linux-toradex.git/arch/um/include, branch v2.6.19.2

[PATCH] uml: make execvp safe for our usage

2006-11-25T21:28:34+00:00

Reimplement execvp for our purposes - after we call fork() it is fundamentally
unsafe to use the kernel allocator - current is not valid there.  So we simply
pass to our modified execvp() a preallocated buffer.  This fixes a real bug
and works very well in testing (I've seen indirectly warning messages from the
forked thread - they went on the pipe connected to its stdout and where read
as a number by UML, when calling read_output().  I verified the obtained
number corresponded to "BUG:").

The added use of __cant_sleep() is not a new bug since __cant_sleep() is
already used in the same function - passing an atomicity parameter would be
better but it would require huge change, stating that this function must not
be called in atomic context and can sleep is a better idea (will make sure of
this gradually).

Signed-off-by: Paolo 'Blaisorblade' Giarrusso 
Acked-by: Jeff Dike 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] uml: fix I/O hang

2006-11-03T20:27:58+00:00

Fix a UML hang in which everything would just stop until some I/O happened
- a ping, someone whacking the keyboard - at which point everything would
start up again as though nothing had happened.

The cause was gcc reordering some code which absolutely needed to be
executed in the order in the source.  When unblock_signals switches signals
from off to on, it needs to see if any interrupts had happened in the
critical section.  The interrupt handlers check signals_enabled - if it is
zero, then the handler adds a bit to the "pending" bitmask and returns.
unblock_signals checks this mask to see if any signals need to be
delivered.

The crucial part is this:
	signals_enabled = 1;
	save_pending = pending;
	if(save_pending == 0)
		return;
	pending = 0;

In order to avoid an interrupt arriving between reading pending and setting
it to zero, in which case, the record of the interrupt would be erased,
signals are enabled.

What happened was that gcc reordered this so that 'save_pending = pending'
came before 'signals_enabled = 1', creating a one-instruction window within
which an interrupt could arrive, set its bit in pending, and have it be
immediately erased.

When the I/O workload is purely disk-based, the loss of a block device
interrupt stops the entire I/O system because the next block request will
wait for the current one to finish.  Thus the system hangs until something
else causes some I/O to arrive, such as a network packet or console input.

The fix to this particular problem is a memory barrier between enabling
signals and reading the pending signal mask.  An xchg would also probably
work.

Looking over this code for similar problems led me to do a few more
things:

- make signals_enabled and pending volatile so that they don't get cached
  in registers

- add an mb() to the return paths of block_signals and unblock_signals so
  that the modification of signals_enabled doesn't get shuffled into the
  caller in the event that these are inlined in the future.

Signed-off-by: Jeff Dike 
Cc: Paolo 'Blaisorblade' Giarrusso 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] uml ubd driver: various little changes

2006-10-31T16:07:00+00:00

Fix a small memory leak in ubd_config, and clearify the confusion which lead
to it.

Then, some little changes not affecting operations -
* move init functions together,
* add a comment about a potential problem in case of some evolution in the block layer,
* mark all initcalls as static __init functions
* mark an used once little function as inline
* document that mconsole methods are all called in process context (was
  triggered when checking ubd mconsole methods).

Signed-off-by: Paolo 'Blaisorblade' Giarrusso 
Cc: Jeff Dike 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] uml: mconsole fixes

2006-10-25T05:01:10+00:00

 * when we have stop/sysrq/go, we get pt_regs of whatever executes
   mc_work_proc().  Would be better to see what we had at the time of
   interrupt that got us stop.

 * stop/stop/stop.....  will give stack overflow.  Shouldn't allow stop
   from mconsole_stop().

 * stop/stop/go leaves us inside mconsole_stop() with
	os_set_fd_block(req->originating_fd, 0);
	reactivate_fd(req->originating_fd, MCONSOLE_IRQ);
   just done by nested mconsole_stop().  Ditto.

 * once we'd seen stop, there's a period when INTR commands are executed
   out of order (as they should; we might have the things stuck badly
   enough to never reach mconsole_stop(), but still not badly enough to
   block mconsole_interrupt(); in that situation we _want_ things like
   "cad" to be executed immediately).  Once we enter monsole_stop(), all
   INTR commands will be executed in order, mixed with PROC ones.  We'd
   better let user see that such change of behaviour has happened.
   (Suggested by lennert).

 * stack footprint of monsole_interrupt() is an atrocity; AFAICS we can
   safely make struct mc_request req; static in function there.

Signed-off-by: Al Viro 
Acked-by: Jeff Dike 
Signed-off-by: Linus Torvalds

[PATCH] uml: split memory allocation prototypes out of user.h

2006-10-20T17:26:36+00:00

user.h is too generic a header name.  I've split out allocation routines from
it.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso 
Cc: Jeff Dike 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] make UML compile (FC6/x86-64)

2006-10-15T21:18:39+00:00

I need this patch to get a UML kernel to compile.  This is with the
kernel headers in FC6 which are automatically generated from the kernel
tree.  Some headers are missing but those files don't need them.  At
least it appears so since the resuling kernel works fine.

Tested on x86-64.

Signed-off-by: Ulrich Drepper 
Signed-off-by: Linus Torvalds

[PATCH] uml: asm offsets duplication removal

2006-10-11T18:14:20+00:00

Unify macros common to x86 and x86_64 kernel-offsets.h files.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso 
Cc: Jeff Dike 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] uml: allow using again x86/x86_64 crypto code

2006-10-11T18:14:20+00:00

Enable compilation of x86_64 crypto code;, and add the needed constant to make
the code compile again (that macro was added to i386 asm-offsets between
2.6.17 and 2.6.18, in 6c2bb98bc33ae33c7a33a133a4cd5a06395fece5).

Cc: Herbert Xu 
Signed-off-by: Paolo 'Blaisorblade' Giarrusso 
Cc: Jeff Dike 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] uml: make UML_SETJMP always safe

2006-10-11T18:14:20+00:00

If enable is moved by GCC in a register its value may not be preserved after
coming back there with longjmp().  So, mark it as volatile to prevent this;
this is suggested (it seems) in info gcc, when it talks about -Wuninitialized.
 I re-read this and it seems to say something different, but I still believe
this may be needed.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso 
Cc: Jeff Dike 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] uml: readd forgot prototype

2006-10-11T18:14:20+00:00

This was forgot in a previous patch so UML does not compile with TT mode
enabled.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso 
Cc: Jeff Dike 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds