AW: 1.9.2.0: Xwin SIGSEGV when font server should be queried
The previous version, that was stable for xorg-server was 1.8.2-1, according to setup.log. The gdb stack trace is not very talkative: ... Loaded symbols for /usr/bin/cygXrender-1.dll [Switching to thread 8788.0x2740] (gdb) c Continuing. (no debugging symbols found) Program received signal SIGSEGV, Segmentation fault. [Switching to thread 8788.0x1178] 0x005508b5 in ?? () (gdb) bt full #0 0x005508b5 in ?? () No symbol table info available. #1 0x000c in ?? () No symbol table info available. #2 0x005fa120 in ?? () No symbol table info available. #3 0x in ?? () No symbol table info available. I will download current source and compile it with -g . Something funny about it: I had the "xset fp+ tcp/mystique:7100" commented out in the .startxwinrc. Then I started XWin, attached the gdb and issued the xset command from within a AIX aixterm shell. Then I started a window that has to use a font server font and it worked. Then I activated the xset command in the .startxwinrc again and the XWin crashed again. I keep on researching. Burkhard _ Geschaeftsfuehrer/Managing Directors: Christoph Hahn-Woernle, Frank Apel HRB 17335, Amtsgericht Stuttgart (Commercial Register District Court Stuttgart) St.-Nr. 99064/06051, USt-IdNr./VAT Reg.No.: DE 203036780 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Updated: xorg-server-1.9.2-1
Dear Jon, Since the update I have been unable to start XWin under [1.7]. Following my usual sequence at a bash prompt: mount n:/tmp /tmp # drive n: is NTFS as required run XWin -multiwindow & Normally these two would be followed by a command line for xterm, but I am not getting that far! The second line repeatedly generates a Cygwin/X error message box, where the reason given is very familiar: Can't read lock file /tmp/.X0-lock I used to get this when /tmp was FAT32 but as you see from line 1 I am making an explicit mount to a NTFS drive, and in the recent past this step has been all that is necessary to make progress. When I open /var/log/xwin/XWin.0.log for more information I read ~> cat /var/log/xwin/XWin.0.log Welcome to the XWin X Server Vendor: The Cygwin/X Project Release: 1.9.2.0 (10902000) Build Date: 2010-11-03 XWin was started with the following command line: /usr/bin/XWin -multiwindow ddxProcessArgument - Initializing default screens winInitializeScreenDefaults - primary monitor w 1280 h 1024 winInitializeDefaultScreens - native DPI x 96 y 96 _XSERVTransSocketOpenCOTSServer: Unable to open socket for inet6 _XSERVTransOpen: transport open failed for inet6/mandible:0 _XSERVTransMakeAllCOTSServerListeners: failed to open listener for inet6 Fatal server error: Can't read lock file /tmp/.X0-lock [1]+ Donerun XWin -multiwindow Can anything be determined from the 3 lines in this file beginning _XSERV ("Unable" .. "failed" .. "failed")? Do I need to change my command sequence? BTW it's interesting (is it?) that the error message refers to 1.9.2.0 not 1.9.2-1. I'm surprised that there are no other anguished reports to you following the update, but I guess that's because the fault/ problem/ difficulty is mine rather than general. ?? Any insights much appreciated. Thank you, Fergus PS It kills me not to copy this to cygwin at cygwin dot com which I feel will have a much wider readership and skill set amongst that readership, but anybody sinning in this regard is always referred straight to Cygwin-X. A strange separation of reporting. Not that I am doubting your specific capability, just the potential for "Me Too" amongst readers. (Or even "Not Me, but I Know What's Going Wrong".) -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: AW: 1.9.2.0: Xwin SIGSEGV when font server should be queried
On 08/11/2010 10:21, Weeber, Burkhard wrote: > The previous version, that was stable for xorg-server was 1.8.2-1, > according to setup.log. > > The gdb stack trace is not very talkative: > ... > Loaded symbols for /usr/bin/cygXrender-1.dll > [Switching to thread 8788.0x2740] > (gdb) c > Continuing. > (no debugging symbols found) > > Program received signal SIGSEGV, Segmentation fault. > [Switching to thread 8788.0x1178] > 0x005508b5 in ?? () > (gdb) bt full > #0 0x005508b5 in ?? () > No symbol table info available. > #1 0x000c in ?? () > No symbol table info available. > #2 0x005fa120 in ?? () > No symbol table info available. > #3 0x in ?? () > No symbol table info available. > > I will download current source and compile it with -g . Sorry, I forgot that it was stripped. To save you a bit of time, I've uploaded an unstripped build at [1] [1] ftp://cygwin.com/pub/cygwinx/XWin.20101108-git-eb4deaf9bfbab023.exe.bz2 > Something funny about it: > I had the "xset fp+ tcp/mystique:7100" commented out in the > .startxwinrc. > Then I started XWin, attached the gdb and issued the xset command from > within a AIX aixterm shell. > Then I started a window that has to use a font server font and it > worked. > > Then I activated the xset command in the .startxwinrc again and the XWin > crashed again. > I keep on researching. Thanks. -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Updated: xorg-server-1.9.2-1
On 08/11/2010 11:48, Fergus wrote: > Since the update I have been unable to start XWin under [1.7]. > Following my usual sequence at a bash prompt: > > mount n:/tmp /tmp # drive n: is NTFS as required > run XWin -multiwindow & > > Normally these two would be followed by a command line for xterm, but I am not > getting that far! The second line repeatedly generates a Cygwin/X error > message box, where the reason given is very familiar: > > Can't read lock file /tmp/.X0-lock > > I used to get this when /tmp was FAT32 but as you see from line 1 I am making > an explicit mount to a NTFS drive, and in the recent past this step has been > all that is necessary to make progress. When I open /var/log/xwin/XWin.0.log > for more information I read > > ~> cat /var/log/xwin/XWin.0.log > Welcome to the XWin X Server > Vendor: The Cygwin/X Project > Release: 1.9.2.0 (10902000) > Build Date: 2010-11-03 > > XWin was started with the following command line: > > /usr/bin/XWin -multiwindow > > ddxProcessArgument - Initializing default screens > winInitializeScreenDefaults - primary monitor w 1280 h 1024 > winInitializeDefaultScreens - native DPI x 96 y 96 > _XSERVTransSocketOpenCOTSServer: Unable to open socket for inet6 > _XSERVTransOpen: transport open failed for inet6/mandible:0 > _XSERVTransMakeAllCOTSServerListeners: failed to open listener for inet6 > > Fatal server error: > Can't read lock file /tmp/.X0-lock I don't think there have been any significant changes in the lockfile code since the previous version, so this is mysterious. Does /tmp/.X0-lock exists? Is it readable by you? ('ls -al /tmp/.X0-lock' ) Are you sure /tmp is getting mounted successfully where you expect it to be? Have you tried '-nolock', as suggested by [1] > Can anything be determined from the 3 lines in this file beginning _XSERV > ("Unable" .. "failed" .. "failed")? Do I need to change my command sequence? The lines about inet6 are expected if you don't have IPv6 installed. If they have just started appearing with 1.9.2, that would be odd. > BTW it's interesting (is it?) that the error message refers to 1.9.2.0 not > 1.9.2-1. XWin only reports the upstream version (1.9.2.0), not our packaging version (the -1), for various uninteresting reasons. > I'm surprised that there are no other anguished reports to you following the > update, but I guess that's because the fault/ problem/ difficulty is mine > rather than general. ?? Any insights much appreciated. Either that or you are the only person who uses XWin :-) I never hear from anyone saying it just works fine, so I assume they don't exist :-) [1] http://x.cygwin.com/docs/faq/cygwin-x-faq.html#q-cant-read-lock-file -- Jon TURNEY Volunteer Cygwin/X X Server maintainer -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
AW: 1.9.2.0: Xwin SIGSEGV when font server should be queried
I did a bit of further reseach. It looks like it does not have anything to do with the way the font server is queried, but rather the way the startup is done. When I have the font server added during the XWin startup - either in the startxwin.bat, or a bash-script I call from there, the XWin SIGSEGVs. When I let the XWin come up with a few of my fav screens, and I then add the font server either from a cygwin shell or from a client shell the font server is subsequently queried correctly and the SIGSEGV does not appear. Strange. Anyway. Thanks for the exe to create a full stack trace. Here it is: (gdb) bt full #0 0x005b56a2 in doListFontsWithInfo (client=0x10981dc8, c=0x108693f0) at dixfonts.c:920 fpe = (FontPathElementPtr) 0x109816a0 err = 85 name = 0x5b3f36 "\203E▒\001▒h▒e" namelen = 2280120 numFonts = 1628938183 fontInfo = {firstCol = 51800, lastCol = 34, firstRow = 1, lastRow = 0, defaultCh = 1, noOverlap = 0, terminalFont = 0, constantMetrics = 0, constantWidth = 0, inkInside = 0, inkMetrics = 0, allExist = 0, drawDirection = 0, cachable = 0, anamorphic = 0, maxOverlap = 0, pad = 0, maxbounds = {leftSideBearing = 0, rightSideBearing = 0, characterWidth = 0, ascent = 0, descent = 5840, attributes = 4248}, minbounds = {leftSideBearing = 16384, rightSideBearing = 0, characterWidth = 0, ascent = 0, descent = 100, attributes = 0}, ink_maxbounds = {leftSideBearing = -24633, rightSideBearing = 24855, characterWidth = 12266, ascent = 4168, descent = -29488, attributes = 65}, ink_minbounds = { leftSideBearing = -13640, rightSideBearing = 34, characterWidth = -29439, ascent = 65, descent = -13668, attributes = 34}, fontAscent = 0, fontDescent = 0, nprops = 0, props = 0x0, isStringProp = 0x1 } pFontInfo = (FontInfoRec *) 0x1097d4d8 reply = (xListFontsWithInfoReply *) 0x0 ---Type to continue, or q to quit--- length = 0 pFP = (xFontProp *) 0x1 i = 278402720 aliascount = 0 finalReply = {type = 0 '\0', nameLength = 0 '\0', sequenceNumber = 0, length = 0, minBounds = {leftSideBearing = 0, rightSideBearing = 0, characterWidth = -10208, ascent = 4207, descent = 24928, attributes = 102}, walign1 = 0, maxBounds = {leftSideBearing = -16392, rightSideBearing = 24843, characterWidth = 0, ascent = 0, descent = 0, attributes = 0}, walign2 = 0, minCharOrByte2 = 0, maxCharOrByte2 = 0, defaultChar = 0, nFontProps = 0, drawDirection = 0 '\0', minByte1 = 0 '\0', maxByte1 = 0 '\0', allCharsExist = 0 '\0', fontAscent = 0, fontDescent = 0, nReplies = 0} #1 0x00579a0f in ProcessWorkQueue () at dixutils.c:527 q = (WorkQueuePtr) 0x108698a8 p = (WorkQueuePtr *) 0x665de4 #2 0x005eb4f5 in WaitForSomething (pClientsReady=0x106d0ff8) at WaitFor.c:169 i = 1 waittime = {tv_sec = 30, tv_usec = 0} wt = (struct timeval *) 0x22cba0 timeout = 570906 clientsReadable = {fds_bits = {0, 0, 0, 0, 0, 0, 0, 0}} clientsWritable = {fds_bits = {277256200, 277256384, 0, 2280408, 6197569, 278385880, 2280328, 1}} ---Type to continue, or q to quit--- curclient = 292 selecterr = 0 nready = 0 devicesReadable = {fds_bits = {0, 0, 0, 0, 0, 0, 0, 0}} now = 447551595 someReady = 0 #3 0x0057093e in Dispatch () at dispatch.c:369 clientReady = (int *) 0x106d0ff8 result = 0 client = (ClientPtr) 0x10981dc8 nready = -1 icheck = (HWEventQueuePtr *) 0x665aa0 start_tick = 0 #4 0x0056af00 in main (argc=8, argv=0x10482fa8, envp=0x104480e0) at main.c:298 i = 1 alwaysCheckForInput = {0, 1} (gdb) Thanks for your help Burkhard _ Geschaeftsfuehrer/Managing Directors: Christoph Hahn-Woernle, Frank Apel HRB 17335, Amtsgericht Stuttgart (Commercial Register District Court Stuttgart) St.-Nr. 99064/06051, USt-IdNr./VAT Reg.No.: DE 203036780 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/
Re: Updated: xorg-server-1.9.2-1
On 11/8/2010 6:48 AM, Fergus wrote: PS It kills me not to copy this to cygwin at cygwin dot com which I feel will have a much wider readership and skill set amongst that readership, but anybody sinning in this regard is always referred straight to Cygwin-X. Because this is where those interested in Cygwin-X will be "hanging out" and looking for answers. For issues that are clearly X-related (or suspected to be so), it makes sense to query the list that supports that functionality, even if the readership is smaller than the main list. From the perspective of those with a problem, I would think there would be a preference for quality (of responses from this list) vs quantity (of eyes on some other list). And I'm sure those on the main list who aren't interested in X issues much prefer the lack of X traffic (and vice-versa). If you look in the email archives for discussions about merging the lists, you'll see these two points were well represented in the feedback, if my memory serves. -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 216 Dalton Rd. (508) 893-9889 - FAX Holliston, MA 01746 _ A: Yes. Q: Are you sure? A: Because it reverses the logical flow of conversation. Q: Why is top posting annoying in email? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/