Firefox problems with arm-debian Etch distribution
Yegor Yefremov
I try to execute Firefox under arm-debian Etch. At the moment I have
Firefox 1.5.0.4 from
https://sourceforge.net/project/showfiles.php?group_id=8846.
The problem is when I start firefox I don't get any window. I use Xming
to connect to arm host. The ps -A shows among others following processes:
PID TTY TIME CMD
1 ? 00:00:02 init
405 ? 00:00:00 inetd
448 ? 00:00:00 rpc.statd
458 ? 00:00:00 cron
480 ? 00:00:00 xdm
492 ? 00:00:00 getty
511 ? 00:00:00 xdm
519 ? 00:00:00 WindowMaker
545 ? 00:00:00 ssh-agent
546 ? 00:00:24 WindowMaker
551 ? 00:00:03 xterm
552 pts/0 00:00:04 bash
973 pts/7 00:00:00 bash
979 pts/0 00:00:00 firefox
983 pts/0 00:00:00 run-mozilla.sh
988 pts/0 00:00:14 firefox-bin
989 pts/0 00:00:00 firefox-bin
990 pts/0 00:00:00 firefox-bin
997 pts/0 00:00:00 firefox-bin
1013 pts/1 00:00:00 ps
The process 979 is doing (strace output):
Process 979 attached - interrupt to quit
wait4(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0, NULL) = 983
--- SIGINT (Interrupt) @ 0 (0) ---
sigreturn() = ? (mask now [CHLD RTMIN])
rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0
--- SIGCHLD (Child exited) @ 0 (0) ---
wait4(-1, 0xbffff584, WNOHANG, NULL) = -1 ECHILD (No child processes)
sigreturn() = ? (mask now [RTMIN])
rt_sigaction(SIGINT, {SIG_DFL}, {0x3b068, [], 0x4000000 /* SA_??? */},
8) = 0
rt_sigprocmask(SIG_BLOCK, NULL, [RTMIN], 8) = 0
read(255, "exitcode=$?\n\n## Stop addon scrip"..., 5247) = 91
rt_sigprocmask(SIG_BLOCK, NULL, [RTMIN], 8) = 0
The process 988:
Process 988 attached - interrupt to quit
gettimeofday({1177580871, 613903}, NULL) = 0
gettimeofday({1177580871, 614826}, NULL) = 0
ioctl(3, FIONREAD, [0]) = 0
poll([{fd=3, events=POLLIN}, {fd=16, events=POLLIN}, {fd=5,
events=POLLIN, revents=POLLIN}], 3, -1) = 1
gettimeofday({1177580886, 548929}, NULL) = 0
gettimeofday({1177580886, 549791}, NULL) = 0
write(8, "8", 1) = 1
gettimeofday({1177580886, 551962}, NULL) = 0
gettimeofday({1177580886, 552755}, NULL) = 0
kill(997, SIGRTMIN) = 0
read(5, "\372", 1) = 1
gettimeofday({1177580886, 555468}, NULL) = 0
The process 989:
Process 989 attached - interrupt to quit
getppid() = 988
poll([{fd=9, events=POLLIN}], 1, 2000) = 0
getppid() = 988
poll([{fd=9, events=POLLIN}], 1, 2000) = 0
getppid() = 988
poll([{fd=9, events=POLLIN}], 1, 2000) = 0
The process 990:
Process 990 attached - interrupt to quit
poll([{fd=7, events=POLLIN, revents=POLLIN}], 1, -1) = 1
gettimeofday({1177580916, 577781}, NULL) = 0
read(7, "8", 1024) = 1
gettimeofday({1177580916, 579625}, NULL) = 0
poll([{fd=7, events=POLLIN, revents=POLLIN}], 1, -1) = 1
gettimeofday({1177580931, 641323}, NULL) = 0
read(7, "8", 1024) = 1
gettimeofday({1177580931, 644438}, NULL) = 0
poll([{fd=7, events=POLLIN, revents=POLLIN}], 1, -1) = 1
gettimeofday({1177580946, 662661}, NULL) = 0
read(7, "8", 1024) = 1
The process 997:
Process 997 attached - interrupt to quit
gettimeofday({1177580926, 257602}, NULL) = 0
nanosleep({5, 313792000}, NULL) = 0
rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0
gettimeofday({1177580931, 589379}, NULL) = 0
write(6, "\372", 1) = 1
gettimeofday({1177580931, 626669}, NULL) = 0
gettimeofday({1177580931, 627528}, NULL) = 0
rt_sigprocmask(SIG_BLOCK, NULL, [RTMIN], 8) = 0
rt_sigprocmask(SIG_UNBLOCK, [RTMIN], [RTMIN], 8) = 0
gettimeofday({1177580931, 630094}, NULL) = 0
nanosleep({387, 893434000}, 0) = -1 EINTR (Interrupted system call)
--- SIGRTMIN (Unknown signal 32) @ 0 (0) ---
rt_sigprocmask(SIG_SETMASK, [RTMIN], NULL, 8) = 0
The fd=3 is a socket, that is connected to Xming. The other fds are
pipes. Any ideas what's wrong? Where should I search in source for
finding these activities? Some information about my system:
Kendin/Micrel KS8695 system (ARM9)
Kernel: 2.4.32
RAM: 59Mb
Best regards,
Yegor
Tony Mechelynck
> Hi!
>
> I try to execute Firefox under arm-debian Etch. At the moment I have
> Firefox 1.5.0.4 from
> https://sourceforge.net/project/showfiles.php?group_id=8846.
Have you tried getting Firefox from Mozilla? 1.5.0.4 is obsolete anyway.
1.5.0.11 (previous release, will be supported for a few months, then dropped):
http://releases.mozilla.org/pub/mozilla.org/firefox/releases/1.5.0.11/
2.0.0.3 (current release): http://getfirefox.com/
Best regards,
Tony.
--
"I'll carry your books, I'll carry a tune, I'll carry on, carry over,
carry forward, Cary Grant, cash & carry, Carry Me Back To Old Virginia,
I'll even Hara Kari if you show me how, but I will *not* carry a gun."
-- Hawkeye, M*A*S*H
Yegor Yefremov
> Yegor Yefremov wrote:
>
>> Hi!
>>
>> I try to execute Firefox under arm-debian Etch. At the moment I have
>> Firefox 1.5.0.4 from
>> https://sourceforge.net/project/showfiles.php?group_id=8846.
>>
> [...]
>
> Have you tried getting Firefox from Mozilla? 1.5.0.4 is obsolete anyway.
>
> 1.5.0.11 (previous release, will be supported for a few months, then dropped):
> http://releases.mozilla.org/pub/mozilla.org/firefox/releases/1.5.0.11/
>
> 2.0.0.3 (current release): http://getfirefox.com/
>
>
>
> Best regards,
> Tony.
>
ARM platform. I want to avoid compiling myself due to a lack of
performance on the ARM9 system. I tried Firefox 2.0.0.3 from the
arm-debian repository, but it crashes during the loading (see
http://lists.debian.org/debian-arm/2007/03/msg00011.html).
Yegor
L. David Baron
> I try to execute Firefox under arm-debian Etch. At the moment I have
> Firefox 1.5.0.4 from
> https://sourceforge.net/project/showfiles.php?group_id=8846.
Does "./run-mozilla.sh ./TestXPTCInvoke" work? Or are those files
you got binaries that come without tests?
-David
--
L. David Baron <URL: http://dbaron.org/ >
Technical Lead, Layout & CSS, Mozilla Corporation
Tony Mechelynck
The versions that can be downloaded at the URLs I gave you above have been
compiled and tested by Mozilla; they are supposed to work out of the box on
any Linux distribution.
If you get IceWeasel 2.0.0.3 from Debian, it may include patches not approved
by Mozilla. Debian is known for having added *destabilizing* modifications to
the original Mozilla code in the past.
Any crashes due to any patches added by Debian or by Sourceforge are of course
not Mozilla's responsibility. That's why I suggested that you try an official
Mozilla release build to see if you had the same problems.
Best regards,
Tony.
--
Nothing astonishes men so much as common sense and plain dealing.
Boris Zbarsky
> The versions that can be downloaded at the URLs I gave you above have
> been compiled and tested by Mozilla; they are supposed to work out of
> the box on any Linux distribution.
You left out "x86". Those builds will work on "any" x86 Linux (modulo it being
new enough, etc, etc). They will not work on Linux/PPC or Linux/SPARC, or
Linux/MIPS, or Linux/ARM, etc. This becomes very clear when you note that the
directory the "Linux" builds are in on the FTP server is called "linux-i686".
The original poster, however, is using an ARM processor. Mozilla.org does not
provide any ARM builds that I know of. So the advice to try a Mozilla.org build
is completely useless in this case.
-Boris
P.S. There are issues in the Gecko code that only manifest themselves on ARM
processors (search bugzilla if you care).
Yegor Yefremov
> On Thursday 2007-04-26 12:05 +0200, Yegor Yefremov wrote:
>
>> Firefox 1.5.0.4 from
>> https://sourceforge.net/project/showfiles.php?group_id=8846.
>>
>
> Does "./run-mozilla.sh ./TestXPTCInvoke" work? Or are those files
> you got binaries that come without tests?
>
> -David
>
>
>
> _______________________________________________
> dev-general mailing list
> dev-g...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-general
>
Unfortunately I don't have any TestXPTCInvoke (it was not in the binary
package), but I have TestGtkEmbed that produces the following output:
new_gtk_browser
menu bar
tool bar
location bar
status bar
and then crashes with:
./run-mozilla.sh: line 131: 620 Segmentation fault "$prog" ${1+"$@"}
Before crash I can see gtk-Windows for 1-2 seconds.
Yegor
Tony Mechelynck
ah, that's the hitch. Didn't know ARM is a different kind of processor. Thanks
Boris for your reply, which made me (find and) read
http://en.wikipedia.org/wiki/ARM_architecture -- interesting even if "somewhat
technical" reading.
Best regards,
Tony.
--
The longer I am out of office, the more infallible I appear to myself.
-- Henry Kissinger