Bug 136185 - disInstr(ppc32): unhandled instruction: 0xC5ABFFFC
Summary: disInstr(ppc32): unhandled instruction: 0xC5ABFFFC
Status: RESOLVED FIXED
Alias: None
Product: SUSE Linux 10.1
Classification: openSUSE
Component: Development (show other bugs)
Version: unspecified
Hardware: PowerPC Other
: P5 - None : Normal (vote)
Target Milestone: ---
Assignee: Dirk Mueller
QA Contact: E-mail List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-11-30 21:26 UTC by Andreas Schwab
Modified: 2006-01-05 09:41 UTC (History)
0 users

See Also:
Found By: Other
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---


Attachments
valgrind log (34.57 KB, text/plain)
2005-11-30 21:27 UTC, Andreas Schwab
Details
valgrind log (42.88 KB, text/plain)
2005-11-30 21:27 UTC, Andreas Schwab
Details
valgrind log (20.16 KB, text/plain)
2005-12-01 21:25 UTC, Andreas Schwab
Details
patch (785 bytes, patch)
2005-12-01 23:27 UTC, Dirk Mueller
Details | Diff
fsqrt.diff (1.20 KB, patch)
2005-12-02 23:52 UTC, Andreas Schwab
Details | Diff
implement clock_getres (785 bytes, patch)
2005-12-05 10:10 UTC, Dirk Mueller
Details | Diff
Missing syscalls (2.08 KB, patch)
2005-12-05 21:46 UTC, Andreas Schwab
Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Andreas Schwab 2005-11-30 21:26:42 UTC
valgrind can't handle amarok.
Comment 1 Andreas Schwab 2005-11-30 21:27:16 UTC
Created attachment 59315 [details]
valgrind log
Comment 2 Andreas Schwab 2005-11-30 21:27:39 UTC
Created attachment 59316 [details]
valgrind log
Comment 3 Stephan Kulow 2005-11-30 21:39:16 UTC
hmm, we can patch out the ppc support of valgrind if the arch team has no resources to fix it
Comment 4 Dirk Mueller 2005-12-01 08:32:58 UTC
no, we want ppc support. I've asked Andreas to create this bugreport. 

Comment 6 Dirk Mueller 2005-12-01 09:07:14 UTC
it seems there was already an attempt to support this instruction: 

VEX/priv/guest-ppc32/toIR.c: 

//zz    case 0x31: // lfsu (Load Float Single with Update, PPC32 p442)
//zz       if (rA_addr == 0) {
//zz          vex_printf("dis_fp_load(PPC32)(instr,lfsu)\n");
//zz          return False;
//zz       }
//zz       DIP("lfsu fr%u,%d(r%u)\n", frD_addr, simm16, rA_addr);
//zz       assign( EA, ea_rA_simm(rA_addr, simm16) );
//zz       putFReg( frD_addr, unop(Iop_F32toF64, loadBE(Ity_F32, mkexpr(EA))) );
//zz       putIReg( rA_addr, mkexpr(EA) );
//zz       break;

Andreas, can you try with commenting in this? 

Also, use --workaround-gcc296-bugs=yes. for some reason our compiler seems to write below the stack, thats a different bug we have to fix. 
Comment 7 Dirk Mueller 2005-12-01 13:04:24 UTC
http://bugs.kde.org/show_bug.cgi?id=117419
Comment 8 Andreas Schwab 2005-12-01 21:24:18 UTC
dis_fp_arith(PPC32)(3F: opc2)
disInstr(ppc32): unhandled instruction: 0xFDA0082C
                 primary 63(0x3F), secondary 44(0x2C)
Comment 9 Andreas Schwab 2005-12-01 21:25:25 UTC
Created attachment 59625 [details]
valgrind log
Comment 10 Dirk Mueller 2005-12-01 23:26:21 UTC
hmm, thats fsqrt. unsure if its 64bit or 32bit. guessing 64bit. 
Comment 11 Dirk Mueller 2005-12-01 23:27:08 UTC
Created attachment 59645 [details]
patch

try this patch.
Comment 12 Andreas Schwab 2005-12-02 23:52:10 UTC
Created attachment 59764 [details]
fsqrt.diff
Comment 13 Dirk Mueller 2005-12-05 10:10:24 UTC
Created attachment 59793 [details]
implement clock_getres
Comment 14 Andreas Schwab 2005-12-05 16:14:05 UTC
m_syswrap/syswrap-ppc32-linux.c:1748: error: '__NR_clock_getres' undeclared here (not in a function)
Comment 15 Dirk Mueller 2005-12-05 16:21:54 UTC
ok, 

--- coregrind/vki_unistd-ppc32-linux.h  (revision 5283)
+++ coregrind/vki_unistd-ppc32-linux.h  (working copy)
@@ -276,7 +276,7 @@
 //#define __NR_timer_delete            244
 //#define __NR_clock_settime           245
 #define __NR_clock_gettime             246
-//#define __NR_clock_getres            247
+#define __NR_clock_getres              247
 //#define __NR_clock_nanosleep         248
 //#define __NR_swapcontext             249
 #define __NR_tgkill                    250
Comment 16 Andreas Schwab 2005-12-05 21:46:00 UTC
Created attachment 59869 [details]
Missing syscalls
Comment 17 Dirk Mueller 2005-12-05 22:12:05 UTC
thanks, committed to upstream
Comment 18 Dirk Mueller 2006-01-05 09:41:48 UTC
everything upstream, package updated.