Question about the io_context of I/O scheduler

[Roy Lee]

Hi,

I'm currently tracing the I/O path of the kernel (2.6.19 release)

for the I/O scheduler, as far I as know, io_context is currently based on
per-process basis.
so, when an user process issued an I/O, the request generated in the kernel 
will be associated to the io_context of the "current" process, right?

but, isn't there some request generated in the kernel code which is not
relevant to the I/O pattern of the "current" user process?
for example, when the request generated to handle the page fault
or the kernel is paging out other process' pages due to memory pressure.

is this an issue or did I miss something?


thanks in advance for clearifying this.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/