> - Can address 17.5 terabytes of memory, 9TB files
In theory, so can linux (on the alpha and ultra)
> - Arithmetic computations get to use 64-bit registers and operations
So can linux
> - UFS logging (mount -o logging /dev/dsk/c0t0d0s0 /mnt)
linux 2.3.x will probably have this
> - mount "noatime" support
something linux has had for eons
> - TCP performance enhancement with SACK, RFC 2018
something linux has had for a few months now
> - New crash dump features
something we don't have - and might be useful (I know we've had this
discussion a long time back when Linus said he didn't find them
useful or something)
> - BIND 8.1.2
so does linux (most recent. distrubutions anyhow)
> - Sendmail 8.9.1b
go it too
> - improved poll(2) system call (used in many applications)
not sure what imporved means, but poll(2) we got and have had for a
while now
> - increase in number of filedescriptors
actually, this is a killer for 2.1.x / 2.2.x - Bill Hawes did some
stuff to make this number dynamic, but I could never get it working
properly with gobs of FDs. We need gobs of FDs, really we do.
> - Full Unicode 2.1 support
dont have it
> - Better threading subsystem
we don't have lwp of threads, 2.3.x maybe
> - Lots of new CDE apps/improvements
gnome
> - Netscape 4.05
got it
> - directory name lookup cache optimized and now dynamically allocated on
> demand (static before)
umm... got this I think
> - significantly improved paging algorithim, not "on" by default
> add the line: set priority_paging = 1 to /etc/system, and reboot
> - Note, backported to 2.6 kernel patch "-09"
> http://www.sun.com/sun-on-net/performance/priority_paging.html
> IMPORTANT NOTE: Ensure that data files do not have the executable
> bit set. This can fool the VM into thinking that these are really
> executables, and will not engage priority paging on these files.
no idea how we stack up here... last time I did any comparisons Sunos
4.1.3U2 seemed to page _much_ better than linux on the same box, but
I think this was a hacked 2.0.x or early 2.1.x kernel...
> - some new commands like
> plimit(1): Set or get resource limits of a process by taking a
> process ID (PID) as an argument (limit and ulimit only work with the
> current process)
don't have this... but it wouldn't be hard to add
> pgrep(1): Find a process by name and print the PID
pidof (had for a long time)
> pkill(1): Find a process by name and send it a signal specified on
> the command line
kill `pidof process`
?
a 2 line script maybe...
> traceroute
had it for eons
> - File Access Control Lists "facls" (man setfacl) added with Solaris 2.5.1
> are damn usefull.
another linux 2.3.x thing (maybe?)
I guess I'm just not easily impressed...
-cw
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/