Documentation/powerpc/eeh-pci-error-recovery.txt

   1
   2
   3                       PCI Bus EEH Error Recovery
   4                       --------------------------
   5                            Linas Vepstas
   6                        <linas@austin.ibm.com>
   7                           12 January 2005
   8
   9
  10 Overview:
  11 ---------
  12 The IBM POWER-based pSeries and iSeries computers include PCI bus
  13 controller chips that have extended capabilities for detecting and
  14 reporting a large variety of PCI bus error conditions.  These features
  15 go under the name of "EEH", for "Enhanced Error Handling".  The EEH
  16 hardware features allow PCI bus errors to be cleared and a PCI
  17 card to be "rebooted", without also having to reboot the operating
  18 system.
  19
  20 This is in contrast to traditional PCI error handling, where the
  21 PCI chip is wired directly to the CPU, and an error would cause
  22 a CPU machine-check/check-stop condition, halting the CPU entirely.
  23 Another "traditional" technique is to ignore such errors, which
  24 can lead to data corruption, both of user data or of kernel data,
  25 hung/unresponsive adapters, or system crashes/lockups.  Thus,
  26 the idea behind EEH is that the operating system can become more
  27 reliable and robust by protecting it from PCI errors, and giving
  28 the OS the ability to "reboot"/recover individual PCI devices.
  29
  30 Future systems from other vendors, based on the PCI-E specification,
  31 may contain similar features.
  32
  33
  34 Causes of EEH Errors
  35 --------------------
  36 EEH was originally designed to guard against hardware failure, such
  37 as PCI cards dying from heat, humidity, dust, vibration and bad
  38 electrical connections. The vast majority of EEH errors seen in
  39 "real life" are due to either poorly seated PCI cards, or,
  40 unfortunately quite commonly, due to device driver bugs, device firmware
  41 bugs, and sometimes PCI card hardware bugs.
  42
  43 The most common software bug, is one that causes the device to
  44 attempt to DMA to a location in system memory that has not been
  45 reserved for DMA access for that card.  This is a powerful feature,
  46 as it prevents what; otherwise, would have been silent memory
  47 corruption caused by the bad DMA.  A number of device driver
  48 bugs have been found and fixed in this way over the past few
  49 years.  Other possible causes of EEH errors include data or
  50 address line parity errors (for example, due to poor electrical
  51 connectivity due to a poorly seated card), and PCI-X split-completion
  52 errors (due to software, device firmware, or device PCI hardware bugs).
  53 The vast majority of "true hardware failures" can be cured by
  54 physically removing and re-seating the PCI card.
  55
  56
  57 Detection and Recovery
  58 ----------------------
  59 In the following discussion, a generic overview of how to detect
  60 and recover from EEH errors will be presented. This is followed
  61 by an overview of how the current implementation in the Linux
  62 kernel does it.  The actual implementation is subject to change,
  63 and some of the finer points are still being debated.  These
  64 may in turn be swayed if or when other architectures implement
  65 similar functionality.
  66
  67 When a PCI Host Bridge (PHB, the bus controller connecting the
  68 PCI bus to the system CPU electronics complex) detects a PCI error
  69 condition, it will "isolate" the affected PCI card.  Isolation
  70 will block all writes (either to the card from the system, or
  71 from the card to the system), and it will cause all reads to
  72 return all-ff's (0xff, 0xffff, 0xffffffff for 8/16/32-bit reads).
  73 This value was chosen because it is the same value you would
  74 get if the device was physically unplugged from the slot.
  75 This includes access to PCI memory, I/O space, and PCI config
  76 space.  Interrupts; however, will continued to be delivered.
  77
  78 Detection and recovery are performed with the aid of ppc64
  79 firmware.  The programming interfaces in the Linux kernel
  80 into the firmware are referred to as RTAS (Run-Time Abstraction
  81 Services).  The Linux kernel does not (should not) access
  82 the EEH function in the PCI chipsets directly, primarily because
  83 there are a number of different chipsets out there, each with
  84 different interfaces and quirks. The firmware provides a
  85 uniform abstraction layer that will work with all pSeries
  86 and iSeries hardware (and be forwards-compatible).
  87
  88 If the OS or device driver suspects that a PCI slot has been
  89 EEH-isolated, there is a firmware call it can make to determine if
  90 this is the case. If so, then the device driver should put itself
  91 into a consistent state (given that it won't be able to complete any
  92 pending work) and start recovery of the card.  Recovery normally
  93 would consist of resetting the PCI device (holding the PCI #RST
  94 line high for two seconds), followed by setting up the device
  95 config space (the base address registers (BAR's), latency timer,
  96 cache line size, interrupt line, and so on).  This is followed by a
  97 reinitialization of the device driver.  In a worst-case scenario,
  98 the power to the card can be toggled, at least on hot-plug-capable
  99 slots.  In principle, layers far above the device driver probably
 100 do not need to know that the PCI card has been "rebooted" in this
 101 way; ideally, there should be at most a pause in Ethernet/disk/USB
 102 I/O while the card is being reset.
 103
 104 If the card cannot be recovered after three or four resets, the
 105 kernel/device driver should assume the worst-case scenario, that the
 106 card has died completely, and report this error to the sysadmin.
 107 In addition, error messages are reported through RTAS and also through
 108 syslogd (/var/log/messages) to alert the sysadmin of PCI resets.
 109 The correct way to deal with failed adapters is to use the standard
 110 PCI hotplug tools to remove and replace the dead card.
 111
 112
 113 Current PPC64 Linux EEH Implementation
 114 --------------------------------------
 115 At this time, a generic EEH recovery mechanism has been implemented,
 116 so that individual device drivers do not need to be modified to support
 117 EEH recovery.  This generic mechanism piggy-backs on the PCI hotplug
 118 infrastructure,  and percolates events up through the userspace/udev
 119 infrastructure.  Following is a detailed description of how this is
 120 accomplished.
 121
 122 EEH must be enabled in the PHB's very early during the boot process,
 123 and if a PCI slot is hot-plugged. The former is performed by
 124 eeh_init() in arch/powerpc/platforms/pseries/eeh.c, and the later by
 125 drivers/pci/hotplug/pSeries_pci.c calling in to the eeh.c code.
 126 EEH must be enabled before a PCI scan of the device can proceed.
 127 Current Power5 hardware will not work unless EEH is enabled;
 128 although older Power4 can run with it disabled.  Effectively,
 129 EEH can no longer be turned off.  PCI devices *must* be
 130 registered with the EEH code; the EEH code needs to know about
 131 the I/O address ranges of the PCI device in order to detect an
 132 error.  Given an arbitrary address, the routine
 133 pci_get_device_by_addr() will find the pci device associated
 134 with that address (if any).
 135
 136 The default arch/powerpc/include/asm/io.h macros readb(), inb(), insb(),
 137 etc. include a check to see if the i/o read returned all-0xff's.
 138 If so, these make a call to eeh_dn_check_failure(), which in turn
 139 asks the firmware if the all-ff's value is the sign of a true EEH
 140 error.  If it is not, processing continues as normal.  The grand
 141 total number of these false alarms or "false positives" can be
 142 seen in /proc/ppc64/eeh (subject to change).  Normally, almost
 143 all of these occur during boot, when the PCI bus is scanned, where
 144 a large number of 0xff reads are part of the bus scan procedure.
 145
 146 If a frozen slot is detected, code in
 147 arch/powerpc/platforms/pseries/eeh.c will print a stack trace to
 148 syslog (/var/log/messages).  This stack trace has proven to be very
 149 useful to device-driver authors for finding out at what point the EEH
 150 error was detected, as the error itself usually occurs slightly
 151 beforehand.
 152
 153 Next, it uses the Linux kernel notifier chain/work queue mechanism to
 154 allow any interested parties to find out about the failure.  Device
 155 drivers, or other parts of the kernel, can use
 156 eeh_register_notifier(struct notifier_block *) to find out about EEH
 157 events.  The event will include a pointer to the pci device, the
 158 device node and some state info.  Receivers of the event can "do as
 159 they wish"; the default handler will be described further in this
 160 section.
 161
 162 To assist in the recovery of the device, eeh.c exports the
 163 following functions:
 164
 165 rtas_set_slot_reset() -- assert the  PCI #RST line for 1/8th of a second
 166 rtas_configure_bridge() -- ask firmware to configure any PCI bridges
 167    located topologically under the pci slot.
 168 eeh_save_bars() and eeh_restore_bars(): save and restore the PCI
 169    config-space info for a device and any devices under it.
 170
 171
 172 A handler for the EEH notifier_block events is implemented in
 173 drivers/pci/hotplug/pSeries_pci.c, called handle_eeh_events().
 174 It saves the device BAR's and then calls rpaphp_unconfig_pci_adapter().
 175 This last call causes the device driver for the card to be stopped,
 176 which causes uevents to go out to user space. This triggers
 177 user-space scripts that might issue commands such as "ifdown eth0"
 178 for ethernet cards, and so on.  This handler then sleeps for 5 seconds,
 179 hoping to give the user-space scripts enough time to complete.
 180 It then resets the PCI card, reconfigures the device BAR's, and
 181 any bridges underneath. It then calls rpaphp_enable_pci_slot(),
 182 which restarts the device driver and triggers more user-space
 183 events (for example, calling "ifup eth0" for ethernet cards).
 184
 185
 186 Device Shutdown and User-Space Events
 187 -------------------------------------
 188 This section documents what happens when a pci slot is unconfigured,
 189 focusing on how the device driver gets shut down, and on how the
 190 events get delivered to user-space scripts.
 191
 192 Following is an example sequence of events that cause a device driver
 193 close function to be called during the first phase of an EEH reset.
 194 The following sequence is an example of the pcnet32 device driver.
 195
 196     rpa_php_unconfig_pci_adapter (struct slot *)  // in rpaphp_pci.c
 197     {
 198       calls
 199       pci_remove_bus_device (struct pci_dev *) // in /drivers/pci/remove.c
 200       {
 201         calls
 202         pci_destroy_dev (struct pci_dev *)
 203         {
 204           calls
 205           device_unregister (&dev->dev) // in /drivers/base/core.c
 206           {
 207             calls
 208             device_del (struct device *)
 209             {
 210               calls
 211               bus_remove_device() // in /drivers/base/bus.c
 212               {
 213                 calls
 214                 device_release_driver()
 215                 {
 216                   calls
 217                   struct device_driver->remove() which is just
 218                   pci_device_remove()  // in /drivers/pci/pci_driver.c
 219                   {
 220                     calls
 221                     struct pci_driver->remove() which is just
 222                     pcnet32_remove_one() // in /drivers/net/pcnet32.c
 223                     {
 224                       calls
 225                       unregister_netdev() // in /net/core/dev.c
 226                       {
 227                         calls
 228                         dev_close()  // in /net/core/dev.c
 229                         {
 230                            calls dev->stop();
 231                            which is just pcnet32_close() // in pcnet32.c
 232                            {
 233                              which does what you wanted
 234                              to stop the device
 235                            }
 236                         }
 237                      }
 238                    which
 239                    frees pcnet32 device driver memory
 240                 }
 241      }}}}}}
 242
 243
 244     in drivers/pci/pci_driver.c,
 245     struct device_driver->remove() is just pci_device_remove()
 246     which calls struct pci_driver->remove() which is pcnet32_remove_one()
 247     which calls unregister_netdev()  (in net/core/dev.c)
 248     which calls dev_close()  (in net/core/dev.c)
 249     which calls dev->stop() which is pcnet32_close()
 250     which then does the appropriate shutdown.
 251
 252 ---
 253 Following is the analogous stack trace for events sent to user-space
 254 when the pci device is unconfigured.
 255
 256 rpa_php_unconfig_pci_adapter() {             // in rpaphp_pci.c
 257   calls
 258   pci_remove_bus_device (struct pci_dev *) { // in /drivers/pci/remove.c
 259     calls
 260     pci_destroy_dev (struct pci_dev *) {
 261       calls
 262       device_unregister (&dev->dev) {        // in /drivers/base/core.c
 263         calls
 264         device_del(struct device * dev) {    // in /drivers/base/core.c
 265           calls
 266           kobject_del() {                    //in /libs/kobject.c
 267             calls
 268             kobject_uevent() {               // in /libs/kobject.c
 269               calls
 270               kset_uevent() {                // in /lib/kobject.c
 271                 calls
 272                 kset->uevent_ops->uevent()   // which is really just
 273                 a call to
 274                 dev_uevent() {               // in /drivers/base/core.c
 275                   calls
 276                   dev->bus->uevent() which is really just a call to
 277                   pci_uevent () {            // in drivers/pci/hotplug.c
 278                     which prints device name, etc....
 279                  }
 280                }
 281                then kobject_uevent() sends a netlink uevent to userspace
 282                --> userspace uevent
 283                (during early boot, nobody listens to netlink events and
 284                kobject_uevent() executes uevent_helper[], which runs the
 285                event process /sbin/hotplug)
 286            }
 287          }
 288          kobject_del() then calls sysfs_remove_dir(), which would
 289          trigger any user-space daemon that was watching /sysfs,
 290          and notice the delete event.
 291
 292
 293 Pro's and Con's of the Current Design
 294 -------------------------------------
 295 There are several issues with the current EEH software recovery design,
 296 which may be addressed in future revisions.  But first, note that the
 297 big plus of the current design is that no changes need to be made to
 298 individual device drivers, so that the current design throws a wide net.
 299 The biggest negative of the design is that it potentially disturbs
 300 network daemons and file systems that didn't need to be disturbed.
 301
 302 -- A minor complaint is that resetting the network card causes
 303    user-space back-to-back ifdown/ifup burps that potentially disturb
 304    network daemons, that didn't need to even know that the pci
 305    card was being rebooted.
 306
 307 -- A more serious concern is that the same reset, for SCSI devices,
 308    causes havoc to mounted file systems.  Scripts cannot post-facto
 309    unmount a file system without flushing pending buffers, but this
 310    is impossible, because I/O has already been stopped.  Thus,
 311    ideally, the reset should happen at or below the block layer,
 312    so that the file systems are not disturbed.
 313
 314    Reiserfs does not tolerate errors returned from the block device.
 315    Ext3fs seems to be tolerant, retrying reads/writes until it does
 316    succeed. Both have been only lightly tested in this scenario.
 317
 318    The SCSI-generic subsystem already has built-in code for performing
 319    SCSI device resets, SCSI bus resets, and SCSI host-bus-adapter
 320    (HBA) resets.  These are cascaded into a chain of attempted
 321    resets if a SCSI command fails. These are completely hidden
 322    from the block layer.  It would be very natural to add an EEH
 323    reset into this chain of events.
 324
 325 -- If a SCSI error occurs for the root device, all is lost unless
 326    the sysadmin had the foresight to run /bin, /sbin, /etc, /var
 327    and so on, out of ramdisk/tmpfs.
 328
 329
 330 Conclusions
 331 -----------
 332 There's forward progress ...
 333
 334