vstm
Vector Store Multiple (VFP)
VSTM<c><mode> <Rn>{!}, <list>
Stores multiple VFP registers to memory.
Details
Vector Store Multiple stores multiple consecutive double-precision or single-precision floating-point registers to memory using the address in a general-purpose base register. The base register can be optionally auto-incremented by the total bytes stored. Condition flags (N, Z, C, V) are unaffected unless an exception occurs.
Pseudocode Operation
address ← Rn; for i = 0 to (list_count - 1) do; [address] ← Vriestlist[i]; address ← address + (register_size / 8); end; if !(!) then Rn ← address;
Example
VSTMia r1!, {r0-r3
Encoding
Binary Layout
cond
110
0
1
D
W
0
Rn
Vd
10
10
imm8
Operands
-
Rn
First source / base general-purpose register -
list
Registers
Reference (Arm AArch32 ISA)
Instruction Forms
| Encoding | Instruction | ISA | Bit pattern | ||
|---|---|---|---|---|---|
| 0x0C800B00 | VSTM{<c>}{<q>}{.<size>} <Rn>{!}, <dreglist> | A32 | cond | 110 | 0 | 1 | D | W | 0 | Rn | Vd | 10 | 11 | imm8<7:1> | 0 | ||
| 0x0C800A00 | VSTM{<c>}{<q>}{.<size>} <Rn>{!}, <sreglist> | A32 | cond | 110 | 0 | 1 | D | W | 0 | Rn | Vd | 10 | 10 | imm8 | ||
| 0xEC800B00 | VSTM{<c>}{<q>}{.<size>} <Rn>{!}, <dreglist> | T32 | 1110110 | 0 | 1 | D | W | 0 | Rn | Vd | 10 | 11 | imm8<7:1> | 0 | ||
| 0xEC800A00 | VSTM{<c>}{<q>}{.<size>} <Rn>{!}, <sreglist> | T32 | 1110110 | 0 | 1 | D | W | 0 | Rn | Vd | 10 | 10 | imm8 |
Description
Store multiple SIMD&FP registers stores multiple registers from the Advanced SIMD and floating-point register file to consecutive memory locations using an address from a general-purpose register.
Depending on settings in the CPACR, NSACR, HCPTR, and FPEXC registers, and the Security state and PE mode in which the instruction is executed, an attempt to execute the instruction might be undefined, or trapped to Hyp mode. For more information, see Enabling Advanced SIMD and floating-point support.
Operation
if ConditionPassed() then
EncodingSpecificOperations(); CheckVFPEnabled(TRUE);
address = if add then R[n] else R[n]-imm32;
for r = 0 to regs-1
if single_regs then
MemA[address,4] = S[d+r];
address = address+4;
else
// Store as two word-aligned words in the correct order for current endianness.
if BigEndian(AccessType_ASIMD) then
MemA[address,4] = D[d+r]<63:32>;
MemA[address+4,4] = D[d+r]<31:0>;
else
MemA[address,4] = D[d+r]<31:0>;
MemA[address+4,4] = D[d+r]<63:32>;
address = address+8;
if wback then R[n] = if add then R[n]+imm32 else R[n]-imm32;