Commit Graph

529 Commits

Author SHA1 Message Date
thorpej 534e7d4454 Several changes, which get us generally further along with
multiprocessor support:
- Implement MP-safe halt.
- Make the FPU saving code more like Bill's on the i386 MP branch.
  XXX This code will no doubt be revisited again.
- Pass the cpu_info and trapframe to IPI handlers, saving some work
  in the handlers themselves, and also making it possible for the
  "pause" handler to reference register state for DDB.
- Add "machine cpu" to DDB, making it possible to reference other
  CPUs registers (and thus get e.g. a traceback) from whichever
  CPU is actually running the debugger.
- Garbage-collect "machine halt" and "machine reboot" DDB commands.
  They don't have a prayer of working properly in multiprocessor
  kernels, and didn't really work all that well in uniprocessor kernels.
2000-11-22 08:39:46 +00:00
thorpej 6cbdf24c45 Add a debugging printf, commented out by default. 2000-11-22 07:44:01 +00:00
thorpej 7231516137 Just access the trapframe directly for DDB registers. 2000-11-22 02:25:52 +00:00
thorpej f595d8e741 Ooops, do the last only if MULTIPROCESSOR. 2000-11-20 21:18:07 +00:00
thorpej e12f731db5 On the Alpha, interprocessor interrupts come in at the same priority
as normal device interrupts.  Because of this, we won't get IPIs while
servicing such interrupts.  This can lead to the following deadlock
scenario as reported by Bill Sommerfeld:
- Process runs on cpu1, but has FP state on cpu0.
- Process executes FP-using insn, causing an FP trap, which causes
  the kernel lock to be acquired.
- At roughly the same time, cpu0 receives a device interrupt, and attempts
  to acquire the kernel lock, which blocks since cpu1 already has it.
- cpu1 sends cpu0 a SYNCH FPU IPI, and waits for cpu0 to release its
  FP state.
- Since cpu0 cannot notice the IPI until it has processed the device
  interrupt, which it cannot do because it cannot acquire the kernel
  lock, we have deadlock.

Solve the problem by adding a spinlock interlock release hook which
checks for pending IPIs and processes them.

Idea from Bill Sommerfeld.
2000-11-20 20:17:13 +00:00
thorpej fdfb847e3e Make ci_ipis volatile. 2000-11-20 19:25:43 +00:00
thorpej 7f059c8e69 Move IPI processing into a separate function. 2000-11-20 19:24:36 +00:00
thorpej b9ad1f75b9 Make VM_MAX_KERNEL_ADDRESS end on a page boundary. 2000-11-19 03:16:55 +00:00
thorpej 561db1fb7e Implement pmap_growkernel(). 2000-11-19 03:16:34 +00:00
thorpej 137c030d5b Count individual interprocessor interrupts -- it's good to know where
they all come from.
2000-11-18 19:25:35 +00:00
nisimura 48ff19c98b u_char -> u_int for function argument and return of Zilog register
manipulation helper functions; no loss, no gain, for Alpha processor.
2000-09-09 06:08:42 +00:00
thorpej 9c5afcb2f0 Move the Alpha _mcount() into a .S file so that it can be
pre-processed, like other Alpha assembly files.
2000-09-05 16:28:30 +00:00
thorpej 0778dad08f Define the biglock perimeter for the Alpha port, and generally clean
up a bunch of the MP code.  Still doens't work properly yet, this is
just a snapshot of work-in-progress.
2000-09-04 00:31:58 +00:00
thorpej 0e04909346 Snapshot of TLB shootdown bugfixes. 2000-08-26 03:27:44 +00:00
thorpej 4db6fc7542 Make need_resched() take a "struct cpu_info *" argument. This
causes gives a primitive form of processor affinity.  Its use in
roundrobin() still needs some work.
2000-08-25 01:04:06 +00:00
thorpej 58e7a6954b Add spllock(). See spl(9) for details. 2000-08-22 19:46:26 +00:00
thorpej 2648536e0c Add experimental code for pausing other CPUs upon a CPU's
entry into the debugger.  While I'm here, add splsched()
as per spl(9).
2000-08-21 02:03:12 +00:00
thorpej 5a7793edd9 Implement MP-safe lazy FP context switching, modeled on the
way Bill Sommerfeld implemented it for x86 (and bug fixes
fed back to Bill :-)
2000-08-15 22:16:17 +00:00
thorpej f10ccf00b1 Remove an unnecessary MB in __cpu_simple_unlock(), per sections
5.3.3 and 5.3.4 of the Green Book, and a discussion w/ cgd@netbsd.org.
2000-08-14 20:28:28 +00:00
thorpej b80f38bf8b Issue a memory barrier before clearing a lock, as per section 5.5.3
of the Green Book and a discussion on tech-smp and port-alpha.
2000-08-14 17:18:50 +00:00
thorpej 325f9ccafc Add alpha_multicast_ipi(). 2000-08-13 18:20:55 +00:00
thorpej e207174eb0 Add functions to read EISA configuration data for MEM, IRQ, DMA, and IO.
XXX Just error stubs on the i386 right now -- someone needs to write
XXX EISA BIOS code for i386.
2000-08-11 00:43:18 +00:00
tshiozak 51a53de0d5 Preparation for the future introduction of multibyte locale.
- MB_LEN_MAX is increased to 32.
 - To ensure binary compatibility for old executables
   under multibyte locale, versioned setlocale is added.
 - __mb_len_cur definision is added in setlocale.c
   and enable it in stdlib.h .
   It is also important for multibyte locale stuffs,
   but I just forgot.
2000-08-08 22:31:13 +00:00
briggs 3f1e17a340 Define bus_space_*_stream_N functions as their non-stream counterparts. 2000-07-31 22:43:46 +00:00
thorpej d760e0b407 Add code to read the EISA configuration NVRAM as set up by
an EISA Configuration Utility.  Code to access this data
is forthcoming.

XXX This could probably be made MI at some point.
2000-07-29 23:18:46 +00:00
thorpej 3eedcb3009 Whitespace police. 2000-07-13 05:47:39 +00:00
thorpej 2c9058efbb Oops, g/c some stuff from previous. 2000-07-06 23:29:13 +00:00
thorpej 65780061bd Define some more CTB formats. 2000-07-06 22:46:54 +00:00
thorpej 01bd2a3496 The `rpb_phys_address_size' field is really 2 32-bit fields,
`rpb_phys_address_size' and `rpb_extended_va_size', per Brown Book.
2000-07-03 01:47:48 +00:00
simonb edf0a0d67e Pass the abst to alpha_pci_mem_unmap() so that the real size of the
mapped area can be unmapped.
2000-06-30 18:19:28 +00:00
mrg 80bb8049ba remove include of <vm/vm.h>. <vm/vm.h> -> <uvm/uvm_extern.h> 2000-06-29 09:02:51 +00:00
kleink bb2ed0f487 G/c _BSD_INTPTR_T_ and _BSD_UINTPTR_T_. 2000-06-27 05:53:22 +00:00
kleink 47b5c5e3b1 Resolve some formatting nits; add __intptr_t and __uintptr_t. 2000-06-27 04:58:51 +00:00
kleink e695f72a2e Add <machine/int_types.h>, which provides namespace-pure definitions
of exact-width integer types.
2000-06-26 15:42:16 +00:00
simonb 889c658b5b Change the kernel mmap interface so that the offset to map is an
"off_t" and the return value is a "paddr_t" to allow mappings
at offsets past 2^31 bytes.  Somewhat inspired by FreeBSD, which
only changed the offset to a "vm_offset_t".

Includes updates for the i386, pc532 and sh3 mmmmap from Jason Thorpe.
2000-06-26 04:55:19 +00:00
kleink 133ea38323 Add a WEAK_ALIAS() macro. 2000-06-23 12:18:45 +00:00
simonb 8e0ec9a961 s/ouwl/outl/ in macro definition. 2000-06-19 12:19:38 +00:00
thorpej 71cd989b8f Only include <machine/intr.h> if _KERNEL. Noted by enami tsugutomo. 2000-06-09 16:03:04 +00:00
cgd c2ebb05a74 make spl0() and spllowersoftclock() return void. Also, move spl0()
prototype from param.h to intr.h.  (there were some big XXXs in param.h
that said to do that, and intr.h is included by param.h, so...)
2000-06-09 01:40:12 +00:00
thorpej 81afcb3940 And more ANSI'ification! 2000-06-08 03:10:06 +00:00
thorpej 966b3da402 ANSI'ify. 2000-06-08 02:53:47 +00:00
thorpej 0bf91362b9 Add cpu_counter() for rnd(4), using the rpcc insn. 2000-06-08 02:19:20 +00:00
simonb e59356ecdd In __cpu_simple_lock_try(), use "return (v0 != 0);" instead of
"return (v0);" where v0 is an unsigned long and the function return
type is int.  Stops lint warnings when this file is included.

Fix from Jason Thorpe.
2000-06-07 01:49:13 +00:00
thorpej 2668e3b213 Switch to the new `evcnt' mechanism for counting interrupts. Maintain
a per-CPU interrupt counter for clock, device, and interprocessor
interrupts.
2000-06-05 21:47:10 +00:00
cgd cffb580806 Implement the more flexiable `evcnt' interface as discussed (briefly) on
tech-kern and now documented in evcnt(9).
2000-06-04 19:14:14 +00:00
thorpej 70dc4d8718 Un-__P'ify kernel prototypes. 2000-06-04 05:23:18 +00:00
thorpej df4a5a7f43 Increment an interrupt depth counter in non-clock interrupt cases
so that we can get time spent in interrupt statistics.
2000-06-04 03:40:03 +00:00
thorpej cd423732b8 - Clean up clock interrupt code a bit, and provide a CPU_IS_PRIMARY()
macro in the MULTIPROCESSOR case (hardclock() wants it).
- Implement __GENERIC_SOFT_INTERRUPTS, and redefine the legacy
  software interrupts in terms of it.  Garbage-collect setsoftserial().
2000-06-03 20:47:36 +00:00
thorpej 392948ced8 Add EV57 and EV67 processor variants.
cvs: ----------------------------------------------------------------------
2000-06-03 14:12:28 +00:00
thorpej 79744e1563 Add a global patchable variable to enable debugging printfs in the
code that finds the boot device.
2000-06-01 17:12:38 +00:00