fastfix for 1.23

That network-stop problem I have it very frequently as well. Just before I tried to visit amigans.net and amigaworld.net from my bookmark, but I got an error message; had to quit and restart Odyssey to resume.

@Phantom

Sounds like same symptoms here.

I get error message eventually, if I wait enought long.

If I try to refresh page, it stays just same, network is dead.

In same time other network related progs works ok and i can ping outside with names and IP addresses..

I can find some java reated error messages from windows menuitem under messages. Many of them seems wait network.

Now I disabled javascript and works much better. It was same in 1.16.

I just enable javascript to pages, wich really needs them, network banking, facebook, gmail.

I usually surf quite lot news sites and they use javascript heavily.

I test tomorrow without javascript and let see, if it still behaves well.

ATM 1.23 fix2

With original 1.23 I had network freezing but overall not so many system or Odyssey hangs.

This stalling connection issue is most likely due to the version kas1e used for his build (7,34,0). There's no such issue with 7,28,1, and it's likely a regression in curl itself.

@Fab

Thanks.

@fab
but version on os4depot have 7.34.0 too, the same as fastfix1, so there should be no differences in terms of network or freezes. fastfix2 have 7.32.0. i can understand differences with version on os4depot and fix2 (where curl changed),but betwen initial and fix1 its kind of mictic, as all changes in fix1 are cosmetic + rexx port change

and its strange that i not have those problems, as well as didnt have any "error messages" about, as well as i read on other forums it works ok

@phantom
dud you fix your "open tabs in initial window" ? can you also try debug kernel with munge to catch log when you have those random freezes ?

@mlehto
by stals and hangs you mean network freezes or odyssey crashes ? have you any craslogs ? may you try debug kerlen with munge ? did you install everything from scratch to new directory and then reboot ?

Really no problem here with os4depot version and now with FF2 version.

I use it a lot, no freeze, just 1 or 2 DSI with embedded MPlayer, nothing serious.

(I have the same config than Phantom).

@Phantom & Mlehto

There is just a binary of odyssey build with curl7.28.1.
http://www.sendspace.com/file/56klqp

To test it normally, plz, just install fastfix2 to fresh directory (and that include everything, bookmarks, passwords, doesn't matter what : all should be just fresh without your old files from any previous installs, to exclude any possible and impossible problems) , and then, just overwrite binary by that new binary. Reboot must too. Preferable to not have anything in background which may cause problems too

@zzd10h
I just fear that there will be no reasons to build with curl7.28.1 , because problems with their freezes will be somewhere else (like some beta network driver for example).

Can you also test plz, if for you when you run multiply instances of odyssey, "open in new tabs" open tabs where need it, and not as for Phantom in first running window ?

@Phantom and Mlehto
Don't you use some beta versions of network drivers ? Like maybe some beta of on-board driver for ? Also, plz (plz) when do tests 101% to not mix any files with previous versions. Everything should be fresh, in one directory, that include everything, and does not matter if it "ok to have my old bookmarks" or whatever : just clean stuff in fresh directory , all from scratch, hard reboot and co.

@kas1e
Like Phantom, I'm not an OS4 betatester.

Just tested with 3 differents instances of Odyssey, when I "open in a new tab" some links in instance 2, it opens well on instance 2 window. The same for the 2 others Odyssey instances
=> no problem for me.

@zzd10h

Quote:

Probably different networks cards ? Or maybe some app in background which may trash something in memory

Quote:

Ok, so problem with "open tabs in wrong window" seems happens for Phantom only, and my bet its something in odyssey/mui settings then.

@kas1e
"Probably different networks cards ? "

RTL8139, stock x1000 network card.

@all

After updating to fastfix2, Backspace or Delete key doesn't work on address bar. They work on other applications. Does anybody have the same problem ?

@Sinan

I have a guess: numlock or mui settings in the Text/Keyboard (if it works in other mui apps, then its not global, but local for odyssey, if it didn't works in another mui apps too, then global).

All what fastfix2 have which maybe-somehow can related, its new catalog from Phantom (it can be related only if you are Greek user of course) , but probably nope, and its just your setting.

As you can see in first post, there is no code changes in fastfix2 at all (except different version of curl) and problems you have its only can be related to mui settings of keyboard/text areas, or depend on what kind of locales/keymaps you setup.

To test you can just run mui prefs, go to "text", make "Example" string be active , and there test your backspace/delete keys. If all works, then run odyssey, go to mui prefs from odyssey, and do the same test. And if then all is ok still, but in adress bar still not, and you are greek user, then delete greek catalog and try again.

And if still, then delete everything, install fastfix1 and try all the same stuff.

@All
Be sure 101% you install everything from scratch, in different directory, without any old files from previous ones. Also keep in mind, that name of baseapp changed from OWB to Odyssey, what mean, that all mui settings you setup for will need new retouch.

Just to add to the network freezing questions.

I am not a beta tester and have not experienced any network freezes. I actually notice faster page loading with 1.23, FF1 and FF2.

X1000, 4GB RAM, HD6850, 9250, RTL8139 based NIC

A minor and very low priority suggestion for a fix. The window title still hows OWB rather than Odyssey. It would be nice to have consistency with the programme name and Arexx port name.

@BillE
fixed

@kas1e

I use Turkish locale setting. I will test as you requested and report back to you.

Thanks.

@kas1e
Since yesterday, I have an error using Odyssey. This error arises regularly but I can ignore and go on.
I can't tell you how to reproduce the error, so I hope with the crashlog you will guess the origin of the error.
.Crash log for task "pthread id 7"
Generated by GrimReaper 53.16
Crash occured in module bsdsocket.library at address 0x6FB68C54
Type of crash: DSI (Data Storage Interrupt) exception

Register dump:
GPR (General Purpose Registers):
0: 00000004 4D7CADD0 00000000 00000000 54E008F0 00000001 51EFC954 00000400
8: 54E008F0 50D67020 51EFCD5C 021BDC66 00000330 00000000 00000000 532901B0
16: 02270000 7FD13580 02270000 50E52DB0 001E8FF8 00000000 0183B608 6FD9D17C
24: 00000400 00000001 00000001 51EFC954 00000000 4D9DD518 00000000 54E008F0


FPR (Floating Point Registers, NaN = Not a Number):
0: nan 218 1112 0
4: 35 105.8 51.8 2.14748e+09
8: 17 128 1 255
12: 1 1129 0 0
16: 0 1.08646e-311 0 0
20: 0 0 0 0
24: 2 -1 1 -100
28: 1e+06 4.5036e+15 1 0

FPSCR (Floating Point Status and Control Register): 0x82004000


SPRs (Special Purpose Registers):
Machine State (msr) : 0x0002F030
Condition (cr) : 0x524F9608
Instruction Pointer (ip) : 0x6FB68C54
Xtended Exception (xer) : 0x534ECEBC
Count (ctr) : 0x5FF924D0
Link (lr) : 0x6FB0E918
DSI Status (dsisr) : 0x524F93C0
Data Address (dar) : 0x019AAF8C



680x0 emulated registers:
DATA: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
ADDR: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
FPU0: 0 0 0 0
FPU4: 0 0 0 0



Symbol info:
Instruction pointer 0x6FB68C54 belongs to module "bsdsocket.library" (HUNK/Kickstart)

Stack trace:
module LIBS:bsdsocket.library at 0x6FB68C54 (section 5 @ 0xEC30)
module LIBS:bsdsocket.library at 0x6FB8C9A0 (section 5 @ 0x3297C)
module LIBS:bsdsocket.library at 0x6FB7FD04 (section 5 @ 0x25CE0)
Curl_ipv4_resolve_r()+0xd4 (section 1 @ 0x1975190)
gethostbyname_thread()+0x20 (section 1 @ 0x1954F3C)
curl_thread_create_thunk()+0x38 (section 1 @ 0x19802F4)
run()+0x278 (section 1 @ 0x410C)
ThreadCode()+0x35c (section 1 @ 0x44B4)
native kernel module dos.library.kmod+0x00024cd0
native kernel module kernel+0x0003b4b0
native kernel module kernel+0x0003b530

PPC disassembly:
6fb68c4c: 4bffff84 b 0x6FB68BD0
6fb68c50: 8124004c lwz r9,76(r4)
*6fb68c54: 90690000 stw r3,0(r9)
6fb68c58: 4bffff78 b 0x6FB68BD0
6fb68c5c: 8124004c lwz r9,76(r4)

@yescop
os4depot one, fastfix1 or fastfix2 ? crash looks exactly about threaded resolver, but if it happens for you still, it mean fixed bsdsocket still need it (dunno when it will be released, i have one from Olaf whete he maid some fixes, but when it will be in public dunno)

@kas1e
I am using fastfix2