debugging.rst 1.7 KB

12345678910111213141516171819202122232425262728293031323334353637383940414243444546
  1. =================
  2. PA-RISC Debugging
  3. =================
  4. okay, here are some hints for debugging the lower-level parts of
  5. linux/parisc.
  6. 1. Absolute addresses
  7. =====================
  8. A lot of the assembly code currently runs in real mode, which means
  9. absolute addresses are used instead of virtual addresses as in the
  10. rest of the kernel. To translate an absolute address to a virtual
  11. address you can lookup in System.map, add __PAGE_OFFSET (0x10000000
  12. currently).
  13. 2. HPMCs
  14. ========
  15. When real-mode code tries to access non-existent memory, you'll get
  16. an HPMC instead of a kernel oops. To debug an HPMC, try to find
  17. the System Responder/Requestor addresses. The System Requestor
  18. address should match (one of the) processor HPAs (high addresses in
  19. the I/O range); the System Responder address is the address real-mode
  20. code tried to access.
  21. Typical values for the System Responder address are addresses larger
  22. than __PAGE_OFFSET (0x10000000) which mean a virtual address didn't
  23. get translated to a physical address before real-mode code tried to
  24. access it.
  25. 3. Q bit fun
  26. ============
  27. Certain, very critical code has to clear the Q bit in the PSW. What
  28. happens when the Q bit is cleared is the CPU does not update the
  29. registers interruption handlers read to find out where the machine
  30. was interrupted - so if you get an interruption between the instruction
  31. that clears the Q bit and the RFI that sets it again you don't know
  32. where exactly it happened. If you're lucky the IAOQ will point to the
  33. instruction that cleared the Q bit, if you're not it points anywhere
  34. at all. Usually Q bit problems will show themselves in unexplainable
  35. system hangs or running off the end of physical memory.