NonStop S-Series Server Description Manual (G06.27+)

Instruction Processing Environments

HP NonStop S-Series Server Description Manual—520331-004

5-6

Code and Data Allocations

As explained in Section 4, Memory Addressing and Access, each process has its own

separate allocation of user space in virtual memory. Within that allocation, segments

for code are assigned in the last few regions, and segments for data are assigned in

several lower address regions. The code segments contain program instructions and

constants in one or more of three possible forms (native code, Accelerator-generated

RISC code, or TNS code), and, for accelerated programs, mapping tables. The data

segments contain program variables. Each process also has a logical segment called

the process file segment (PFS), containing operating system data for the process.

The code segments of a process can be thought of as read-only storage, because no

user code instructions can write into them. Accordingly, because code segments

cannot be modified, they can be shared by any number of processes. Information in

the data segments can be both read and written, using load and store instructions—

except for those data segments that are declared to be read-only.

Every process has at least two data segments: the privileged stack segment and the

process file segment. If the process is a TNS process or if it is unprivileged, it also has

a main stack segment. A TNS process also always has a TNS user data segment (for

TNS stack and global data). A native process also has a global data segment.

Figure 5-3 shows a simple example of three processes (A, B, and C), each executing

its own set of code segments and accumulating data in its own private data segment

(correspondingly labeled).

The process A is executing instructions fetched from code segment A, which is located

near the end of the user space in virtual memory. The process is loading data from

stack segment A, operating on that data in processor registers, and storing the

resulting data back to the data stack segment. Processes B and C, not currently in

execution, have their own code segment sets and their own data stack segments.