Closed Bug 241757 Opened 21 years ago Closed 21 years ago

Crash when "forward" mail is used

Categories

(Core Graveyard :: GFX: OS/2, defect)

x86
OS/2
defect
Not set
critical

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: paul, Assigned: mkaply)

References

Details

(Keywords: crash, verified1.7)

Attachments

(1 file)

User-Agent: Mozilla/5.0 (OS/2; U; Warp 4.5; en-US; rv:1.7b) Gecko/20040422 Build Identifier: Mozilla/5.0 (OS/2; U; Warp 4.5; en-US; rv:1.7b) Gecko/20040422 Forwarding a message results in browser crash. The following message appears in POPUPLOG.OS2 04-26-2004 08:46:31 SYS3175 PID 0081 TID 0001 Slot 0081 D:\BIN\MOZILLA.ORG\MOZILLA\MOZILLA.EXE c0000005 1ddfd6d6 P1=00000001 P2=00000014 P3=XXXXXXXX P4=XXXXXXXX EAX=00000000 EBX=0012d1e8 ECX=16dc6d04 EDX=00000000 ESI=00000000 EDI=0012d370 DS=0053 DSACC=f0f3 DSLIM=ffffffff ES=0053 ESACC=f0f3 ESLIM=ffffffff FS=150b FSACC=00f3 FSLIM=00000030 GS=0000 GSACC=**** GSLIM=******** CS:EIP=005b:1ddfd6d6 CSACC=f0df CSLIM=ffffffff SS:ESP=0053:0012d058 SSACC=f0f3 SSLIM=ffffffff EBP=0012d170 FLG=00012202 LIBC05.DLL 0001:0001d6d6 Reproducible: Always Steps to Reproduce: 1. Highlight mail 2. Click "Forward" button 3. Actual Results: Complete browser crash Expected Results: New compose window with mail to be forwarded embedded Build platform target i386-pc-os2-emx Build tools Compiler Version Compiler flags gcc gcc version 3.2.2 (Innotek Build 2004-02-23 14:15) -Wall -W -Wno-unused -Wpointer-arith -Wcast-align -Wno-long-long -pedantic -Zomf -pipe g++ gcc version 3.2.2 (Innotek Build 2004-02-23 14:15) -fno-rtti -fno-exceptions -Wall -Wconversion -Wpointer-arith -Wcast-align -Woverloaded-virtual -Wsynth -Wno-ctor-dtor-privacy -Wno-non-virtual-dtor -Wno-long-long -pedantic -Zomf -pipe Configure arguments --enable-crypto --disable-tests --disable-debug --enable-optimize
Does it happen with new or just with mail?
(In reply to comment #1) > Does it happen with new or just with mail? "Compose" and "Reply" work fine. Just when using the "Forward". So far it happens every time (well at least three times on several messages).
Sorry, I meant does it happen with news :) Or just mail
(In reply to comment #3) > Sorry, I meant does it happen with news :) > > Or just mail Interesting... it works in newsgroups, but not in mail.
Yep it crashes
Assignee: general → mkaply
Status: UNCONFIRMED → NEW
Ever confirmed: true
The crash is in fclose() and is caused by passing a NULL pointer (esi) to it. There is to my knowlegde no defined behaviour for this scenario, but crashing is the normal action (IIRC LIBC, VAC and FreeBSD does so, while GLIBC doesn't).
Mike: If you want LIBC symbols in idbug you should just rename libc05.dbg to libc05.dll. libc05.dbg is the unstripped library and since we on purpose does not enable strack frame optimizations you'll have a stacktrace out of libc at least. (You will get perhaps get the source file dialog if crashing inside LIBC but that's normally something you can dismiss without crashing the debugger...) Also mark that there is a libc05.map in the usr/lib directory of the GCC installation. This mapfile also includes linenumbers offsets which is very handy for figuring out what's wrong.
Attached patch FixSplinter Review
You may wonder what on earth a fix to the icon code has to do with this. For some reason, when you forward an email, it tries to get an icon for: .com/Inbox#0 My new icon code tries to create this as a file to get an icon and of course it fails. I wasn't checking the error code from creating the file, and the fclose was crashing.
Attachment #147158 - Flags: review?(pedemont)
My build is from the 25th and it doesn't crash when I click forward so I don't know if this is related but if I have Mozilla set to use a socks server then Mozilla crashes when I click send on a mail, regardless of forward, reply or compose. I don't see a relation but don't want to open a new bug if it is related. ------------------------------------------------------------ 04-27-2004 07:05:23 SYS3175 PID 0047 TID 0003 Slot 00a4 E:\BROWSERS\WARPZILLA\MOZILLA.EXE c0000005 00000000 P1=00000001 P2=00000000 P3=XXXXXXXX P4=XXXXXXXX EAX=00000134 EBX=00000000 ECX=00000000 EDX=00000134 ESI=01513780 EDI=027afb8c DS=0053 DSACC=f0f3 DSLIM=ffffffff ES=0053 ESACC=f0f3 ESLIM=ffffffff FS=150b FSACC=00f3 FSLIM=00000030 GS=0000 GSACC=**** GSLIM=******** CS:EIP=005b:00000000 CSACC=f0df CSLIM=ffffffff SS:ESP=0053:027afb7c SSACC=f0f3 SSLIM=ffffffff EBP=01127e68 FLG=00012202 Andy
Attachment #147158 - Flags: review?(pedemont) → review+
Comment on attachment 147158 [details] [diff] [review] Fix sr=blizzard (platform specific) a=mkaply (OS/2 only)
Attachment #147158 - Flags: superreview+
Attachment #147158 - Flags: approval1.7+
Fix checked in.
Status: NEW → RESOLVED
Closed: 21 years ago
Resolution: --- → FIXED
Severity: normal → critical
Keywords: crash
Component: Browser-General → GFX: OS/2
*** Bug 242384 has been marked as a duplicate of this bug. ***
Keywords: fixed1.7
Verified as fix on latest 1.7 branch 06-24 build for Windows, Mac & Linux. Changing keywords from fixed1.7 to verified1.7. Leave this bug status "as is" until this bug be verified on trunk again... Reporter, please reopen this bug if problem is still existing on OS/2.
Keywords: fixed1.7verified1.7
Product: Core → Core Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: