What exactly is microcode and how does it differ from firmware

cpufirmwareterminology

Insofar as terminology, what exactly is "microcode" and if it can be updated, how does it differ from firmware?

This question is not a duplicate of this question (as far as I can tell) which I've also asked about modifying microcode. Here I'm strictly wanting to know how to use these terms properly.

Update

I have an answer selected, but I'm not particularly satisfied with it. I've engaged a lot of the answers and I find a lot of those responses equally unsatisfying. So let me present to you my two frames,

"Processor microcode is akin to processor firmware." As I read this more and more, that's how I take it. "Microcode" is in this context is just marketing over "processor firmware." Bear with me,..
Or, I'm wrong, and I know it happens! In this case, I need a much more thorough idea of why I am wrong. In that answers that I read to point to me being wrong, I am struggling understanding them,
- "Execution vs data" a lot of answers use this paradigm, but for the CPU that doesn't make much sense to me. Some assert firmware is executed, but by what? When it comes to the CPU, are programs instructions or data?
- If the firmware bridges software and hardware (read: the electrical engineering stuff of the Gods), then how is microcode not also satisifying that distinction.
- "Interpreting" as time goes on, this even makes less and less sense. What does it mean to say "hardware instructions are interpreted" with Microcode. If that was true, would something be as performant if it wasn't interpretted but precompiled to different hardware instructions and merely "executed"? Also, how is General MIDI not interpreted in the same light? It's a language that gets interpreted by "MIDI microcode" and runs on the hardware. Or, dumb terminals that interpret teleprinter instructions for visual display?
- Does "microcode" apply to the code that runs on sound cards, and video cards (GPUs)?

Best Answer

The origin of the word firmware is a mid-point between hardware and software - software embedded on hardware. It refers to software that it is stored in non-volatile memory (such as ROM, EEPROM or Flash memory) on a hardware device, and is used by the device itself.

It's becoming more common in some types of hardware for its "firmware" to be stored in driver software and loaded onto the device when it's booted/initialised, instead of leaving it permanently on the device. It's not a big deal nowadays, for example, to store a few hundred KB of firmware code in a software driver loaded onto the host OS, and to send that down to the device as it's initialised by the driver.

This is still commonly referred to as "firmware" even though it is not resident on the hardware (if you detach the hardware and put it in another system, it won't retain that version of "firmware").

Microcode is a subset of this. Microcode is not a generic term for all firmware that is loaded onto a device at boot. Instead, it's specific to CPUs, where the microcode forms the translation layer between higher level standard CPU instructions and the lower-level operations specific to that CPU. It may be loaded onto the CPU at boot, by the BIOS, and replaced later in the boot stage by the OS as well.

An update to microcode can allow a CPU's low-level behaviour to be modified to work around certain yet to be discovered bugs, without needing to replace the CPU hardware. Microcode usually contains the most efficient mapping from higher to lower level instructions as possible for the best speed and energy efficiency, so sometimes when a change to microcode is necessary to fix some bug, it can result in reduced performance.

Note that unlike Spectre, Meltdown (the vulnerability affecting only Intel chips) cannot be fixed with microcode updates alone and requires changes to core OS functionality, which may reduce performance even further.

To answer some of your specific questions since your edit:

Yes, microcode is basically firmware that runs on the processor. The special term "microcode" specifically refers to the firmware on a processor that contains the blueprint for translating from standard machine language to low level processor instructions. So it is a more specific term than firmware.

Note that as I discuss above it isn't stored on the CPU while it's off but loaded onto it each time you boot.
I don't think you're wrong. Firmware doesn't have to be written in a certain machine language and its execution doesn't have to be triggered in a certain way. At a certain low level all machine code is "data" that is "read" by a processor and interpreted in a certain way.

The term "microcode" is typically only used for main CPUs and not graphics cards or other hardware, even though those other devices may have code loaded onto them in the same way.

Related Solutions

How to recover a Linksys WRT54GL router that has a blinking green power LED and no response from the web interface

Have you tried failsafe mode. http://wiki.openwrt.org/oldwiki/openwrtdocs/hardware/linksys/wrt54gl

Check your arp cache. I found the openwrt pages useful for a similar problem.

Also note references to problems with the power supply. Everything seems ok but system doesn't come up. I swapped in an old power supply temporarily until I could order a new one.

How to find a list of all SSE instructions? What happens if a CPU doesn’t support SSE

I've saw the names of some instructions that we're added on SSE, however there's no explain about all of them (Maybe SSE4? They're not even listed on Wikipedia). Where i can read about what they do?

The best source would be straight from the people who designed the extensions: Intel. The definitive references are the Intel® 64 and IA-32 Architectures Software Developer Manuals; I would recommend that you download the combined Volumes 1 through 3C (first download link on that page). You may want to look at Vol. 1, Ch. 12 - Programming with SSE3, SSSE3, SSE4 and AESNI. To refer to specific instructions, see Vol. 2, Ch. 3-4. (Appendix B is also helpful)

How do i know which of these instructions are being used?

The instructions are only used if a program you're running actually uses them (i.e. the bytecode corresponding to the various SSE4 instructions are being called). To find out what instructions a program uses, you need to use a disassembler.

If we do know which are being used, let's say i'm doing a comparation, (This may be the stupidest question i've ever done, i don't know about assembly, though) It's possible to directly use the instruction on an assembly code? (I've been looking at this: http://asm.inightmare.org/opcodelst/index.php?op=CMP)

How does the processor interpret the instructions?

You may want to have a look at my answer to the question, "How does a CPU 'know' what commands and instructions actually mean?". When you write out assembly code by hand, to make an executable, you pass the "human readable" assembly code to an assembler, which turns the instructions into the actual 0's and 1's the processor executes.

What would happen if i have a processor without any of the SSE instructions? (I suppose if in the case we want to do a comparation, we wouldn't be able, right?)

Since your computer is Turing complete, it can execute any arbitrary mathematical function using a software algorithm if it does not have the dedicated hardware to do so. Obviously, doing intense parallel or matrix mathematics in hardware is much faster than in software (requiring many loops of instructions), so this would cause a slow-down for the end user. Depending on how the program was created, it's possible that it may require a particular instruction (i.e. one from the SSE4 set), although given it's possible to do the same thing in software (and thus useable on more processors), this practice is rare.

As an example of the above, you may recall when processors first came out with the MMX instruction set extension. Let's say we want to add two 8-element, signed 8-bit vectors together (so each vector is 64-bits, equal to a single MMX register), or in other words, A + B = C. This could be done with a single MMX instruction called paddsb. For brevity, let's say our vectors are held at memory locations A, B, and C as well. Our equivalent assembly code would be:

movq   MM0, [A]
paddsb MM0, [B]
movq   [C], MM0

However, this operation could also easily be done in software. For example, the following C code performs the equivalent operation (since a char is 8-bits wide):

#define LEN 8
char A[LEN], B[LEN], C[LEN];

/* Code to initialize vectors A and B... */

for (i = 0; i < LEN; i++)
{
    C[i] = A[i] + B[i];
}

You can probably guess how the assembly code of the above loop would look, but it's clear that it would contain significantly more instructions (as we now need a loop to handle adding the vectors), and thus, we would need to perform that many more fetches. This is similar to how the word length of a processor affects a computer's performance (the purpose of MMX/SSEx is to provide both larger registers, as well as the ability to perform the same instruction on multiple pieces of data).

Update

Best Answer

Related Solutions

How to recover a Linksys WRT54GL router that has a blinking green power LED and no response from the web interface

How to find a list of all SSE instructions? What happens if a CPU doesn’t support SSE

Related Question