Skip to content

L
linux
Switch branch/tag
  1. 02 Nov, 2008 1 commit
  3. 31 Oct, 2008 1 commit
  5. 29 Oct, 2008 4 commits
  7. 23 Oct, 2008 6 commits
    • David S. Miller's avatar
      math-emu: Fix thinko in _FP_DIV · 60b82673
      David S. Miller authored 
      In commit 48d6c643 ("math-emu: Add
support for reporting exact invalid exception") code was added to
set the new FP_EX_INVALID_{IDI,ZDZ} exception flag bits.

However there is a missing break statement for the
_FP_CLS_COMBINE(FP_CLS_INF,FP_CLS_INF) switch case, the
code just falls into _FP_CLS_COMBINE(FP_CLS_ZERO,FP_CLS_ZERO)
which then proceeds to overwrite all of the settings.

Fix by adding the missing break.
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      60b82673
    • Kumar Gala's avatar
      math-emu: Fix signalling of underflow and inexact while packing result. · 930cc144
      Kumar Gala authored 
      I'm trying to move the powerpc math-emu code to use the include/math-emu bits.

In doing so I've been using TestFloat to see how good or bad we are
doing.  For the most part the current math-emu code that PPC uses has
a number of issues that the code in include/math-emu seems to solve
(plus bugs we've had for ever that no one every realized).

Anyways, I've come across a case that we are flagging underflow and
inexact because we think we have a denormalized result from a double
precision divide:

000.FFFFFFFFFFFFF / 3FE.FFFFFFFFFFFFE
	soft: 001.0000000000000 .....  syst: 001.0000000000000 ...ux

What it looks like is the results out of FP_DIV_D are:

D:
sign:	  0
mantissa: 01000000 00000000
exp:	 -1023 (0)

The problem seems like we aren't normalizing the result and bumping the exp.

Now that I'm digging into this a bit I'm thinking my issue has to do with
the fix DaveM put in place from back in Aug 2007 (commit
40584961):

[MATH-EMU]: Fix underflow exception reporting.

    2) we ended up rounding back up to normal (this is the case where
       we set the exponent to 1 and set the fraction to zero), this
       should set inexact too
...

    Another example, "0x0.0000000000001p-1022 / 16.0", should signal both
    inexact and underflow.  The cpu implementations and ieee1754
    literature is very clear about this.  This is case #2 above.

Here is the distilled glibc test case from Jakub Jelinek which prompted that
commit:

--------------------
#include <float.h>
#include <fenv.h>
#include <stdio.h>

volatile double d = DBL_MIN;
volatile double e = 0x0.0000000000001p-1022;
volatile double f = 16.0;
int
main (void)
{
  printf ("%x\n", fetestexcept (FE_UNDERFLOW));
  d /= f;
  printf ("%x\n", fetestexcept (FE_UNDERFLOW));
  e /= f;
  printf ("%x\n", fetestexcept (FE_UNDERFLOW));
  return 0;
}
--------------------

It looks like the case I have we are exact before rounding, but think it
looks like the rounding case since it appears as if "overflow is set".

000.FFFFFFFFFFFFF / 3FE.FFFFFFFFFFFFE = 001.0000000000000

I think the following adds the check for my case and still works for the
issue your commit was trying to resolve.
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      930cc144
    • Martin Habets's avatar
      sparc: Add checkstack support · d41e2d73
      Martin Habets authored 
      Add sparc support to checkstack.
Signed-off-by: default avatarMartin Habets <errandir_news@mph.eclipse.co.uk>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      d41e2d73
    • Frederic Weisbecker's avatar
      sparc: correct section of current_pc() · 8dd94537
      Frederic Weisbecker authored 
      Latest mainline gives this section mismatch on sparc:

The function current_pc() references
the variable __init no_sun4u_here.
This is often because current_pc lacks a __init
annotation or the annotation of no_sun4u_here is wrong.

Since current_pc() is used only in early time, it is correct to
put it in .init section.
Signed-off-by: default avatarFrederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      8dd94537
    • Frederic Weisbecker's avatar
      sparc: correct section of apc_no_idle · a1731e5b
      Frederic Weisbecker authored 
      The latest mainline gives this section mismatch on sparc:

The function __devinit apc_probe() references
a variable __initdata apc_no_idle.
If apc_no_idle is only used by apc_probe then
annotate apc_no_idle with a matching annotation.

Since the commit 7e7e2f03,
apc_probe() is on __devinit so we have to correct apc_no_idle
which is referenced by this function.
Signed-off-by: default avatarFrederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      a1731e5b
    • Andrea Shepard's avatar
      sparc64: Fix race in arch/sparc64/kernel/trampoline.S · e0037df3
      Andrea Shepard authored 
      Make arch/sparc64/kernel/trampoline.S in 2.6.27.1 lock prom_entry_lock
when calling the PROM.  This prevents a race condition that I observed
causing a hang on startup on a 12-CPU E4500.

I am not subscribed to this list, so please CC me on replies.
Signed-off-by: default avatarAndrea Shepard <andrea@persephoneslair.org>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      e0037df3
  9. 21 Oct, 2008 1 commit
  11. 20 Oct, 2008 27 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7