Memory, caches and atomics
MOVDIR64B
MOVDIR64B copies 64 bytes from memory to a 64-byte aligned address as one direct store, the way to hand a work descriptor to a device.
- CPUID
- MOVDIR64B
CPUID.(EAX=07H,ECX=0):ECX[28] - Instructions
- 1 mnemonic
- Processors
- Tiger Lake, Alder Lake, Arrow Lake, Panther Lake, Nova Lake, Sapphire Rapids, Emerald Rapids, Granite Rapids, Diamond Rapids, Snow Ridge, Sierra Forest, Clearwater Forest, Zen 5, Zen 6
- Linux flag
movdir64b
MOVDIR64B reads 64 bytes from memory and writes them to a 64-byte aligned destination, whose address is in a register, as one direct store: the device receives all 64 bytes in one write, not combined with younger stores and not through the caches. The read need not be atomic. Drivers use it to hand a work descriptor to a device queue that one program owns; for queues that many programs share there is ENQCMD, which also reports whether the device took the command.
Instructions
| Mnemonic | Summary | Forms |
|---|---|---|
| MOVDIR64B | Copies 64 bytes from memory to a 64-byte-aligned destination as one direct store that is written atomically and bypasses the caches. | 1 |
CPUID and processors
XED splits MOVDIR64B into 1 ISA set. Software can use a form when the processor reports every CPUID bit of one alternative of the form's ISA set.
| ISA set | CPUID | Processors |
|---|---|---|
MOVDIR64B | MOVDIR64B | Tiger Lake, Alder Lake, Arrow Lake, Panther Lake, Nova Lake, Sapphire Rapids, Emerald Rapids, Granite Rapids, Diamond Rapids, Snow Ridge, Sierra Forest, Clearwater Forest, Zen 5, Zen 6 |
Sources
- Intel XED v2026.08.23 (commit
0bcb6237345c): forms, encodings, 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)
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.