Instruction
FLDL2E
Pushes the constant log2(e) onto the x87 register stack.
- Extensions
- Forms
- 1 (1 Legacy)
- Data
- Intel XED v2026.08.23
FLDL2E pushes log2(e), about 1.4427, onto the x87 register stack. It is two bytes, D9 EA, and one of the seven constant loads that FLD1’s page describes together. The x87 computes powers and logarithms in base 2, and this constant is the factor that turns base e into base 2: e^x = 2^(x * log2(e)).
Its job in exp
F2XM1 and FSCALE only raise 2 to a
power, so an exponential of e starts with FLDL2E. With -ffast-math,
GCC 14 at -O2 compiles expl(x) on a long double to FLD of x,
fldl2e and fmulp, then the base-2 sequence of F2XM1’s page; its
expm1l begins the same way, and exp10l loads
FLDL2T, log2(10), instead. The natural logarithm goes
the other way with another constant: logl(x) is
FLDLN2, FLD of x and FYL2X, which
computes ln(2) * log2(x). Without -ffast-math, and with Clang 23 in
either case, these are calls of the library.
The product x * log2(e) is rounded to 64 bits before the exponent is split off, so its rounding error becomes an error in the power of 2 that grows with |x|: for a product near 2^14, half an ulp is 2^-50, and the result keeps only about 50 of its 64 bits.
Rounding of the constant
Zen 5 gave 3FFF B8AA3B29 5C17F0BC when rounding to nearest or up, about
0.26 ulp above log2(e), and …F0BB, about 0.74 ulp below, when rounding
down or toward zero, without PE and with C1 cleared. PC at 24 or 53 bits
did not shorten it. GCC 14 also returns a long double constant equal
to log2(e) with FLDL2E; Clang 23 loads it from memory.
A full stack
As a push, FLDL2E needs ST(7) empty. When it was not, Zen 5 pushed the QNaN indefinite with IE masked, setting IE, SF and C1; with IE unmasked it changed no register, and #MF came at the next waiting instruction.
AMD’s Zen 5 guide lists FLDL2E as one macro-op on pipe FP1, one per cycle.
Forms
| Instruction | Encoding | Requires |
|---|---|---|
FLDL2E
FLDL2E | D9 EA | FPU |
Operation
What the forms do, in pseudocode that our tests run against results recorded on real processors. How to read.
if not empty(st7):
overflow()
push(INDEFINITE)
else:
push(L2E) # log2(e)
c1 = 0
c0 = c2 = c3 = UNDEFINEDRun on AMD Zen 5 (EPYC 9645): 8 cases.
Operands
| Operand | Kind | Action |
|---|---|---|
FLDL2E | ||
ST(0) | implicit | write |
x87 register stack | implicit | push |
x87 status word | implicit | write |
Flags
| C0 | C1 | C2 | C3 |
|---|---|---|---|
| All forms | |||
| U | M | U | U |
Attributes
What else XED records about the forms. How to read.
- All forms
- may abort a TSX transaction
CPUID and processors
Each form belongs to an XED ISA set. A form can be used when the processor reports every CPUID bit of one of its ISA set's alternatives.
| ISA set | CPUID | Processors |
|---|---|---|
X87 | FPU | Every x86-64 processor in XED’s model and Zen 3 to Zen 6 |
More x87 instructions
Sources
- Intel XED v2026.08.23 (commit
0bcb6237345c): forms, encodings, operands, flags, exception classes, attributes, ISA sets, CPUID bits and chip model. - The CPU data of x86-64.net: Zen 3 to Zen 6, from GCC’s
-marchfeature sets and AMD’s Processor Programming References, which the page of each processor cites. - Intel 64 and IA-32 Architectures Software Developer's Manual, Volume 2 (325383-093, September 2026): Volume 2A, page 3-354 (FLD1/
FLDL2T/ FLDL2E/ FLDPI/ FLDLG2/ FLDLN2/ FLDZ) - Intel 64 and IA-32 Architectures Software Developer's Manual, Volume 1 (253665-093, September 2026)
- AMD64 Architecture Programmer's Manual, Volume 5: 64-Bit Media and x87 Floating-Point Instructions (26569, rev. 3.16, November 2021), Wayback Machine copy of AMD's PDF
- Software Optimization Guide for the AMD Zen5 Microarchitecture (58455, rev. 1.00, August 2024), Wayback Machine copy of AMD's ZIP
The tables are derived from Intel XED, Copyright Intel Corporation, licensed under the Apache License 2.0; x86-64.net converted and reformatted them. The text is our own.