[Bug 1216] New: PAGESIZE/PAGE_SIZE/sysconf/getpagesize conflicts

[rtems-bugs at rtems.org]

http://www.rtems.org/bugzilla/show_bug.cgi?id=1216

           Summary: PAGESIZE/PAGE_SIZE/sysconf/getpagesize conflicts
           Product: RTEMS
           Version: 4.7
          Platform: All
        OS/Version: RTEMS
            Status: NEW
          Severity: critical
          Priority: P3
         Component: cpukit
        AssignedTo: joel.sherrill at oarcorp.com
        ReportedBy: ralf.corsepius at rtems.org
                CC: jwillemsen at remedy.nl


http://www.rtems.org/bugzilla/show_bug.cgi?id=1215

has caused me to investigate a bit on "page size" related things in RTEMS.

The result is close to a desaster:

* limits.h contains a hardcoded #define PAGESIZE 1
* sys/params.h contains some magic to 
#define PAGE_SIZE ....
which, after some computation results into PAGE_SIZE 4096

According to POSIX PAGESIZE and PAGE_SIZE must be identical

* RTEMS's getpagesize.c contains a hard-coded 4096

* sysconf could easily return PAGE_SIZE, PAGESIZE, getpagesize() or else, but
doesn't so, ATM.

As a temporary band-aid, I am proposing to (ASAP)
+ change getpagesize() to return PAGE_SIZE
+ let sysconf( _SC_PAGESIZE|_SC_PAGE_SIZE) return PAGE_SIZE

Fixing PAGESIZE also should be put "on schedule", but doing so would require
some non-trivial surgery to newlib which needs to be tested carefully.
=> IMO, a post 4.7 release issue.

Related questions I can't answer "out-of-the-box"/without further
investigations:
- Does PAGESIZE|PAGE_SIZE have an impact on GCC?
- Does RTEMS apply page-sizes? (AFAICT: no)


-- 
Configure bugmail: http://www.rtems.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.