Closed Bug 37841 Opened 24 years ago Closed 24 years ago

Quitting after using Java crashes Mozilla

Categories

(Core Graveyard :: Java: OJI, defect, P3)

Product:
Core Graveyard
Component:
Java: OJI
Platform:
PowerPC
All
Type:
defect

Tracking

(Not tracked)

Status:
VERIFIED DUPLICATE of bug 36381

People

(Reporter: mozilla, Assigned: drapeau)

References

Details

(Keywords: crash, Whiteboard: [dogfood-] no one available to resolve this :-() 

Build ID: 2000050111

Description: If you open a page that has Java on it, and you then quit Mozilla, 
Mozilla crashes.

Reproducability: Every Time

Steps to Reproduce:
	1. Open a page with Java.
	2. Quit Mozilla.

Actual Results: It crashes.

Expected Results: It shouldn't.

Additional Information:

Partial stdlog. If you want the whole thing, let me know

PowerPC unmapped memory exception at 14AB0668 sysThreadExit+00560
 
  1-May-2000 6:20:05 PM (since boot = 2 days, 1 hour, 33 minutes)
  Current application is “Mozilla”
  Machine = #406 (NewWorldMac), System $0904, sysu = $01008000
  ROM version $077D, $45F6, $0001 (ROMBase $FFC00000)
  VM is on; paging is currently safe (and it probably isn't VM's fault)
  NIL^ = $FFC10000
  Stack space used = +5078236
 Address 14AB0668 is in VM file-mapped logical memory space
 It is in the CFM fragment “AppleJava” at 14A7D460
 It is 00033208 bytes from the start of the fragment
  and 0002B6D8 bytes into a non-writeable code section at 14A84F90
 PowerPC 7400 (G4) Registers
                         CR0  CR1  CR2  CR3  CR4  CR5  CR6  CR7
  PC  = 14AB0668     CR  0100 0100 0000 0000 0000 0010 0100 0100
  LR  = 14AB0608         <>=O XEVO
  CTR = 14AD767C
  MSR = 00000000         SOC Compare Count
  Int = 0            XER 000   00     03                     MQ  = 00000000
  
  R0  = 00000000     R8  = 00000009      R16 = 000000E2      R24 = 04D9D880
  SP  = 07B41B50     R9  = 06C87C64      R17 = 06C988E4      R25 = 04D9EB18
  TOC = 04D9A230     R10 = 00000002      R18 = 04D9D7B8      R26 = 04D9D890
  R3  = 04D9D7B8     R11 = 00000401      R19 = 04D9D7B8      R27 = 04D9D720
  R4  = 00009110     R12 = 04D95878      R20 = 04D9D7B8      R28 = 04D9D8A0
  R5  = 00000000     R13 = 04D98C7C      R21 = 04D9ECE0      R29 = 04D9ED08
  R6  = 7D244B78     R14 = 04D93564      R22 = 04D9D898      R30 = 04D9D7B0
  R7  = 00000001     R15 = 06C76B58      R23 = 04D9D554      R31 = 04D9D7A0


 (CurStackBase does not seem to apply...dumping 4K.)
  Calling chain using A6/R1 links
  Back chain  ISA  Caller
  07B433AB    PPC  0047249C  EmToNatEndMoveParams+00014
  07B43340    PPC  FFCF7778  FragRelease+0014C
  07B431C0    PPC  FFCED8AC  GetIndSymbol+006E0
  07B43170    PPC  FFCEE43C  GetIndSymbol+01270
  07B43120    PPC  FFCEF708  GetIndSymbol+0253C
  07B430B0    PPC  160C951C  MRJPlugin__terminate+0004C
  07B43070    PPC  160CB02C  MRJPlugin::Release()+00018
  07B43030    PPC  160D38C8  SupportsMixin::release()+0003C
  07B42FF0    PPC  160C9728  MRJPlugin::~MRJPlugin()+0009C
  07B42FB0    PPC  160CFF34  MRJSession::~MRJSession()+00078
  07B42F70    PPC  14B88F68  JMCloseSession+0008C
  07B42F30    PPC  14AA737C  ShutdownMacJava+00050
  07B42EF0    PPC  14AA6CE4  dprintf+001D0
  07B42EA0    PPC  14AB64F8  jni_AddRefCell+04254
  07B42E50    PPC  14AB31C8  jni_AddRefCell+00F24
  07B42DA0    PPC  14AA0814  DumpMonitors+02C0C
  07B42A10    PPC  149E14F8  Java_SymantecJITCompilationThread_DoCompileMethod+
11378
  07B42950    PPC  149E1790  Java_SymantecJITCompilationThread_DoCompileMethod+
11610
  07B42810    PPC  149E1BFC  Java_SymantecJITCompilationThread_DoCompileMethod+
11A7C
  07B42600    PPC  149E2288  Java_SymantecJITCompilationThread_DoCompileMethod+
12108
  07B42480    PPC  149E24BC  Java_SymantecJITCompilationThread_DoCompileMethod+
1233C
  07B42400    PPC  14A96BC8  jio_snprintf+05908
  07B423B0    PPC  14A96B70  jio_snprintf+058B0
  07B42360    PPC  14A9709C  jio_snprintf+05DDC
  07B422F0    PPC  14A97180  jio_snprintf+05EC0
  07B42180    PPC  14A9A914  makeslottable+00980
  07B42130    PPC  14A9B0DC  makeslottable+01148
  07B420E0    PPC  14A9A914  makeslottable+00980
  07B42090    PPC  14A9B1DC  makeslottable+01248
  07B42040    PPC  14A87028  
  07B42000    PPC  14A56C58  java_lang_Compiler_start+044DC
  07B41F80    PPC  149E038C  Java_SymantecJITCompilationThread_DoCompileMethod+
1020C
  07B41E50    PPC  14AA4A6C  monitorEnter+0005C
  07B41E10    PPC  14AAB3CC  sysMonitorEnter+00064
Geetha.Vaidyanaathan@sun.com, do you know where this should go?
Component: Browser-General → Java-Implemented Plugins
This is not a Pluglet (Java Implemented Plugins) bug. Change category to OJI.

-Geetha
Assignee: asadotzler → drapeau
Status: UNCONFIRMED → NEW
Component: Java-Implemented Plugins → OJI
Ever confirmed: true
QA Contact: jelwell → paw
->OJI, reassign and confirm.
Adding crash to keywords
Keywords: crash
could you give me the version from the getInfo on the MRJInABox plugin that 
you're running?  thanks.

ps. free george.
Assignee: drapeau → loki-sun
MRJInABox says that it's version 1.1eqd1, last modified on March 24th.

ya - this is still the old plugin (i'm surprised it works at all).. sometime soon 
they're to start compiling the plugin with the build.. if you'd not mind, i could 
send you new compiles so you can try them out to see if it's still occurring?

MRJInABox 1.1eqd2 did not resolve the problem, however, the sc changed:

PowerPC illegal instruction at FFC10008 _CursorDeviceDispatch+00A68
 (CurStackBase does not seem to apply...dumping 4K.)
  Calling chain using A6/R1 links
  Back chain  ISA  Caller
  07B433AB    PPC  0047E80C  EmToNatEndMoveParams+00014
  07B43340    PPC  FFCF7778  FragRelease+0014C
  07B431C0    PPC  FFCED8AC  GetIndSymbol+006E0
  07B43170    PPC  FFCEE43C  GetIndSymbol+01270
  07B43120    PPC  FFCEF708  GetIndSymbol+0253C
  07B430B0    PPC  14E7351C  MRJPlugin__terminate+0004C
  07B43070    PPC  14E7502C  MRJPlugin::Release()+00018
  07B43030    PPC  14E7D8C8  SupportsMixin::release()+0003C
  07B42FF0    PPC  14E73728  MRJPlugin::~MRJPlugin()+0009C
  07B42FB0    PPC  14E79F34  MRJSession::~MRJSession()+00078
  07B42F70    PPC  14DCAF9C  JMCloseSession+000C0
  07B42F30    PPC  14DCA26C  
Java_com_apple_mrj_JManager_JMSessionImpl_1jni_jniPopulateCached
HashtableForMSIE401+0060C
  07B42EF0    PPC  FFCED078  CloseConnection+00054
  07B42EB0    PPC  FFCF78C0  FragRelease+00294
  07B42D30    PPC  FFCED8AC  GetIndSymbol+006E0
  07B42CE0    PPC  FFCEE43C  GetIndSymbol+01270
  07B42C90    PPC  FFCEF708  GetIndSymbol+0253C
  07B42C20    PPC  14CE9B84  SetMainStackPtr+0024C
  07B42BE0    PPC  14CE942C  ShutdownMacJava+00100
  07B42BA0    PPC  14CEC70C  closeOurResourceFile+004AC
  07B42B50    PPC  FFCED078  CloseConnection+00054
  07B42B10    PPC  FFCF78C0  FragRelease+00294
  07B42990    PPC  FFCED8AC  GetIndSymbol+006E0
  07B42940    PPC  FFCEE43C  GetIndSymbol+01270
  07B428F0    PPC  FFCEF708  GetIndSymbol+0253C
  07B42880    PPC  14D975B4  
  07B42840    PPC  14D9A924  InitJavaLibrary+031E0
  07B427F0    PPC  14D98C88  InitJavaLibrary+01544
  07B42790    PPC  14DA1174  OTCloseProvider+00010
  07B42750    PPC  17BE914C  OTCloseProviderPriv+00054

merkwürdig.. :-(
will investigate...


hey - if you have a second, could you try this: open a page with java. close that 
page by clicking the close button on the title bar. then quit mozilla.

could you see if that gives the crash and if so whether it's the mostly same 
stack trace?

thanks, loki

I tried that. Stack trace is the same except that it ends like this:

  07B428F0    PPC  FFCEF708  GetIndSymbol+0253C
  07B42880    PPC  151AA5B4  
  07B42840    PPC  151AD924  InitJavaLibrary+031E0
  07B427F0    PPC  151ABC7C  InitJavaLibrary+01538

Java crasher-->nsbeta2. Java on b2 list.
Keywords: nsbeta2
Removing nsbeta2 nomination (PDT team hadn't responded yet) and replacing with 
dogfood nomination. beard told me verbally that the presence of this bug is 
blocking him on dogfood+ bug 13848, "build MRJ plugin as part of the standard 
mozilla build process." Because quitting after using Java crashes Mozilla, he 
doesn't want to add in MRJ plug-in as part of the standard mozilla build 
process. Hence, we're not getting test coverage of Java on the Mac.

Marking as blocking 13848. Patrick, please correct me if I misunderstood 
anything.
Blocks: 13848
Keywords: nsbeta2 → dogfood
Putting on [dogfood+] radar.
Whiteboard: [dogfood+]
Moving the milestone for this bug to "Future" to indicate that there's nobody 
currently working on this, so the estimated time to fix is unknown.
Also, re-assigning this bug to the module owner (c'est moi).
Assignee: loki-sun → drapeau
QA Contact: paw → shrir
Target Milestone: --- → Future
Updated status whiteboard per Drapeau's comment.  This could certainly be looked 
at by other folks, and hopefully the status comment will atttract help.
Whiteboard: [dogfood+] → [dogfood+] no one available to resolve this :-(
Just for the record, sent email to Pete Steinauer at Apple (pas@apple.com) to 
ask for help from Apple on MRJ Plugin.  I'll update status on this bug with his 
response.
Changing milestone to "--" to express that the due date is unknown, as opposed 
to "Future" which means "after FCS, a future release".
Target Milestone: Future → ---
No way to get this in to PR2.  Adding relnote2 keyword, moving [dogfood-].
Keywords: relnote2
Whiteboard: [dogfood+] no one available to resolve this :-( → [dogfood-] no one available to resolve this :-(

*** This bug has been marked as a duplicate of 36381 ***
Status: NEW → RESOLVED
Closed: 24 years ago
Resolution: --- → DUPLICATE
Verified dupe.
Status: RESOLVED → VERIFIED
OS: All
Product: Core → Core Graveyard
You need to log in before you can comment on or make changes to this bug.