Commit graph

142 commits

Author SHA1 Message Date
Frans Kaashoek c3dcf47966 Clean up memlayout.h
Get rid of last instances of linear address and "la"
Get ready for detecting physical memory dynamically
2011-08-16 15:47:22 -04:00
Frans Kaashoek a4b213cf49 Avoid "boot" in xv6 2011-08-15 20:11:13 -04:00
Frans Kaashoek 66ba8079c7 Use static page table for boot, mapping first 4Mbyte; no more segment trick
Allocate proper kernel page table immediately in main using boot allocator
Remove pginit
Simplify address space layout a tiny bit
More to come (e.g., superpages to simplify static table)
2011-08-09 21:37:35 -04:00
Frans Kaashoek 9aa0337dc1 Map kernel high
Very important to give qemu memory through PHYSTOP :(
2011-07-29 07:31:27 -04:00
Frans Kaashoek 13a96baefc Dirt simple logging
Passes usertests and stressfs
Seems to recover correctly in a number of simple cases
2011-07-27 20:35:46 -04:00
Russ Cox 417c37115e more trivial cleanup 2011-01-11 13:51:40 -05:00
Russ Cox 1a81e38b17 make new code like old code
Variable declarations at top of function,
separate from initialization.

Use == 0 instead of ! for checking pointers.

Consistent spacing around {, *, casts.

Declare 0-parameter functions as (void) not ().

Integer valued functions return -1 on failure, 0 on success.
2011-01-11 13:01:13 -05:00
Robert Morris 4655d42e3b copyout() copies data to a va in a pagetable, for exec() &c
usertest that passes too many arguments, break exec
2010-09-27 16:14:33 -04:00
Robert Morris faad047ab2 change some comments, maybe more informative
delete most comments from bootother.S (since copy of bootasm.S)
ksegment() -> seginit()
move more stuff from main() to mainc()
2010-09-13 15:34:44 -04:00
Austin Clements 588644f472 Consistent style in defs.h 2010-09-02 19:01:25 -04:00
Austin Clements 79cd8b3eed Simplify allocuvm/deallocuvm to operate in a contiguous memory model. This makes their interface match up better with proc->sz and also simplifies the callers (it even gets the main body of exec on one page). 2010-09-02 18:28:36 -04:00
Austin Clements f53e6110be Simplify inituvm and userinit by assuming initcode fits on a page 2010-09-02 15:42:25 -04:00
Austin Clements c7c21467c3 Oops. Broke the build when I rearranged proc.c 2010-09-02 14:30:06 -04:00
Robert Morris 8d774afb2d no more pminit, or ELF header at 0x10000
kinit() knows about end and PHYSTOP
map all of kernel read/write (rather than r/o instructions)
thanks, austin
2010-08-31 15:39:25 -04:00
Robert Morris 7d7dc9331b kalloc/kfree now only a page at a time
do not keep sorted contiguous free list
2010-08-31 12:54:47 -04:00
Robert Morris ac090078c6 xx 2010-08-30 10:13:49 -04:00
Robert Morris 789b508d53 uptime() sys call for benchmarking
increase PHYSTOP
2010-08-11 14:34:45 -04:00
Robert Morris 83d2db91f7 allow sbrk(-x) to de-allocate user memory 2010-08-10 17:08:41 -04:00
Robert Morris c4cc10da7e fix corner cases in exec of ELF
put an invalid page below the stack
have fork() handle invalid pages
2010-08-06 11:12:18 -04:00
Robert Morris c99599784e remove some unused vm #defines
fix corner cases with alignment when mapping kernel ELF file
2010-08-05 16:00:59 -04:00
Robert Morris 2cf6b32d4d move jkstack to main.c
replace jstack with asm()s
2010-08-05 14:15:03 -04:00
Robert Morris eb18645f17 fix allocuvm() to handle sbrk() with non-page-granularity argument
(maybe this never worked, but it works now)
2010-08-05 12:10:54 -04:00
Frans Kaashoek b738a4f1a2 kill TLB shoot down code 2010-07-28 14:38:05 -04:00
Frans Kaashoek 30f5bf0548 some cleanup 2010-07-25 20:30:21 -04:00
Frans Kaashoek 4714c20521 Checkpoint page-table version for SMP
Includes code for TLB shootdown (which actually seems unnecessary for xv6)
2010-07-23 07:41:13 -04:00
Frans Kaashoek 74c82bc158 nits 2010-07-02 17:45:37 -04:00
Frans Kaashoek 40889627ba Initial version of single-cpu xv6 with page tables 2010-07-02 14:51:53 -04:00
Austin Clements 2c536bff67 Remove memcpy prototypes at Russ' request to prevent code from calling
memcpy directly.
2009-10-07 13:06:55 -04:00
Austin Clements 6639ce56d9 Provide memcpy for compatibility with older versions of gcc 2009-10-07 12:05:56 -04:00
Russ Cox 48755214c9 assorted fixes:
* rename c/cp to cpu/proc
 * rename cpu.context to cpu.scheduler
 * fix some comments
 * formatting for printout
2009-08-30 23:02:08 -07:00
Russ Cox 0aef891495 shuffle and tweak for formatting.
pdf has very good page breaks now.
would be a good copy for fall 2009.
2009-08-08 01:07:30 -07:00
Russ Cox 2c5f7aba38 initproc, usegment, swtch tweaks 2009-07-11 19:28:29 -07:00
rsc 215738336a move fork into proc.c 2009-05-31 00:38:51 +00:00
rsc 19333efb9e Some proc cleanup, moving some of copyproc into allocproc.
Also, an experiment: use "thread-local" storage for c and cp
instead of the #define macro for curproc[cpu()].
2009-05-31 00:28:45 +00:00
rsc 2157576107 be consistent: no underscores in function names 2009-03-08 22:07:13 +00:00
kolya c100d9ee2d cleaner swtch.S 2008-10-15 05:14:10 +00:00
rsc f97f0d2b3d cleaner 2007-09-27 21:02:03 +00:00
rsc 3807c1f20b rename splhi/spllo to pushcli/popcli 2007-09-27 20:09:40 +00:00
rsc c8919e6537 kernel SMP interruptibility fixes.
Last year, right before I sent xv6 to the printer, I changed the
SETGATE calls so that interrupts would be disabled on entry to
interrupt handlers, and I added the nlock++ / nlock-- in trap()
so that interrupts would stay disabled while the hw handlers
(but not the syscall handler) did their work.  I did this because
the kernel was otherwise causing Bochs to triple-fault in SMP
mode, and time was short.

Robert observed yesterday that something was keeping the SMP
preemption user test from working.  It turned out that when I
simplified the lapic code I swapped the order of two register
writes that I didn't realize were order dependent.  I fixed that
and then since I had everything paged in kept going and tried
to figure out why you can't leave interrupts on during interrupt
handlers.  There are a few issues.

First, there must be some way to keep interrupts from "stacking
up" and overflowing the stack.  Keeping interrupts off the whole
time solves this problem -- even if the clock tick handler runs
long enough that the next clock tick is waiting when it finishes,
keeping interrupts off means that the handler runs all the way
through the "iret" before the next handler begins.  This is not
really a problem unless you are putting too many prints in trap
-- if the OS is doing its job right, the handlers should run
quickly and not stack up.

Second, if xv6 had page faults, then it would be important to
keep interrupts disabled between the start of the interrupt and
the time that cr2 was read, to avoid a scenario like:

   p1 page faults [cr2 set to faulting address]
   p1 starts executing trapasm.S
   clock interrupt, p1 preempted, p2 starts executing
   p2 page faults [cr2 set to another faulting address]
   p2 starts, finishes fault handler
   p1 rescheduled, reads cr2, sees wrong fault address

Alternately p1 could be rescheduled on the other cpu, in which
case it would still see the wrong cr2.  That said, I think cr2
is the only interrupt state that isn't pushed onto the interrupt
stack atomically at fault time, and xv6 doesn't care.  (This isn't
entirely hypothetical -- I debugged this problem on Plan 9.)

Third, and this is the big one, it is not safe to call cpu()
unless interrupts are disabled.  If interrupts are enabled then
there is no guarantee that, between the time cpu() looks up the
cpu id and the time that it the result gets used, the process
has not been rescheduled to the other cpu.  For example, the
very commonly-used expression curproc[cpu()] (aka the macro cp)
can end up referring to the wrong proc: the code stores the
result of cpu() in %eax, gets rescheduled to the other cpu at
just the wrong instant, and then reads curproc[%eax].

We use curproc[cpu()] to get the current process a LOT.  In that
particular case, if we arranged for the current curproc entry
to be addressed by %fs:0 and just use a different %fs on each
CPU, then we could safely get at curproc even with interrupts
disabled, since the read of %fs would be atomic with the read
of %fs:0.  Alternately, we could have a curproc() function that
disables interrupts while computing curproc[cpu()].  I've done
that last one.

Even in the current kernel, with interrupts off on entry to trap,
interrupts are enabled inside release if there are no locks held.
Also, the scheduler's idle loop must be interruptible at times
so that the clock and disk interrupts (which might make processes
runnable) can be handled.

In addition to the rampant use of curproc[cpu()], this little
snippet from acquire is wrong on smp:

  if(cpus[cpu()].nlock == 0)
    cli();
  cpus[cpu()].nlock++;

because if interrupts are off then we might call cpu(), get
rescheduled to a different cpu, look at cpus[oldcpu].nlock, and
wrongly decide not to disable interrupts on the new cpu.  The
fix is to always call cli().  But this is wrong too:

  if(holding(lock))
    panic("acquire");
  cli();
  cpus[cpu()].nlock++;

because holding looks at cpu().  The fix is:

  cli();
  if(holding(lock))
    panic("acquire");
  cpus[cpu()].nlock++;

I've done that, and I changed cpu() to complain the first time
it gets called with interrupts disabled.  (It gets called too
much to complain every time.)

I added new functions splhi and spllo that are like acquire and
release but without the locking:

  void
  splhi(void)
  {
    cli();
    cpus[cpu()].nsplhi++;
  }

  void
  spllo(void)
  {
    if(--cpus[cpu()].nsplhi == 0)
      sti();
  }

and I've used those to protect other sections of code that refer
to cpu() when interrupts would otherwise be disabled (basically
just curproc and setupsegs).  I also use them in acquire/release
and got rid of nlock.

I'm not thrilled with the names, but I think the concept -- a
counted cli/sti -- is sound.  Having them also replaces the
nlock++/nlock-- in trap.c and main.c, which is nice.


Final note: it's still not safe to enable interrupts in
the middle of trap() between lapic_eoi and returning
to user space.  I don't understand why, but we get a
fault on pop %es because 0x10 is a bad segment
descriptor (!) and then the fault faults trying to go into
a new interrupt because 0x8 is a bad segment descriptor too!
Triple fault.  I haven't debugged this yet.
2007-09-27 12:58:42 +00:00
rsc d5596cd61d Apparently the initial interrupt count lapic[TICR]
must be set *after* initializing the lapic[TIMER] vector.

Doing this, we now get clock interrupts on cpu 1.
(No idea why we always got them on cpu 0.)

Don't write to TCCR - it is read-only.
2007-09-26 20:34:12 +00:00
rsc cffa954301 nits 2007-08-28 19:25:04 +00:00
rsc 5573c8f296 delete proc_ on proc_exit, proc_wait, proc_kill 2007-08-28 19:14:43 +00:00
rsc eb52c7de1d comments; rename irq_ to pic_ 2007-08-28 19:04:36 +00:00
rsc 818fc0125e replace setjmp/longjmp with swtch 2007-08-28 12:48:33 +00:00
rsc 9e82bfb04c rename 8253pit.c to timer.c 2007-08-28 04:40:58 +00:00
rsc 7834cca604 remove _ from pipe; be like file 2007-08-28 04:22:35 +00:00
rsc f0d11fea82 Move keyboard code into kbd.c; add backspace handling. 2007-08-28 03:28:13 +00:00
rsc 19b42cc078 Rename main0 to main. 2007-08-27 23:32:16 +00:00
rsc b63bb0fd00 Clean up lapic code.
One initialization function now, not three.
Use #defines instead of enums (consistent with other code, but sigh).

Still boots in Bochs in SMP mode.
2007-08-27 16:57:13 +00:00
rsc 1ccff18b24 fileincref -> filedup (consistent with idup) 2007-08-27 14:35:09 +00:00