diff options
| author | Ralf Baechle <ralf@linux-mips.org> | 1998-08-25 09:12:35 +0000 |
|---|---|---|
| committer | Ralf Baechle <ralf@linux-mips.org> | 1998-08-25 09:12:35 +0000 |
| commit | c7fc24dc4420057f103afe8fc64524ebc25c5d37 (patch) | |
| tree | 3682407a599b8f9f03fc096298134cafba1c9b2f /arch/i386/math-emu/README | |
| parent | 1d793fade8b063fde3cf275bf1a5c2d381292cd9 (diff) | |
o Merge with Linux 2.1.116.
o New Newport console code.
o New G364 console code.
Diffstat (limited to 'arch/i386/math-emu/README')
| -rw-r--r-- | arch/i386/math-emu/README | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/arch/i386/math-emu/README b/arch/i386/math-emu/README index c6c774092..04434b618 100644 --- a/arch/i386/math-emu/README +++ b/arch/i386/math-emu/README @@ -131,7 +131,7 @@ code is: movl %esp,[%ebx] fld1 The FPU instruction may be (usually will be) loaded into the pre-fetch -queue of the cpu before the mov instruction is executed. If the +queue of the CPU before the mov instruction is executed. If the destination of the 'movl' overlaps the FPU instruction then the bytes in the prefetch queue and memory will be inconsistent when the FPU instruction is executed. The emulator will be invoked but will not be @@ -163,7 +163,7 @@ Speed. The speed of floating point computation with the emulator will depend upon instruction mix. Relative performance is best for the instructions which require most computation. The simple instructions are adversely -affected by the fpu instruction trap overhead. +affected by the FPU instruction trap overhead. Timing: Some simple timing tests have been made on the emulator functions. |