diff options
author | David Miller <davem@davemloft.net> | 2007-08-21 21:04:07 -0700 |
---|---|---|
committer | Greg Kroah-Hartman <gregkh@suse.de> | 2007-08-30 23:01:33 -0700 |
commit | 14d5c15ac1e080baaa861fbcde342fda3c549bdd (patch) | |
tree | d2baa9ab98a83aef6214a1d93081500cd2a01fbb /include | |
parent | f7d75b68abd28e63dd982bb6748bc5d32a289c42 (diff) |
Fix soft-fp underflow handling.
The underflow exception cases were wrong.
This is one weird area of ieee1754 handling in that the underflow
behavior changes based upon whether underflow is enabled in the trap
enable mask of the FPU control register. As a specific case the Sparc
V9 manual gives us the following description:
--------------------
If UFM = 0: Underflow occurs if a nonzero result is tiny and a
loss of accuracy occurs. Tininess may be detected
before or after rounding. Loss of accuracy may be
either a denormalization loss or an inexact result.
If UFM = 1: Underflow occurs if a nonzero result is tiny.
Tininess may be detected before or after rounding.
--------------------
What this amounts to in the packing case is if we go subnormal,
we set underflow if any of the following are true:
1) rounding sets inexact
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
3) underflow is set in FPU control register trap-enable mask
The initially discovered example was "DBL_MIN / 16.0" which
incorrectly generated an underflow. It should not, unless underflow
is set in the trap-enable mask of the FPU csr.
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.
However, if underflow is set in the trap enable mask, only underflow
should be set and reported as a trap. That is handled properly by the
prioritization logic in
arch/sparc{,64}/math-emu/math.c:record_exception().
Based upon a report and test case from Jakub Jelinek.
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Diffstat (limited to 'include')
-rw-r--r-- | include/asm-sparc/sfp-machine.h | 6 | ||||
-rw-r--r-- | include/asm-sparc64/sfp-machine.h | 2 | ||||
-rw-r--r-- | include/math-emu/op-common.h | 5 | ||||
-rw-r--r-- | include/math-emu/soft-fp.h | 7 |
4 files changed, 19 insertions, 1 deletions
diff --git a/include/asm-sparc/sfp-machine.h b/include/asm-sparc/sfp-machine.h index ecfc86a4a72..266a42b8f99 100644 --- a/include/asm-sparc/sfp-machine.h +++ b/include/asm-sparc/sfp-machine.h @@ -203,4 +203,10 @@ extern struct task_struct *last_task_used_math; #define FP_INHIBIT_RESULTS ((last_task_used_math->thread.fsr >> 23) & _fex) #endif +#ifdef CONFIG_SMP +#define FP_TRAPPING_EXCEPTIONS ((current->thread.fsr >> 23) & 0x1f) +#else +#define FP_TRAPPING_EXCEPTIONS ((last_task_used_math->thread.fsr >> 23) & 0x1f) +#endif + #endif diff --git a/include/asm-sparc64/sfp-machine.h b/include/asm-sparc64/sfp-machine.h index 89d42431efb..c9331b02d9c 100644 --- a/include/asm-sparc64/sfp-machine.h +++ b/include/asm-sparc64/sfp-machine.h @@ -88,4 +88,6 @@ #define FP_INHIBIT_RESULTS ((current_thread_info()->xfsr[0] >> 23) & _fex) +#define FP_TRAPPING_EXCEPTIONS ((current_thread_info()->xfsr[0] >> 23) & 0x1f) + #endif diff --git a/include/math-emu/op-common.h b/include/math-emu/op-common.h index 93780abd01b..bb46e7645d5 100644 --- a/include/math-emu/op-common.h +++ b/include/math-emu/op-common.h @@ -145,13 +145,16 @@ do { \ { \ X##_e = 1; \ _FP_FRAC_SET_##wc(X, _FP_ZEROFRAC_##wc); \ + FP_SET_EXCEPTION(FP_EX_INEXACT); \ } \ else \ { \ X##_e = 0; \ _FP_FRAC_SRL_##wc(X, _FP_WORKBITS); \ - FP_SET_EXCEPTION(FP_EX_UNDERFLOW); \ } \ + if ((FP_CUR_EXCEPTIONS & FP_EX_INEXACT) || \ + (FP_TRAPPING_EXCEPTIONS & FP_EX_UNDERFLOW)) \ + FP_SET_EXCEPTION(FP_EX_UNDERFLOW); \ } \ else \ { \ diff --git a/include/math-emu/soft-fp.h b/include/math-emu/soft-fp.h index d02eb64a865..a6f873b45f9 100644 --- a/include/math-emu/soft-fp.h +++ b/include/math-emu/soft-fp.h @@ -97,12 +97,19 @@ #define FP_INHIBIT_RESULTS 0 #endif +#ifndef FP_TRAPPING_EXCEPTIONS +#define FP_TRAPPING_EXCEPTIONS 0 +#endif + #define FP_SET_EXCEPTION(ex) \ _fex |= (ex) #define FP_UNSET_EXCEPTION(ex) \ _fex &= ~(ex) +#define FP_CUR_EXCEPTIONS \ + (_fex) + #define FP_CLEAR_EXCEPTIONS \ _fex = 0 |