(2011/12/16 5:17), Dave Anderson wrote: > > > ----- Original Message ----- >> >> >> ----- Original Message ----- >>> >>> Come to think of it, this is not important for analysis very much >>> as against making possible destructive impacts at the same time. >>> So I decide to cancel this patch set at here. >> >> OK good -- that patch was way too complex for just being able display >> those additional section addresses. Plus, for example, it resulted in >> this when run on RHEL5: > >> ... [ cut ] ... > >> But I will keep your first module-section-address-display patch >> queued for crash-6.0.2. > > Hello Toshi, > > Actually, after using this patch a bit, I've found that sometimes > it can be more confusing than helpful, especially with kernel modules > that only have a small number of symbols. Plus it does things like > this, which is not particularly helpful: Hello Dave, At first, thanks for your proper arrangement. I could not care this broken impact of the symbol search outputs, it looked just making cumberers. I'm going to use "mod -gs" when I need to load section pseudo symbols. Thanks a lot, Toshi. > crash> sym -q odata > ffffffff8007f2a4 (t) .text.mark_rodata_ro > ffffffff8007f2a4 (T) mark_rodata_ro > ffffffff80274000 (r) .rodata > ffffffff80274000 (R) __start_rodata > ffffffff802ecbac (A) __end_rodata > ffffffff88004f20 (S) _MODULE_SECTION_START [.rodata] [ehci_hcd] > ffffffff88005140 (S) _MODULE_SECTION_END [.rodata] [ehci_hcd] > ffffffff88014ee0 (S) _MODULE_SECTION_START [.rodata] [ohci_hcd] > ffffffff88015100 (S) _MODULE_SECTION_END [.rodata] [ohci_hcd] > ffffffff88024160 (S) _MODULE_SECTION_START [.rodata] [uhci_hcd] > ffffffff880243e0 (S) _MODULE_SECTION_END [.rodata] [uhci_hcd] > ffffffff880382c0 (S) _MODULE_SECTION_START [.rodata] [jbd] > ffffffff8803885e (S) _MODULE_SECTION_END [.rodata] [jbd] > ffffffff8805c920 (S) _MODULE_SECTION_START [.rodata] [ext3] > ffffffff8805da02 (S) _MODULE_SECTION_END [.rodata] [ext3] > ffffffff8807fe20 (S) _MODULE_SECTION_START [.rodata] [scsi_mod] > ... > > But presumably you had your reasons for implementing it, so I'll > keep your patch in place -- but only as an option to the "mod" command. > > So in order to have module symbol lists intermingled with section > start/end address pseudo-symbols, I've added a "mod -g" flag that > can be used in conjunction with "-s" or "-S". Hopefully that works > OK for you. > > Thanks, > Dave > -- Crash-utility mailing list Crash-utility@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/crash-utility
