Amiga.org
Amiga computer related discussion => Amiga Software Issues and Discussion => Topic started by: utri007 on May 08, 2009, 09:22:08 AM
-
Where it is???
Makers of OS4 version has said multiple times that they had made Netsurf for OS4 so that OS3.9/3.5 version should be easy to make. Unsupported featured could simply commented out from source etc.
-
The source is there, just waiting for you to compile it. ;-)
If it's just a question of commenting out the unused features, then it's easy, right?
That appears to be wrong, unfortunately. :-( - or at least to my very quick trip through the source it does...
There's bits of it which give me hope it might be straightforward, such as checking which version of graphics.library is returned before assuming compositing is available, but then others which worry me, such as some of the dependencies.
Having a quick look, it seems that things like libcurl (first dependency I chose) require OS4+... So then you're into backporting all the dependencies too.
I don't know how much time it would take to port this to 3.9 or previous, but it's not a question of commenting out the bits that don't work as that would be everything.
Unfortunately my development for AmigaOS sorta stopped 16 years ago, and I wasn't doing anything nearly as complex as this.
But I would love to try to help get this ported. I think an analysis of what is needed would be a wise first step, and then a bit of a plan...
-
Just found this when digging a bit deeper:
http://www.ppa.pl/artykul-Chris.Young.November.2008.-3_32_1169.html
Are there any chances for AmigaOS 3.X, MorphOS or AROS versions of NetSurf?
I originally started out writing the NetSurf Amiga specific code in such a way that it would work on OS3.x, if somebody had the time to port all the dependecies. That quickly went out the window when features I needed were either only available on OS4 or there was a better way of doing them on OS4. I still think it can be done, but it would require a lot more work. Certainly the speed of NetSurf is comparable to AWeb and should be ideal for the sort of hardware OS3 is running on. I have had contact from people interested in helping port to OS3 but nothing has come of it yet.
MorphOS and AROS have the additional problem of using MUI as their native UI. Somebody did contact me regarding porting NetSurf to AROS and I'll help out where I can, but if you're using MUI (or Zune or whatever) then most of the work I've done needs to be repeated again for MUI. MorphOS users may be better off sticking with Sputnik unless somebody particulary wants to put the effort in.
-
http://aminet.net/package/dev/c/netsurf-morphos-src
It might be easier to compile this for 3.x.
-
Piru wrote:
http://aminet.net/package/dev/c/netsurf-morphos-src
It might be easier to compile this for 3.x.
You mean you're gonna do it ? ;-)
-
No ;-)
-
what if say please ? pretty please with icecream on top ? :-P
-
I'll take a look at the morphos codebase, but realistically I've got a huge amount to do to bring myself up to speed here. For starters, I never even used MUI... I'd be better starting off on a smaller project!
If you're interested and have some development experience, please let the world know . Of course, the luxury of time is another requirement!
Piru, do you know if Illka has considered the 3.X side of life at all when writing this?
(Just noticed his email address - my youngest son has a pikku veli t-shirt )
First problem: creating a suitable build environment...
Does anyone have any experience of building with gcc and other gnu tools on the miggy, or am I best off cross compiling?
Looks like the Amiga native option would be using Geek Gadgets, which seems to be rather moribund...
Hints? :-)
(failing at first hurdle - amazing! )
-
I give a try and at first I get error
makefile:320: *** missing `endif'.
ifeq ($(TARGET),gtk)
LDFLAGS += $(shell $(PKG_CONFIG) --libs libxml-2.0 libcurl openssl)
# define additional CFLAGS and LDFLAGS requirements for pkg-configed libs here
NETSURF_FEATURE_RSVG_CFLAGS := -DWITH_RSVG
NETSURF_FEATURE_ROSPRITE_CFLAGS := -DWITH_NSSPRITE
NETSURF_FEATURE_HUBBUB_CFLAGS := -DWITH_HUBBUB
NETSURF_FEATURE_BMP_CFLAGS := -DWITH_BMP
NETSURF_FEATURE_GIF_CFLAGS := -DWITH_GIF
NETSURF_FEATURE_PNG_CFLAGS := -DWITH_PNG
***** [line 320] Why here ? *****
# add a line similar to below for each optional pkg-configed lib here
$(eval $(call pkg_config_find_and_add,RSVG,librsvg-2.0,SVG rendering))
$(eval $(call pkg_config_find_and_add,ROSPRITE,librosprite,RISC OS sprite rendering))
$(eval $(call pkg_config_find_and_add,HUBBUB,libhubbub,Hubbub HTML parser))
$(eval $(call pkg_config_find_and_add,BMP,libnsbmp,NetSurf BMP decoder))
$(eval $(call pkg_config_find_and_add,GIF,libnsgif,NetSurf GIF decoder))
$(eval $(call pkg_config_find_and_add,PNG,libpng,PNG support))
GTKCFLAGS := -std=c99 -Dgtk -Dnsgtk \
-DGTK_DISABLE_DEPRECATED \
-D_BSD_SOURCE \
-D_XOPEN_SOURCE=600 \
-D_POSIX_C_SOURCE=200112L \
-D_NETBSD_SOURCE \
-DGTK_RESPATH=\"$(NETSURF_GTK_RESOURCES)\" \
$(WARNFLAGS) -I. -g \
$(shell $(PKG_CONFIG) --cflags libglade-2.0 gtk+-2.0) \
$(shell xml2-config --cflags)
GTKLDFLAGS := $(shell $(PKG_CONFIG) --cflags --libs libglade-2.0 gtk+-2.0 gthread-2.0 gmodule-2.0 lcms)
CFLAGS += $(GTKCFLAGS)
LDFLAGS += $(GTKLDFLAGS)
# ----------------------------------------------------------------------------
# Windows flag setup
# ----------------------------------------------------------------------------
ifeq ($(HOST),Windows_NT)
CFLAGS += -U__STRICT_ANSI__
endif
endif
-
isnt that the stuff after line 320 is optional? what if you really set endif on 320?
-
makefile:325: *** Unable to find library for: PNG support (libpng).
I have libpng in PATH .
-
let it out for first, i would say
-
Package libglade-2.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `libglade-2.0.pc'
to the PKG_CONFIG_PATH environment variable
No package 'libglade-2.0' found
Package gtk+-2.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `gtk+-2.0.pc'
to the PKG_CONFIG_PATH environment variable
No package 'gtk+-2.0' found
Package libglade-2.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `libglade-2.0.pc'
to the PKG_CONFIG_PATH environment variable
No package 'libglade-2.0' found
Package gtk+-2.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `gtk+-2.0.pc'
to the PKG_CONFIG_PATH environment variable
No package 'gtk+-2.0' found
Package gthread-2.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `gthread-2.0.pc'
to the PKG_CONFIG_PATH environment variable
No package 'gthread-2.0' found
Package gmodule-2.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `gmodule-2.0.pc'
to the PKG_CONFIG_PATH environment variable
No package 'gmodule-2.0' found
Package lcms was not found in the pkg-config search path.
Perhaps you should add the directory containing `lcms.pc'
to the PKG_CONFIG_PATH environment variable
No package 'lcms' found
makefile:350: *** extraneous `endif'. Stop.
I found this on aminet
http://aminet.net/package/dev/lib/libgtk-mui.i386-aros
But is this needed? I thought netsurf uses Mui already.
-
Try using TARGET=morphos.
-
Take a look at the documentation (http://source.netsurf-browser.org/trunk/netsurf/Docs/BUILDING-AmigaOS). There are a *lot* of dependencies.
-
@__artur:
Which version are u trying to compile ?
The OS4 version (http://www.netsurf-browser.org/downloads/source/ ) does NOT use MUI, if I'm correct, while the MorphOS version does.
You can try to port the OS4 version (read the build instructions, here: http://source.netsurf-browser.org/trunk/netsurf/Docs/BUILDING-AmigaOS )
-
@itix:
TARGET=morphos give the same result.
@0amigan0:
I'm trying morphos version.
How about one make one link lib?
-
@__artur:
The OS4 version should be easier to port.
PS. Do u have the dependancy libraries that netsurf for amiga requires ??
-
1st dependancy, lemon, compiled :)
-
i tryed netsurf on linux and it was orrible.
i think someone could work on OWB for 68k or to help jorg in this version.
-
@artur
What command "uname" returns for you?
-
@itix:
CYGWIN-NT-5.1
-
__artur wrote:
1st dependancy, lemon, compiled :)
Good to know!
Keep us informed on the progress.
-
@seyia: then again probably would be better to take aros or at least mos owb source. os4 owb depends eventually on system functions not present in 3.x and on new reaction. both mui 3.8 and zune is available for 68k.
-
@__artur
Ah, try using HOST=morphos now. Also edit morphos entry in makefile and replace gcc4 by gcc and remove PPC flags CFLAGS. You can also disable some junk from makefile.config and pay attention to link libs.
There are some MorphOS API calls in my MUI backend but if you manage to get it compiling I can help you to fix MorphOS dependencies in the code.
-
Nice to know that there are still capable coders and lots of helpfull people in amiga scene.
It make sense that MorphOS version would be easier to port OS3.9/3.5
And it allways nice to know that long term amiga coders like piru and itix are intersted to help
-
COuld someone advise on the best way to set up a compilation environment - should I try cross compiling, or go native? Either way, which toolchain is recommended? I know DICE and a little bit of the SAS/C environment, but I don't feel that that'll do the job :-)
Any advice appreciated on this before I waste a load of time just on the build environment! :-D
-
bumb
Wonder why people haven't found this more interested. Lack of modern web browser is big problem for OS3.9@ / 3.5
-
You'd probably want geekgadgets with the latest version of gcc you can find for m68K.
-
I've been pretty busy yesterday.
What I did:
I have all dependencies now...yay
What to do:
I'll see if netsurf compiles.
-
GOGOGO!!!
-
Great :D Netsurf would be more usable with 68k cpus than owb
-
wawrzon wrote:
GOGOGO!!!
What he said! :bow:
-
Netsurf for 68k/3.x machines? Yes please and thank you :-)
-
chiark wrote:
COuld someone advise on the best way to set up a compilation environment - should I try cross compiling, or go native? Either way, which toolchain is recommended? I know DICE and a little bit of the SAS/C environment, but I don't feel that that'll do the job :-)
Any advice appreciated on this before I waste a load of time just on the build environment! :-D
I use the GCC environment of Cubic IDE (http://devplex.awardspace.biz/) which is a really good start.
-
Fantastic, thank you DJBase - just setting that up now!
-
"I'll see if netsurf compiles"
:D what did happen?
-
is not that easy 4 sure. lets be patient.
-
I've taken a look at the source a while ago but the make file is a bit complicated for me, it uses a lot of advanced Make syntax. The 1st hurdle I found was what seems to be a Perl script. There's various versions of Perl available on Aminet but I didn't take a look on setting them up. There probably would have to be modifications of the make file on the way it calls the perl script on 3.x too.
-
Perl is awful too. I'd probably stop there...
-
I found THIS (http://search.cpan.org/dist/perl/README.amiga) page for perl on AmigaOS 3.x, if anyone wants to take I look.
I just know they start with "#!" or something:)
-
Jose wrote:
I just know they start with "#!" or something:)
That's the shebang (http://en.wikipedia.org/wiki/Shebang_(Unix)), used to indicate to unix-like shells how to invoke the rest of the script.
#!/usr/bin/perl
for instance, tells the shell to invoke the perl command.
-
Perl is great! Long live perl <3
-
Shebang, wow, I like the name :-D
-
@__artur:
Let us know how u're progressing.
-
I think I won't handle this task.
For example
COMPILE: amiga/font.c
amiga/font.c: In function 'nsfont_position_in_string':
amiga/font.c:113: warning: implicit declaration of function 'ESetInfo'
amiga/font.c:113: error: dereferencing pointer to incomplete type
amiga/font.c:117: warning: implicit declaration of function 'EObtainInfo'
amiga/font.c:117: error: dereferencing pointer to incomplete type
amiga/font.c:118: error: 'OT_GlyphMap8Bit' undeclared (first use in this functio
n)
amiga/font.c:118: error: (Each undeclared identifier is reported only once
amiga/font.c:118: error: for each function it appears in.)
amiga/font.c:125: error: dereferencing pointer to incomplete type
amiga/font.c:135: error: dereferencing pointer to incomplete type
amiga/font.c:137: warning: implicit declaration of function 'EReleaseInfo'
amiga/font.c:137: error: dereferencing pointer to incomplete type
amiga/font.c: In function 'nsfont_split':
amiga/font.c:195: error: dereferencing pointer to incomplete type
amiga/font.c:199: error: dereferencing pointer to incomplete type
amiga/font.c:200: error: 'OT_GlyphMap8Bit' undeclared (first use in this functio
n)
amiga/font.c:209: error: dereferencing pointer to incomplete type
amiga/font.c:214: error: dereferencing pointer to incomplete type
amiga/font.c:216: error: dereferencing pointer to incomplete type
amiga/font.c: In function 'ami_open_outline_font':
amiga/font.c:289: error: dereferencing pointer to incomplete type
amiga/font.c: In function 'ami_unicode_text':
amiga/font.c:322: error: 'RPTAG_APenColor' undeclared (first use in this functio
n)
amiga/font.c:328: error: dereferencing pointer to incomplete type
amiga/font.c:328: error: request for member 'olf_EEngine' in something not a str
ucture or union
amiga/font.c:332: error: dereferencing pointer to incomplete type
amiga/font.c:332: error: request for member 'olf_EEngine' in something not a str
ucture or union
amiga/font.c:333: error: 'OT_GlyphMap8Bit' undeclared (first use in this functio
n)
amiga/font.c:336: error: dereferencing pointer to incomplete type
amiga/font.c:336: error: request for member 'glm_BitMap' in something not a stru
cture or union
amiga/font.c:336: warning: assignment from incompatible pointer type
amiga/font.c:341: warning: implicit declaration of function 'BltBitMapTags'
amiga/font.c:341: error: 'BLITA_SrcX' undeclared (first use in this function)
amiga/font.c:341: error: dereferencing pointer to incomplete type
amiga/font.c:341: error: request for member 'glm_BlackLeft' in something not a s
tructure or union
amiga/font.c:342: error: 'BLITA_SrcY' undeclared (first use in this function)
amiga/font.c:342: error: dereferencing pointer to incomplete type
amiga/font.c:342: error: request for member 'glm_BlackTop' in something not a st
ructure or union
amiga/font.c:343: error: 'BLITA_DestX' undeclared (first use in this function)
amiga/font.c:344: error: 'BLITA_DestY' undeclared (first use in this function)
amiga/font.c:344: error: dereferencing pointer to incomplete type
amiga/font.c:344: error: request for member 'glm_Y1' in something not a structur
e or union
amiga/font.c:344: error: invalid operands to binary - (have 'ULONG' and 'ULONG *
')
amiga/font.c:345: error: 'BLITA_Width' undeclared (first use in this function)
amiga/font.c:345: error: dereferencing pointer to incomplete type
amiga/font.c:345: error: request for member 'glm_X1' in something not a structur
e or union
amiga/font.c:346: error: 'BLITA_Height' undeclared (first use in this function)
amiga/font.c:346: error: dereferencing pointer to incomplete type
amiga/font.c:346: error: request for member 'glm_BlackHeight' in something not a
structure or union
amiga/font.c:347: error: 'BLITA_Source' undeclared (first use in this function)
amiga/font.c:348: error: 'BLITA_SrcType' undeclared (first use in this function)
amiga/font.c:348: error: 'BLITT_ALPHATEMPLATE' undeclared (first use in this fun
ction)
amiga/font.c:349: error: 'BLITA_Dest' undeclared (first use in this function)
amiga/font.c:350: error: 'BLITA_DestType' undeclared (first use in this function
)
amiga/font.c:350: error: 'BLITT_RASTPORT' undeclared (first use in this function
)
amiga/font.c:351: error: 'BLITA_SrcBytesPerRow' undeclared (first use in this fu
nction)
amiga/font.c:351: error: dereferencing pointer to incomplete type
amiga/font.c:351: error: request for member 'glm_BMModulo' in something not a st
ructure or union
amiga/font.c:355: error: dereferencing pointer to incomplete type
amiga/font.c:355: error: request for member 'glm_X1' in something not a structur
e or union
amiga/font.c:355: warning: assignment makes integer from pointer without a cast
amiga/font.c:357: error: dereferencing pointer to incomplete type
amiga/font.c:357: error: request for member 'olf_EEngine' in something not a str
ucture or union
amiga/font.c: In function 'ami_init_fonts':
amiga/font.c:377: warning: implicit declaration of function 'OpenOutlineFont'
amiga/font.c:377: error: 'OFF_OPEN' undeclared (first use in this function)
amiga/font.c:377: warning: assignment makes pointer from integer without a cast
amiga/font.c:378: warning: assignment makes pointer from integer without a cast
amiga/font.c:379: warning: assignment makes pointer from integer without a cast
amiga/font.c:380: warning: assignment makes pointer from integer without a cast
amiga/font.c:381: warning: assignment makes pointer from integer without a cast
amiga/font.c:382: warning: assignment makes pointer from integer without a cast
amiga/font.c:383: warning: assignment makes pointer from integer without a cast
amiga/font.c:389: error: dereferencing pointer to incomplete type
amiga/font.c:389: error: request for member 'olf_OTagList' in something not a st
ructure or union
amiga/font.c:389: warning: initialization from incompatible pointer type
amiga/font.c:391: warning: assignment makes pointer from integer without a cast
amiga/font.c:398: error: dereferencing pointer to incomplete type
amiga/font.c:398: error: request for member 'olf_OTagList' in something not a st
ructure or union
amiga/font.c:398: warning: initialization from incompatible pointer type
amiga/font.c:400: warning: assignment makes pointer from integer without a cast
amiga/font.c:407: error: dereferencing pointer to incomplete type
amiga/font.c:407: error: request for member 'olf_OTagList' in something not a st
ructure or union
amiga/font.c:407: warning: initialization from incompatible pointer type
amiga/font.c:409: warning: assignment makes pointer from integer without a cast
amiga/font.c: In function 'ami_close_fonts':
amiga/font.c:424: warning: implicit declaration of function 'CloseOutlineFont'
-
It wasn't ever going to compile without modifications. The port will require removing/replacing the code (in either the os4 or morphos version) for stuff that isnt available on os3.x
-
__artur:
why, i see you have made some progress with itix backup. just do-not-surrender.
-
@artur
I have a ttfengine.library version of font.c if you want it?
It's horribly slow as it opens the fonts each time rather than once at the start, but that's easily fixable - it probably won't compile until this is fixed anyway as I've abstracted the font code a little more since then.
Otherwise it's a case of opening bullet.library and replacing the E..() calls with the equivalents from bullet, and reading and plotting the glyphs manually.
Chris
-
@chris: great to have all fractions helping here.
@artur:
an actual discussion on amiga-news.de regarding a restricted os3.x unicode rendering implementation can be found here, if that is the problem, that is:
http://www.amiga-news.de/forum/thread.php?id=31281&BoardID=7
(unfortunatelly in german, but skilled coders involved)
-
@chris
I noticed this on my MorphOS port of Netsurf where I had to implement font caching wih TTEngine. Works very well and there are also some advantages in using TTEngine rather than plotting glyphs manually.
-
writing a font output stub and getting it to compile would be a big step - I'll see if I can get the x-dev environment set up properly...
-
What is hardware/atomic.h and is for os3.x ?
-
atomi.h is only in MorphOS SDK and implements atomic operations (sub, add and so on) in PPC assembly. I have implemented generic replacement for that header file now, just ask on IRC.
-
itix when will you be on IRC?
I need help.
-
What's the best IRC channel for this sort of discussion? :D
-
Chiark, #amigadev at amigaworld.net ?
-
@__artur:
Any news ??
-
MHO: Too much talking about a web broswer that doesn't support unicode or even javascript and the only advantage of using it instead of Ibrowse or AWeb is to have CSS layout and NOTHING else... Unless i'm missing something.
-
Why not use for 68k porting the latest snapshot here and then put all in the main CVS ?
http://www.netsurf-browser.org/downloads/source/
It contains the latest Core/OS4 development too and it seems quite stable, surely better than the MorphOS release :-)
-
Yes, its too no full browser but now that you begin and manage to compile it, i like to see how fast it is ;-)
My most time consuming problem is to get something compilable.Find a problem in a compilable program and fix is much easier for me and need not much time.i spend not much time on programming.
currently i cant compile your source, but because of your near complete files its more easy.
You get big progress.
And i think to have something workable is not much work.most you have done.i think some mui classes need deactivate, but to get it working so that a name can in and surf should not be need much MUI4 features.and if this later need can add to zune.
Problem now is that newcreateproc need a special MOS command, but i think this need not add, because multithread seem in the OS4 Version too not done.OWB too use not multithread
with multithreaded define off its more easy to get it working first.
i am postive suprised about netsurf.
the good news is, it most use no C++, and compile time is fast, also when do a full compile.
And thats most important to me.OWB need more than 30 sec to find out what file is changed, and wait such a long time, make no fun.
when a header file is changed in OWB, then a compile to 50% need over 20 minutes also on my fast AMD64 3000+ (1,8 GHZ real clock)on crosscompilers and 1 gigabyte ram.make too no fun.
btw i like to see a video how OWB compile on a OS4 system or MOS system, i cant believe that this is usable fast, and i think OWB is develop on a more core X86 system with over 3 GHZ. ;-)
I wonder if it is not possible to add with few work a better than nothing java script.when i look on pages, i guess 99% need to be usable only few simple commands.
aweb have a java script, and but miss more, so it seem get Java working not too hard, so wy should not get working a java script that fit 99% the need to be usefull in short time ?
the OS4 Version with reaction i dont like, because reaction is far behind MUI in any case.
it should be possible to have a MUI class from the browser, a GUI can be much more easy code in amiblitz at least for me, and also a MUI class can use in many other programs, not sure if MOS or OS4 GUI is fully working in all features, or some features have no GUI elements on MOS or OS4 Version but on GTK Version.
-
MHO: Too much talking about a web broswer that doesn't support unicode or even javascript and the only advantage of using it instead of Ibrowse or AWeb is to have CSS layout and NOTHING else... Unless i'm missing something.
It does support Unicode.
-
Why not use for 68k porting the latest snapshot here and then put all in the main CVS ?
http://www.netsurf-browser.org/downloads/source/
It contains the latest Core/OS4 development too and it seems quite stable, surely better than the MorphOS release :-)
I don't wish to sound rude, but have you read the discussion in this thread? Back porting OS4 to OS3 is seriously non-trivial: surprisingly, porting MorphOS to OS3.X looks a lot more straightforward.
-
I have compiled Netsurf (thanks Bernd_afa for help) !
It has bugs like wrong collors on pictures or problems with text input.
But it is faster than OWB.
Netsurf uses truetype fonts so you have to install AfaOS or other truetype handler.
Fonts are included. Copy them to Fonts:TRUETYPE/ttf-bitsream-vera folder.
Use 32bpp screen depth.
I release it for testing .
Please tell how it works.
http://sites.google.com/site/amigacorner/Home/Netsurf2.zip?attredirects=0
-
It crashes instantly on a 16-bit screen and quits instantly if the fonts aren't available at the exact path specified in the readme, but I did get it functioning on a 24-bit screen. And it's quite fast on my 060.
Well done!
Suggestions:
Change the env-vars to tooltypes or at least use the path ENV:Netsurf/[env-var] instead of ENV:[env-var].
Use Fonts:_ttf for the fonts path, or allow it to be specified by the user
General cleanup.
[I also just noticed that this thread is in the Marketplace forum - could one of the moderators move it to Software?]
-
It starts but I don't get any connection.
-
very well! display is trashed unusable on most color depths, it works with partly wrong colors on32bit though (as described). complicated pages do not render quite correctly but neither completely wrong, so its going in the right direction (amiga.org renders in about 28s) there is always an exeption on exit pointing to ixemul.library (61.1). if i comment smart crash out in my ss i might maybe catch more details about it, but supposedly you know yourself. no hits while execution. seems quite stable despite still slow, but doesnt block the system.
ps:many thanks to you and bernd!!
ps: on 16bit it seems to crash the system. on16bit_pc it throws a lot of hits. i recorded hits that occure on quit. (the offsets seem always to be the same) as follows:
USP : 0920792C SR: 0004 (U0)(F)(-) TCB: 09167470
Data: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
Addr: 08000ADA 08BB1520 00000000 00000000 00000000 00000000 3A67616D 080023B4
----> 0853221C - "Work:C/Assign" Hunk 0000 Offset 00000214
Stck: 65747375 72663200 6E736662 2D73646C 00000000 00028600 0241523F 70FF4E75
Stck: 4AFC0920 794C0920 7966803D 09000920 79660920 79750920 79B06978 656D756C
Stck: 2E6C6962 72617279 00697865 6D756C20 36312E31 205B6E6F 74726170 2C203638
Stck: 3032302C 20363838 38312C20 616D6967 616F735D 20283139 2E332E32 30303929
Stck: 0D0A0000 000002E4 092079C0 09208468 09208494 092084AA 092084CA 092084EA
Stck: 092084E6 09224CA8 09224CD4 09208EFC 092206C4 09224D64 09224D88 09224D98
Stck: 09224DB0 0921F584 0921F5D0 09224DC8 0921F64C 09224E18 0922081E 09220814
Stck: 09220858 09224E4E 09220836 09220828 09220844 0920902C 09209328 092095C0
----> 09816474 - "Leer1:games/Netsurf2/nsfb-sdl" Hunk 0000 Offset 000002B4
----> 092079C0 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00000078
----> 09208468 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00000B20
----> 09208494 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00000B4C
----> 092084AA - "Work:Libs/ixemul.library" Hunk 0000 Offset 00000B62
----> 092084CA - "Work:Libs/ixemul.library" Hunk 0000 Offset 00000B82
----> 092084EA - "Work:Libs/ixemul.library" Hunk 0000 Offset 00000BA2
----> 092084E6 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00000B9E
----> 09224CA8 - "Work:Libs/ixemul.library" Hunk 0000 Offset 0001D360
----> 09224CD4 - "Work:Libs/ixemul.library" Hunk 0000 Offset 0001D38C
----> 09208EFC - "Work:Libs/ixemul.library" Hunk 0000 Offset 000015B4
----> 092206C4 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00018D7C
----> 09224D64 - "Work:Libs/ixemul.library" Hunk 0000 Offset 0001D41C
----> 09224D88 - "Work:Libs/ixemul.library" Hunk 0000 Offset 0001D440
----> 09224D98 - "Work:Libs/ixemul.library" Hunk 0000 Offset 0001D450
----> 09224DB0 - "Work:Libs/ixemul.library" Hunk 0000 Offset 0001D468
----> 0921F584 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00017C3C
----> 0921F5D0 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00017C88
----> 09224DC8 - "Work:Libs/ixemul.library" Hunk 0000 Offset 0001D480
----> 0921F64C - "Work:Libs/ixemul.library" Hunk 0000 Offset 00017D04
----> 09224E18 - "Work:Libs/ixemul.library" Hunk 0000 Offset 0001D4D0
----> 0922081E - "Work:Libs/ixemul.library" Hunk 0000 Offset 00018ED6
----> 09220814 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00018ECC
----> 09220858 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00018F10
----> 09224E4E - "Work:Libs/ixemul.library" Hunk 0000 Offset 0001D506
----> 09220836 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00018EEE
----> 09220828 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00018EE0
----> 09220844 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00018EFC
----> 0920902C - "Work:Libs/ixemul.library" Hunk 0000 Offset 000016E4
----> 09209328 - "Work:Libs/ixemul.library" Hunk 0000 Offset 000019E0
----> 092095C0 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00001C78
PC-8: 2F40000C 2C780004 226F000C 4EAEFE86 70142F40 0010202F 00102C5F 4FEF001C
PC *: 4E75514F 42A74879 09110A56 48790911 089C4879 09110898 4AAF001C 670A2F7C
09816454 : 2f40 000c move.l d0,$c(a7)
09816458 : 2c78 0004 movea.l $4.w,a6
0981645c : 226f 000c movea.l $c(a7),a1
09816460 : 4eae fe86 jsr -$17a(a6)
09816464 : 7014 moveq.l #$14,d0
09816466 : 2f40 0010 move.l d0,$10(a7)
0981646a : 202f 0010 move.l $10(a7),d0
0981646e : 2c5f movea.l (a7)+,a6
09816470 : 4fef 001c lea.l $1c(a7),a7
09816474 : *4e75 rts
09816476 : 514f subq.w #$8,a7
09816478 : 42a7 clr.l -(a7)
0981647a : 4879 0911 0a56 pea.l $9110a56
09816480 : 4879 0911 089c pea.l $911089c
09816486 : 4879 0911 0898 pea.l $9110898
0981648c : 4aaf 001c tst.l $1c(a7)
09816490 : 670a beq.s $981649c
09816492 : 2f7c 0911 0a5e 0014 move.l #$9110a5e,$14(a7)
Name: "nsfb-sdl" Hunk 0000 Offset 000002B4
@djbase: you sure have tcp running?
-
@__artur:
Thanks for your effort!
NetSurf is really fast on my Workbench 3.9 AFA_OS with 32 Meg RAM !
It works well, although the images appear in negative; anyone else experienced that ??
Also it seems that the Slashdot layout is not stellar.
Anyway, I like it; keep up on improving it !!
-
Also it seems that the Slashdot layout is not stellar.
That's a general bug - also in the OS4 version.
-
you sure have tcp running?
I tested on AmiKit with WinUAE. Ibrowse and AWeb works, but Netsurf gives me always unresolve errors. I there a secret setting I forgot?
-
I tested on AmiKit with WinUAE. Ibrowse and AWeb works, but Netsurf gives me always unresolve errors. I there a secret setting I forgot?
It's the same as with OWB under winuae.
Download AmiTCP from aminet. Copy all install files to a drawer on your
harddisk and assign amitcp: to this drawer.
Then it will work.
Edit: There's no need to really install AMITCP. Just copy all files to a drawer on your hd. And yes, you still have to use bsdsocket emulation.
-
@artur and bernd
Thank you very much for your efforts. Hope you will continue to improve it.
-
I have compiled Netsurf (thanks Bernd_afa for help) !
It has bugs like wrong collors on pictures or problems with text input.
But it is faster than OWB.
Netsurf uses truetype fonts so you have to install AfaOS or other truetype handler.
Fonts are included. Copy them to Fonts:TRUETYPE/ttf-bitsream-vera folder.
Use 32bpp screen depth.
I release it for testing .
Please tell how it works.
http://sites.google.com/site/amigacorner/Home/Netsurf2.zip?attredirects=0
Wow, massively ace.
Ofcourse this means I'm going to have to go round my parents house and dig out my A1200T.
Keep bashing away at it lads!
-
>It's the same as with OWB under winuae.
>Download AmiTCP from aminet. Copy all install files to a drawer on >your harddisk and assign amitcp: to this drawer.
>Then it will work.
yes, netsurf use too ixnet.library.but its possible to use bsdsocket later, when libcurl is fix.
netsurf work too with libcurl for bsdsocket, but give on my winuae system not more speed and some enforcer hits.
on the left below side, when the mouse is not over a link, the time is show a page is load.
what we need to know are there delay times on classic or is it maybe a winuae only socket problem.or work netsurf on this pages also on other system extrem slow.
can somebody please post the time value a page rebuild need show in left corner when click on netsurf_web_site link.
my winuae amd64 3000+ need for this 1,5 -1,6 sec.
when i disable the jit, then i get same CPU mips as a 68040 25.
then time take 6,4 sec.
also try press on reload look if time differ much.i get time from 8,2 to 6,4 sec.
now go back to start site and click on wikipedia.with JIT site is show in 3,6 sec.
without JIT, there is about 18 sec show
0.1sec load 0.00 kb from 20 kb.on some pages i notice how the time increase.but not here, then after 22 sec teh page is show.
i am not sure if there are hangs and internet access take unecessary delay.the delay and hang problem i have on OWB 68k too.but OWB show no page load time.
then the last test, click on the reuters link.
this is also with JIT on my system a unusable page.it load 5 sec, thats ok, but seem load more images after page is load and show.
so i cant scroll (with cursore keys press very often nice) and get always this 5 sec hangs when a new image is load.and it load again and again images.
does this happen on other Systems too ?
also reports from MOS or OS4 systems are intresting, what time values show here.
and after tests its always good when you reboot, this is only a quick sdl port and should show what netsurf can
the version for MOS and source can Artur compile, but it dont work, and wait.
Now we get the idea to try out with the mos sources the TARGET=framebuffer for SDL.this show only a black screen.
but netsurf work well with the checked out sources from netsurf homepage and thats the testversion here.
but the future(if it make sense to port newtsurf complete)is to get the MOS version working but first need find out wy the framebuffer not work with MOS sources, and use only most important MUI code and classes.
-
@bernd:yes there are sites that load extremly long or do not load at all, like bild.de you pointed me once to. i wondered if this is maybe because netsurf doesnt get enough memory allocated or it requests further objects too lame, but this sounds bs.
to load netsurf_web_site my 060/50 setup needs 4.9 s, this is on a clean first attempt, but if you have surfed around and then tried the link first the actual time can vary around 5-6sec. same for reload (around 5sek)
wikipedia loads in 18,x s at best, but sometimes it takes up to 30s. idlea68k shows constant full cpu load while processing page.
reuters page loads basically some 14s (it only shows 4.9s shortly after) an then it loads further content. i have some hard drive activity like something is being cashed. i can scroll a little but it takes ages and netsurf seem to update the layout all the time. but maybe its due to my impatience. its a huge page too.
edit: well no, i have it open for some time now, it is rendered but being updated again and again
btw:i dont understand why you care for mos so much out of the sudden:P there is mos native port worked on by itix afaik, mos is surely better off with that. not that you sholdnt join the forces though.
-
to load netsurf_web_site my 060/50 setup needs 4.9 s, this is on a clean first attempt, but if you have surfed around and then tried the link first the actual time can vary around 5-6sec. same for reload (around 5sek)
Great, then your system and netsurf is faster as my OWB 68k on winuae.it need with jit 6 sec and OWB 68k do not antialias.without JIT this page need 92 sec.this page use no Java so can good compare.
Text look ugly on OWB 68k but netsurf have nice antialiased text.this page contain many text.
A build without freetype and use internal netsurf font system give speedup, because do no antialias, but it do not work, because no text is show.i dont know wy it not work.
also gcc add for antializing font bitfield operations, not shure how freetype is compiled, but -mnobitfield option need set for faster speed
does somebody know how can easy disable antialiasing in freetype so Artur can build a antialias free version too ?
the hubbub HTML engine used by this netsurf build print out also many lines to console, or when start from wb in nil.but some unneeded code is execute, here its also possible that it work faster.
The reuters page can be work ok, if the stop button work correct(blue x).but when i get some time the mouse pointer can move, and i have luck so i can click on it, it do not help, page load show load show.but when i scroll it seem all elements are here.
netsurf do very often memallocs.this is see because when run memtracker as wipeout give huge slowdown(4*), also on my fast winuae with 800 megabyte memtransfer rate.on amiga apps there is not such slowdown see
so better for tests on a classic not use wipeout or other memtracker
>btw:i dont understand why you care for mos so much out of the >sudden:P there is mos native port worked on by itix afaik, mos is >surely better off with that. not that you sholdnt join the forces >though.
yes and so i like to get 1 sourcetree.itix is very friendly, he help Artur and have some code in his source add that make it easier to work on OS3.x and MOS1.4.He have written in netsurf source, there seem working together possible.
but because we was not sure what problem is we want see netsurf running to see libcurl libxml libhubbub and the many other libs run well.and now can see the browser work and the way to a full browsers is GUI code only.
the sdl version you see have only very few elements.
-
so better for tests on a classic not use wipeout or other memtracker
i disabled all debug in background now but it doesnt make much difference if any at all.
yes and so i like to get 1 sourcetree.itix is very friendly, he help Artur and have some code in his source add that make it easier to work on OS3.x and MOS1.4.He have written in netsurf source, there seem working together possible.
this attitude is much appreciated.
but because we was not sure what problem is we want see netsurf running to see libcurl libxml libhubbub and the many other libs run well.and now can see the browser work and the way to a full browsers is GUI code only.
the sdl version you see have only very few elements.
there are still bugs to iron out. colors. text input. rendering (check http://www.bahn.de). not loading pages. i do not know how much of this is general fault of current netsurf distribution. replacing sdl with native functions might bring some speedup but i dont take it for granted. gui would be nice too, of course.
-
there are still bugs to iron out. colors. text input. rendering (check http://www.bahn.de). not loading pages. i do not know how much of this is general fault of current netsurf distribution. replacing sdl with native functions might bring some speedup but i dont take it for granted. gui would be nice too, of course.
oh yes, the bahn.de show wrong.i deactivate on my Internet explorer always java script and active x.its my no Java browser ;-)
The http://www.bahn.de page work without java script ok(village name completion is not done but on OWb 68k too no village completion), later when you add start and destination, is show a message activate java script or click here.when click on here, then all work too without java script.
what happen on this page with a netsurf OS4 or MOS Version ?
that the colors are wrong is because endian problems.
its really bad that OWB is program in C++.i thinks thats main reason that it is so slow and compile time is so extrem long.
and i need also much longer to find Bugs in a C++ source than in a fast c source.because when singlestep in debugger c++ source need lots more instruction for same functions
-
what happen on this page with a netsurf OS4 or MOS Version ?
that the colors are wrong is because endian problems.
that is clear. it shouldt be too hard to fix it, right?;-D
its really bad that OWB is program in C++.i thinks thats main reason that it is so slow and compile time is so extrem long.
and i need also much longer to find Bugs in a C++ source than in a fast c source.because when singlestep in debugger c++ source need lots more instruction for same functions
i see. bad thing: it is seems c has been abandoned already. most available sources are >c++ for what ive seen.
-
@bernd_afa & @__artur:
How difficult is it to make a MUI/Zune version, instead of SDL ??
-
@bernd_afa & @__artur:
How difficult is it to make a MUI/Zune version, instead of SDL ??
cant say, as i told, the MUI version with MOS source can compile with MUI3.8 headers, so there seem no MUI4 parts in.but it still not work.
now must begin examine how netsurf work and what go wrong wy MUI sources not work.the good news is, it can compare in debugger with the working sdl version.
the bad thing is, i never use, and must learn from beginning how MUI work.
I dont like GUI without GUI editor.in the world outside amiga there are GUI editors used, (called RAD Tools rapid application development)
So i get the idea to test the usability of the gtk to zune wrapper and try a gtk netsurf compile.this help also to get more Unix programs easier port.
Help is welcome.if that work, then i like to have the glade GUI Editor, do somebody know a more easy portable GUI Editor ?
http://glade.gnome.org/
and at end of all, i also want that there can GUI for amiblitz programs build by using libglade
stormwizard i use for my programs, can easy build with editor, but they look not nice, MUI have no GUI Editor so i see the modernst solution to use GTK MUI wrapper and a GTK GUI Editor
-
I compiled with -mnobitfield like Bernd suggested.
Some dependency libs are compiled with opimisations.
Maybe it is faster now, please test.
Also it probaly doesn't crash at exit now.
-
it isnt noticeably faster alas.
the sashimi output on exit looks like that now:
09-Jun-09 19:30:11
WORD READ from 73757266 (INST) PC: 73757266
USP : 0A885068 SR: 0004 (U0)(F)(-) TCB: 0A7E4BA8
Data: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
Addr: 08000ADA 08BC3178 00000000 00000000 00000000 00000000 616D6573 080023B0
Stck: 325F3100 6E736662 2D73646C 00000000 0963BF38 00000000 00000003 00000001
Stck: 00000000 00000000 096A37B8 00000000 00000000 096A382C 00000B18 09609F2C
Stck: 00000000 00000000 00000000 00000000 00000000 00000A88 50C80000 00000000
Stck: 0A8850B4 00000000 00000000 00000000 00000000 00000000 00000000 00000000
Stck: 00000000 00000000 00000000 00000000 00000000 00000000 00000A88 510C0000
Stck: 00000000 0A8850F8 00000000 00000000 00000000 00000000 00000000 00000000
Stck: 00000000 00000000 00000000 00000000 00000000 005C0871 123441C0 0002FFFE
Stck: FFFE0000 00000B18 4A2EB0CA 00000000 4A2EB0CA 00000000 4A2EB0CA 0871005C
----> 096A37B8 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00017B60
----> 096A382C - "Work:Libs/ixemul.library" Hunk 0000 Offset 00017BD4
PC Address invalid
Name: "nsfb-sdl"
09-Jun-09 19:30:11
Exception !! 00000002 TCB: 0A7E4BA8 CTX: 08178A90 SSP: 080023B0
USP : 0A885068 SR: 0004 (U0)(F)(-) TCB: 0A7E4BA8
Data: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
Addr: 08000ADA 08BC3178 00000000 00000000 00000000 00000000 616D6573 080023B0
Stck: 325F3100 6E736662 2D73646C 00000000 0963BF38 00000000 00000003 00000001
Stck: 00000000 00000000 096A37B8 00000000 00000000 096A382C 00000B18 09609F2C
Stck: 00000000 00000000 00000000 00000000 00000000 00000A88 50C80000 00000000
Stck: 0A8850B4 00000000 00000000 00000000 00000000 00000000 00000000 00000000
Stck: 00000000 00000000 00000000 00000000 00000000 00000000 00000A88 510C0000
Stck: 00000000 0A8850F8 00000000 00000000 00000000 00000000 00000000 00000000
Stck: 00000000 00000000 00000000 00000000 00000000 005C0871 123441C0 0002FFFE
Stck: FFFE0000 00000B18 4A2EB0CA 00000000 4A2EB0CA 00000000 4A2EB0CA 0871005C
----> 096A37B8 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00017B60
----> 096A382C - "Work:Libs/ixemul.library" Hunk 0000 Offset 00017BD4
PC Address invalid
Name: "nsfb-sdl"
so, it seems its been improved. it isnt deadly anyway. if you ask me the stability of the system is not affected at all.
-
I can't believe it. This netsurf also uses ixemul?
Artur: just don't use ixemul, please (add -noixemul at compilation and link). You're just asking for deep trouble, otherwise.
Bernd: I really question your development/porting advices. :)
-
I can't believe it. This netsurf also uses ixemul?
Artur: just don't use ixemul, please (add -noixemul at compilation and link). You're just asking for deep trouble, otherwise.
Bernd: I really question your development/porting advices. :)
@fab
with libnix netsurf crashes immidiate test with the MOS source in that func in gui_init.
void gui_init(int argc, char** argv)
{
....
messages_load(lang); // check locale language and read appropriate file
but forget to test with framebuffer.maybe Artur can compile and see if it work.
The MOS libnix source is not released.
maybe you release libnix source from MOS, then i compile that and look if it work better.
and btw.the netsurf sdl port use only 1 thread so it should not be critical.
-
>i am not sure if there are hangs and internet access take unecessary delay.
One of the basic rules of benchmarking anything is to remove the uncertainty or your results will be totally useless. Everyone and their cousin knows this...
Doing any kind of guestimates which can be affected by the state of the internet connection is silly.
Solution: Remove the internet connection from the equation, use a locally cached copy of the page you want to benchmark. Only then you will be able to draw any kind of conclusions.
After that, the very first move is to get rid of ixemul usage. It just will not work reliably with AmigaOS code.
-
@fab
with libnix netsurf crashes immidiate test with the MOS source in that func in gui_init.
Have you actually figured out why it crashes? Surely developer of your skill is able to do this?
-
with libnix netsurf crashes immidiate test with the MOS source in that func in gui_init.
void gui_init(int argc, char** argv)
{
....
messages_load(lang); // check locale language and read appropriate file
but forget to test with framebuffer.maybe Artur can compile and see if it work.
The MOS libnix source is not released.
maybe you release libnix source from MOS, then i compile that and look if it work better.
Why not figure out what instructions crash and replace them with the same ones from vbcc's posixlib ???
These are freely available.
x303 :D :D :D
-
Arent there working mui gui builders?
Wasn't there one called MUIBuilder (http://aminet.net/package/dev/mui/MUIBuilder22)? :)
-
One of the basic rules of benchmarking anything is to remove the uncertainty or your results will be totally useless. Everyone and their cousin knows this...
it count not the exact time, because in real world internet load time can differ around 1-2 sec on fast server(as the netsurf and other sites i name are.)
Everybody knows that when a page on a amiga/ winbrowser can load in 1-2 sec and OWB need for this 90sec(without java script), then the influence of internet acess is about 1-2% at overall time, or something is bad in code.
and only this value is usefull how much slower it is in compare to aweb or OWB and if it is usefull
>Have you actually figured out why it crashes? Surely developer of >your skill is able to do this?
i have no time and fun to do so, Artur build first libnix and ixemul version too that crash.I look on both, but soon i notice libnix crash earlier.
When i port a Unix program i choose always the best solution.and this is not use libnix and add the missing functions, this is use a bsd kernel on amiga os, (ixemul)
>After that, the very first move is to get rid of ixemul usage. It just >will not work reliably with AmigaOS code.
and wy not ?
you cant compare MOS ixemul with 68k ixemul.
MOS ixemul have 2 very bad bugs, that make it very unstable, i find and fix that.i can tell you what places make problems.
when you use pngcrush and a picture that ami_stuff send me, and the MOS largeread code is enable, then there is extrem slowdown (more than 10*).it happen only on 1 picture.i deactivate that(ixmeul V48 have this code not) and all work ok
#if LARGEREADS
r = fp->_r;
/* TODO: should limit to specific buffering modes? - Piru */
if (resid >= r + fp->_bf._size) {
/* empty current buffer first, if a
--------------------
then next is,
in __write.c last MOS source have disable IXTTY_RAW.
line in V48 was this.
#define TTY_NLCR_ENABLE (IXTTY_RAW | IXTTY_OPOST | IXTTY_ONLCR)
when do a fh = open(audio:xxxxxxx) and do write(fh,buf,len)
give crap audio data hear on ahi.
use the V48 code fix that.
also i get now information wy ixemul V48 cant work on MOS even if MOS is called that it can execute 68k software.
its due to emulation bug.yes i call it a Bug, because when the AOS supervisor call is not support by 68k then MOS should nicely stop the program and tell supervisor is not support on 68k.
here is log output i get from a user and symbolinfo is in ixemul in.Here you can verify yourself hits come from supervsior if you dont believe that.
-> mlnet Hunk 1 Offset 0x000d0fe8
PPCStackFrame History:
StackFrame[ 0].LR-> Address 0x25660c70 -> mlnet Hunk 1 Offset 0x000c46d0
StackFrame[ 1].LR-> Address 0x25661860 -> mlnet Hunk 1 Offset 0x000c52c0
StackFrame[ 2].LR-> Address 0x2566d5bc -> mlnet Hunk 1 Offset 0x000d101c
StackFrame[ 3].LR-> Address 0x25668efc -> mlnet Hunk 1 Offset 0x000cc95c
StackFrame[ 4].LR-> Address 0x25668f50 -> mlnet Hunk 1 Offset 0x000cc9b0
StackFrame[ 5].LR-> Address 0x2565c2dc -> mlnet Hunk 1 Offset 0x000bfd3c
StackFrame[ 6].LR-> Address 0x2565c3e0 -> mlnet Hunk 1 Offset 0x000bfe40
StackFrame[ 7].LR-> Address 0x22f0bd58 -> MOSSYS:LIBS/ixemul.library Hunk 1 Offset 0x000088f0
StackFrame[ 8].LR-> Address 0x22f156b8 -> MOSSYS:LIBS/ixemul.library Hunk 1 Offset 0x00012250
StackFrame[ 9].LR-> Address 0x22f0bc1c -> MOSSYS:LIBS/ixemul.library Hunk 1 Offset 0x000087b4
StackFrame[10].LR-> Address 0x10121970 -> Module Hunk 0 Offset 0x00021970
StackFrame[11].LR-> Address 0xdeadfee1 **Not Valid**
MOS ixemul use that code and i guess its a bad hack and can break compatibility, when just set sr to 0 instead of getting the correct value.
#ifdef NATIVE_MORPHOS
sr = 0;
#else
/*#ifdef MORPHOS
if (has_morphos)
sr = 0;
else
#endif*/
{
asm volatile (" \n\
movel a5,a0 \n\
lea Lget_sr,a5 \n\
movel 4:w,a6 \n\
jsr a6@(-0x1e) \n\
movel a1,%0 \n\
bra Lskip \n\
Lget_sr: \n\
movew sp@,a1 | get sr register from the calling function \n\
rte \n\
Lskip: \n\
movel a0,a5 \n\
" : "=g" (sr) : : "a0", "a1", "a6");
}
#endif
I change ixemul now on two places to use AOS command.setsr.
important is that use gcc inline LPT stubs to not change flags on call
sr=SetSR(0,0);
and some tests show this place make no problems.ixemul V61.2 contain now no supervisor calls.
now problem when run under MOS is ixpipe.the MOS version have add code and fake a wbstartupmessage, because MOS ixemul want read on ixpipe-handler wbstartupmessage.
fake a message i think is a unclean solution and i dont want 2 diffrent ixpipes to test more easy with V48 and V50 and above.
/* pr_CLI is NULL since we are a handler, so ix_open will expect a wb message. Fake one. */
port.mp_Flags = PA_IGNORE;
NEWLIST(&port.mp_MsgList);
msg.sm_Message.mn_ReplyPort = &port;
msg.sm_Process = &me->pr_MsgPort;
msg.sm_NumArgs = 0;
msg.sm_ToolWindow = NULL;
msg.sm_ArgList = 0;
PutMsg(&me->pr_MsgPort, &msg.sm_Message);
ixbase = (APTR)OpenLibrary ("ixemul.library", NEEDED_IX_VERSION);
------
so new ixemul version need also correct ixpipe 68k for now.
more progress reports i dont get, but MOS OS4 is promotet to have a 68k emul and so it should execute 68k code and nothing else is ixemul.
-
Arent there working mui gui builders?
Wasn't there one called MUIBuilder (http://aminet.net/package/dev/mui/MUIBuilder22)? :)
yes but this are too old and not furtherdevelop and add too code in programs
i think it btw not usefull that a GUI is in program code.it should be loadable so when change something in GUI the program need not new compile.a large GUI compile need lot time.
on other systems there are gui editors and change GUI data need no recompile.
storm wizard offer this modern features too on amiga since long time, but sad, that it is not furtherdevelop
-
it count not the exact time, because in real world internet load time can differ around 1-2 sec on fast server(as the netsurf and other sites i name are.)
I guess you missed this before so I shall repeat it:
Solution: Remove the internet connection from the equation, use a locally cached copy of the page you want to benchmark. Only then you will be able to draw any kind of conclusions.
Have you actually figured out why it crashes? Surely developer of your skill is able to do this?
i have no time and fun to do so
Yet you have time to give bad advice and talk smack? How nice of you. You know it would be much more constructive to actually locate the bug and fix/workaround it rather than talk bull manure about ixemul and libnix.
When i port a Unix program i choose always the best solution.and this is not use libnix and add the missing functions, this is use a bsd kernel on amiga os, (ixemul)
Netsurf isn't a unix program anymore, at least if you want to have any kind of usable GUI with it (SDL one doesn't quite count as usable). ixemul is only usable for things that you don't actually port: for recompiles (configure && make). As soon as you start actually porting you get rid of ixemul as the first step.
After that, the very first move is to get rid of ixemul usage. It just will not work reliably with AmigaOS code.
and wy not ?
I've explained this before: AmigaOS code doesn't mix with ixemul. Regardless what you've supposedly done to "fix" this in your V600 ixemul versions does not really fix this fundamental design choice.
you cant compare MOS ixemul with 68k ixemul.
Yes I can. I've actually developed ixemul for years. I know the internals better than most. I can say for a fact that ixemul and AmigaOS code do not mix. Nothing I've seen from you has changed this fact, either.
--- Irrelevant drivel not worth commenting cut off ---
-
Yes I can. I've actually developed ixemul for years. I know the internals better than most. I can say for a fact that ixemul and AmigaOS code do not mix. Nothing I've seen from you has changed this fact, either.
yes you have develop ixemul for years, but what you have done is unclear.
The code in ixemul for MOS have code to build also a amigaos 68k versions.there are lots of ifdef morphos in.for example see in file uname.c
#if defined(__MORPHOS__)
static char sysname[] = "MorphOS";
#else
static char sysname[] = "AmigaOS";
#endif
but at end of all, the amiga os build from MOS source do not work with many programs on 68k.
So when you say you have done of ixemul MOS for years and not only do some few small fixes and test only with the few programs run on MOS or if one of the the few users report a Bug on MOS , i cant believe that you understand ixemul well, because the amiga os build run very bad, when do a simple compile. your current sourcetree with changelog is not visible for all, so cant see what you have done and what not.
>Yet you have time to give bad advice and talk smack? How nice of >you. You know it would be much more constructive to actually >locate the bug and fix/workaround it rather than talk bull manure >about ixemul and libnix.
yes i know that, but the MOS extension libnix source is not release for all as the old source, seem another weapon of red versus blue war.
I think MOS libnix source have this fix.
and i dont want have something to do with that stupid amiga PPC guy fanatic, that every side must reinvite the wheel again and no working together is possible.
this bad behaviour i dislike and i think was the main reason wy most of the millions amiga users are gone, and one good side of it, is that i never get the idea to buy such a PPC system, not that they use my money to build a cannon in the red versus blue war.I dont want support low quality system that want force users to buy the system with unclean methods.
and what some PPC OS devs say about ixemul is unclean.
So bring an argument wy it is impossible that ixemul run with amiga funcs ?
I like perfect systems and if its not perfect, i like when i can fix it to be perfect.
so if i need something fix allone i do that on the best solution that is here.and thats ixemul for 68k.
>I guess you missed this before so I shall repeat it:
as i told, eveybody can check with a fast windows PC and a fast windows browser(safari firefox) how fast it can be on his internet connect.and when he use amiga software and do the same test 5-10 times again then he get a average value and can say around 50% accurate how fast amiga browsers in compare to firefox etc are.
-
the amiga os build from MOS source do not work with many programs on 68k.
Now there's you problem: It was never intended to. You asked for the MorphOS ixemul source. This is what you got.
--- removed tons of irrevevant drivel, again ---
the MOS extension libnix source is not release
libnix is not released because it isn't open source.
--- removed tons of irrevevant drivel, again ---
eveybody can check with a fast windows PC and a fast windows browser(safari firefox) how fast it can be on his internet connect.and when he use amiga software and do the same test 5-10 times again then he get a average value and can say around 50% accurate how fast amiga browsers in compare to firefox etc are.
No they really don't. First of all the browsers run on a different system (different CPU, operating system, HDD and memory speed). Second the performance of the internet connection varies from time to time. Third the load of the remote web server varies from time to time. Regardless how many times you repeat the test you will not get any kind of comparable results that could be used to compare the browsers.
You can get better results by removing as much variables as you can. The obvious solution is to load the same web page from the local RAM disk. Even if you do this, you still won't be able to do any kind of comparison between different platforms (but you CAN compare the browser to another inside amigaos for example). To fix this you need to build Netsuft for the same system where the other browsers are tested (windows in this case).
I would have thought this is kind of obvious, but I you never know it seems.
-
@piru: we dont need to be that picky about the results atm. it might become important while the port will be optimized, but at the moment approximate results will do too. the most important is how netsurf does in praxis. and i have to say even with this simple sdl solution it is beyond my expectations. im testing the os4 post as well on the same system and have to say though more advanced os4 version seems a little slower on simple websites and less stable, which is but the fault of the system itself i suppose. os4 netsurf had stability issues from the start. on the other hand the current version renders correctly quite complex websites, which 68k port is not able to do yet.
-
on the other hand the current version renders correctly quite complex websites, which 68k port is not able to do yet.
can you tell me what pages OS4 netsurf render correct and 68k not ?
the netsurf build use the source from saterday, so it seem newer than MOS or OS4 build.
i see, in older netsurf use of new hubbub html parser was optional.
maybe on OS4 is used not hubbub, and when i report the site, netsurf hubbub can fix.
i dont believe that a wrong display website depend on sdl port or not.
the http://www.bahn.de is same wrong on OS4 ?
@Piru
>You can get better results by removing as much variables as you can.
yes of course everbody know that, but its good enough to get a quick overview about speed.for example on my winuae when i start netsurf 10* and for pageload measure 6 sec without JIT
now i start OWB several times and it take more than 90 sec without JIT.when i activate JIT, page is load always in 1,5 sec in netsurf
what do you think, is the inet connection a big error problem ?.the rendering because of slow CPU have lots of % in overall performance.
and btw, i have some time ago test with a cache proxy (set env var in libcurl)and i notice with that proxy not faster page load.
and i think not that users now for test install a proxy.We can be happy to get any value.
you see from OS4/MOS side no value is come or what happen on reuters page, so i guess there is same problem.
>libnix is not released because it isn't open source.
Yes i see, there is no need to release the sources, because its no GPL.everybody can use it without releasing source.
how many % have MOS enhance in this libnix source ?
i think its a thing of fair play to make also availible what you get for free, when you use 20% of this code.since Genesi build no PPC Hardware or something other there is also no fight for bigger market possible.MOS is a retro system without new Hardware, wy do not share sources and have fun on amiga without much costs, and the fight for more users or bashing again features that the loved OS not have end ?
I think you never test ixemul V61 on UAE and notice yourself, it run stable enough and far better than MOS native Version.
I for myself have test MOS, and i notice that MOS ixemul is not so compatible, and it stand also on MOS ixemul readme.
btw: MOS use too much AROS source, is since MOS2.x release some code give back to AROS ?.
here is libnix licence.
"""""
This package is public domain. That means that you can copy, use and modify it
without any problems and that you can get it for free. If you actually
paid for getting it this is completely your fault - I didn't see a cent
of that money. It also means that I cannot be made responsible for any damage
resulting out of the use of it - you simply shouldn't trust anything you didn't
pay for :-).
""""
-
Now there's you problem: It was never intended to. You asked for the MorphOS ixemul source. This is what you got.
Yes of course but when the MOS source contain so many ifdefs morphos and 68k asm code, so it seem it should run on amiga OS.
you never answer my question i ask severla times.
Have you really do much work and add this #ifdef morphos or have you only done some small fixes during last years ?
-
i dont believe that a wrong display website depend on sdl port or not.
the http://www.bahn.de is same wrong on OS4 ?
i have already reported it to you yesterday by mail. the problem looks the same. it is not a sdl issue.
-
libnix is not released because it isn't open source.
how many % have MOS enhance in this libnix source ?
Impossible to say. In overall it has been tweaked a LOT.
MOS use too much AROS source
Pardon? There's some kind of limit to as how much AROS source code you may use? The AROS source usage in MorphOS is limited to parts of dos, intuition, gadtools, diskfont, locale, commodities and some c commands. You can download the MorphOS 2.x ISO and check the documentation for the full list.
is since MOS2.x release some code give back to AROS ?
The license is obeyed, yes. MorphOS devs have been contributing back to AROS since 2001.
here is libnix licence.
"""""
This package is public domain.
""""
That is not the MorphOS libnix license. MorphOS libnix is not public domain. Some parts of the MorphOS libnix are based on the public domain work, however.
PS. Certain individual distributed MorphOS libnix sources with the wrong license at some point. This distribution was never authorized by the copyright holders.
---
Yes of course but when the MOS source contain so many ifdefs morphos and 68k asm code, so it seem it should run on amiga OS.
Assumption is the mother of all f***ups.
Have you really do much work?
Yes. I have. Now give it a rest already.
-
i have already reported it to you yesterday by mail. the problem looks the same. it is not a sdl issue.
yes i know and also you report that reuters page after some time do system crash on OS4 right ?
from my SDL Version tests netsurf dont crash system, but have the problems with some pages.i stress netsurf with reuter page showing go eating, after 20 minutes come back and it still want reload page.no crash see.
So i think ixemul work good enough.but its of course also intresting how performance is with libnix if it can run easy on it.
i see you have report the bahn problem chris in amigans Forum, so i need not report this on netsurf page.
so i guess this freeze problems come from the OS4 addons and its not netsurf fault.
intresting is now to know, if the reuter page on MOS do too freezes system.
-
now i start OWB several times and it take more than 90 sec without JIT.when i activate JIT, page is load always in 1,5 sec in netsurf
[...]
you see from OS4/MOS side no value is come or what happen on reuters page, so i guess there is same problem.
Despite all the arguments previously given regarding link speed/latency and whatever that can greatly ruin a "benchmark", here's a value for you:
OWB 1.3 MorphOS loads reuters in 10s on my peg2.
FireFox 3.something loads it in 15s on a pc/xp in the same network (and that pc is quite faster than my peg2).
So you see, it's easy to throw random values. :)
-
I hope the handbags at dawn don't put you off this, and you keep going with it.
-
Despite all the arguments previously given regarding link speed/latency and whatever that can greatly ruin a "benchmark", here's a value for you:
OWB 1.3 MorphOS loads reuters in 10s on my peg2.
FireFox 3.something loads it in 15s on a pc/xp in the same network (and that pc is quite faster than my peg2).
So you see, it's easy to throw random values. :)
how do you measure that ?
My OWB 68k need 20 sec to show http://www.reuters.com page(800*600 window).but i want see how long it take until the complete page text is load so the buttom text can read (impressum).if there is a picture load later is not important, but whats important for usability of a browser is that the layout is correct as soon as possible and the text can read full early as possible.if then later the images are load take again 10 sec is not important.
owb 68k need for this until i can read buttom text 85 sec
firefox need for this 4 sec.safari have same speed and ie too
-
how do you measure that ?
My measure is done when the page is fully loaded (images included). If I take your argument that not everything has to be loaded, then I can already scroll to bottom at 7s or so.
Anyway, silly test if it's not done locally, like Piru told.
-
My measure is done when the page is fully loaded (images included). If I take your argument that not everything has to be loaded, then I can already scroll to bottom at 7s or so.
Anyway, silly test if it's not done locally, like Piru told.
but when not all images need load to scroll to buttom, then not much data is need from internet and it is check if there are lockups or slowdowns and the browser is code good.
the amiga browser are in this case very clever programed, text is see very soon, no scroll block and images load later.
i still wonder wy there is no OWB on windows, there are patches since long time in OWB, only idea i have its far not so good as firefox or safari or other browser.
same is of course with netsurf too, but many pages are really same fast show as firefox and safari and IE on winuae.
-
owb on os4 is in meantime using the clever amiga-way to render pages too. lol.
reuters is really quite an unusable benchmark if:
1.netsurf doesnt load it correctly anyway.
2.there is perdefinied missunderstanding as to when the page is loaded completely.
3...
go, find a better one. but anyways to me its clear already that netsurf is quite fast even though the engine seems to be still a little limited.
-
owb on os4 is in meantime using the clever amiga-way to render pages too. lol.
reuters is really quite an unusable benchmark if:
after what time you can move the slider or scroll with cursor keys and you need not wait ?
-
after what time you can move the slider or scroll with cursor keys and you need not wait ?
First, the rendering is not OWB specific. It renders the same way on any webkit-based browser.
And then, you can scroll after about 2s (the page is already partially loaded at that point). At 5s the whole document height is computed and you can scroll to bottom as well.
But again, silly test.
And by the way, to compare it on a fair basis, if javascript is disabled in OWB, then the whole document loads in 4s (images included), and you can already scroll to bottom after 1s.
So, happy now? :)
-
First, the rendering is not OWB specific. It renders the same way on any webkit-based browser.
And then, you can scroll after about 2s (the page is already partially loaded at that point). At 5s the whole document height is computed and you can scroll to bottom as well.
But again, silly test.
And by the way, to compare it on a fair basis, if javascript is disabled in OWB, then the whole document loads in 4s (images included), and you can already scroll to bottom after 1s.
So, happy now? :)
i like to know on OS4 cybppc OWB speed values.
its really strange wy the 68k OWB is so extrem slow.the OWB 1.2 use SDL and is more 2* faster.OWB1.4 use no SDL and is slower.
are there important changes in wekkit or OWB so it work now so fast ?
-
current versions of owb on os4 are also faster as the initial release. i cant say how much and why exactly. surely due to both engine improvements and joerg os4 specific enchantments.
-
current versions of owb on os4 are also faster as the initial release. i cant say how much and why exactly. surely due to both engine improvements and joerg os4 specific enchantments.
the 68k OWB must speedup bei Factor 10 to be usefull.have OWB this speedup get ?
please post some values of your OS4 OWB.netsurf page, reuters page
-
The internal font version is working now.
It doesn't use freetype so there may be a speed gain.
You can download it here:
http://sites.google.com/site/amigacorner/Home/Netsurf2-nofreetype.zip .
-
The internal font version is working now.
It doesn't use freetype so there may be a speed gain.
You can download it here:
http://sites.google.com/site/amigacorner/Home/Netsurf2-nofreetype.zip .
good but thats a 16 bit version and work only on 16 bit screen.
-
good but thats a 16 bit version and work only on 16 bit screen.
Both 16bit and 32bit binaries are in archive now.
-
i see, in older netsurf use of new hubbub html parser was optional.
maybe on OS4 is used not hubbub, and when i report the site, netsurf hubbub can fix.
OS4 version has used Hubbub since forever.
If there are any rendering problems please report them to the NetSurf mailing list or in the bugtracker as per the instructions on the NetSurf website.
Chris
-
Both 16bit and 32bit binaries are in archive now.
Have tired both versions. The 32bit one crashes, with a recoverable alert. The 16 bit one runs and I can access a range of sites - Amiga.org, Bloomberg.com etc but I seem to lose the mouse after abit - can't see the pointer outside of the browser.
I'm running the 16 bit version on my A4000D with CSMII 060, 128MB Ram, Z3 Fastlane with 48MB Ram, Cybervision 3d with Scandoubler, P96 drivers, X-Surf, Miami, Deneb 2.0. OS3.9 BB1 & BBII, 3.1 Roms.
-
Have tired both versions. The 32bit one crashes, with a recoverable alert. The 16 bit one runs and I can access a range of sites - Amiga.org, Bloomberg.com etc but I seem to lose the mouse after abit - can't see the pointer outside of the browser.
I'm running the 16 bit version on my A4000D with CSMII 060, 128MB Ram, Z3 Fastlane with 48MB Ram, Cybervision 3d with Scandoubler, P96 drivers, X-Surf, Miami, Deneb 2.0. OS3.9 BB1 & BBII, 3.1 Roms.
the 32 bit version run only on a 32 bit screen.so you need to select correct version, for your workbench screendepth.
the pointer is a sdl shape, and when sdl window is active it can only move in sdl window.when you move the pointer over the edge, then the sdl pointer not move, but when you then press left mousebutton to activate another window, your amiga mousepointer is here.
-
I've selected a "32bit BGRA" screenmode, but it refused to open a window. Only the 16 bit version works for me after I switch the screenmode to "16bit PC".
The site www.nu.nl (http://www.nu.nl) (dutch news site) results in a hangup of netsurf.
-
I've selected a "32bit BGRA" screenmode, but it refused to open a window. Only the 16 bit version works for me after I switch the screenmode to "16bit PC".
The site www.nu.nl (http://www.nu.nl) (dutch news site) results in a hangup of netsurf.
i think your system have no 800*600 32 bit screen ?
i get too the problem with http://www.nu.nl.no crash, netsurf only not continue to load and quit of netsurf is too not possible.
does this site work on MOS or OS4 netsurf ?
""""
Google Summer of Code
NetSurf was accepted as a mentoring organisation into Google Summer of Code 2009[19]. The projects they are running include development of LibDOM, the project's Document Object Model, and improvement of NetSurf's user interface.[20]
""""
http://en.wikipedia.org/wiki/NetSurf
I dont know if all notice that, but netsurf is a sponsered project of google summer code, so hope there is manpower to fix it
if somebody have a videocam, its also nice to see how fast netsurf is, on a 50 MHZ system. but wait until next release, its faster then.
http://socghop.appspot.com/student_project/show/google/gsoc2009/netsurf/t124022821160
I see in google summer the project, that netsurf seem change the GUI way soon.SO its also not sure if amiga or MOS GUI need much changes, so i think use gtk to MUI wrapper is easest way
"""
Title: Core UI enhancements and keyboard navigation
Abstract: At the time being, all parts of the UI are created by the platform dependent code. This approach has some major drawbacks.
NetSurf is designed for resource constrained hardware but it doesn't work with restricted input methods. At the moment a mouse is necessary for browsing. This can be a barrier for hand held devices and a serious inconvenience for advanced users.
My intention is to remove the above limitations.
"""
-
i get too the problem with http://www.nu.nl crash, netsurf only not continue to load and quit of netsurf is too not possible.
does this site work on MOS or OS4 netsurf ?
No, it busy-loops trying to retrieve something-or-other (doesn't crash though, I can browse to a different page). Report it to the ml please.
Chris
-
This is great :D You get it work extremely fast :D
This sounds stupid from me but you should trust more your skills as a programmer _Arthur
Thanks
-
@__arti:
Any chance of a MUI version ??
-
@__arti:
Any chance of a MUI version ??
in the google code of summer project, there seem GUI implementation DOM large change.seem there is time to wait until it is finished if MUI or amiga OS work with new netsurf.
-
in the google code of summer project, there seem GUI implementation DOM large change.seem there is time to wait until it is finished if MUI or amiga OS work with new netsurf.
AmigaOS4 will work with the changes, as I will argue against them being merged in until I get the necessary modifications in place :) Ultimately it will mean less work per feature for new ports, and a more consistent interface across all platforms.
DOM is irrelevant as far as the platform code is concerned, as is the new CSS parser, except for possibly the odd minor change.
Chris
-
ok, im back in business (for the time being).:)
i had a terrible problem to get my dsl router up und running after i switched it off for two weeks. it has lost all data and the damn pc didnt wanted to reprogram the thing neither with iexpl nor firefox. hung around, some firewall prob or who knows.:/ in the end i throwed os4 with owb at that and voila. it worked immediately. so it seems aos4 is still good for something, right?
back to subject:
i looked at the nofreetype version. it is slightly faster, around 5% i would say. but i think its not worth it. some pages look trashed with internal type and the speedup is not significant.
edit: well, not quite, sometimes its pretty much faster, like +1/3
it would be good to get hold of this endiannes problem first i think. and it still runs out of memory. it has nothing to do with stack apparently.
i see the bugtracker has already been taken care of too :D
btw, on os4 i have an impression that the current owb is quite a lot faster than netsurf so maybe it still would be worth a try.
-
ok, im back in business (for the time being).:)
i had a terrible problem to get my dsl router up und running after i switched it off for two weeks. it has lost all data and the damn pc didnt wanted to reprogram the thing neither with iexpl nor firefox. hung around, some firewall prob or who knows.:/ in the end i throwed os4 with owb at that and voila. it worked immediately. so it seems aos4 is still good for something, right?
back to subject:
i looked at the nofreetype version. it is slightly faster, around 5% i would say. but i think its not worth it. some pages look trashed with internal type and the speedup is not significant.
edit: well, not quite, sometimes its pretty much faster, like +1/3
it would be good to get hold of this endiannes problem first i think. and it still runs out of memory. it has nothing to do with stack apparently.
i see the bugtracker has already been taken care of too :D
btw, on os4 i have an impression that the current owb is quite a lot faster than netsurf so maybe it still would be worth a try.
try the page
Edit:Page was before wrong
http://www.t-online.de
i notice on netsurf it is only show when it is full load.take 10 sec on my system
on OWB 68k same i can only scroll to down and read full text until all images are load.
but on firefox, safari there can scroll around and read the text after 3 sec.
but of course the images are not load complete.
what happen on OWB OS4 with that page ?
I dont know if netsurf use threads, i only see MOS netsurf use a GUI thread.but if this help to get better reaction when a page is not full load i dont know.
maybe somebody can tell what happen with the site on netsurf MOS.can there read text and scroll much time before all images are load ?
-
the page loads ages on 68k: around 80sec, both versions of ns, nofreetype was even some sec slower in this case.
approximate results:
68k ns-nofreetype ~68-83.s -has to load full, slight layout glitches doesnt scroll immediately after page is complete
68k ns ~77.s -has to load full
owb 68k 1.3 - difficult to benchmark, definitely much longer, but you can scroll a little after a minute or so, doesnt need to load full, rendered perfectly
owb 3.13 os4 - loads about a minute, able to scroll after ~20s (604/150), it opens a popup and default window size is much bigger of course.
netsurf2.1 on os4 starts to display but locks the system before the images load
for comparison my firefox needs ~13s on a athlon dual 4200+, my dsl is only 2k btw
-
the page loads ages on 68k: around 80sec, both versions of ns, nofreetype was even some sec slower in this case.
approximate results:
68k ns-nofreetype ~68-83.s -has to load full, slight layout glitches doesnt scroll immediately after page is complete
68k ns ~77.s -has to load full
owb 68k 1.3 - difficult to benchmark, definitely much longer, but you can scroll a little after a minute or so, doesnt need to load full, rendered perfectly
owb 3.13 os4 - loads about a minute, able to scroll after ~20s (604/150), it opens a popup and default window size is much bigger of course.
netsurf2.1 on os4 starts to display but locks the system before the images load
for comparison my firefox needs ~13s on a athlon dual 4200+, my dsl is only 2k btw
the OWB OS4 version have after 20 sec load, full page Text load, so can full scroll down to read last line ?
and your firefox result, is the 13 sec full page i guess ?
what time you get on your firefox until you can scroll full down to read last text line ?
-
owb os4 20 sec for scrolling 1 min for full page, just tried again and firefox rendered t-online in some 5 sec now, scrolling is possible immediately. as piru said the results vary so much it is difficult to make something out of this. the fact is on a pages like amiga.org netsurf sdl is already faster than ibrowse and renders quite well even if the scrolling is jerky. maybe there is a way to optimize this wit hwsurface or what it was?
i see the hit after closing the app has been disposed of, is it only on nofreetype or did you find the bug?
what is the policy now? to wait till the next release, because of endiannes prob and the gui? do this port stays on sdl? i suppose further features like window resizing will not be possible in this case? running out of memory on big pages is still a problem, regardless the stack. if the reason is unknown maybe it could be asked on ml?
-
owb os4 20 sec for scrolling 1 min for full page, just tried again and firefox rendered t-online in some 5 sec now, scrolling is possible immediately. as piru said the results vary so much it is difficult to make something out of this. the fact is on a pages like amiga.org netsurf sdl is already faster than ibrowse and renders quite well even if the scrolling is jerky. maybe there is a way to optimize this wit hwsurface or what it was?
i see the hit after closing the app has been disposed of, is it only on nofreetype or did you find the bug?
what is the policy now? to wait till the next release, because of endiannes prob and the gui?
The endianess problem is fixed. I have submited Netsurf to Aminet. It should be ready tomorrow.
do this port stays on sdl? i suppose further features like window resizing will not be possible in this case?
Window resize can be done be editing Resources/Options file.
running out of memory on big pages is still a problem, regardless the stack. if the reason is unknown maybe it could be asked on ml?
Can you tell on which page do you get out of memory?
-
owb os4 20 sec for scrolling 1 min for full page, just tried again and firefox rendered t-online in some 5 sec now, scrolling is possible immediately. as piru said the results vary so much it is difficult ?
i think the fast firefox result is
because firefox have a cache you need a page reload do(press reload button).
but even if firefox is fast on cache , OWB and netsurf should of course too use the cache and show page faster.
its right that the internet time make things differ.But not very much.I test myself with proxy
If you have your amiga connect to a PC you can run a cache proxy here.
when you set a env var on amiga OS
http_proxy
127.0.0.1:0080
with content then it can use the proxy on the PC when you set the address and port to above values
this env var is used by all browsers that use libcurl.so OWB and netsurf use a proxy cache
I use allegro surf(its kn own as fast) and i notice no speedup of amiga browsers/firefox on pages with fastserver as the t-online site and other big sites i tell are.
>maybe there is a way to optimize this wit hwsurface or what it was?
yes, newest version is faster and use the gfx card blitter for scroll because it use HWsurface.
We send test versions to Thomas with sdl_software and SDL_HWsurface and on his slow Z2 system, sdl_HWsurface give same render speed, but scroll lots faster
also new is 16 and 32 bit mode is automatic detect, key input with shift and alt work, and speed should be faster,because to get correct endian, code can remove
but there are still lots of known bugs that are netsurf SDL Problems, i have report them in ML, but because all work on integrate the google summer code, nobody have time, so need to wait for SDL and new DOM what happen here.
but what can say, netsurf run stable on 68k and the OS4 crashes are no fault of netsurf main code.
-
Still don't know why netsurf 68k runs on some (winuae) systems, but not on mine. I've assigned amitcp as suggested earlier, but netsurf still doesn't see the connection. Pretty strange, as aweb and ibrowse do work perfectly.
Does anyone have some more suggestions ???
x303 :D :D :D
-
Here, with OWB 1.3 for MorphOS, it takes about 20s to load fully, and you can already scroll to bottom after 4s (note that on MorphOS, OWB network manager is threaded).
FireFox gave a similar result, loading in about 20s too, and allowing to scroll after about 4-5s.
But i wonder once again why i replied about that kind of benchmark that depends on way too many factors. :)
-
Great work Artur & Bernd. :)
-
Still don't know why netsurf 68k runs on some (winuae) systems, but not on mine. I've assigned amitcp as suggested earlier, but netsurf still doesn't see the connection. Pretty strange, as aweb and ibrowse do work perfectly.
Does anyone have some more suggestions ???
x303 :D :D :D
If you use AmiKit:
The default assignment for AMITCP: in AmiKit is the Ibrowse drawer, so make sure to replace this assignment.
edit: The assignment is made in user-startup. Either replace the current assignment, or after this assignment add the following:
Assign AMITCP: ADD
Check in a CLI window by entering "assign" to which drawer AMITCP is currently assigned.
Also check if you've installed all the fonts that come with Netsurf if your running the TTF version.
Edit: If the nofreetype version runs ok, you haven't installed the fonts correctly. You must install them in Fonts:truetype/ttf-bitstream-vera
Also check if you have the latest ixemul lib.
-
Great work Artur and Bernd. It's pretty fast under WinUAE.
Still have to test it on my a4000.
Thank you !!!
-
Here, with OWB 1.3 for MorphOS, it takes about 20s to load fully, and you can already scroll to bottom after 4s (note that on MorphOS, OWB network manager is threaded).
FireFox gave a similar result, loading in about 20s too, and allowing to scroll after about 4-5s.
But i wonder once again why i replied about that kind of benchmark that depends on way too many factors. :)
You are free to run a cache proxy and bring more accurate results.
But because most are too lazy to do this extra work, then this values are better than nothing to see if amiga browsers work good in compare to firefox, to see if it is usefull to have firefox too
also you can post values when you do a site back to show how long time need to show full page but in cache.
its also intresting to see how the speed is when MOS OWB use no threads.if it is possible to disable by define.
as far i know OS4 OWB use no threads and 68k too not.
Maybe it help users to choose the better OWB.
I think using thread is important, to get same speed as firefox
but if not really know if it have noticable speedup with threa
-
netsurf2.1 on os4 starts to display but locks the system before the images load
Works on the latest dev version - 19.3s reported for t-online.de
-
@x303:
Does anyone have some more suggestions ???
get out the real hw without w$ underneath:P
@bernd:
If you have your amiga connect to a PC you can run a cache proxy here.
when you set a env var on amiga OS
http_proxy
yeah im lazy, i know. :P but in this case my amigas and the pc are connected independently to the net via telecom router. i dont now if i can establish a proxy somewhere within my home network but i dont think this speed comparizons are so important except nmaybe for the threading you have spoken of
@chris
Quote:
Originally Posted by wawrzon
netsurf2.1 on os4 starts to display but locks the system before the images load
Works on the latest dev version - 19.3s reported for t-online.de
i have some probs that might be due to im using elbox voodoo drivers, i have to revert to the hyperion ones if im able to get the network card working that way. again was way too lazy to do it till now. maybe then os4 will get more stable
-
@artur:
Quote:
do this port stays on sdl? i suppose further features like window resizing will not be possible in this case?
Window resize can be done be editing Resources/Options file.
i mean resize on the fly. but sure it demands an amiga-gui. anyways nice for the full initioal release. i just dl it and everything seems to work nice.
Quote:
running out of memory on big pages is still a problem, regardless the stack. if the reason is unknown maybe it could be asked on ml?
Can you tell on which page do you get out of memory?
bbc news link on netsurf start page fi. i will find some other sites and post it here. i think generally every complex site runs eventually out of mem. just testing...
edit:
i just noticed a minor issue: alpha channel still doesnt work in overleaping mages like the arrow in the top right corner of netsurf-browser.org. if i recall right this renders ok on the os4 port.
check ns bug tracker page too. there is some color corruption in images. the top bar (sf.net) backgroundis sapphire should be black. the top bar of the list has the same taint, and the list background gradient is rendered i two fields of color. i suppose this is still some overseen minor endiannes issue somwhere.
32bit argb argb on my voodoo the whole sdl window is tainted yellow too, -i think this is adressed to bernd and his sdl port.
edit: wow! ttf text input has been improved! very nice! arrow keys still not working though. any chance window bars can be dragged with the mouse? (mousewheel is working cool)
running out of mem on bild.de (bernds beloved test site, lol)
edit:
oughh!! i just had a hit. and my first netsurf caused complete system lockup under 3.x!! that while reloading t-online.de. the page rendered quite correctly the first time but i had an impression of some images missing. it took 92s btw. now i go take my sunday bike tour. take care!
-
@artur:
bbc news link on netsurf start page fi. i will find some other sites and post it here. i think generally every complex site runs eventually out of mem. just testing...
i get too this out of memory error in netsurf status page that netsurf report when press on BBC Link.
sometimes the error is not show and BBC Page is also not draw.
the last page is show.
I think thats too a netsurf problem or does it work on OS4 ?
btw if you change the window size pref of netsurf to your screen res can cause problems.best is you use 20 pixel less so border can fit.
>32bit argb argb on my voodoo the whole sdl window is tainted >yellow too, -i think this is adressed to bernd and his sdl port.
netsurf SDL have problem to use SDL pixelformats.
it work only when you choose a bgra or bgr screen bgr 16pc
but now this version detect if run on 16 or 32 bit screen and use correct depth.
-
i get too this out of memory error in netsurf status page that netsurf report when press on BBC Link.
since the problem doesnt show up on some other big sites im not sure if this is really out of mem problem. maybe it is wrongly reported as such. i dont know.
bernd,i have reproduced the failure on reloading t-online.de and this timethere was no hit nor crash but the request of ixemul.library: abort trap. after that netsurf closed quietly.
something strange happens on amigans.net (maybe due to java script) and on utilitybase.com. you noticed? i havnt confirmed it using os4 version yet, will try it later.
edit: no, netsurf os4 loads news.bbc.co.uk succesfully (29.1s) reporting one object of a wrong type.
also no probs on utilitybase.com and amigans.net (corruption around the cursor)
well, but this might be due to sdl version limitations.
-
since the problem doesnt show up on some other big sites im not sure if this is really out of mem problem. maybe it is wrongly reported as such. i dont know.
bernd,i have reproduced the failure on reloading t-online.de and this timethere was no hit nor crash but the request of ixemul.library: abort trap. after that netsurf closed quietly.
something strange happens on amigans.net (maybe due to java script) and on utilitybase.com. you noticed? i havnt confirmed it using os4 version yet, will try it later.
edit: no, netsurf os4 loads news.bbc.co.uk succesfully (29.1s) reporting one object of a wrong type.
also no probs on utilitybase.com and amigans.net (corruption around the cursor)
well, but this might be due to sdl version limitations.
can you make screenshot of this page ?
what screen res you have on workbench and what setting you have in netsurf screenwidth option ?
i test the version from aminet, it create a shadow surface and use HWsurface on my system.please look what happen on your system.
normaly output on sashimi should be no shadowsurface
-
can you make screenshot of this page ?
what screen res you have on workbench and what setting you have in netsurf screenwidth option ?
workbench: 1680x1050 netsurf:1200x1000
i test the version from aminet, it create a shadow surface and use HWsurface on my system.please look what happen on your system.
normaly output on sashimi should be no shadowsurface
[/QUOTE]
me too. i have SDL_HWSURFACE set to 1 in the env if this is what you mean. sashimi reports: create shadow surface.
-
screenshot
-
i just had a netsurf crash on loading t-online.de. i have managed to catch the sashimi output:
-
workbench: 1680x1050 netsurf:1200x1000
thats a memtrash error.
I try your resolution and after i press reload i get a abort trap error and netsurf quit.
then i try out 1024*768
this too dont work get after second reload too problem.allis reproducable
I test on 800*600 more than 10 reloads and all work ok.
please try out to use a resolution of 800*600 or below
seem netsurf have a >800 Bug.
I try out resolutions 761*571 to see maybe sdl have a allign memtrash problem.but this resolution too work ok with more than 10 reloads no crash
maybe you can try out the OS4 version and set window to below 800*600.
maybe on OS4 its same the large window cause the crash.
-
using 1024x768 on another system i had a hangup and now repeating hit without a crash just like that:
BUFFER WAS OVERRUN - Data may have been lost
BUFFER WRAPPED - This is the most recent captured data
movea.l a1,a0
08dc5408 : *10c0 move.b d0,(a0)+
08dc540a : 5381 subq.l #$1,d1
08dc540c : 66fa bne.s $8dc5408
08dc540e : 2009 move.l a1,d0
08dc5410 : 4e75 rts
08dc5412 : 0000 226f ori.b #$6f,d0
08dc5416 : 0004 122f ori.b #$2f,d4
08dc541a : 000b 7000 ori.b #$0,a3 ;illegal opcode
08dc541e : b211 cmp.b (a1),d1
08dc5420 : 6602 bne.s $8dc5424
08dc5422 : 2009 move.l a1,d0
08dc5424 : 4a19 tst.b (a1)+
08dc5426 : 66f6 bne.s $8dc541e
Name: "Netsurf"
20-Apr-08 21:43:16
BYTE WRITE to 00000458 data=13 PC: 08DC5408
USP : 08BE986C SR: 0000 (U0)(-)(-) TCB: 08B45D30
Data: 00000013 FFFFFBA6 00008536 00000001 00000000 00000000 00000000 00000000
Addr: 00000458 FFFFFFFF 08D88488 092511B8 08F6308E 08BE9880 08037B1C 0800232C
Stck: 09198FF0 FFFFFFFF 00000013 FFFFFFFF FFFFFFFB 08BE9898 0918C10C FFFFFFFF
Stck: 000002B7 0918BAAC FFFFFFFF 08BE98AC 091934F8 FFFFFFFF 095AE838 08C3F598
Stck: 08BE98C0 091A41C4 08C3F598 095AE838 FFFFFFFF 08BE98E8 09193AA8 08C3F610
Stck: 095AF9B8 08C3F598 095AE838 00000001 08C3F610 09AAA338 095AF9B8 08BE9904
Stck: 0918B9A4 08C3F598 08BE98FC 0918B938 0000003C 480BB8B2 08BE991C 0918BA0A
Stck: 08C3F598 0000003C 480BB8B2 480BB8B2 08BE9938 0918F482 095F9480 FFFFFFFF
Stck: 00000000 08BE9960 095F9480 08BE9960 0918FB6A 09AD8918 08BE9960 09191D50
Stck: 08BE999C 00010000 00000000 09AD8918 00000000 08BE9BAC 09192A4E 095F9480
----> 08DC5408 - "Work:Libs/ixemul.library" Hunk 0000 Offset 00017F28
----> 09198FF0 - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 002A8358
----> 0918C10C - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 0029B474
----> 0918BAAC - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 0029AE14
----> 091934F8 - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 002A2860
----> 091A41C4 - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 002B352C
----> 09193AA8 - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 002A2E10
----> 0918B9A4 - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 0029AD0C
----> 0918B938 - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 0029ACA0
----> 0918BA0A - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 0029AD72
----> 0918F482 - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 0029E7EA
----> 0918FB6A - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 0029EED2
----> 09191D50 - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 002A10B8
----> 09192A4E - "Work:Internet/NetSurf/Netsurf" Hunk 0000 Offset 002A1DB6
PC-8: 61FFFFFF FE0A4FEF 000C2002 241F4E75 226F0004 202F0008 222F000C 67082049
PC *: 10C05381 66FA2009 4E750000 226F0004 122F000B 7000B211 66022009 4A1966F6
08dc53e8 : 61ff ffff fe0a bsr.l $8dc51f4 ;extended opcode
08dc53ee : 4fef 000c lea.l $c(a7),a7
08dc53f2 : 2002 move.l d2,d0
08dc53f4 : 241f move.l (a7)+,d2
08dc53f6 : 4e75 rts
08dc53f8 : 226f 0004 movea.l $4(a7),a1
08dc53fc : 202f 0008 move.l $8(a7),d0
08dc5400 : 222f 000c move.l $c(a7),d1
08dc5404 : 6708 beq.s $8dc540e
08dc5406 : 2049 movea.l a1,a0
08dc5408 : *10c0 move.b d0,(a0)+
08dc540a : 5381 subq.l #$1,d1
08dc540c : 66fa bne.s $8dc5408
08dc540e : 2009 move.l a1,d0
08dc5410 : 4e75 rts
08dc5412 : 0000 226f ori.b #$6f,d0
08dc5416 : 0004 122f ori.b #$2f,d4
08dc541a : 000b 7000 ori.b #$0,a3 ;illegal opcode
08dc541e : b211 cmp.b (a1),d1
08dc5420 : 6602 bne.s $8dc5424
08dc5422 : 2009 move.l a1,d0
08dc5424 : 4a19 tst.b (a1)+
08dc5426 : 66f6 bne.s $8dc541e
Name: "Netsurf"
while using the 800x600 did work on both. even if reload was sometimes rediculously slow (obout 5 min). might be due to my home network with 2 miggys and one pc connected to the router.
edit: bernd, were you able to reproduce these strange pointer effects on utilitybase and amigans? i had something similar while editing image on arteffect with the netsurf window in the background. the image disaperared and i was able to get it back by resizing window or just "painting" with cursor over it. i think there is still some sdl bug.
edit: under os4 i get a crash while rendering t-online no matter what resolution
-
You are free to run a cache proxy and bring more accurate results.
But because most are too lazy to do this extra work, then this values are better than nothing to see if amiga browsers work good in compare to firefox, to see if it is usefull to have firefox too
also you can post values when you do a site back to show how long time need to show full page but in cache.
its also intresting to see how the speed is when MOS OWB use no threads.if it is possible to disable by define.
as far i know OS4 OWB use no threads and 68k too not.
Maybe it help users to choose the better OWB.
I think using thread is important, to get same speed as firefox
but if not really know if it have noticable speedup with threa
Actually, the way it's currently done, threads don't speed up the process, since all the connections run in a single (and separated from main context) thread.
But like i already said previously, all CURL handles are handled asynchronously (non blocking sockets, process stuff when data arrives,...). The problem is some parts of the process are still a bit long (data reception) or very long (DNS resolution which is a major blocking call, unless c-ares or an additional lookup thread is spawned).
So, even tho CURL is mostly asynchronous, it still makes sense to run it in a dedicated thread to avoid blocking the gui during DNS lookups or long data reception phase.
-
edit: bernd, were you able to reproduce these strange pointer effects on utilitybase and amigans? i had something similar while editing
Edit:
this i see too.when scroll with cursor then a width rectangle is see.
But when wait until the bitmaps(OS4 AROS classic images) are draw before mouse move or scroll, then all is ok.but sometimes i get no images show only when scroll full down and then up
strange what SDL can do with that
-
you write you get with the BBC page on OS4 wrong object type message ?
Now that i increase window size to 1200*1000 i get too this message, and no out of memory but page is not show.
or can you post a screenshot of the BBC Page on OS4 is show ?
-
since the problem doesnt show up on some other big sites im not sure if this is really out of mem problem. maybe it is wrongly reported as such. i dont know.
Out of memory is a bit of a default error message on NetSurf. I wouldn't trust it to actually mean out of memory, I used to get it here due to a bug in newlib's iconv() functions.
@bernd
Since I'm here, here's a screenshot of BBC News site. I don't get that error though.
-
Out of memory is a bit of a default error message on NetSurf. I wouldn't trust it to actually mean out of memory, I used to get it here due to a bug in newlib's iconv() functions.
@bernd
Since I'm here, here's a screenshot of BBC News site. I don't get that error though.
the 68k version use the libiconv and not the iconv on netsurf page.We notice with internal fonts problems with the CP437 codepage.(No fonts show)
Artur change internal font to CP1250.this support most chars
but maybe try to compile netsurf iconv maybe that help
-
there is a option for option file to disable antialiazing of truetype fonts.now netsurf is faster with freetype.
fb_font_monochrome:true
maybe there is no need to do a internal font build anymore ?