New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.
Starred by 1 user

Issue metadata

Status: WontFix
Owner:
Email to this user bounced
Closed: Aug 2009
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux
Pri: 1
Type: Bug

Restricted
  • Only users with Commit permission may comment.



Sign in to add a comment

thecamels.org site doesn't load

Reported by kkarpies...@gmail.com, Aug 17 2009

Issue description

Chrome Version       : 3.0.198.1
OS + version : Ubuntu 9.04
CPU architecture (32-bit):
window manager : Gnome
URLs (if applicable) : http://thecamels.org/2009/08/17/gimp-2-7-0/
Behavior in Firefox 3.x (if applicable): works normal
Behavior in Chrome for Windows (optional): i don't know

What steps will reproduce the problem?
1. go to url http://thecamels.org/2009/08/17/gimp-2-7-0/
2. site load not full, throbber shows that site is still loading
3. you cannot reaload this site, you cannot stop loading, you cannot load 
any other page in this tab
4. sometimes you cannot kill/close this tab

What is the expected result?
sie should load normally


 
also, when this page is loading, you cannot open any other page in other tabs

Comment 2 by est...@chromium.org, Aug 17 2009

Labels: -Area-Misc Area-BrowserBackend
Status: Available
confirmed, don't know what might be causing this
It doesn't happen in Chrome on Windows on a ToT build according to eroman, and it 
doesn't happen on the Chrome on Windows dev channel release.

If no one else gets to it, I can take a look at it later.

Comment 4 by evan@chromium.org, Aug 18 2009

Labels: Pri-1
Glanced at this on a recent trunk build; from looking at top it looks like we may be 
getting into an infinite loop.
Adding ager for v8 expertise.

I loaded the page up and took a look at the renderer process.  I get all v8 
stacktraces:

[Switching to thread 1 (Thread 0xf708e720 (LWP 2972))]#0  0x091df860 in 
v8::internal::String::IsAsciiRepresentation (this=0xf5456b29) at v8/src/objects-
inl.h:246
246       if ((type & kStringRepresentationMask) == kConsStringTag &&
(gdb) ba
#0  0x091df860 in v8::internal::String::IsAsciiRepresentation (this=0xf5456b29) at 
v8/src/objects-inl.h:246
#1  0x09228311 in v8::internal::String::Set (this=0xf5456b29, index=20, value=119) at 
v8/src/objects-inl.h:1647
#2  0x09277680 in ConvertCaseHelper<unibrow::ToLowercase> (s=0xf5456acd, length=37, 
input_string_length=37, mapping=0xb83f700) at v8/src/runtime.cc:3428
#3  0x092789d0 in ConvertCase<unibrow::ToLowercase> (args={<v8::internal::Embedded> = 
{<No data fields>}, length_ = 1, arguments_ = 0xffd3e2bc}, mapping=0xb83f700)
    at v8/src/runtime.cc:3504
#4  0x09278a4e in Runtime_StringToLowerCase (args={<v8::internal::Embedded> = {<No 
data fields>}, length_ = 1, arguments_ = 0xffd3e2bc}) at v8/src/runtime.cc:3514
#5  0xf4d0018b in ?? ()
#6  0x00000001 in ?? ()
#7  0xffd3e2bc in ?? ()
#8  0xf5873fa9 in ?? ()
#9  0x00000000 in ?? ()
(gdb) c
Continuing.

Program received signal SIGINT, Interrupt.
0x091dc75b in MapWord (this=0xffd3e154, value=4106748385) at v8/src/objects.h:1134
1134      explicit MapWord(uintptr_t value) : value_(value) {}
(gdb) ba
#0  0x091dc75b in MapWord (this=0xffd3e154, value=4106748385) at 
v8/src/objects.h:1134
#1  0x091dcc80 in v8::internal::HeapObject::map_word (this=0xf54581b9) at 
v8/src/objects-inl.h:981
#2  0x091dcca2 in v8::internal::HeapObject::map (this=0xf54581b9) at v8/src/objects-
inl.h:971
#3  0x091de27a in v8::internal::Object::IsString (this=0xf54581b9) at v8/src/objects-
inl.h:140
#4  0x091df5ef in v8::internal::String::cast (object=0xf54581b9) at v8/src/objects-
inl.h:1517
#5  0x09220871 in v8::internal::Heap::AllocateRawAsciiString (length=37, 
pretenure=v8::internal::NOT_TENURED) at v8/src/heap.cc:2513
#6  0x09252c51 in v8::internal::String::TryFlatten (this=0xf545818d) at 
v8/src/objects.cc:702
#7  0x091e45c2 in v8::internal::String::TryFlattenIfNotFlat (this=0xf545818d) at 
v8/src/objects-inl.h:1609
#8  0x0927898e in ConvertCase<unibrow::ToLowercase> (args={<v8::internal::Embedded> = 
{<No data fields>}, length_ = 1, arguments_ = 0xffd3e2bc}, mapping=0xb83f700)
    at v8/src/runtime.cc:3497
#9  0x09278a4e in Runtime_StringToLowerCase (args={<v8::internal::Embedded> = {<No 
data fields>}, length_ = 1, arguments_ = 0xffd3e2bc}) at v8/src/runtime.cc:3514
#10 0xf4d0018b in ?? ()
#11 0x00000001 in ?? ()
#12 0xffd3e2bc in ?? ()
#13 0xf5873fa9 in ?? ()
#14 0x00000000 in ?? ()
(gdb) c
Continuing.

Program received signal SIGINT, Interrupt.
v8::internal::String::Set (this=0xf50128c9, index=2, value=111) at v8/src/objects-
inl.h:1647
1647          : SeqTwoByteString::cast(this)->SeqTwoByteStringSet(index, value);
(gdb) c
Continuing.

Program received signal SIGINT, Interrupt.
0x091df5b7 in StringShape (this=0xffd3e19c, str=0xf5a92bb5) at v8/src/objects-
inl.h:217
217       ASSERT((type_ & kIsNotStringMask) == kStringTag);
(gdb) ba
#0  0x091df5b7 in StringShape (this=0xffd3e19c, str=0xf5a92bb5) at v8/src/objects-
inl.h:217
#1  0x091df79e in v8::internal::String::length (this=0xf5a92bb5) at v8/src/objects-
inl.h:1580
#2  0x0924963c in v8::internal::String::SlowEquals (this=0xf509b369, 
other=0xf5a92bb5) at v8/src/objects.cc:4367
#3  0x091e4706 in v8::internal::String::Equals (this=0xf509b369, other=0xf5a92bb5) at 
v8/src/objects-inl.h:1569
#4  0x0928f531 in Runtime_StringEquals (args={<v8::internal::Embedded> = {<No data 
fields>}, length_ = 2, arguments_ = 0xffd3e2b8}) at v8/src/runtime.cc:3895
#5  0xf4d0018b in ?? ()
#6  0x00000002 in ?? ()
#7  0xffd3e2b8 in ?? ()
#8  0xffd3e2b4 in ?? ()
#9  0xf4d0018b in ?? ()
#10 0x00000001 in ?? ()
#11 0x00000000 in ?? ()

I'm not sure what's going on, but v8 seems to be stuck.  The renderer process has a 
cpu pegged at 100%.  The browser isn't doing anything, and there aren't any more IPC 
messages being sent to/from the renderer handling this site instance.

Comment 6 by ager@chromium.org, Aug 19 2009

Status: Assigned
I tried this on almost identical versions on Windows and Linux and only the Linux 
version hangs.  Lasse, could you have a look and attempt to figure out why V8 is 
looping on Linux on this page?
The page works if you enable the Flash plugin (using --enable-plugins on a recent 
Chromium build), and it hangs if the plugin is not enabled. I'm guessing that the page 
is trying to interact with a non-existing, and probably doesn't get the response its 
expecting.

That said, it does work in, e.g., Firefox with the flash plugin disabled, so something 
isn't working as expected in Chromium.
Status: WontFix
The bug is in the page script. 
The problem is that Chromium without any plugins has a navigator.plugins and 
navigator.mimeTypes, but navigator.mimeTypes.length is zero. The script does not 
account for this and tries to use the IE specific code and apparently fails 
spectacularly (guess is it tries to find an ActiveXObject by incrementing a number 
until it finds a match).

If I remove all plugins in Firefox, thereby lowering its navigator.mimeTypes.length 
to zero, it exhibits the same problem.
Labels: -Area-BrowserBackend Area-Internals
Project Member

Comment 10 by bugdroid1@chromium.org, Oct 12 2012

Labels: Restrict-AddIssueComment-Commit
This issue has been closed for some time. No one will pay attention to new comments.
If you are seeing this bug or have new data, please click New Issue to start a new bug.
Project Member

Comment 11 by bugdroid1@chromium.org, Mar 10 2013

Labels: -Area-Internals Cr-Internals

Sign in to add a comment