src/conio sdl_con.c 1.183 1.184
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv12454

Modified Files:
sdl_con.c
Log Message:
Coverity bug when a vidmode event fails to malloc() enough memory, and there are expose or updatecrect events in the pipe already. Allows the program
to exit cleanly instead of crashing.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.184 1.185
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv12582

Modified Files:
sdl_con.c
Log Message:
Per Coverity, check return value of malloc().




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.185 1.186
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv24820

Modified Files:
sdl_con.c
Log Message:
From IBM:
"On keyboards that support the Caps Lock key, Caps Lock affects only those
keys whose Shift state yields the uppercase character (A, B, C) of the Base state lowercase character (a, b, c) of the key."

http://www-01.ibm.com/support/knowledgecenter/ssw_aix_61/com.ibm.aix.keyboardtechref/doc/kybdtech/Understanding.htm%23dbffbda939dagi



---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.186 1.187
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv25099

Modified Files:
sdl_con.c
Log Message:
Fix copy pasta in the last commit.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.187 1.188
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv4634

Modified Files:
sdl_con.c
Log Message:
Fix bug in Overlay mode... upd_rects can be NULL.
Fixes "Black Window" error in Overlay modes.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.188 1.189
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv8790

Modified Files:
sdl_con.c
Log Message:
We only need to lock the YUV overlay when diddling pixles.
Also, ensure we only free the overlay when it's allocated.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.189 1.190
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv19020

Modified Files:
sdl_con.c
Log Message:
Remove some obsolete paranoia and move the redraw forcing into the
function after which it is always triggered.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.190 1.191
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv20051

Modified Files:
sdl_con.c
Log Message:
Work around bug present in (at least) SDL 1.2.15 which can prevent SDL_VIDEORESIZE events from occuring under X11 after SDL_SetVideoMode() is called for a second time.

Simply send yourself the "unsitck" message and you're golden.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.191 1.192
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv27130

Modified Files:
sdl_con.c
Log Message:
Only work around X11 SDL bugs when X11 is enabled.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.192 1.193
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv4655

Modified Files:
sdl_con.c
Log Message:
Remove the timed wait added in Rev. 1.181 and replace with an ACK system.
This is still an ugly hack, but at least now I know how events are getting lost.

When SDL is configured to pass WM events (enabled when using X11 for copy/ paste support), a WM event can overflow the event queue removing the oldest entry. Most of the user-defined events occur in under 100ms, so the old timeout was usually fine. However, SDL_USEREVENT_INIT absolutely does take longer when initializing SDL Overlay mode in fullscreen. It's likely that other init cases could take longer as well.

This commit adds an ACK semaphore so that the caller knows that the event
is being handled. If it doesn't get this ACK within 1s, it resends as
before, but if it does get the ACK, it now blocks indefinitely knowing that
the event is being handled.

This one second delay can be noticable at times, but making it lower risks double-free errors. This delay should only occur on X11 systems, and seems
to mostly occur during startup. To work around this, I don't enable WM
events until after startup is mostly done.

If the problem gets too bad, some other approach may be needed. Hopefully, this will be good enough until I port the whole thing to SDL2 and can
eliminate this entire problem.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.193 1.194
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv19103

Modified Files:
sdl_con.c
Log Message:
Add missing file from the last copy/paste commit... this is the bit that actually makes it work. ;-)



---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.195 1.196
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv16611

Modified Files:
sdl_con.c
Log Message:
Actually, enabling the SYSWMEVENTS in the INIT event results in them
happening earlier... so early in fact, that we don't even know if we should
at that point.



---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.196 1.197
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv18479

Modified Files:
sdl_con.c
Log Message:
SDL_WaitEvent() waits at least 10ms after not getting an event. Only sleep
one ms to pump the event queue faster.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.198 1.199
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv30047

Modified Files:
sdl_con.c
Log Message:
Add the cheery_reaper() function which waits 500ms then calls exit(0). This ensures the program quits when the user presses the exit button and the program doesn't handle CIO_KEY_QUIT.

This still isn't the best way to do things.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.199 1.200
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv30282

Modified Files:
sdl_con.c
Log Message:
Derp, the reaper is supposed to be a thread, not a function.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.201 1.202
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv14346

Modified Files:
sdl_con.c
Log Message:
Remove the reaper thread and exit(0) immediately if reaping is enabled
(the default). If the program doesn't already handle CIO_KEY_EXIT, waiting
a half-second isn't going to help, and if it does, it should turn off the reaper.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.202 1.203
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv11242

Modified Files:
sdl_con.c
Log Message:
Remove the funcret_mutex since it's functionality is completely duplicated
by sdl_ufunc_mtx. Fix up the sdl_ufunc_mtx usage to help prevent errors.

Also, work around a deadlock on Win32 where a video resize event while a
redraw event is later in the queue would deadlock on the vstatlock mutex.

This problem is *not* completely resolved yet. I think the new scaling
will need to be cached, then applied the next time the queue is empty.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.203 1.204
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv11553

Modified Files:
sdl_con.c
Log Message:
Fix deadlock on vstatlock.

When a resize event is serviced, cache the new scaling factor. Next time
the event queue is empty, attempt to lock vstatlock. If that succeeds,
apply the new scaling. If not, wait for the queue to drain again.

This may cause the scaling to be slightly delayed, but should completely
solve this deadlock.

This deadlock likely never affected FreeBSD since FreeBSD has deadlock detection, and the return value of pthread_mutex_lock() isn't tested. The result would have been a silent failure to lock, and an update of the scaling factor regardless, followed by a silent (again untested) failure of the matching
pthread_mutex_unlock() call.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.205 1.206
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv12103

Modified Files:
sdl_con.c
Log Message:
Remove debug output.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.206 1.207
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv21843

Modified Files:
sdl_con.c
Log Message:
In a blind attempt to fix the reported XP crahses, mutex protect the last couple global pointers (win and new_rect), and move upd_rect, rectspace, and rectsused into automatic variables in the event thread function.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.207 1.208
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv1623

Modified Files:
sdl_con.c
Log Message:
When calling setup_surfaces() from a SDL_USEREVENT_SETVIDMODE event, set new_scaling and old_scaling to prevent setup_surfaces() from being called
again while the window is initially being set up in a mode other than the default 80x25.

Fixes a Windows crash when starting in other modes.



---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.209 1.210
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv12015

Modified Files:
sdl_con.c
Log Message:
Fix memory leak when quitting (should not really fix anything except a
Valgrind report).




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.211 1.212
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv23002

Modified Files:
sdl_con.c
Log Message:
Fix overlay mode resizing. With the recent race condition work-around, setup_surfaces() was never called for overlay mode, resulting in the window contents never changing size.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.213 1.214
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv17124

Modified Files:
sdl_con.c
Log Message:
Don't lock the vstat mutex in SDL_USEREVENT_UPDATERECT event since it's
already held by the bitmap layer.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.216 1.217
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv9940

Modified Files:
sdl_con.c
Log Message:
Add a Windows-specific hack to read the state of the numlock key and override the mod state. This should fix num-lock on Windows.

Other mod states (CTRL, SHIFT, ALT) are likely still messed up though.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.216 1.217
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv9940

Modified Files:
sdl_con.c
Log Message:
Add a Windows-specific hack to read the state of the numlock key and override the mod state. This should fix num-lock on Windows.

Other mod states (CTRL, SHIFT, ALT) are likely still messed up though.

fix :) but found a new problem, my xp computer can't run the 7/9/15 build, it coming up and then goes back down. now on my win. 7 (same build) it ok


---

Rob Starr
Lord Time SysOp of
Time Warp of the Future BBS
Telnet://Time.Darktech.Org:24 or
Telnet://Time.Synchro.Net:24 (qwk or ftn & e-mail)
ICQ # 11868133 or # 70398519 Jabber : lordtime2000@gmail.com
Yahoo : lordtime2000 AIM : LordTime20000 MSN : Lord Time
Astra : lord_time X-Box : Lord Time 2000 oovoo : lordtime2000
---
þ Synchronet þ Time Warp of the Future BBS - Home of League 10 IBBS Games

Re: Re: src/conio/sdl_con.c
By: Lord Time to deuce on Thu Jul 09 2015 10:12 pm

it coming up and then goes back down. now on my win. 7 (same build) it ok

I have no idea what this means.

I'm not likely to fix anything for Windows XP support though. Best hope is that it's a real bug that affects all Windows builds.

---
http://DuckDuckGo.com/ a better search engine that respects your privacy.
Mro is an idiot. Please ignore him, we keep hoping he'll go away.
þ Synchronet þ My Brand-New BBS (All the cool SysOps run STOCK!)

Re: Re: src/conio/sdl_con.c
By: Lord Time to deuce on Thu Jul 09 2015 10:12 pm

it coming up and then goes back down. now on my win. 7 (same build) it ok

I have no idea what this means.

I'm not likely to fix anything for Windows XP support though. Best hope is that it's a real bug that affects all Windows builds.

I happens to me on Win7 and Win8.1 (both x64) sometimes when launching from a shortcut on the desktop. However, if I open it from the same shortcut one more time, it opens correctly. Hasn't been enough of an annoyance for me to to even think twice about it.

Symptoms: Launch Syncterm, it loads up momentarily, and suddenly closes in a fraction of a second (I'd estimate about 0.5 seconds, enough to just see it happening). Launching it a second time (and very rarely 3rd) allows it to launch correctly.

I thought it was happening on my Win7 at work (where I noticed it first) because it's a laptop that sometimes gets docked with dual monitors, and sometimes does not, and I figured that was the problem. However, I eventually saw it happen (much less often) on my Win8.1 machine at home (running a Win7-like shell, called Start is Back - startisback.com).

Wish I could explain it a bit more, but it happens very quickly. And also I have never tried it on WinXP, so I don't know what's up with that and if launching it a second time would help any.

~KenDB3

---
þ Synchronet þ KD3net-Rhode Island's only BBS about nothing. http://bbs.kd3.us

I happens to me on Win7 and Win8.1 (both x64) sometimes when launching from a
shortcut on the desktop. However, if I open it from the same shortcut one more
time, it opens correctly. Hasn't been enough of an annoyance for me to to even
think twice about it.

Symptoms: Launch Syncterm, it loads up momentarily, and suddenly closes in a fraction of a second (I'd estimate about 0.5 seconds, enough to just see it happening). Launching it a second time (and very rarely 3rd) allows it to launch correctly.

I've noticed the same issue. I'm currently using Windows 7 64-bit.

Eric

---
þ Synchronet þ Digital Distortion: digitaldistortionbbs.com

Re: Re: src/conio/sdl_con.c
By: KenDB3 to Deuce on Fri Jul 10 2015 01:39 pm

I'm not likely to fix anything for Windows XP support though. Best
hope is that it's a real bug that affects all Windows builds.

I happens to me on Win7 and Win8.1 (both x64) sometimes when launching from a shortcut on the desktop. However, if I open it from the same shortcut one more time, it opens correctly. Hasn't been enough of an annoyance for me to to even think twice about it.

For fun I installed it on the XP machine that runs my BBS. I've never run Syncterm on it, so it was a fresh install. It behaves nearly the same as Win7/8 except it's worse. Basically, it fails to start more often. But it works.

~KenDB3

---
þ Synchronet þ KD3net-Rhode Island's only BBS about nothing. http://bbs.kd3.us

Re: Re: src/conio/sdl_con.c
By: KenDB3 to Deuce on Fri Jul 10 2015 01:39 pm

Symptoms: Launch Syncterm, it loads up momentarily, and suddenly closes in a fraction of a second (I'd estimate about 0.5 seconds, enough to just see it happening). Launching it a second time (and very rarely 3rd) allows it to launch correctly.

Cool thanks, I'll take a look.

---
http://DuckDuckGo.com/ a better search engine that respects your privacy.
Mro is an idiot. Please ignore him, we keep hoping he'll go away.
þ Synchronet þ My Brand-New BBS (All the cool SysOps run STOCK!)

src/conio sdl_con.c 1.217 1.218
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv28906

Modified Files:
sdl_con.c
Log Message:
Wait for bitmap_init() to finish before locking vstat at the beginning of sdl_video_event_thread().

This should fix the crash on startup bug seen on Windows systems.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

src/conio sdl_con.c 1.218 1.219
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv30779

Modified Files:
sdl_con.c
Log Message:
Fix bug in last commit... don't SLEEP() with the mutex held.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ telnet://vert.synchro.net

Re: Re: src/conio/sdl_con.c
By: Deuce to KenDB3 on Sun Jul 12 2015 03:36 pm

Re: Re: src/conio/sdl_con.c
By: KenDB3 to Deuce on Fri Jul 10 2015 01:39 pm

Symptoms: Launch Syncterm, it loads up momentarily, and suddenly
closes in a fraction of a second (I'd estimate about 0.5 seconds,
enough to just see it happening). Launching it a second time (and very
rarely 3rd) allows it to launch correctly.

Cool thanks, I'll take a look.

Looking good since the update you did. Thanks Deuce!

---
þ Synchronet þ KD3net-Rhode Island's only BBS about nothing. http://bbs.kd3.us

src/conio sdl_con.c 1.224 1.225
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv31425

Modified Files:
sdl_con.c
Log Message:
Use 32-bit RGB surfaces.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.225 1.226
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv32111

Modified Files:
sdl_con.c
Log Message:
Request the current display depth rather than an emulated screen.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.226 1.227
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv1191

Modified Files:
sdl_con.c
Log Message:
Only remap colours that have been added, not all colours. Fixes issue
with palette redefinitions when the window scaling changes in SDL Surface mode.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.227 1.228
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv1454

Modified Files:
sdl_con.c
Log Message:
Always call SDL_SetColors() from sdl_set_colours()... that's kinda the
point of the function.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.228 1.229
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv1561

Modified Files:
sdl_con.c
Log Message:
We need to call SDL_SetColors() before SDL_MapRGB() or we won't get our
new colour.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.232 1.233
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv15704

Modified Files:
sdl_con.c
Log Message:
Don't wait for sdl_ufinc_rec semaphore on UPDATERECT calls...

It's WAY too slow... the average latency per function is set by the SLEEP(1)
on line 1641... I'm not sure we have a reliable cross-platform sleep more granular, and 1ms is too slow if we're doing individual pixels.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.233 1.234
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv18089

Modified Files:
sdl_con.c
Log Message:
Make a copy of vstat for the hot paths, avoid locking vstatlock there.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.234 1.235
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv19417

Modified Files:
sdl_con.c
Log Message:
SDL_USEREVENT_FLUSH is in the hot path too... and it really doesn't matter
if it gets lost. The UPDATERECT does matter, but we'll see if lost events start happening again.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.241 1.242
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv20200

Modified Files:
sdl_con.c
Log Message:
Enable lost event detection for DRAWRECT and FLUSH again...

Looks like we need to eat that performance cost.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.242 1.243
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv12494

Modified Files:
sdl_con.c
Log Message:
Keep rectangles in a linked list until a flush is called. This keeps
pressure off the event loop, avoids waiting for a response for each update,
and keeps the existing event retry logic in place. It will also never
re-send a rectangle any more, regardless of what happens.

HOPEFULLY this will fix the SDL crap once and for all.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.243 1.244
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv20406

Modified Files:
sdl_con.c
Log Message:
Tighten up the sdl_ufunc_ret semaphore, ensure it's not already posted
before waiting for it.

Flush doesn't need to sdl_ufunc_ret semaphore anymore since it places rectangles back in the list on failure

Only enable SDL_SYSWMEVENT (a main cause of issues) during copy/paste operations. This should mostly elimiate timeouts and odd 1-second pasues during initialization on Linux.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.244 1.245
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv22970

Modified Files:
sdl_con.c
Log Message:
Fix various classes of use after free bugs...

The biggest one is if a lost event is re-sent, we now LEAK whatever's in data1/data2 rather than the pervious case where we would use it after
free if it was delivered more than once.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.245 1.246
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv10943

Modified Files:
sdl_con.c
Log Message:
Only use the ufunc_rec semaphore when we're going to need it... fixes
long delays on Win32.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.246 1.247
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv13026

Modified Files:
sdl_con.c
Log Message:
Since we're batching updates now, stop tracking all the updated rects and
just use SDL_Flip().




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.247 1.248
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv7682

Modified Files:
sdl_con.c
Log Message:
Fix all kinds of palette related unlikely bugs.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.248 1.249
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv12222

Modified Files:
sdl_con.c
Log Message:
Fix SDL palette setting.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.249 1.250
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv13170

Modified Files:
sdl_con.c
Log Message:
Heh, index isn't a uint16_t




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.250 1.251
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv27217

Modified Files:
sdl_con.c
Log Message:
vstat.scaling MUST always equal 2 for overlay mode to work.

Also fix some crazy dereference gibberish.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.252 1.253
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv22178

Modified Files:
sdl_con.c
Log Message:
Fix error releasing the headlock.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.253 1.254
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv22445

Modified Files:
sdl_con.c
Log Message:
Don't free rectlist in the driver.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.254 1.255
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv24984

Modified Files:
sdl_con.c
Log Message:
Ensure tail is set when the head is initialized.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.257 1.258
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv14491

Modified Files:
sdl_con.c
Log Message:
Use integer math for RGB -> YUV conversion.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.258 1.259
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv16960

Modified Files:
sdl_con.c
Log Message:
Some more minor optimizations




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.260 1.261
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv3875

Modified Files:
sdl_con.c
Log Message:
Some minor issues
--coverity



---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.261 1.262
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv20144

Modified Files:
sdl_con.c
Log Message:
Fix GCC warning:
sdl_con.c: In function ‘sdl_mouse_thread’:
sdl_con.c:1424:1: warning: no return statement in function returning non-void




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.262 1.263
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv11840

Modified Files:
sdl_con.c
Log Message:
The key normally used for "Backspace" (on a PC keyboard) is labeled "Delete"
on a Mac keyboard and although that key sends the ASCII DEL char (127), we're converting it to backspace (^H) here. But if the user actually wants the "delete character" key functionality, there is no dedicated key, so treat Ctrl-Delete (on a Mac) as Delete (not backspace).



---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.263 1.264
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv31657/conio

Modified Files:
sdl_con.c
Log Message:
Fix 40-col doubling to modify vstat with the vstatlock held rather than just cvstat.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.264 1.265
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv4577

Modified Files:
sdl_con.c
Log Message:
Blank existing window before calling SetVideoMode() again.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.267 1.268
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv655

Modified Files:
sdl_con.c
Log Message:
Remove unused variable.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.270 1.271
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv21111

Modified Files:
sdl_con.c
Log Message:
Fixup termination, and always lock the mutex before fiddling with the window

Previously, an atexit() function could call exit() (for non-SyncTERM utils) causing the program to occasionally crash rather than exit cleanly.



---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.271 1.272
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv4292

Modified Files:
sdl_con.c
Log Message:
Set the window size when setup_surfaces() is called.
This allows the saved scaling to work again, and makes the mouse not weird
when scaling is secretly set differently.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.272 1.273
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv15252

Modified Files:
sdl_con.c
Log Message:
The update rectangle applies to both the texture and the renderer, not just
the renderer. Fixes crash when there's fewer screen pixels than in the
window.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.278 1.279
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv11577

Modified Files:
sdl_con.c
Log Message:
Always destroy the texture before creating a new one.
Derp.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.280 1.281
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv9227

Modified Files:
sdl_con.c
Log Message:
Double both height and width when switching to 40 column mode.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.281 1.282
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv16882

Modified Files:
sdl_con.c
Log Message:
Update input translation for SDL2...
Hopefully this fixes the issue Ragnorok was reporting with a Spanish
keyboard.



---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.282 1.283
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv18737

Modified Files:
sdl_con.c
Log Message:
I take that last back... this likely won't fix anything... SDL2 just has
a major regression for key input. :(




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.283 1.284
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv7299

Modified Files:
sdl_con.c
Log Message:
Overhaul input yet again... this *should* fix Ragnorok's problem... but
*may* break all sorts of other things.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

El 15/4/20 a las 21:02, deuce escribi¢:

src/conio sdl_con.c 1.283 1.284
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv7299

Modified Files:
sdl_con.c
Log Message:
Overhaul input yet again... this *should* fix Ragnorok's problem... but
*may* break all sorts of other things.

realy break!
i move o up/down/let/right and segfault

Starting program: /home/ragnarok/src/ftoledo/sbbs/cvs/src/syncterm/gcc.linux.x64.exe.debug/syncterm

[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7ffff6479700 (LWP 3417)]
[Thread 0x7ffff6479700 (LWP 3417) exited]
[New Thread 0x7fffe8015700 (LWP 3418)]
[New Thread 0x7fffe7814700 (LWP 3419)]
[New Thread 0x7fffe7013700 (LWP 3420)]
[New Thread 0x7fffe6812700 (LWP 3421)]

Thread 4 "Main Thread" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fffe7814700 (LWP 3419)]
0x00005555556f3811 in sdl_video_event_thread (data=0x0) at sdl_con.c:836
836 if (block_text || !isprint(ev.key.keysym.sym))
(gdb) bt
#0 0x00005555556f3811 in sdl_video_event_thread (data=0x0) at sdl_con.c:836
#1 0x00007ffff705e4a4 in start_thread (arg=0x7fffe7814700) at pthread_create.c:456
#2 0x00007ffff6da0d0f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:97

also (previous this last commits) i fond that hang when try to play bell
sound i think:

(gdb) bt
#0 __lll_lock_wait () at
../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135
#1 0x00007ffff7060bb5 in __GI___pthread_mutex_lock
(mutex=0x555555c230c0 <sample_mutex>) at ../nptl/pthread_mutex_lock.c:80
#2 0x000055555570f805 in xp_play_sample (
sample=0x555555bf1440 <wave> "~\177\177\200\200\201\201\200\200\177}zywvuuuuvvuuutrqqppoooppponljhfdcbbcfhlpx|z\202\210\211\210\203zria[VSSSSUUUSQMIE@><;;<>>>;81*!\031\020\n\a\a\f\023\037+;^ly\202\210\212\207\203|skb\\VTRSSUUTTQNIEB><<;<>>=<81*\"\031\021\v\b\a\v\022\034+8[kw\202\210\212\210\204}tkc]WSRRSUUUTRNJFA><;;<=>?<92+$\033\022\f\b\006\n\020\034("...,
size=2205, background=1) at xpbeep.c:924
#3 0x00005555556f4237 in sdl_beep () at sdl_con.c:1052
#4 0x00005555556d7403 in ciolib_beep () at ciolib.c:1716
#5 0x00005555556d6af9 in ciolib_putch (ch=7) at ciolib.c:1515
#6 0x00005555556eb1c0 in cterm_write (cterm=0x555555e61870, vbuf=0x7fffffff4f10, buflen=4, retbuf=0x555555bc3300 <ansi_replybuf> "", retsize=2048,
speed=0x7fffffff4efc) at cterm.c:5197
#7 0x0000555555584da4 in doterm (bbs=0x555555bbc040 <retlist>) at
term.c:2579
#8 0x00005555555912cc in main (argc=1, argv=0x7fffffffdea8) at
syncterm.c:1673
(gdb) c
Continuing.
[Thread 0x7fffe56d0700 (LWP 31011) exited]
c^C
Thread 1 "Main Thread" received signal SIGINT, Interrupt.
__lll_lock_wait () at ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135
135 in ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S
(gdb) bt
#0 __lll_lock_wait () at
../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135
#1 0x00007ffff7060bb5 in __GI___pthread_mutex_lock
(mutex=0x555555c230c0 <sample_mutex>) at ../nptl/pthread_mutex_lock.c:80
#2 0x000055555570f805 in xp_play_sample (
sample=0x555555bf1440 <wave> "~\177\177\200\200\201\201\200\200\177}zywvuuuuvvuuutrqqppoooppponljhfdcbbcfhlpx|z\202\210\211\210\203zria[VSSSSUUUSQMIE@><;;<>>>;81*!\031\020\n\a\a\f\023\037+;^ly\202\210\212\207\203|skb\\VTRSSUUTTQNIEB><<;<>>=<81*\"\031\021\v\b\a\v\022\034+8[kw\202\210\212\210\204}tkc]WSRRSUUUTRNJFA><;;<=>?<92+$\033\022\f\b\006\n\020\034("...,
size=2205, background=1) at xpbeep.c:924
#3 0x00005555556f4237 in sdl_beep () at sdl_con.c:1052
#4 0x00005555556d7403 in ciolib_beep () at ciolib.c:1716
#5 0x00005555556d6af9 in ciolib_putch (ch=7) at ciolib.c:1515
#6 0x00005555556eb1c0 in cterm_write (cterm=0x555555e61870, vbuf=0x7fffffff4f10, buflen=4, retbuf=0x555555bc3300 <ansi_replybuf> "", retsize=2048,
speed=0x7fffffff4efc) at cterm.c:5197
#7 0x0000555555584da4 in doterm (bbs=0x555555bbc040 <retlist>) at
term.c:2579
#8 0x00005555555912cc in main (argc=1, argv=0x7fffffffdea8) at
syncterm.c:1673
(gdb) c
Continuing.


Saludos! and thanks!

---
¨ Synchronet ¨ Vertrauen ¨ Home of Synchronet ¨ [vert/cvs/bbs].synchro.net

---
þ Synchronet þ Dock Sud BBS TLD 24 HS - bbs.docksud.com.ar

Re: Re: src/conio/sdl_con.c
By: Ragnarok to deuce on Wed Apr 15 2020 10:57 pm

El 15/4/20 a las 21:02, deuce escribi¢:

src/conio sdl_con.c 1.283 1.284
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv7299

Modified Files:
sdl_con.c
Log Message:
Overhaul input yet again... this *should* fix Ragnorok's problem... but *may* break all sorts of other things.

realy break!
i move o up/down/let/right and segfault

Starting program: /home/ragnarok/src/ftoledo/sbbs/cvs/src/syncterm/gcc.linux.x6 4.exe.debug/syncterm

[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". [New Thread 0x7ffff6479700 (LWP 3417)]
[Thread 0x7ffff6479700 (LWP 3417) exited]
[New Thread 0x7fffe8015700 (LWP 3418)]
[New Thread 0x7fffe7814700 (LWP 3419)]
[New Thread 0x7fffe7013700 (LWP 3420)]
[New Thread 0x7fffe6812700 (LWP 3421)]

Thread 4 "Main Thread" received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x7fffe7814700 (LWP 3419)]
0x00005555556f3811 in sdl_video_event_thread (data=0x0) at sdl_con.c:836
836 if (block_text ||
!isprint(ev.key.keysym.sym))
(gdb) bt
#0 0x00005555556f3811 in sdl_video_event_thread (data=0x0) at sdl_con.c:836 #1 0x00007ffff705e4a4 in start_thread (arg=0x7fffe7814700) at pthread_create.c:456
#2 0x00007ffff6da0d0f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:97

also (previous this last commits) i fond that hang when try to play bell sound i think:

(gdb) bt
#0 __lll_lock_wait () at ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135
#1 0x00007ffff7060bb5 in __GI___pthread_mutex_lock
(mutex=0x555555c230c0 <sample_mutex>) at ../nptl/pthread_mutex_lock.c:80
#2 0x000055555570f805 in xp_play_sample (
sample=0x555555bf1440 <wave> "~\177\177\200\200\201\201\200\200\177}zywvuuu uvvuuutrqqppoooppponljhfdcbbc fhlpx|z\202\210\211\210\203zria[VSSSSUUUSQMIE@><;; <>>>;81*!\031\020\n\a\a\f \023\037+;^ly\202\210\212\207\203|skb\\VTRSSUUTTQNIEB> <<;<>>=<81*\"\031\021 \v\b\a\v\022\034+8[kw\202\210\212\210\204}tkc]WSRRSUUUTRNJ FA><;;<=>?<92+$\033\022\f\b\006\n\020\034("...,
size=2205, background=1) at xpbeep.c:924
#3 0x00005555556f4237 in sdl_beep () at sdl_con.c:1052
#4 0x00005555556d7403 in ciolib_beep () at ciolib.c:1716
#5 0x00005555556d6af9 in ciolib_putch (ch=7) at ciolib.c:1515
#6 0x00005555556eb1c0 in cterm_write (cterm=0x555555e61870, vbuf=0x7fffffff4f10, buflen=4, retbuf=0x555555bc3300 <ansi_replybuf> "", retsize=2048,
speed=0x7fffffff4efc) at cterm.c:5197
#7 0x0000555555584da4 in doterm (bbs=0x555555bbc040 <retlist>) at term.c:2579
#8 0x00005555555912cc in main (argc=1, argv=0x7fffffffdea8) at syncterm.c:1673
(gdb) c
Continuing.
[Thread 0x7fffe56d0700 (LWP 31011) exited]
c^C
Thread 1 "Main Thread" received signal SIGINT, Interrupt.
__lll_lock_wait () at ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 135 in ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S
(gdb) bt
#0 __lll_lock_wait () at ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135
#1 0x00007ffff7060bb5 in __GI___pthread_mutex_lock
(mutex=0x555555c230c0 <sample_mutex>) at ../nptl/pthread_mutex_lock.c:80
#2 0x000055555570f805 in xp_play_sample (
sample=0x555555bf1440 <wave> "~\177\177\200\200\201\201\200\200\177}zywvuuu uvvuuutrqqppoooppponljhfdcbbc fhlpx|z\202\210\211\210\203zria[VSSSSUUUSQMIE@><;; <>>>;81*!\031\020\n\a\a\f \023\037+;^ly\202\210\212\207\203|skb\\VTRSSUUTTQNIEB> <<;<>>=<81*\"\031\021 \v\b\a\v\022\034+8[kw\202\210\212\210\204}tkc]WSRRSUUUTRNJ FA><;;<=>?<92+$\033\022\f\b\006\n\020\034("...,
size=2205, background=1) at xpbeep.c:924
#3 0x00005555556f4237 in sdl_beep () at sdl_con.c:1052
#4 0x00005555556d7403 in ciolib_beep () at ciolib.c:1716
#5 0x00005555556d6af9 in ciolib_putch (ch=7) at ciolib.c:1515
#6 0x00005555556eb1c0 in cterm_write (cterm=0x555555e61870, vbuf=0x7fffffff4f10, buflen=4, retbuf=0x555555bc3300 <ansi_replybuf> "", retsize=2048,
speed=0x7fffffff4efc) at cterm.c:5197
#7 0x0000555555584da4 in doterm (bbs=0x555555bbc040 <retlist>) at term.c:2579
#8 0x00005555555912cc in main (argc=1, argv=0x7fffffffdea8) at syncterm.c:1673
(gdb) c
Continuing.

Saludos! and thanks!

I don't think Deuce is going to see your message here without being directed to it. I'll try to direct him to it.

digital man

Synchronet/BBS Terminology Definition #17:
DCD = Data Carrier Detect
Norco, CA WX: 64.4øF, 45.0% humidity, 5 mph ESE wind, 0.00 inches rain/24hrs

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.284 1.285
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv353

Modified Files:
sdl_con.c
Log Message:
isprint() segfaults if the argument is greater than some locale-dependent value.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.285 1.286
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv1002

Modified Files:
sdl_con.c
Log Message:
Right, but we still need isprint() (even on Linux)




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

Re: Re: src/conio/sdl_con.c
By: Digital Man to Ragnarok on Wed Apr 15 2020 09:07 pm

I don't think Deuce is going to see your message here without being directed to it. I'll try to direct him to it.

<@Deuce> DigitalMan, if you could point him at the bug tracker on SourceForge, that is the best place to report SyncTERM bugs.

digital man

This Is Spinal Tap quote #28:
We've got Armadillos in our trousers. It's really quite frightening.
Norco, CA WX: 68.3øF, 38.0% humidity, 3 mph SSE wind, 0.00 inches rain/24hrs

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.288 1.289
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv19397

Modified Files:
sdl_con.c
Log Message:
Return the UTF-8 clipboard test.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

El 16/4/20 a las 01:07, Digital Man escribi¢:

Re: Re: src/conio/sdl_con.c
By: Ragnarok to deuce on Wed Apr 15 2020 10:57 pm

El 15/4/20 a las 21:02, deuce escribi¨:

src/conio sdl_con.c 1.283 1.284
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv7299

Modified Files:
sdl_con.c
Log Message:
Overhaul input yet again... this *should* fix Ragnorok's problem... but *may* break all sorts of other things.

realy break!
i move o up/down/let/right and segfault

Starting program: /home/ragnarok/src/ftoledo/sbbs/cvs/src/syncterm/gcc.linux.x6 4.exe.debug/syncterm

[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". [New Thread 0x7ffff6479700 (LWP 3417)]
[Thread 0x7ffff6479700 (LWP 3417) exited]
[New Thread 0x7fffe8015700 (LWP 3418)]
[New Thread 0x7fffe7814700 (LWP 3419)]
[New Thread 0x7fffe7013700 (LWP 3420)]
[New Thread 0x7fffe6812700 (LWP 3421)]

Thread 4 "Main Thread" received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x7fffe7814700 (LWP 3419)]
0x00005555556f3811 in sdl_video_event_thread (data=0x0) at sdl_con.c:836 836 if (block_text ||
!isprint(ev.key.keysym.sym))
(gdb) bt
#0 0x00005555556f3811 in sdl_video_event_thread (data=0x0) at sdl_con.c:836
#1 0x00007ffff705e4a4 in start_thread (arg=0x7fffe7814700) at pthread_create.c:456
#2 0x00007ffff6da0d0f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:97

also (previous this last commits) i fond that hang when try to play bell sound i think:

(gdb) bt
#0 __lll_lock_wait () at ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135
#1 0x00007ffff7060bb5 in __GI___pthread_mutex_lock
(mutex=0x555555c230c0 <sample_mutex>) at ../nptl/pthread_mutex_lock.c:80 #2 0x000055555570f805 in xp_play_sample (
sample=0x555555bf1440 <wave> "~\177\177\200\200\201\201\200\200\177}zywvuuu uvvuuutrqqppoooppponljhfdcbbc
fhlpx|z\202\210\211\210\203zria[VSSSSUUUSQMIE@><;; <>>>;81*!\031\020\n\a\a\f
\023\037+;^ly\202\210\212\207\203|skb\\VTRSSUUTTQNIEB> <<;<>>=<81*\"\031\021
\v\b\a\v\022\034+8[kw\202\210\212\210\204}tkc]WSRRSUUUTRNJ FA><;;<=>?<92+$\033\022\f\b\006\n\020\034("...,
size=2205, background=1) at xpbeep.c:924
#3 0x00005555556f4237 in sdl_beep () at sdl_con.c:1052
#4 0x00005555556d7403 in ciolib_beep () at ciolib.c:1716
#5 0x00005555556d6af9 in ciolib_putch (ch=7) at ciolib.c:1515
#6 0x00005555556eb1c0 in cterm_write (cterm=0x555555e61870, vbuf=0x7fffffff4f10, buflen=4, retbuf=0x555555bc3300 <ansi_replybuf> "", retsize=2048,
speed=0x7fffffff4efc) at cterm.c:5197
#7 0x0000555555584da4 in doterm (bbs=0x555555bbc040 <retlist>) at term.c:2579
#8 0x00005555555912cc in main (argc=1, argv=0x7fffffffdea8) at syncterm.c:1673
(gdb) c
Continuing.
[Thread 0x7fffe56d0700 (LWP 31011) exited]
c^C
Thread 1 "Main Thread" received signal SIGINT, Interrupt.
__lll_lock_wait () at ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 135 in ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S
(gdb) bt
#0 __lll_lock_wait () at ../sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135
#1 0x00007ffff7060bb5 in __GI___pthread_mutex_lock
(mutex=0x555555c230c0 <sample_mutex>) at ../nptl/pthread_mutex_lock.c:80 #2 0x000055555570f805 in xp_play_sample (
sample=0x555555bf1440 <wave> "~\177\177\200\200\201\201\200\200\177}zywvuuu uvvuuutrqqppoooppponljhfdcbbc
fhlpx|z\202\210\211\210\203zria[VSSSSUUUSQMIE@><;; <>>>;81*!\031\020\n\a\a\f
\023\037+;^ly\202\210\212\207\203|skb\\VTRSSUUTTQNIEB> <<;<>>=<81*\"\031\021
\v\b\a\v\022\034+8[kw\202\210\212\210\204}tkc]WSRRSUUUTRNJ FA><;;<=>?<92+$\033\022\f\b\006\n\020\034("...,
size=2205, background=1) at xpbeep.c:924
#3 0x00005555556f4237 in sdl_beep () at sdl_con.c:1052
#4 0x00005555556d7403 in ciolib_beep () at ciolib.c:1716
#5 0x00005555556d6af9 in ciolib_putch (ch=7) at ciolib.c:1515
#6 0x00005555556eb1c0 in cterm_write (cterm=0x555555e61870, vbuf=0x7fffffff4f10, buflen=4, retbuf=0x555555bc3300 <ansi_replybuf> "", retsize=2048,
speed=0x7fffffff4efc) at cterm.c:5197
#7 0x0000555555584da4 in doterm (bbs=0x555555bbc040 <retlist>) at term.c:2579
#8 0x00005555555912cc in main (argc=1, argv=0x7fffffffdea8) at syncterm.c:1673
(gdb) c
Continuing.

Saludos! and thanks!

I don't think Deuce is going to see your message here without being directed to it. I'll try to direct him to it.

digital man

ok!! my apologies! where is the better path to report bugs to syncterm?

---
þ Synchronet þ Dock Sud BBS TLD 24 HS - bbs.docksud.com.ar

Re: Re: src/conio/sdl_con.c
By: Ragnarok to Digital Man on Fri Apr 17 2020 08:43 pm

I don't think Deuce is going to see your message here without being directed to it. I'll try to direct him to it.

ok!! my apologies! where is the better path to report bugs to syncterm?

https://sourceforge.net/p/syncterm/tickets/

digital man

Synchronet "Real Fact" #41:
Synchronet's cross-platform library is called XPDEV (named before Windows XP). Norco, CA WX: 60.8øF, 67.0% humidity, 8 mph ESE wind, 0.00 inches rain/24hrs

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.290 1.291
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv5914

Modified Files:
sdl_con.c
Log Message:
Translate non-printing input characters using the codepage, don't assume anything that doesn't print is "fine".




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.292 1.293
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv5303

Modified Files:
sdl_con.c
Log Message:
Fix height calculation when changing modes.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.294 1.295
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv14138

Modified Files:
sdl_con.c
Log Message:
Move sdl.SetWindowMinimumSize() into appropropriate thread.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.295 1.296
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv4331

Modified Files:
sdl_con.c
Log Message:
Nope, putting it there doesn't work either... just do it before we
change the window size and it should work out alright.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.297 1.298
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv30415

Modified Files:
sdl_con.c
Log Message:
Don't treat keysyms as though they are unicode (they aren't).




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.298 1.299
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv9464

Modified Files:
sdl_con.c
Log Message:
First attempt at getting AltGr working for ragnarok in SDL mode.
The SDL wiki lists KMOD_MODE as AltGR... if that flag is flying in a
keypress, do not block text input.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.299 1.300
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv20611

Modified Files:
sdl_con.c
Log Message:
Commit this for no good reason.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

El 28/4/20 a las 02:25, deuce escribi¢:

src/conio sdl_con.c 1.298 1.299
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv9464

Modified Files:
sdl_con.c
Log Message:
First attempt at getting AltGr working for ragnarok in SDL mode.
The SDL wiki lists KMOD_MODE as AltGR... if that flag is flying in a keypress, do not block text input.

YOU ARE the fucking king of the world of bbs terminals!


. * .. . * *

* * @()Ooc()* o .

(Q@*0CG*O() ___

|\_________/|/ _ \

| | | | | / | |

| | | | | | | |

| | | | | | | |

| | | | | | | |

| | | | | | | |

| | | | | \_| |

| | | | |\___/
|\_|__|__|_/|

\_________/



work perfect!!!!!!!!!!!!

---
þ Synchronet þ Dock Sud BBS TLD 24 HS - bbs.docksud.com.ar

src/conio sdl_con.c 1.302 1.303
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv28032

Modified Files:
sdl_con.c
Log Message:
Allow DEL key to work again.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.304 1.305
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv11032

Modified Files:
sdl_con.c
Log Message:
Remove debug output from last commit.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.307 1.308
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv10188

Modified Files:
sdl_con.c
Log Message:
Fix error which caused keypad with NumLock on to send both the arrow and
number to the remote.
Fixes bug https://sourceforge.net/p/syncterm/tickets/15/




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.310 1.311
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv17007

Modified Files:
sdl_con.c
Log Message:
Support building with ancient versions of SDL2 from six years ago.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.311 1.312
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv413

Modified Files:
sdl_con.c
Log Message:
Work around issue in SDL2 (or OpenGL? Or nvidia driver?) where creating
a new texture with the same address as a destroyed texture causes crashes.
Just allocate the new texture before freeing the old one to prevent that
from occuring.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

src/conio sdl_con.c 1.312 1.313
Update of /cvsroot/sbbs/src/conio
In directory cvs:/tmp/cvs-serv21104

Modified Files:
sdl_con.c
Log Message:
Clamp the rectangle height to the texture height when copying pixel data.
May fix https://sf.net/p/syncterm/tickets/19/

Please try to log in to SourceForge when creating issues or I won't be able
to contact you with follow-up questions or get confirmation when an issue
is resolved.




---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/0634130f732643afcefe1b34
Modified Files:
src/conio/sdl_con.c
Log Message:
Fix SF #44 problem keypad non-numbers

Make an exception for all keypad stuff.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/248eb2b590cfc5dee387ad2c
Modified Files:
src/conio/sdl_con.c
Log Message:
Special-case TAB so "backtab" works properly.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/1495774a066cb027a3d96cf7
Modified Files:
src/conio/sdl_con.c
Log Message:
Fix AltGr with SDL2

This is pretty insane since SDL2 completely ignores the keysym value
of the X11 keyboard event, so here's what we do...

1) When there's a keypress that includes right-alt, store the sym/mod
2) When we get text input, if it's the same as we would get if ALT
wasn't pressed for the last keydown, parse through the mapping
(ie: handle ALT keys)
3) If it's different, use that (gets AltGr modified value)

I get that AltGr is hard in a cross-platform way, but just pretending
it doesn't exist at all for key input is a very weird choice.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/5f027c495b76aabf182c4905
Modified Files:
src/conio/sdl_con.c
Log Message:
Keep correct aspect ratio when resizing window.

Also, the 8x16 font at 80x25 is *not* a square pixel aspect ratio.
As a result, it now stays "proper", which is likely not what you
want.

This was discovered when I was trying to make DigitalMan happy
because he wanted the apsect ratio retained on scaling.

Sorry for the impending imminent scaling hassle DigitalMan!

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/26def07178f2affe11442bf5
Modified Files:
src/conio/sdl_con.c
Log Message:
More fallout from non-square pixels...

This fixes all the window adjustment "stuff", but the new aspect
ratio enforcement means that ALT-UP and ALT-DOWN are useless for
resizing the window unless it's not already snapped... use ALT-LEFT
and ALT-RIGHT to decrease/increase the window size.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/9aca9bf4b603144ffb46a09f
Modified Files:
src/conio/sdl_con.c
Log Message:
sdl_con.c can have stdbool.h too.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/49756aa756b0570337da16e7
Modified Files:
src/conio/sdl_con.c
Log Message:
Use lround(), not roundl().

Also, include math.h

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/e7100fcbd1aae6959c38ee40
Modified Files:
src/conio/sdl_con.c
Log Message:
Specify HighDPI mode when creating a window.

This may solve issue where using Windows "DPI scaling" forces the
window to be scaled.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/fccd7c1dd05309a672690d0d
Modified Files:
src/conio/sdl_con.c
Log Message:
Use the API to insist SyncTERM is "High DPI aware"

Microsoft wants us to use a manifest for this, but I really don't
feel like digging into however one uses manifest files when not
using the Microsoft build files.

Details here: https://docs.microsoft.com/en-us/windows/win32/hidpi/setting-the-default-dpi-awareness-for-a-process

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/a194a15736de98517f48a9b3
Modified Files:
src/conio/sdl_con.c
Log Message:
Various SDL fixes:

Pass window size in SDL_USEREVENT_SETVIDMODE
- Fixes race condition where SDL_WINDOWEVENT_RESIZED or
SDL_WINDOWEVENT_SIZE_CHANGED arraves while
SDL_USEREVENT_SETVIDMODE is in the queue

Set new scaling based on old scaling multiplier
- Will resize window on mode changes, but tries to keep integer
multiplier similar. May break fullscreen modes, dunno.

Have aspect_fix() return higher rather than lower
- This may make it the same as aspect_correct(), but I'm too lazy
to figure that out.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/2fa700696c34fb298fd24c7b
Modified Files:
src/conio/sdl_con.c
Log Message:
Now that SDL_USEREVENT_SETVIDMODE takes a size, we can clean up sdl_setwinsize() and eliminate another race.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/f47c60792c13c8bd9c4e67c4
Modified Files:
src/conio/sdl_con.c
Log Message:
Fix a couple issues found by Coverity

1) Unreachable return without a lock in sdl_useR_func_ret() (harmless)
2) LOR in handling SDL_USEREVENT
- vstatlock was obtained inside of win_mutex which should never
be done... this could (and likely does) result in a deadlock
when using SDL mode.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/6369925c868a8a9485e6627f
Modified Files:
src/conio/sdl_con.c
Log Message:
Use a local vstat copy instead of cvstat to avoid holding the lock
for an extended time. Should fix SDL mode slowness.

Only call RenderPresent() if we called RenderCopy(). Should fix
SDL mode flickering.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/f06029a3889734491909ebe9
Modified Files:
src/conio/sdl_con.c
Log Message:
Add TODO comment for cvstat.

When originally created, it was intended to be accessed only by a
single thread, and therefore not need mutex protection, allowing
less code to run with vstatlock held. Unfortunately, this is not
how it turned out, so cvstat should either be removed, or it should
actually be cleaned up to work as intended.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/fdce00a456a33dc858067aef
Modified Files:
src/conio/sdl_con.c
Log Message:
Push cvstat into sdl_video_event_thread() stack

This makes cvstat work as intended, so we no longer need to lock
vstatlock when accessing it since it's only accessed from the
video event thread.

There's still an irritating dependency on vstat for ALT-Enter handling.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/22c8b09f824496b7c68e599e
Modified Files:
src/conio/sdl_con.c
Log Message:
Get vstat out of sdl_add_key() path

This should almost completely remove vstatlock contention in SyncTERM,
which should bring SDL output performance close to X11 performance.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/832871707fc3f2d37ff1891d
Modified Files:
src/conio/sdl_con.c
Log Message:
On a window resize event, only set the window size if it's different

Fixes issue on Linux where the window size is constantly updated,
and window rabidly steals focus.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/defc638c9f1cc95ca234a780
Modified Files:
src/conio/sdl_con.c
Log Message:
Check SDL window size when checking if size changed.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/47977472de4bfce66f1bd4d1
Modified Files:
src/conio/sdl_con.c
Log Message:
Don't treat numpad /*-+ differently based on NumLock in SDL

This caused double-keys on those when numlock was off. Only the
number keys and period need special NumLock handling.

Fixes bug 108 on SourceForge, thanks nelgin!

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/9137e82c9afa91f0c0647c70
Modified Files:
src/conio/sdl_con.c
Log Message:
Fix warnings reported by Digitalman

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/875c77ea0bc774be2c2a4f67
Modified Files:
src/conio/sdl_con.c
Log Message:
Resolve GCC warnings: cast from pointer to integer of different size

Pretty sure this is what intptr_t was created for.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/4af4effb3148c0c695e56ad1
Modified Files:
src/conio/sdl_con.c
Log Message:
Fix deadlock in SDL mode.

sdl_getscaling() does not require vstatlock. Move it out.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/978d4d0b7b64a39a30092e57
Modified Files:
src/conio/sdl_con.c
Log Message:
Use cast instead of pointer math to fix Win32 build

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/268893cc9b77f7e96343bb11
Modified Files:
src/conio/sdl_con.c
Log Message:
When switching from fullscreen mode, recalculate win size.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/7c44187868e59355f598180a
Modified Files:
src/conio/sdl_con.c
Log Message:
For SDL mode, ensure rect matches vstat as well.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/960cee2703cd4650f2805d4f
Modified Files:
src/conio/sdl_con.c
Log Message:
Do not attempt to clean up SDL unless it was initialized.

Should fix js test suite on darwin.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/d281862b290f8ee309892e63
Modified Files:
src/conio/sdl_con.c
Log Message:
Call exit() from exit_sdl_con if ciolib_reaper is enabled and SDL is not initialized.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

https://gitlab.synchro.net/main/sbbs/-/commit/ea900928153cef278cf404bc
Modified Files:
src/conio/sdl_con.c
Log Message:
Revert last change reaper is explicitly disabled here.

---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net