Sophie

Sophie

distrib > Mandriva > 2010.0 > i586 > media > contrib-release > by-pkgid > d5fac68fcd1f96abc531b332ec12db94 > files > 8

mpt-status-1.2.0-4mdv2010.0.i586.rpm

2006-11-01, ratz [ratz@drugphish.ch]:
	o man/mpt-status.8
		- merged contributions from Jean Delvare
		- updated with missing sections
	o doc/*
		- merged typo/spelling fixes from Jean Delvare
		- tidy up 64bit information since this problem is solved now
	o mpt-status.c
		- removed MPT_STATUS_DEBUG, since this was an unfinished
		  duplicate version of --debug.

2006-10-26, ratz [ratz@drugphish.ch]:
	o man/{README,mpt-status.8}
		- added man page written by Steffen Joeris for the Debian
		  project.
	o Makefile
		- integrated preliminary support for the man page installation
		  and uninstallation through make install, resp. make uninstall
		- fixed once again the KERNEL_PATH stuff after Jean Delvare
		  from Suse mentioned that it broke Suse builds. This time it
		  should work for both, Suse and Redhat (Fedora).
	o mpt-status.c
		- added detection if no RAID is configured on HBA. This only
		  works with the "--newstyle" or "-n" option for safety reasons
		  and has not yet been tested.

2006-10-25, ratz [ratz@drugphish.ch]:
	o Makefile
		- fixed brokenness due to previously applied patch from Rich :)
		  The point is that LSI-related headers are in the source and
		  not the include directory of the build symlink.
	o mpt-status.c
		- Suse packages are compiled with additional flags to detect
		  possibly problematic code. The checker complained about a call
		  to system() with unchecked return value in mpt-status. Applied
		  patch from Jean Delvare.

2006-06-29, ratz [ratz@drugphish.ch]:
	o Makefile
		- applied patch from Rich Edelman to set KERNEL_PATH more
		  intelligently 
	o contrib/contrib/mpt-status.spec
		- generic spec file from Rich Edelman with patch support
		- changed quite a lot again actually regarding wording
		  and also removed the patch section
	o contrib/*diff
		- remove obsolete diff from tree regarding sanitation

2006-04-20-
2006-05-02, ratz [ratz@drugphish.ch]:
	* Version: 1.2.0-RC7 (not released), after 3 almost complete rewrites
		   too many changes happened in between the last two weeks to
		   list them properly. Kernel header sanitizing might not be
		   possible after all.
	o mpt-status.c
		- proper support for 64-bit nodes
		- added support for spare disk information
		- added preliminary support for the sync percentage information
	o doc/TODO
		- updated

2006-04-20, ratz [ratz@drugphish.ch]:
	o mpt-status.c
		- reworked physical disk section
		- integrated the exit status patch submitted by Leandro Santi

2006-04-16, ratz [ratz@drugphish.ch]:
	o mpt-status.c
		- merged Steffen Joeris patch to support integrating mpt-status
		  into the KFreeBSD project.

2006-03-09, ratz [ratz@drugphish.ch]:
	o mpt-status.c
		- Martin Hamant reported an issue regarding the IOC unit, which
		  did not work for more than 1 controller. Fix it, by setting
		  ioc_unit as part of the address unit passed over to readpage.

2006-03-06, ratz [ratz@drugphish.ch]:
	o mpt-status.c
		- removed SGE debugging lines
		- added checkForLibraryMode() which reads the __MPT_STATUS_LIB
		  environment variable and if set will initialize the library
		  functionality of mpt-status.
		- changed the output to please Martin Hamant and to also ease
		  up the scriptability.

2006-03-05, ratz [ratz@drugphish.ch]:
	o doc/Changelog
		- fixed overly long line
	o mpt-status.c
		- added human readable output for IM, IME, IS, which is nothing
		  else than RAID0, RAID1E and RAID1.

2006-03-04, ratz [ratz@drugphish.ch]:
	* Version: 1.1.5 [released] (Codename: Sebnem)
	o doc/README
		- major overhaul regarding distribution availability and
		  installation process
	o doc/FAQ
		- added a possible FAQ document containing parts of an email
		  I wrote to explain the output of mpt-status
	o doc/ReleaseNotes
		- wrote Problems, Upgrade, Outlook sections
	o incl/sles.h
		- added preliminary work to kernel header sanitation done by
		  SuSE
	o mpt-status.c
		- made the struct divisior 32/64 bit safe
		- removed irritating compile warning for type signedness since
		  too many people complained about it and before the kernel
		  headers are not sanitized, there is no point in trying to
		  address this deficiency
		- preliminary work done for IOCpage5_t support (hotspare)
		- fixed missing signedness casts
	o contrib/sizes.c
		- added a little tool to check the machine's type sizes
	o contrib/*.spec*
		- added various untested spec files

2006-02-11, ratz [ratz@drugphish.ch]:
	o Makefile
		- Added support for DESTDIR, patch kindly submitted by Petter
		  Reinholdtsen

2006-01-22, ratz [ratz@drugphish.ch]:
	o mpt-status.{ch}
		- s/log_id/vol_id/ in the first line if -s is submitted
		- added --probe_id which will probe for the first scsi id
		- fixed a long standing bug resulting in a coredump when using
		  the -d parameter
		- moved print_status_only, id_of_primary_device, ioc_unit into
		  global context, so there is no need to pass those variables
		  through the stack all the time
		- reorganized long_options and usage to be in alphabetical
		  order
		- fixed read_page to actually work on LSI-SAS1064 controllers
	o doc/*
		- major cleanup and improved documentation

2005-11-14, ratz [ratz@drugphish.ch]:
	o mpt-status.c
		- removed Nietsche, added "--autoload" which will try to exec
		  /sbin/modprobe mptctl.

2005-11-12, ratz [ratz@drugphish.ch]:
	* Version: 1.1.4 [released]

2005-11-11, ratz [ratz@drugphish.ch]:
	o doc/DeveloperNotes
		- Added information from the mpt(4) man page of OpenBSD
	o mpt-status.h
		- Fixed VERSION string
	o mpt-status.c
		- Merged patch by Thomas Richer to allow setting the ioc unit
		- Added Copyright note
		- Removed the modprobe call since this is braindead. Not all
		  people have mpt loaded as module. Making it easy to use does
		  not mean making it dumb.
		- Rewrote the primitive parameter passing on command line to
		  more sophisticated GNU style option parsing.
		- Added some more switches, check them out with --help
		  The --nietzsche_kills switch is only so I won't forget how
		  to use long options without short options.
		- Reorganized include headers to have a legitimate C code ;).
		  --> Check news://comp.std.c for some crazy stuff on C
		- Style cleanup
		- Function prototypes are static now, since we're using it only
		  in one context.
		- Added the '-v' parameter which can be used after a Houston
		  problem to get some more output on the warnings.
		- Added the '-q' flag to really quiet down things.

2005-07-16, ratz [ratz@drugphish.ch]:
	o Makefile
		- Incorporated patch to auto-detect redhat, done by Matti
		  Hiljanen
	o doc/INSTALL
		- Removed RH specific entry, since RH is now auto-detected
	o doc/THANKS
		- Matti earned his stars ;)
	o contrib/mpt-status.spec
		- Added preliminary support for RPM building, work done by
		  Jean-Philippe Civade.

2005-07-15, ratz [ratz@drugphish.ch]:
	* Version: 1.1.3 [released]
	o mpt-status.h
		- incorporated compile fixes for redhat based distros from
		  Jean-Philippe Civade. Added the REDHAT define because only
		  redhat distros seem to expose the undefined __user and
		  __kernel problem. The impact on the correct ioctl passing
		  is unknown. I should sanitize the LSI related headers and
		  create user space headers. It's not like they change their
		  API like their underwear.
	o Makefile
		- added DFLAGS for redhat parameter passing
		- fixed install target, bug found by Jean-Philippe Civade
		- fixed uninstall target
		- added preliminary support for compiling on 64bit platforms.
		  Tested only on sparc64.
	o doc/README
		- proper split between README and INSTALL
	o doc/INSTALL
		- added redhat related compilation instructions
		- added "howto install redhat kernel sources rpm" document
		  written by Jean-Philippe Civade.
		- explained the install and uninstall process
	o doc/THANKS
		- initial import: this is to thank all the people contributing
		  to this software.
	o doc/ReleaseNotes
		- initial import: I consider them important, even though it's
		  just a very simple tool ... so far ;)

2005-06-16, ratz [ratz@drugphish.ch]:
	* Version: 1.1.3
	o doc/Changelog
		- corrected date format
	o mpt-status.h
		- added <linux/compiler.h> for better 2.6.x compilation
		  support
		- Lindent'ed
	o mpt-status.c
		- Lindent'ed
		- added comment regarding compilation warning

2005-06-09, ratz [ratz@drugphish.ch]:
	* Version: 1.1.2
	o mpt-status.c
		- removed some local headers and put into mpt-status.h
	o mpt-status.h
		- added local headers
	o Makefile
		- introduced KERNEL_PATH which can be submitted to the make
		  command, so no more editing is needed. Use as follows:

			make KERNEL_PATH=/path/to/kernel/source

		- introduced PREFIX which is a pseudo prefix semantics as seen
		  with autoconf tools for install and uninstall targets
		- added install and uninstall targets.

	o README -> doc/README
		- moved to doc folder where all documentation should reside
	o doc/INSTALL
		- added to shortly document the installation procedure

2005-05-19, ratz [ratz@tac.ch]:
	* Version: 1.1.1
	o incl
		- created directory and moved pci-utils related header files
	o Makefile
		- adapted to new header location
		- added distclean target which equals to the clean target
	o README
		- tidy up to be presentable
	o Changelog
		- started primitive changelog
	o TODO
		- added some todos I've also sent Matt
	o AUTHORS
		- created and added my name ;).

2005-05-18, ratz [ratz@tac.ch]:
	* Version: 1.1.0
	o mpt-status.c
		- improved output handling
		- documented scsi id issue for address parameter in read_page()
		- made main() ISO C91 standard compliant
		- switched position of state versus flags for easy shell parsing
		- added -i option to specify scsi id of primary disk of array
		- added -h option for little help
		- added UID=0 check
		- added VERSION (1.1.0) info
		- added print_usage(), print_version()
		- added scsi_id output for physical disk status output, so you
		  at least know which disk to rip out of the chassis once it
		  failed.
		- s/exit(1)/exit(EXIT_FAILURE)/g
		- s/bzero/memset/g and of course the 3rd argument
		- s/return 0/exit(EXIT_SUCCESS)/
		- s/vol/vol_id/g
	o pci.h, config.h, header.h
		- added from pci-utils package
	o README
		- started README with some unimportant information
	o Makefile
		- added LDFLAGS for static linking
		- added CC = gcc to be able to compile it under uclibc
		- added -W -O for more warnings