Interrupts related

This is not a completely fair question since the rules of the road (definitons and functions) change regularly. Interrupts originally were hardware oriented, a very immediate need to attend to timely information electronically. Think of a clock that depends on being on time by getting its input in the form of a pulse. If the clock is so busy that it misses some of those pulses then it will gradually lose time. So for the hardware interrupts, they mandate the getting of attention while the information is still available. This could be the gathering of the latest character that came in through a modem before the next character in the stream arrives. Like the clock example, if computer is too busy, the receipt of the information from the modem will end up missing a few characters. Now picture this as coming up from below, the CPU actually changing the address of the next instruction all on its own.

The opposite effect is top-down, where a program decides it needs to interfere with regular processing and hurry on to process something quickly. This can be something like sending anther character to the modem (or monitor etc).

The nature of both is to ensure that no other process will interfere with the exact sequence of code until permitted, which should be promptly after performing a 'save state' operation, which in Assembly Language is essentially preserving the content of the registers that the program intends to use, and any other data item needed. Once that is done then it will permit other processes to interrupt the rest of the routine. One key ingredient here is the use of stacks.

dimitry > Let's use word "services" instead of interrupts.

:-))
Good idea, they are not much more than subroutines, some of which call other subroutines (a waste). Just not to be confused with the Services concept used by XP

> BIOS "interrupts" provide basic Input / Output services.

By definition (I/O is input/output)

bbk_doll > DOS "interrupts" provide OS services.

OK

> What is difference between DOS Interrupts and BIOS Interrupts?

It is actually not all that simple, if we ignore Windows an address older DOS, which was actually further fractionalized, with maybe four such categories, which varied by version level.

There are programs on the BIOS chip. There are also two files used by the OS that have code. One can be called something like IO.SYS or BIOS.SYS etc, and the other as DOS.SYS or MSDOS.SYS etc. Those may be of what you refer to in the question. But their content changed over time (version). In theory these three should have been layered, from bottom up in the order just presented, but MS had ever had problem following rules and organizing in layers. One file eventually became text based (not binary code).

Another issue was whether DOS was resident or not. Also revised from version to  version, when DOS would start up another program, to make more memory available it would remove much of its own code from RAM. In theory this ought to be the highest level parts of DOS, such as interpretting whether or not the user is requesting the processing of a 'DIR' command. That would not be needed any longer since the input from keyboard is transferred to the new program. I am not sure that this applies to the question as asked, but it is key part of same process, such as code layering.

> Please explain 3-4 differences.

Sounds like attempt for a question in a classroom setting. In review, I can only now think of one, like dimitry put it:

> DOS services have file access but BIOS services provide just raw disk

The layering idea of high to low level.

> differences if any between BIOS and DOS interrupts like their calling, working, Interrupt Vector Table etc.

Well, another difference could possibly be the direction of the call, whether from low to high or from high to low. For low-level access, the vector table has to immediately get to a routine to preserve data. With the ones for high to low calling, the DOS ones require user to preserve their own registers and data before calling the interrupt. That helps to minimize the duration of the time that interrupts are prohibited.

Maybe a third difference could be that BIOS is always self contained code, while DOS may be code that call other code that calls other code, including capability of one SW interrupt to invoke another. That is not done for low level IO, which does not take time4 to repetetively save the state of the machine.

How about a 4th difference being that all of the BIOS must be constantly resident, but some of DOS can be transient (either memory resident or not, depending on circumstance).