<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD> <META http-equiv=Content-Type content="text/html; charset=us-ascii"> <META content="MSHTML 6.00.5730.11" name=GENERATOR></HEAD> <BODY> <DIV dir=ltr align=left>Oops, I sent the wrong patch... </DIV> <DIV dir=ltr align=left> </DIV> <DIV dir=ltr align=left>I had found the second problem when reading </DIV> <DIV dir=ltr align=left>through the patch, but somehow grabbed the wrong file. The 'mod -l' thing</DIV> <DIV dir=ltr align=left>seems like it'll take more to fix, but there's another problem which I found and </DIV> <DIV dir=ltr align=left>had fixed.</DIV> <DIV dir=ltr align=left> </DIV> <DIV dir=ltr align=left>Sections which are not '.text' '.data' or '.bss' will get loaded at the wrong</DIV> <DIV dir=ltr align=left>addresses, which is unfortunate. The following patch addresses this</DIV> <DIV dir=ltr align=left>by providing addreses for each section.</DIV> <DIV dir=ltr align=left> </DIV> <DIV dir=ltr align=left>Happy New Year!</DIV> <DIV dir=ltr align=left> </DIV> <DIV dir=ltr align=left> -castor</DIV> <DIV dir=ltr align=left> </DIV> <DIV class=OutlookMessageHeader lang=en-us dir=ltr align=left> <HR tabIndex=-1> From: crash-utility-bounces@redhat.com [mailto:crash-utility-bounces@redhat.com] On Behalf Of Dave Anderson Sent: Tuesday, January 02, 2007 1:04 PM To: Discussion list for crash utility usage, maintenance and development Subject: Re: [Crash-utility] crash-4.0-3.14-sym.3.patch (was: modules and data/bss initialization) </DIV> <DIV></DIV>Castor Fu wrote: <BLOCKQUOTE TYPE="CITE"> <DIV dir=ltr>Hi Dave!</DIV> <DIV dir=ltr></DIV> <DIV dir=ltr>Sorry about the regression.</DIV> <DIV dir=ltr></DIV> <DIV dir=ltr>I looked further through the code, and have modified it to only add the symbols</DIV> <DIV dir=ltr>for which it has actually identified a matching section, and to do it in a way</DIV> <DIV dir=ltr>which supports whatever sections are defined.</DIV>This doesn't explain why a match isn't being found in the ia64 case, but shouldat least avoid the regression in the s390x case.I've also attached a test which one could run which dumps output which mighthelp me fix things by tracing through loading the 'md5' module. By default the new code is off, as requested, and can be forced witha '-l' option to 'mod', e.g.crash> mod -l -s md5Thanks again for the testing, and hopefully this will work for most people.It would be great if someone could send me output from the rarerplatforms like ppc64 or 390 if it doesn't work.Happy New Year! -castor</BLOCKQUOTE> <TT>Somebody forgot to QA this patch... ;-)</TT><TT></TT> <TT>Unfortunately, "mod -l -s module-name" won't work, nor will</TT> <TT>"mod -l -S", because of the unique way that cmd_mod() handles</TT> <TT>-s and -S -- which can optionally take additional "objfile" or "directory"</TT> <TT>arguments respectively:</TT><TT></TT> <TT>crash> mod -l -S</TT> <TT>mod: -S is not a directory</TT> <TT>Usage: mod [ -s module [objfile] | -d module | -S [directory] | -D | -r ]</TT> <TT>Enter "help mod" for details.</TT> <TT>crash> mod -l -s nfs</TT> <TT>mod: /usr/lib/debug/lib/modules/2.4.21-37.ELsmp/kernel/fs/nfs: not an ELF format object file</TT> <TT>Usage: mod [ -s module [objfile] | -d module | -S [directory] | -D | -r ]</TT> <TT>Enter "help mod" for details.</TT> <TT>crash></TT><TT></TT> <TT>So to get around that, you can just enter "mod -l" alone</TT> <TT>prior to any -s or -S commands, which will just set the</TT> <TT>flag.</TT><TT></TT> <TT>But having done that, it actually does just the opposite,</TT> <TT>because of backwards logic here:</TT><TT></TT> <TT>@@ -7288,7 +7410,10 @@</TT> <TT> strcpy(lm->mod_namelist, namelist);</TT> <TT> else</TT> <TT> strncpy(lm->mod_namelist, namelist, MAX_MOD_NAMELIST-1);</TT> <TT>- goto add_symbols;</TT> <TT>+ if (USE_V2_MOD_SYM()) {</TT> <TT>+ goto add_symbols;</TT> <TT>+ }</TT> <TT>+</TT> <TT> }</TT><TT></TT> <TT> if ((mbfd = bfd_openr(namelist, NULL)) == NULL)</TT><TT></TT> <TT>It should be: "if (!USE_V2_MOD_SYM())".</TT><TT></TT> <TT>Anyway, don't bother updating the patch for the above,</TT> <TT>because I can keep testing with the two work-arounds.</TT><TT></TT> <TT>I'd also appreciate any ppc64, s390 and s390x testers</TT> <TT>out there...</TT><TT></TT> <TT>Thanks,</TT> <TT> Dave</TT> </BODY></HTML>