Domovská stránka Prehľadávať Pomoc
Registrovať Prihlásiť sa
phyus
/
linux-vybrid_public
8
0
Fork 0
Súbory Issues 0 Pull requesty 0 Wiki
Strom: a05d7ce36c
Branche Tagy
linux-vybrid_pu...
/
Documentation
/
RCU
/
stallwarn.txt

stallwarn.txt 2.3 KB
História Raw

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758 
  1. Using RCU's CPU Stall Detector
    2. 

  2. 

  3. The CONFIG_RCU_CPU_STALL_DETECTOR kernel config parameter enables
    4. 

  4. RCU's CPU stall detector, which detects conditions that unduly delay
    5. 

  5. RCU grace periods.  The stall detector's idea of what constitutes
    6. 

  6. "unduly delayed" is controlled by a pair of C preprocessor macros:
    7. 

  7. 

  8. RCU_SECONDS_TILL_STALL_CHECK
    9. 

  9. 

  10. 	This macro defines the period of time that RCU will wait from
    11. 

  11. 	the beginning of a grace period until it issues an RCU CPU
    12. 

  12. 	stall warning.	It is normally ten seconds.
    13. 

  13. 

  14. RCU_SECONDS_TILL_STALL_RECHECK
    15. 

  15. 

  16. 	This macro defines the period of time that RCU will wait after
    17. 

  17. 	issuing a stall warning until it issues another stall warning.
    18. 

  18. 	It is normally set to thirty seconds.
    19. 

  19. 

  20. RCU_STALL_RAT_DELAY
    21. 

  21. 

  22. 	The CPU stall detector tries to make the offending CPU rat on itself,
    23. 

  23. 	as this often gives better-quality stack traces.  However, if
    24. 

  24. 	the offending CPU does not detect its own stall in the number
    25. 

  25. 	of jiffies specified by RCU_STALL_RAT_DELAY, then other CPUs will
    26. 

  26. 	complain.  This is normally set to two jiffies.
    27. 

  27. 

  28. The following problems can result in an RCU CPU stall warning:
    29. 

  29. 

  30. o	A CPU looping in an RCU read-side critical section.
    31. 

  31. 	
    32. 

  32. o	A CPU looping with interrupts disabled.
    33. 

  33. 

  34. o	A CPU looping with preemption disabled.
    35. 

  35. 

  36. o	For !CONFIG_PREEMPT kernels, a CPU looping anywhere in the kernel
    37. 

  37. 	without invoking schedule().
    38. 

  38. 

  39. o	A bug in the RCU implementation.
    40. 

  40. 

  41. o	A hardware failure.  This is quite unlikely, but has occurred
    42. 

  42. 	at least once in a former life.  A CPU failed in a running system,
    43. 

  43. 	becoming unresponsive, but not causing an immediate crash.
    44. 

  44. 	This resulted in a series of RCU CPU stall warnings, eventually
    45. 

  45. 	leading the realization that the CPU had failed.
    46. 

  46. 

  47. The RCU, RCU-sched, and RCU-bh implementations have CPU stall warning.
    48. 

  48. SRCU does not do so directly, but its calls to synchronize_sched() will
    49. 

  49. result in RCU-sched detecting any CPU stalls that might be occurring.
    50. 

  50. 

  51. To diagnose the cause of the stall, inspect the stack traces.  The offending
    52. 

  52. function will usually be near the top of the stack.  If you have a series
    53. 

  53. of stall warnings from a single extended stall, comparing the stack traces
    54. 

  54. can often help determine where the stall is occurring, which will usually
    55. 

  55. be in the function nearest the top of the stack that stays the same from
    56. 

  56. trace to trace.
    57. 

  57. 

  58. RCU bugs can often be debugged with the help of CONFIG_RCU_TRACE.
    59.