What is the standard microkernel?
Cut Hour <[email protected]> Mon, 9 Jan 2023 21:49:54 +0900
| Newsgroups | gmane.os.hurd.l4 |
|---|---|
| Message-ID | <[email protected]> |
Hello. Recently I saw a C++ code which implements the recursive descent parser. The code uses own stack variable and own loop. It actually says "the function call is the overhead." I knew Apache web server says "fork() is the overhead." And I thought that the goals of this kind of things must be the dedicated machine, the dedicated OS, or the dedicated language. I came up with an idea that "How about the Apache directly on microkernel." I checked microkernels and I see (the IPC of) Mach was a failure. I also checked Linux on L4 shows only a few percent performance degradation. I think this is a good chance and direction for microkernels. So what's the GNU's plan?