Hi, Here's the first batch of outstanding patches/issues. Please let me know if you'd prefer a different format or type of URL/pointer.
Thank you very much for your review and comments! :-) Carol -------------- 1) Small FREEBSD patch from Dmitry Frolov sent in on 1/25/07 (search page for BSD) - Dmitry, did you still want to add this patch in? http://sourceforge.net/mailarchive/forum.php?forum_name=ipmitool-devel&max_rows=25&style=nested&viewmonth=200701 My comments: looks fine to me. 2) John Blair patches 2/23/07 -- John, did you still want to commit the CR6480372.patch (ipmi_sel.c) for the BSWAP_32(itime) call? http://sourceforge.net/mailarchive/message.php?msg_name=1172213608.9272.20.camel%40localhost.localdomain My comments: BSWAP_32(itime) patch doesn't apply cleanly now but assume it's to the call in ipmi_sel_set_time() to make it the same as the similar call in ipmi_sel_get_time()? That patch looks fine to me. The other patch to LOG_LOCAL4 I think we'd decided against. There were subsequent discussions to perhaps make LOG_DAEMON a run-time config option instead. 3) spec file issue. Discussions on 3/12/07 and 3/13/07. ./contrib dir removed from rpm build. Carol H. put oem_ibm_sel_map file into datadir/ipmitool as per Arkadiusz's suggestion. http://sourceforge.net/mailarchive/forum.php?forum_name=ipmitool-devel&max_rows=25&style=nested&viewmonth=200703&viewday=13 My comments: the datadir/ipmitool approach is fine with me if this works for everyone else. 4) Dan Aloni patch to fix compilation warnings on 3/12/07 http://sourceforge.net/mailarchive/message.php?msg_name=20070312113020.GA29672%40localdomain My comments: Yes, Please! :-) Most of these have never been fixed. The only fix in the patch I'm wondering about is the one to ipmi_picmg_portstate: should we be printing out d->type instead of removing it as an extraneous parameter? 5) James Pearson and Brian Selecki discovered an issue in ipmi_sdr.c traced back to code that was removed in 1.8.9. Discussion and code to be added back found on 4/19/07. http://sourceforge.net/mailarchive/message.php?msg_name=46273906.4000108%40moving-picture.com Subsequent emails on 4/27/07, 4/30/07, and 6/21/07 discuss the origin of the change (Jordan Hargrave 10/25/06) and whether or not to add it back in. James and Brian very much want it added back in. Don't believe we heard back from jharg on whether it can be added back in (or maybe I missed that email)? It has not yet been added back in. http://sourceforge.net/mailarchive/message.php?msg_name=467AAC07.5050601%40moving-picture.com My comments: IMHO, if we don't hear back from jharg or others on why it shouldn't be added back in, we should go ahead and do it. 6) Brian Selecki - fwrite-ing NULL pointers returned from ipmi_sdr_get_record. 4/18/07 and 4/26/07. http://sourceforge.net/mailarchive/message.php?msg_name=20070426115231.R5453%40arbitor.digitalfreaks.org Yes, please include the actual patch for review. BTW: although you've included a handy safety tip for folks that is particularly useful at this time of the year, maybe we should change the LOG_ERR message. Hee hee ;) 7) Jim Sievert sensor issue/possible fix on 4/12/07: http://sourceforge.net/mailarchive/message.php?msg_name=000b01c77cb2% 24f310b5d0%24567011ac%40na.uis.unisys.com My comments: What Jim is suggesting adds a bit more latency but looks like it should work ok in theory -- except that it looks like there's a bug in ipmi_sdr_get_sensor_reading_ipmb() which would probably break other folks (and which seems to have been around awhile): In ipmi_sdr_get_sensor_reading_ipmb(), shouldn't if (strcpy(intf->name, "ipmb")) return ipmi_sdr_get_sensor_reading(intf, sensor); be if ((strncmp(intf->name, "ipmb")) != 0) return ipmi_sdr_get_sensor_reading(intf, sensor); or at least some form of strcmp? It seems like the strcpy should always be true (and will set the interface name to ipmb). 8) ipmi Watchdog discussions starting on 5/24/07. Although we recently added support to get the current watchdog timer state, to shut the timer off, and to reset the timer, as per discussions last year (and again recently), some machines lack the (mandatory) chassis power commands and so a way to alternately make use of the IPMI watchdog reset capability would be of great help (at least until such a time as the deficient firmware had the chassis power commands added back. Anyway, I wanted to ask again if anyone needs to have something along the lines of "ipmitool mc watchdog powercycle" added to this release? 9) Marco Framba's Kontron OEM No Timestamp SEL patch to add OEM strings to the SEL. 5/23/07 http://sourceforge.net/mailarchive/message.php?msg_name=34BDD2A93E5FD84594BB230DD6C18EA20182BC68%40nuova-ex1.hq.nuovaimpresa.com My Comments: IMHO there needs to be man page patch added to go with this patch. Other than that it seems fine to me. ----- More patches/issues to follow... stay tuned... :-) ------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php _______________________________________________ Ipmitool-devel mailing list Ipmitool-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ipmitool-devel
