 loads both the ``amd76x_edac.ko`` memory controller module and the
306 ``edac_mc.ko`` core module.
312 EDAC presents a ``sysfs`` interface for control and reporting purposes. It
313 lives in the /sys/devices/system/edac directory.
315 Within this directory there currently reside 2 components:
317 ======= ==============================
318 mc memory controller(s) system
319 pci PCI control and status system
320 ======= ==============================
324 Memory Controller (mc) Model
325 ----------------------------
327 Each ``mc`` device controls a set of memory modules [#f4]_. These modules
328 are laid out in a Chip-Select Row (``csrowX``) and Channel table (``chX``).
329 There can be multiple csrows and multiple channels.
331 .. [#f4] Nowadays, the term DIMM (Dual In-line Memory Module) is widely
332 used to refer to a memory module, although there are other memory
333 packaging alternatives, like SO-DIMM, SIMM, etc. The UEFI
334 specification (Version 2.7) defines a memory module in the Common
335 Platform Error Record (CPER) section to be an SMBIOS Memory Device
336 (Type 17). Along this document, and inside the EDAC subsystem, the term
337 "dimm" is used for all memory modules, even when they use a
338 different kind of packaging.
340 Memory controllers allow for several csrows, with 8 csrows being a
341 typical value. Yet, the actual number of csrows depends on the layout of
342 a given motherboard, memory controller and memory module characteristics.
344 Dual channels allow for dual data length (e. g. 128 bits, on 64 bit systems)
345 data transfers to/from the CPU from/to memory. Some newer chipsets allow
346 for more than 2 channels, like Fully Buffered DIMMs (FB-DIMMs) memory
347 controllers. The following example will assume 2 channels:
349 +------------+-----------------------+
350 | CS Rows | Channels |
351 +------------+-----------+-----------+
352 | | ``ch0`` | ``ch1`` |
353 +============+===========+===========+
354 | |**DIMM_A0**|**DIMM_B0**|
355 +------------+-----------+-----------+
356 | ``csrow0`` | rank0 | rank0 |
357 +------------+-----------+-----------+
358 | ``csrow1`` | rank1 | rank1 |
359 +------------+-----------+-----------+
360 | |**DIMM_A1**|**DIMM_B1**|
361 +------------+-----------+-----------+
362 | ``csrow2`` | rank0 | rank0 |
363 +------------+-----------+-----------+
364 | ``csrow3`` | rank1 | rank1 |
365 +------------+-----------+-----------+
367 In the above example, there are 4 physical slots on the motherboard
370 +---------+---------+
371 | DIMM_A0 | DIMM_B0 |
372 +---------+---------+
373 | DIMM_A1 | DIMM_B1 |
374 +---------+---------+
376 Labels for these slots are usually silk-screened on the motherboard.
377 Slots labeled ``A`` are channel 0 in this example. Slots labeled ``B`` are
378 channel 1. Notice that there are two csrows possible on a physical DIMM.
379 These csrows are allocated their csrow assignment based on the slot into
380 which the memory DIMM is placed. Thus, when 1 DIMM is placed in each
381 Channel, the csrows cross both DIMMs.
383 Memory DIMMs come single or dual "ranked". A rank is a populated csrow.
384 In the example above 2 dual ranked DIMMs are similarly placed. Thus,
385 both csrow0 and csrow1 are populated. On the other hand, when 2 single
386 ranked DIMMs are placed in slots DIMM_A0 and DIMM_B0, then they will
387 have just one csrow (csrow0) and csrow1 will be empty. The pattern
388 repeats itself for csrow2 and csrow3. Also note that some memory
389 controllers don't have any logic to identify the memory module, see
390 ``rankX`` directories below.
392 The representation of the above is reflected in the directory
393 tree in EDAC's sysfs interface. Starting in directory
394 ``/sys/devices/system/edac/mc``, each memory controller will be
395 represented by its own ``mcX`` directory, where ``X`` is the
405 Under each ``mcX`` directory each ``csrowX`` is again represented by a
406 ``csrowX``, where ``X`` is the csrow index::
415 Notice that there is no csrow1, which indicates that csrow0 is composed
416 of a single ranked DIMMs. This should also apply in both Channels, in
417 order to have dual-channel mode be operational. Since both csrow2 and
418 csrow3 are populated, this indicates a dual ranked set of DIMMs for
421 Within each of the ``mcX`` and ``csrowX`` directories are several EDAC
422 control and attribute files.
427 In ``mcX`` directories are EDAC control and attribute files for
428 this ``X`` instance of the memory controllers.
430 For a description of the sysfs API, please see:
432 Documentation/ABI/testing/sysfs-devices-edac
435 ``dimmX`` or ``rankX`` directories
436 ----------------------------------
438 The recommended way to use the EDAC subsystem is to look at the information
439 provided by the ``dimmX`` or ``rankX`` directories [#f5]_.
441 A typical EDAC system has the following structure under
442 ``/sys/devices/system/edac/``\ [#f6]_::
444 /sys/devices/system/edac/
448 │ │ ├── ce_noinfo_count
450 │ │ │ ├── dimm_ce_count
451 │ │ │ ├── dimm_dev_type
452 │ │ │ ├── dimm_edac_mode
454 │ │ │ ├── dimm_location
455 │ │ │ ├── dimm_mem_type
456 │ │ │ ├── dimm_ue_count
461 │ │ ├── reset_counters
462 │ │ ├── seconds_since_reset
465 │ │ ├── ue_noinfo_count
469 │ │ ├── ce_noinfo_count
471 │ │ │ ├── dimm_ce_count
472 │ │ │ ├── dimm_dev_type
473 │ │ │ ├── dimm_edac_mode
475 │ │ │ ├── dimm_location
476 │ │ │ ├── dimm_mem_type
477 │ │ │ ├── dimm_ue_count
482 │ │ ├── reset_counters
483 │ │ ├── seconds_since_reset
486 │ │ ├── ue_noinfo_count
491 In the ``dimmX`` directories are EDAC control and attribute files for
492 this ``X`` memory module:
494 - ``size`` - Total memory managed by this csrow attribute file
496 This attribute file displays, in count of megabytes, the memory
497 that this csrow contains.
499 - ``dimm_ue_count`` - Uncorrectable Errors count attribute file
501 This attribute file displays the total count of uncorrectable
502 errors that have occurred on this DIMM. If panic_on_ue is set
503 this counter will not have a chance to increment, since EDAC
504 will panic the system.
506 - ``dimm_ce_count`` - Correctable Errors count attribute file
508 This attribute file displays the total count of correctable
509 errors that have occurred on this DIMM. This count is very
510 important to examine. CEs provide early indications that a
511 DIMM is beginning to fail. This count field should be
512 monitored for non-zero values and report such information
513 to the system administrator.
515 - ``dimm_dev_type`` - Device type attribute file
517 This attribute file will display what type of DRAM device is
518 being utilized on this DIMM.
526 - ``dimm_edac_mode`` - EDAC Mode of operation attribute file
528 This attribute file will display what type of Error detection
529 and correction is being utilized.
531 - ``dimm_label`` - memory module label control file
533 This control file allows this DIMM to have a label assigned
534 to it. With this label in the module, when errors occur
535 the output can provide the DIMM label in the system log.
536 This becomes vital for panic events to isolate the
537 cause of the UE event.
539 DIMM Labels must be assigned after booting, with information
540 that correctly identifies the physical slot with its
541 silk screen label. This information is currently very
542 motherboard specific and determination of this information
543 must occur in userland at this time.
545 - ``dimm_location`` - location of the memory module
547 The location can have up to 3 levels, and describe how the
548 memory controller identifies the location of a memory module.
549 Depending on the type of memory and memory controller, it
552 - *csrow* and *channel* - used when the memory controller
553 doesn't identify a single DIMM - e. g. in ``rankX`` dir;
554 - *branch*, *channel*, *slot* - typically used on FB-DIMM memory
556 - *channel*, *slot* - used on Nehalem and newer Intel drivers.
558 - ``dimm_mem_type`` - Memory Type attribute file
560 This attribute file will display what type of memory is currently
561 on this csrow. Normally, either buffered or unbuffered memory.
567 .. [#f5] On some systems, the memory controller doesn't have any logic
568 to identify the memory module. On such systems, the directory is called ``rankX`` and works on a similar way as the ``csrowX`` directories.
569 On modern Intel memory controllers, the memory controller identifies the
570 memory modules directly. On such systems, the directory is called ``dimmX``.
572 .. [#f6] There are also some ``power`` directories and ``subsystem``
573 symlinks inside the sysfs mapping that are automatically created by
574 the sysfs subsystem. Currently, they serve no purpose.
576 ``csrowX`` directories
577 ----------------------
579 When CONFIG_EDAC_LEGACY_SYSFS is enabled, sysfs will contain the ``csrowX``
580 directories. As this API doesn't work properly for Rambus, FB-DIMMs and
581 modern Intel Memory Controllers, this is being deprecated in favor of
582 ``dimmX`` directories.
584 In the ``csrowX`` directories are EDAC control and attribute files for
585 this ``X`` instance of csrow:
588 - ``ue_count`` - Total Uncorrectable Errors count attribute file
590 This attribute file displays the total count of uncorrectable
591 errors that have occurred on this csrow. If panic_on_ue is set
592 this counter will not have a chance to increment, since EDAC
593 will panic the system.
596 - ``ce_count`` - Total Correctable Errors count attribute file
598 This attribute file displays the total count of correctable
599 errors that have occurred on this csrow. This count is very
600 important to examine. CEs provide early indications that a
601 DIMM is beginning to fail. This count field should be
602 monitored for non-zero values and report such information
603 to the system administrator.
606 - ``size_mb`` - Total memory managed by this csrow attribute file
608 This attribute file displays, in count of megabytes, the memory
609 that this csrow contains.
612 - ``mem_type`` - Memory Type attribute file
614 This attribute file will display what type of memory is currently
615 on this csrow. Normally, either buffered or unbuffered memory.
622 - ``edac_mode`` - EDAC Mode of operation attribute file
624 This attribute file will display what type of Error detection
625 and correction is being utilized.
628 - ``dev_type`` - Device type attribute file
630 This attribute file will display what type of DRAM device is
631 being utilized on this DIMM.
640 - ``ch0_ce_count`` - Channel 0 CE Count attribute file
642 This attribute file will display the count of CEs on this
643 DIMM located in channel 0.
646 - ``ch0_ue_count`` - Channel 0 UE Count attribute file
648 This attribute file will display the count of UEs on this
649 DIMM located in channel 0.
652 - ``ch0_dimm_label`` - Channel 0 DIMM Label control file
655 This control file allows this DIMM to have a label assigned
656 to it. With this label in the module, when errors occur
657 the output can provide the DIMM label in the system log.
658 This becomes vital for panic events to isolate the
659 cause of the UE event.
661 DIMM Labels must be assigned after booting, with information
662 that correctly identifies the physical slot with its
663 silk screen label. This information is currently very
664 motherboard specific and determination of this information
665 must occur in userland at this time.
668 - ``ch1_ce_count`` - Channel 1 CE Count attribute file
671 This attribute file will display the count of CEs on this
672 DIMM located in channel 1.
675 - ``ch1_ue_count`` - Channel 1 UE Count attribute file
678 This attribute file will display the count of UEs on this
679 DIMM located in channel 0.
682 - ``ch1_dimm_label`` - Channel 1 DIMM Label control file
684 This control file allows this DIMM to have a label assigned
685 to it. With this label in the module, when errors occur
686 the output can provide the DIMM label in the system log.
687 This becomes vital for panic events to isolate the
688 cause of the UE event.
690 DIMM Labels must be assigned after booting, with information
691 that correctly identifies the physical slot with its
692 silk screen label. This information is currently very
693 motherboard specific and determination of this information
694 must occur in userland at this time.
700 If logging for UEs and CEs is enabled, then system logs will contain
701 information indicating that errors have been detected::
703 EDAC MC0: CE page 0x283, offset 0xce0, grain 8, syndrome 0x6ec3, row 0, channel 1 "DIMM_B1": amd76x_edac
704 EDAC MC0: CE page 0x1e5, offset 0xfb0, grain 8, syndrome 0xb741, row 0, channel 1 "DIMM_B1": amd76x_edac
707 The structure of the message is:
709 +---------------------------------------+-------------+
710 | Content | Example |
711 +=======================================+=============+
712 | The memory controller | MC0 |
713 +---------------------------------------+-------------+
715 +---------------------------------------+-------------+
716 | Memory page | 0x283 |
717 +---------------------------------------+-------------+
718 | Offset in the page | 0xce0 |
719 +---------------------------------------+-------------+
720 | The byte granularity | grain 8 |
721 | or resolution of the error | |
722 +---------------------------------------+-------------+
723 | The error syndrome | 0xb741 |
724 +---------------------------------------+-------------+
725 | Memory row | row 0 |
726 +---------------------------------------+-------------+
727 | Memory channel | channel 1 |
728 +---------------------------------------+-------------+
729 | DIMM label, if set prior | DIMM B1 |
730 +---------------------------------------+-------------+
731 | And then an optional, driver-specific | |
732 | message that may have additional | |
734 +---------------------------------------+-------------+
736 Both UEs and CEs with no info will lack all but memory controller, error
737 type, a notice of "no info" and then an optional, driver-specific error
741 PCI Bus Parity Detection
742 ------------------------
744 On Header Type 00 devices, the primary status is looked at for any
745 parity error regardless of whether parity is enabled on the device or
746 not. (The spec indicates parity is generated in some cases). On Header
747 Type 01 bridges, the secondary status register is also looked at to see
748 if parity occurred on the bus on the other side of the bridge.
754 Under ``/sys/devices/system/edac/pci`` are control and attribute files as
758 - ``check_pci_parity`` - Enable/Disable PCI Parity checking control file
760 This control file enables or disables the PCI Bus Parity scanning
761 operation. Writing a 1 to this file enables the scanning. Writing
762 a 0 to this file disables the scanning.
766 echo "1" >/sys/devices/system/edac/pci/check_pci_parity
770 echo "0" >/sys/devices/system/edac/pci/check_pci_parity
773 - ``pci_parity_count`` - Parity Count
775 This attribute file will display the number of parity errors that
782 - ``edac_mc_panic_on_ue`` - Panic on UE control file
784 An uncorrectable error will cause a machine panic. This is usually
785 desirable. It is a bad idea to continue when an uncorrectable error
786 occurs - it is indeterminate what was uncorrected and the operating
787 system context might be so mangled that continuing will lead to further
788 corruption. If the kernel has MCE configured, then EDAC will never
793 module/kernel parameter: edac_mc_panic_on_ue=[0|1]
797 echo "1" > /sys/module/edac_core/parameters/edac_mc_panic_on_ue
800 - ``edac_mc_log_ue`` - Log UE control file
803 Generate kernel messages describing uncorrectable errors. These errors
804 are reported through the system message log system. UE statistics
805 will be accumulated even when UE logging is disabled.
809 module/kernel parameter: edac_mc_log_ue=[0|1]
813 echo "1" > /sys/module/edac_core/parameters/edac_mc_log_ue
816 - ``edac_mc_log_ce`` - Log CE control file
819 Generate kernel messages describing correctable errors. These
820 errors are reported through the system message log system.
821 CE statistics will be accumulated even when CE logging is disabled.
825 module/kernel parameter: edac_mc_log_ce=[0|1]
829 echo "1" > /sys/module/edac_core/parameters/edac_mc_log_ce
832 - ``edac_mc_poll_msec`` - Polling period control file
835 The time period, in milliseconds, for polling for error information.
836 Too small a value wastes resources. Too large a value might delay
837 necessary handling of errors and might loose valuable information for
838 locating the error. 1000 milliseconds (once each second) is the current
839 default. Systems which require all the bandwidth they can get, may
844 module/kernel parameter: edac_mc_poll_msec=[0|1]
848 echo "1000" > /sys/module/edac_core/parameters/edac_mc_poll_msec
851 - ``panic_on_pci_parity`` - Panic on PCI PARITY Error
854 This control file enables or disables panicking when a parity
855 error has been detected.
858 module/kernel parameter::
860 edac_panic_on_pci_pe=[0|1]
864 echo "1" > /sys/module/edac_core/parameters/edac_panic_on_pci_pe
868 echo "0" > /sys/module/edac_core/parameters/edac_panic_on_pci_pe
875 In the header file, edac_pci.h, there is a series of edac_device structures
876 and APIs for the EDAC_DEVICE.
878 User space access to an edac_device is through the sysfs interface.
880 At the location ``/sys/devices/system/edac`` (sysfs) new edac_device devices
883 There is a three level tree beneath the above ``edac`` directory. For example,
884 the ``test_device_edac`` device (found at the http://bluesmoke.sourceforget.net
885 website) installs itself as::
887 /sys/devices/system/edac/test-instance
889 in this directory are various controls, a symlink and one or more ``instance``
892 The standard default controls are:
894 ============== =======================================================
895 log_ce boolean to log CE events
896 log_ue boolean to log UE events
897 panic_on_ue boolean to ``panic`` the system if an UE is encountered
898 (default off, can be set true via startup script)
899 poll_msec time period between POLL cycles for events
900 ============== =======================================================
902 The test_device_edac device adds at least one of its own custom control:
904 ============== ==================================================
905 test_bits which in the current test driver does nothing but
906 show how it is installed. A ported driver can
907 add one or more such controls and/or attributes
909 One out-of-tree driver uses controls here to allow
910 for ERROR INJECTION operations to hardware
912 ============== ==================================================
914 The symlink points to the 'struct dev' that is registered for this edac_device.
919 One or more instance directories are present. For the ``test_device_edac``
927 In this directory there are two default counter attributes, which are totals of
928 counter in deeper subdirectories.
930 ============== ====================================
931 ce_count total of CE events of subdirectories
932 ue_count total of UE events of subdirectories
933 ============== ====================================
938 At the lowest directory level is the ``block`` directory. There can be 0, 1
939 or more blocks specified in each instance:
945 In this directory the default attributes are:
947 ============== ================================================
948 ce_count which is counter of CE events for this ``block``
949 of hardware being monitored
950 ue_count which is counter of UE events for this ``block``
951 of hardware being monitored
952 ============== ================================================
955 The ``test_device_edac`` device adds 4 attributes and 1 control:
957 ================== ====================================================
958 test-block-bits-0 for every POLL cycle this counter
960 test-block-bits-1 every 10 cycles, this counter is bumped once,
961 and test-block-bits-0 is set to 0
962 test-block-bits-2 every 100 cycles, this counter is bumped once,
963 and test-block-bits-1 is set to 0
964 test-block-bits-3 every 1000 cycles, this counter is bumped once,
965 and test-block-bits-2 is set to 0
966 ================== ====================================================
969 ================== ====================================================
970 reset-counters writing ANY thing to this control will
971 reset all the above counters.
972 ================== ====================================================
975 Use of the ``test_device_edac`` driver should enable any others to create their own
976 unique drivers for their hardware systems.
978 The ``test_device_edac`` sample driver is located at the
979 http://bluesmoke.sourceforge.net project site for EDAC.
982 Usage of EDAC APIs on Nehalem and newer Intel CPUs
983 --------------------------------------------------
985 On older Intel architectures, the memory controller was part of the North
986 Bridge chipset. Nehalem, Sandy Bridge, Ivy Bridge, Haswell, Sky Lake and
987 newer Intel architectures integrated an enhanced version of the memory
988 controller (MC) inside the CPUs.
990 This chapter will cover the differences of the enhanced memory controllers
991 found on newer Intel CPUs, such as ``i7core_edac``, ``sb_edac`` and
992 ``sbx_edac`` drivers.
996 The Xeon E7 processor families use a separate chip for the memory
997 controller, called Intel Scalable Memory Buffer. This section doesn't
998 apply for such families.
1000 1) There is one Memory Controller per Quick Patch Interconnect
1001 (QPI). At the driver, the term "socket" means one QPI. This is
1002 associated with a physical CPU socket.
1004 Each MC have 3 physical read channels, 3 physical write channels and
1005 3 logic channels. The driver currently sees it as just 3 channels.
1006 Each channel can have up to 3 DIMMs.
1008 The minimum known unity is DIMMs. There are no information about csrows.
1009 As EDAC API maps the minimum unity is csrows, the driver sequentially
1010 maps channel/DIMM into different csrows.
1012 For example, supposing the following layout::
1014 Ch0 phy rd0, wr0 (0x063f4031): 2 ranks, UDIMMs
1015 dimm 0 1024 Mb offset: 0, bank: 8, rank: 1, row: 0x4000, col: 0x400
1016 dimm 1 1024 Mb offset: 4, bank: 8, rank: 1, row: 0x4000, col: 0x400
1017 Ch1 phy rd1, wr1 (0x063f4031): 2 ranks, UDIMMs
1018 dimm 0 1024 Mb offset: 0, bank: 8, rank: 1, row: 0x4000, col: 0x400
1019 Ch2 phy rd3, wr3 (0x063f4031): 2 ranks, UDIMMs
1020 dimm 0 1024 Mb offset: 0, bank: 8, rank: 1, row: 0x4000, col: 0x400
1022 The driver will map it as::
1024 csrow0: channel 0, dimm0
1025 csrow1: channel 0, dimm1
1026 csrow2: channel 1, dimm0
1027 csrow3: channel 2, dimm0
1029 exports one DIMM per csrow.
1031 Each QPI is exported as a different memory controller.
1033 2) The MC has the ability to inject errors to test drivers. The drivers
1034 implement this functionality via some error injection nodes:
1036 For injecting a memory error, there are some sysfs nodes, under
1037 ``/sys/devices/system/edac/mc/mc?/``:
1039 - ``inject_addrmatch/*``:
1040 Controls the error injection mask register. It is possible to specify
1041 several characteristics of the address to match an error code::
1043 dimm = the affected dimm. Numbers are relative to a channel;
1044 rank = the memory rank;
1045 channel = the channel that will generate an error;
1046 bank = the affected bank;
1047 page = the page address;
1048 column (or col) = the address column.
1050 each of the above values can be set to "any" to match any valid value.
1052 At driver init, all values are set to any.
1054 For example, to generate an error at rank 1 of dimm 2, for any channel,
1055 any bank, any page, any column::
1057 echo 2 >/sys/devices/system/edac/mc/mc0/inject_addrmatch/dimm
1058 echo 1 >/sys/devices/system/edac/mc/mc0/inject_addrmatch/rank
1060 To return to the default behaviour of matching any, you can do::
1062 echo any >/sys/devices/system/edac/mc/mc0/inject_addrmatch/dimm
1063 echo any >/sys/devices/system/edac/mc/mc0/inject_addrmatch/rank
1065 - ``inject_eccmask``:
1066 specifies what bits will have troubles,
1068 - ``inject_section``:
1069 specifies what ECC cache section will get the error::
1076 specifies the type of error, being a combination of the following bits::
1082 - ``inject_enable``:
1083 starts the error generation when something different than 0 is written.
1085 All inject vars can be read. root permission is needed for write.
1087 Datasheet states that the error will only be generated after a write on an
1088 address that matches inject_addrmatch. It seems, however, that reading will
1089 also produce an error.
1091 For example, the following code will generate an error for any write access
1092 at socket 0, on any DIMM/address on channel 2::
1094 echo 2 >/sys/devices/system/edac/mc/mc0/inject_addrmatch/channel
1095 echo 2 >/sys/devices/system/edac/mc/mc0/inject_type
1096 echo 64 >/sys/devices/system/edac/mc/mc0/inject_eccmask
1097 echo 3 >/sys/devices/system/edac/mc/mc0/inject_section
1098 echo 1 >/sys/devices/system/edac/mc/mc0/inject_enable
1099 dd if=/dev/mem of=/dev/null seek=16k bs=4k count=1 >& /dev/null
1101 For socket 1, it is needed to replace "mc0" by "mc1" at the above
1104 The generated error message will look like::
1106 EDAC MC0: UE row 0, channel-a= 0 channel-b= 0 labels "-": NON_FATAL (addr = 0x0075b980, socket=0, Dimm=0, Channel=2, syndrome=0x00000040, count=1, Err=8c0000400001009f:4000080482 (read error: read ECC error))
1108 3) Corrected Error memory register counters
1110 Those newer MCs have some registers to count memory errors. The driver
1111 uses those registers to report Corrected Errors on devices with Registered
1114 However, those counters don't work with Unregistered DIMM. As the chipset
1115 offers some counters that also work with UDIMMs (but with a worse level of
1116 granularity than the default ones), the driver exposes those registers for
1119 They can be read by looking at the contents of ``all_channel_counts/``::
1121 $ for i in /sys/devices/system/edac/mc/mc0/all_channel_counts/*; do echo $i; cat $i; done
1122 /sys/devices/system/edac/mc/mc0/all_channel_counts/udimm0
1124 /sys/devices/system/edac/mc/mc0/all_channel_counts/udimm1
1126 /sys/devices/system/edac/mc/mc0/all_channel_counts/udimm2
1129 What happens here is that errors on different csrows, but at the same
1130 dimm number will increment the same counter.
1131 So, in this memory mapping::
1133 csrow0: channel 0, dimm0
1134 csrow1: channel 0, dimm1
1135 csrow2: channel 1, dimm0
1136 csrow3: channel 2, dimm0
1138 The hardware will increment udimm0 for an error at the first dimm at either
1139 csrow0, csrow2 or csrow3;
1141 The hardware will increment udimm1 for an error at the second dimm at either
1142 csrow0, csrow2 or csrow3;
1144 The hardware will increment udimm2 for an error at the third dimm at either
1145 csrow0, csrow2 or csrow3;
1147 4) Standard error counters
1149 The standard error counters are generated when an mcelog error is received
1150 by the driver. Since, with UDIMM, this is counted by software, it is
1151 possible that some errors could be lost. With RDIMM's, they display the
1152 contents of the registers
1154 Reference documents used on ``amd64_edac``
1155 ------------------------------------------
1157 ``amd64_edac`` module is based on the following documents
1158 (available from http://support.amd.com/en-us/search/tech-docs):
1160 1. :Title: BIOS and Kernel Developer's Guide for AMD Athlon 64 and AMD
1162 :AMD publication #: 26094
1164 :Link: http://support.amd.com/TechDocs/26094.PDF
1166 2. :Title: BIOS and Kernel Developer's Guide for AMD NPT Family 0Fh
1168 :AMD publication #: 32559
1170 :Issue Date: May 2006
1171 :Link: http://support.amd.com/TechDocs/32559.pdf
1173 3. :Title: BIOS and Kernel Developer's Guide (BKDG) For AMD Family 10h
1175 :AMD publication #: 31116
1177 :Issue Date: September 07, 2007
1178 :Link: http://support.amd.com/TechDocs/31116.pdf
1180 4. :Title: BIOS and Kernel Developer's Guide (BKDG) for AMD Family 15h
1181 Models 30h-3Fh Processors
1182 :AMD publication #: 49125
1184 :Issue Date: 2/12/2015 (latest release)
1185 :Link: http://support.amd.com/TechDocs/49125_15h_Models_30h-3Fh_BKDG.pdf
1187 5. :Title: BIOS and Kernel Developer's Guide (BKDG) for AMD Family 15h
1188 Models 60h-6Fh Processors
1189 :AMD publication #: 50742
1191 :Issue Date: 7/23/2015 (latest release)
1192 :Link: http://support.amd.com/TechDocs/50742_15h_Models_60h-6Fh_BKDG.pdf
1194 6. :Title: BIOS and Kernel Developer's Guide (BKDG) for AMD Family 16h
1195 Models 00h-0Fh Processors
1196 :AMD publication #: 48751
1198 :Issue Date: 2/23/2015 (latest release)
1199 :Link: http://support.amd.com/TechDocs/48751_16h_bkdg.pdf
1204 * Written by Doug Thompson <dougthompson@xmission.com>
1207 - 17 Jul 2007 Updated
1209 * |copy| Mauro Carvalho Chehab
1211 - 05 Aug 2009 Nehalem interface
1212 - 26 Oct 2016 Converted to ReST and cleanups at the Nehalem section
1214 * EDAC authors/maintainers:
1216 - Doug Thompson, Dave Jiang, Dave Peterson et al,
1217 - Mauro Carvalho Chehab
1219 - original author: Thayne Harbaugh