Add -+n option to disable re-testing.
Fixup -+n for throughput mode.
Fix Excel output when -+n is used.
Add support for the IBM S390 running Linux.
Cleanup naming conventions for the S390 and fixup a #define.
Add 64 bit compiles for s390x
Move BIG_ENDIAN to ZBIG_ENDIAN to avoid header conflicts.
Make random offsets always based on 48 bit random values.
Addition for make random offsets always based on 48 bit random values.
Make rands long longs.
Bug fix for 48 bit rands in bsd4_2 and Windows.
Make big_rand a long long.
Inject Erik's changes for Multi-client Windows.
Change proto version due to changes in Windows -+m support.
Add Eric to the contributors list.
Add more Windows support.
Spelling error.
Bug fixes from Erik H.
Reduce usage of shared memory.
Eliminate STUPID warning from the silly compiler.
Changes to remove warnings on BSD. Thanks to Christian Weisgerber
Support for the AMD64
Add -+k for constant aggregate data set size in throughput mode.
Add pread support for the TRU64 target. Department of Defense in Canada.
Add -+q for delay in seconds between tests.
Move variable up, GCC on Solaris was getting a bogus parse error
Add support for -+D (O_DSYNC) mode testing.
Make O_DSYNC conditional.
Add telemetry support for pread/pwrite
Improve macros and add prototypes.
Improve resolution of get_resolution().
Changes to support RedHat 9.0.
Special handling of NAME for broken frontend in Cygwin/Windows env.
Add support for the CrayX1
Remove reference to PAGE_SIZE for linux. This causes problems
with SuSe 8.
Fixup for SCO build.
Add -DHAVE_PREAD for Solaris8-64 target.
Code cleanup for Linux
Improve -+d so that each byte is more unique.
Improve byte level validation.
Provide byte level error detection with Found char and Expecting Char in
-+d mode.
Improve speed of -+d without losing uniqueness of bytes.
Fix so that Windows can use multiple processes. Needed mmap like SCO.
Use malloc() instead of mmap() for threads memory, instead of mmap.
Make CPU utilization use doubles everywhere.
Add support for CPU utilization while in distributed mode.
Make all times relative so multi node can do CPU usage.
Remove unused variables.
Zero compute_val inside of loops.
Add support for O_DIRECT for IRIX and IRIX64
Improve macros and add prototypes.
Improve resolution of get_resolution().
and portability stuff:
Added -z option. This is to be used with the -a option. It
provides more complete testing for small record sizes
when the file sizes are very large.
Added -y and -q to set record size range
Added command line to output
Put auto cross over back to 16 Meg
Add labels to the latency/offset output files.
Prevent mixed modes. Auto and throughput.
Added support for the Plus extended options.
Added support for -+u option. Cpu utilization.
Added network testing mode. -+m (Experimental) Tested: Linux, HP-UX
Added -xflag support for distributed mode.
Handle interrupts when in distributed mode.
Disable CPU utilization in distributed mode.
Add -+m cluster option to the help list and the list of options.
Enable more options in Cluster mode.
Add protocol version checking for distributed messages.
Add -+d file I/O diagnostic mode.
Fix some compiler warnings and implement the -+x option for
setting the multiplier used for file and record size incrementing.
Reduce the message traffic due to master's distribution of STOP. Only
one STOP distribution is needed. More can lead to socket buffer overflows.
Add -+p percentage read option.
Improve the mixed mode distribution algorithm.
Introduce -+r for O_RSYNC.
Add speed check code.
Increase maximum threads/procs to 256
Add contribs and -+t to help splash screen.
Disable fread and fwrite testing if mmap or async is in use.
Add pread/pwrite throughput testing
Permit smaller values for -n and -g
Make initial write in initfile() a page size request.
Stop test if file can not be written.
http://www.iozone.org/src/current/Changes.txt
I have patched the source however, to allow use of the pread/prwite
tests, via the -E extension mode. Also, the benchmark target has been fixed.
package uninstalled, with the exceptionof lmbench. Make non-batch
targets for all the benchmarks, and create a results target which
places all the results in /tmp for easier collection.
