openssh bugs - Apr 2010 - [Bug 1753] New: Use -funroll-loops with umac.c

https://bugzilla.mindrot.org/show_bug.cgi?id=1753

           Summary: Use -funroll-loops with umac.c
           Product: Portable OpenSSH
           Version: -current
          Platform: Itanium
        OS/Version: Other
            Status: NEW
          Severity: enhancement
          Priority: P2
         Component: Build system
        AssignedTo: unassigned-bugs at mindrot.org
        ReportedBy: imorgan at nas.nasa.gov


By default, umac.c is compiled with -O2 and performs well on x86 and
x86_64 architectures. However, on other architectures the performance
can be improved by adding -funroll-loops.

Using Ted Krovetz's original code, the performance for 1KB blocks on
various architectures (clocks per byte) is as follows:

        gcc -O2         gcc -O2 -funroll-loops
x86_64: 0.95            1.04
IA64:   2.31            1.36    -funroll-loops
SPARC:  9.52            9.50
POWER5: 3.88            3.67

The architecture that benefits the most from this is IA64. A
memory-to-mekory test using ssh on a 1.5 GHz Itanium system shows an
improvement of approximately 9 MB/s; 128 MB/s with just -O2 and 137
MB/s
when -funroll-loops is added.

It may be worthwhile adding the following to Makefile.in:

umac.o: umac.c
        $(CC) $(CFLAGS) -funroll-loops $(CPPFLAGS) -c $?

Admittedly, it has to be acknowledged that this would be slightly
detrimental to x86_64 and for architectures other than IA64 the benefit
appears to be marginal.

-- 
Configure bugmail: https://bugzilla.mindrot.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.

https://bugzilla.mindrot.org/show_bug.cgi?id=1753

Darren Tucker <dtucker at zip.com.au> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dtucker at zip.com.au

--- Comment #1 from Darren Tucker <dtucker at zip.com.au> 2010-04-10
10:25:36 EST ---
(In reply to comment #0)
> umac.o: umac.c
>         $(CC) $(CFLAGS) -funroll-loops $(CPPFLAGS) -c $?
Well the first problem with that not all compilers understand
-funroll-loops, so it'll stop it compiling in anything that's not gcc
(or pretending to be gcc).

Can you just do "./configure --with-cflags=-funroll-loops"?  or does
that have a detrimental impact elsewhere?

-- 
Configure bugmail: https://bugzilla.mindrot.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
You are watching someone on the CC list of the bug.

https://bugzilla.mindrot.org/show_bug.cgi?id=1753

--- Comment #2 from Iain Morgan <imorgan at nas.nasa.gov> 2010-04-10
11:22:29 EST ---
Ah, yes, I hadn't considered the (lack of) portability with
-funroll loops. I haven't seen any detrimental effects adding it for
entire build process, but thought it might be better to take a more
surgical approach. Using --with-cflags should do the trick.

I've been using this performance tweak on Itanium for the past two
and thought that it was about time to pass on the observation to the
community. I have to admit that I almost didn't submit this bug when I
saw the marginal benefit on other architectures.

-- 
Configure bugmail: https://bugzilla.mindrot.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching the assignee of the bug.
You are watching someone on the CC list of the bug.