oscam-1.20--ppc-tuxbox-linux-dbox2
- Thread starter iomeali
- Start date
oscam-1.20-svn.7246-ppc-tuxbox-linux-dbox2
Changes from my last oscam/post 7228:
7229: oscam-config.c - correct saving of block_same_ip and block_same_name, default = 1 and should not be saved.
7230: - Auto nagra ecm command filter after 5 ecm requests -> fixes cardlock HD+
- Errorhandling code cardreaders improved
- Protocol T1 Auto card restart after a failed T1 resync command
- Reintroduced EINTR and EAGAIN response (fixes ppcold timeouts)
7231: Last tryfix for EAGAIN problem internal reader of ET-boxes
If this doesnt work please ask clarketech to fix it!
7232: fixed some possible segfaults
7233: WebIf: Fix for listing http_css files in config dir. Remove div around FAILBANNOTIFIER and CLIENTCOUNTNOTIFIER variables as they are unnecessary (see #2656).
7234: fixed a possible segfault
7235: double check: added per-caid configuration option:
[global]
double_check_caid=caid1,caid2,...,caidn
or
double_check_caid=system1,system2,...,systemn (where systemx are the first 2 bytes of the caid, for example 01,17,1
7236: - fixed a loadbalancer issue (too many fallbacks used)
- loadbalancer with ecm format (debug output)
7237: Added documentation for new parameter 'double_check_caid'.
7238: cccam: Correct handling of Bulcrypt card serials. Patch by chepik @ streamboard forum.
7239: Do not log duplicated lines. Instead of filling the log with duplicated lines, count them and when unique line appears or 60 seconds of only duplicated lines have passed inform the user how many duplicated log lines were skipped.
7240: webif: Fix two possible segfaults.
7241: webif: There is no div_notifier CSS class anymore.
7242: fixed some possible segfaults
7243: fixed a memleak
7244: Small documentation update for 'reconnecttimeout' parameter.
7245: webif: More CSS applied to span_notifier. It seems that SPAN.span_notifier should be positioned in certain way.
7246: Tryfix EAGAIN implementation -> tnx to PieterG from Openpli for explanation
Hope this works for ET-9200 owners out there.
Changes from my last oscam/post 7228:
7229: oscam-config.c - correct saving of block_same_ip and block_same_name, default = 1 and should not be saved.
7230: - Auto nagra ecm command filter after 5 ecm requests -> fixes cardlock HD+
- Errorhandling code cardreaders improved
- Protocol T1 Auto card restart after a failed T1 resync command
- Reintroduced EINTR and EAGAIN response (fixes ppcold timeouts)
7231: Last tryfix for EAGAIN problem internal reader of ET-boxes
If this doesnt work please ask clarketech to fix it!
7232: fixed some possible segfaults
7233: WebIf: Fix for listing http_css files in config dir. Remove div around FAILBANNOTIFIER and CLIENTCOUNTNOTIFIER variables as they are unnecessary (see #2656).
7234: fixed a possible segfault
7235: double check: added per-caid configuration option:
[global]
double_check_caid=caid1,caid2,...,caidn
or
double_check_caid=system1,system2,...,systemn (where systemx are the first 2 bytes of the caid, for example 01,17,1
7236: - fixed a loadbalancer issue (too many fallbacks used)
- loadbalancer with ecm format (debug output)
7237: Added documentation for new parameter 'double_check_caid'.
7238: cccam: Correct handling of Bulcrypt card serials. Patch by chepik @ streamboard forum.
7239: Do not log duplicated lines. Instead of filling the log with duplicated lines, count them and when unique line appears or 60 seconds of only duplicated lines have passed inform the user how many duplicated log lines were skipped.
7240: webif: Fix two possible segfaults.
7241: webif: There is no div_notifier CSS class anymore.
7242: fixed some possible segfaults
7243: fixed a memleak
7244: Small documentation update for 'reconnecttimeout' parameter.
7245: webif: More CSS applied to span_notifier. It seems that SPAN.span_notifier should be positioned in certain way.
7246: Tryfix EAGAIN implementation -> tnx to PieterG from Openpli for explanation
Hope this works for ET-9200 owners out there.
oscam-1.20-svn.7267-ppc-tuxbox-linux-dbox2
Changes from my last oscam/post 7246:
7247: fixed lb debug output
7248: lb: fixed statistic evolution
7249: lb: fixed lb_mode=2
7250: Improved functionality of irdeto cards CAID 0624, 0666 and 0648 patch by Black_shi
7251: lb: antoher fix for lb_mode=2
7252: lb_mode=2 another fix
7253: lb_mode=2 tryfix
7254: Do not fill logs with periodic messages.
Reader check is executed every second and generates at least 4 lines per check in the log. That is just noise, that only fills the logs.
7255: webif: Beautify notifier CSS class. Patch by Znuff
7256: lb: fixed lb_mode=3
7257: linux "kill" command (SIGTERM) is forwarded to the main pid and now stat is saved on exit.
7258: logging: Move some messages to higher debug levels.
Moving not important messages to higher debug levels cleans the log and from too much information. Patch by FilipeAmadeuO
7259: config: Fix aulow configuration writing. The default for aulow is 30 and it should not be written in the config if it is less than at least 1.
7260: webif: Do not show monitor settings when monitor module is disabled. The patch was originally from FilipeAmadeuO.
7261: Cleanup choosing of default log file location.
7262: bulcrypt: Detect v2 cards and inform the user about the missing support.
We still have no idea how to deobfuscate the returned code words.
At least after this patch we can distinguish between v1 and v2 cards and inform the user that the card is unsupported.
7263: build: Use LIBDIR and BINDIR variables.
LIBDIR points to the directory where static libraries would be kept while building oscam. The default is "lib".
BINDIR points to the directory where oscam binary would be generated. The default is "Distribution".
7264: build: Add missing quotes in the help text.
7265: config: Fix aulow config var writing.
The previous check introduced in r7259 was idiotic and exactly the opposite of the commit message (which described the correct behaviour).
7266: fixed fallback when not using lb
7267: Fixed CMakeLists.txt files for libusb
No need anymore to use sed command to correct wrong links when crosscompiling. To compile with static links ad parameter -DSTATIC_LIBUSB
The set (LIBUSBDIR ...) Parameter still usable but will be gone in future.
Changes from my last oscam/post 7246:
7247: fixed lb debug output
7248: lb: fixed statistic evolution
7249: lb: fixed lb_mode=2
7250: Improved functionality of irdeto cards CAID 0624, 0666 and 0648 patch by Black_shi
7251: lb: antoher fix for lb_mode=2
7252: lb_mode=2 another fix
7253: lb_mode=2 tryfix
7254: Do not fill logs with periodic messages.
Reader check is executed every second and generates at least 4 lines per check in the log. That is just noise, that only fills the logs.
7255: webif: Beautify notifier CSS class. Patch by Znuff
7256: lb: fixed lb_mode=3
7257: linux "kill" command (SIGTERM) is forwarded to the main pid and now stat is saved on exit.
7258: logging: Move some messages to higher debug levels.
Moving not important messages to higher debug levels cleans the log and from too much information. Patch by FilipeAmadeuO
7259: config: Fix aulow configuration writing. The default for aulow is 30 and it should not be written in the config if it is less than at least 1.
7260: webif: Do not show monitor settings when monitor module is disabled. The patch was originally from FilipeAmadeuO.
7261: Cleanup choosing of default log file location.
7262: bulcrypt: Detect v2 cards and inform the user about the missing support.
We still have no idea how to deobfuscate the returned code words.
At least after this patch we can distinguish between v1 and v2 cards and inform the user that the card is unsupported.
7263: build: Use LIBDIR and BINDIR variables.
LIBDIR points to the directory where static libraries would be kept while building oscam. The default is "lib".
BINDIR points to the directory where oscam binary would be generated. The default is "Distribution".
7264: build: Add missing quotes in the help text.
7265: config: Fix aulow config var writing.
The previous check introduced in r7259 was idiotic and exactly the opposite of the commit message (which described the correct behaviour).
7266: fixed fallback when not using lb
7267: Fixed CMakeLists.txt files for libusb
No need anymore to use sed command to correct wrong links when crosscompiling. To compile with static links ad parameter -DSTATIC_LIBUSB
The set (LIBUSBDIR ...) Parameter still usable but will be gone in future.
oscam-1.20-svn.7316-ppc-tuxbox-linux-dbox2. 03.09.2012
Changeset:Changes from my last oscam posted svn7267
7268: Fix : Now if -DHAVE_LIBUSB=0 is used it never will include any libusb
7269: Restored old ifsc of 251. Max IFSC 254 is causing resyncs -> timeouts with smartreaders
7270: Fixed Broken test about presence of libusb into tools or system
Fixed Link into CMakeLists.txt from smargo utils
7271: module-webif-pages.c:
- correct patch for ticket 2691: moved serialreadertimeout option to Serial config menu - no need for this parameter in global menu.
- grouped parameters in global a bit
- moved suppresscmd08 to global section and show only with cs357x and cs378x. camd33 does not have this parameter.
- some cosmetics and added units (s, ms) for some parameters
oscam-config.c:
- save lb_savepath only when not default
7272: Removed the find_library check to be compatible with standerd oscam toochain files
Adapted possible break for mac and windows on libusb
7273: module-webif-pages.c - serialreadertimeout has nothing to do with module_serial, so i move it back to the timout secition. sorry for the confusion, i was also confused by the patch provided.
7274: fixed a segfault on exit oscam, also fixes missing log entries on exit
7275: oscam-config.c - fix for #2695, tnx to polskafan @ SB!
7276: module-stat: Fix possible segfault related to lb_savepath.
lb_savepath is dynamically allocated and testing for lb_savepath[0]
would most likely result in segfault if the variable is NULL.
7277: lb: fixed reopen_seconds with local readers (see #2682)
7278: vg2: Extract ROM version. Patch by lintendo @ streamboard.
7279: Fix Cmake compile error on mac.
7280: Videoguard: Not all cards support get ROM version, hopefully fixes
7281: fix error previous patch
7282: Introduction to automatic selection of static or dynamic libusb-1.0 in function of the target. Now Activated only for Dreambox dynamic linking can be forced by adding -DSTATIC_LIBUSB=0 parameter
7283: lb: auto-betatunnel enhancement, lb_auto_betatunnel_prefer_beta=x means
0=always beta...100=always nagra
7284: cacheex: fix for "active reader not sending cache but memusage is growing"
7285: cacheex: Another fix for high memory usage and full running "job queues"
7286: Added new options for 'lb_auto_betatunnel_prefer_beta' to documentaion.
7287: cacheex: dirty hack to prevent full running job queues. Maybe someone has an idea why this happens?
7288: Videoguard: Send get ROM version to all cards
7289: Added Viaccess to 'unlockparental' parameter in documentaion.
7290: Tryfix timeouts
7291: After a T1 resync a new retry should be initiated. Since rev 7230 this was never done ret was OK due to succesfull resync. This should fix the retry after resync with T1 cards -> please test and report in forum.
7292: reverted dirty hack vom r7287, tryfix doing it better...
7293: Researched and corrected the implementation of ifsc request and responses for T1 protocol
Old behaviour: the ifsc value was send to the card but the card ifsc response was never evaluated.
If the card responded error oscam still continued to use the ifsc value anyway.
New behaviour: the ifsc value is send to the card and the card ifsc response is evaluated. If the card responds OK, the value of the ifsc response of the card is used
This commit might break the working of T1 cards with certain readers so please test and report -d 65535 logs in the forum!
7294: Do not use pthread_getcpuclockid().
Using pthread_getcpuclockid() is fine for checking if a thread exists, but the problem is that this function may not exist. Using pthread_detach in this case results in the same functionality.
This should fix ticket #2712.
7295: reader-videoguard: Add support for Indovision (09C1). Patch by sat_hunter @ sb.
7296: Avoid filling the log with CACHEEX messages.
7297: Tryfix setting ifsc value
Implemented 3 retries instead of just returning error right away
7298: Seca card reporting:
Many providers use mediaguard to protect their content.
Its misleading to report all mediaguard cards as Canal+ NL
Keep in mind that a mediaguard V10.0 card is actually a nagra card tunneling seca!
7299: reader-videoguard2.c : Add mode detection patch #2706 for ins7e11 , tnx to lintendo.
7300: fixed a segfault
7301: config: Fix detection of OSX_VER.
7302: Implement almost full IPv6 support. The original patch was created by manio and discussed in ticket #2694. I've updated the patch to apply to the latest master and added some cleanups.
7303: build: Compile oscam.c like regular source file.
This removes special case in building of oscam.c. Before this oscam.c was build along with linking the final binary. This was rather strange and served no purpose at all.
7304: reader-seca: Use sensitive logging to hide SA.
7305: Use the correct operator.
It is certain that in this case instead of =+ the operator should be +=.
gcc accepts both but clang correctly warns about using =+.
7306: build: Do not enable HAVE_DVBAPI on OS X.
7307: pthread_self() result should not be casted to unsigned int.
7308: module-stat: Fix warning.
Seen when compiling on OSX with make CC=clang:
module-stat.c:357:47: warning: conversion specifies type 'unsigned short' but the argument has type 'uint32_t' (aka 'unsigned int') [-Wformat]
fprintf(file, "%s,%d,%04hX,%06X,%04hX,%04hX,%d,%d,%ld,%d,%02hX\n",
7309: Videoguard: some cards reply len=0x00 for not supported ins
7310: Videoguard: Improvements for ticket #2706, ins7e11 handling
7311: Mixed Up the setting of IFSD with IFSC -> so reverting 7297 and 7293 + fix of wrong operand use
7312: config: Add -O / --detect-osx-sdk-version option to help text.
7313: config: Update OS X SDK detection to handle 10.7+ versions.
In 10.7 the SDK location have moved. Thanks to stefansp for providing the new paths.
7314: dvbapi: Sent data as one packet.
Some time ago I've sent a patch for generic dvbapi under linux (boxtype=pc). It is working fine but I think there is need for one important enhancement. Currently it is sending each CA_SET_PID and CA_SET_DESCR as two separate packets which are: the request integer and the data. I think it will be much better to sent it in one packet.
At first it is not elegant to sent it this way, besides the UDP packet order is not the think that we should rely on.
About client implementation: as far as I know there is currently two clients using the api: vdr-plugin-dvbapi and ostryck MuMuDVB fork (maybe tvheadend will also use it). I contacted ostryck and he said there will be no problem with this small api change. Obviously also no problem on vdr-plugin-dvbapi side.
Patch by manio
7315: fixed a segfault
7316: Fix cs_inet_addr() to work correctly with IPv4 addresses when IPv6 is enabled.
Patch by manio.
Changeset:Changes from my last oscam posted svn7267
7268: Fix : Now if -DHAVE_LIBUSB=0 is used it never will include any libusb
7269: Restored old ifsc of 251. Max IFSC 254 is causing resyncs -> timeouts with smartreaders
7270: Fixed Broken test about presence of libusb into tools or system
Fixed Link into CMakeLists.txt from smargo utils
7271: module-webif-pages.c:
- correct patch for ticket 2691: moved serialreadertimeout option to Serial config menu - no need for this parameter in global menu.
- grouped parameters in global a bit
- moved suppresscmd08 to global section and show only with cs357x and cs378x. camd33 does not have this parameter.
- some cosmetics and added units (s, ms) for some parameters
oscam-config.c:
- save lb_savepath only when not default
7272: Removed the find_library check to be compatible with standerd oscam toochain files
Adapted possible break for mac and windows on libusb
7273: module-webif-pages.c - serialreadertimeout has nothing to do with module_serial, so i move it back to the timout secition. sorry for the confusion, i was also confused by the patch provided.
7274: fixed a segfault on exit oscam, also fixes missing log entries on exit
7275: oscam-config.c - fix for #2695, tnx to polskafan @ SB!
7276: module-stat: Fix possible segfault related to lb_savepath.
lb_savepath is dynamically allocated and testing for lb_savepath[0]
would most likely result in segfault if the variable is NULL.
7277: lb: fixed reopen_seconds with local readers (see #2682)
7278: vg2: Extract ROM version. Patch by lintendo @ streamboard.
7279: Fix Cmake compile error on mac.
7280: Videoguard: Not all cards support get ROM version, hopefully fixes
7281: fix error previous patch
7282: Introduction to automatic selection of static or dynamic libusb-1.0 in function of the target. Now Activated only for Dreambox dynamic linking can be forced by adding -DSTATIC_LIBUSB=0 parameter
7283: lb: auto-betatunnel enhancement, lb_auto_betatunnel_prefer_beta=x means
0=always beta...100=always nagra
7284: cacheex: fix for "active reader not sending cache but memusage is growing"
7285: cacheex: Another fix for high memory usage and full running "job queues"
7286: Added new options for 'lb_auto_betatunnel_prefer_beta' to documentaion.
7287: cacheex: dirty hack to prevent full running job queues. Maybe someone has an idea why this happens?
7288: Videoguard: Send get ROM version to all cards
7289: Added Viaccess to 'unlockparental' parameter in documentaion.
7290: Tryfix timeouts
7291: After a T1 resync a new retry should be initiated. Since rev 7230 this was never done ret was OK due to succesfull resync. This should fix the retry after resync with T1 cards -> please test and report in forum.
7292: reverted dirty hack vom r7287, tryfix doing it better...
7293: Researched and corrected the implementation of ifsc request and responses for T1 protocol
Old behaviour: the ifsc value was send to the card but the card ifsc response was never evaluated.
If the card responded error oscam still continued to use the ifsc value anyway.
New behaviour: the ifsc value is send to the card and the card ifsc response is evaluated. If the card responds OK, the value of the ifsc response of the card is used
This commit might break the working of T1 cards with certain readers so please test and report -d 65535 logs in the forum!
7294: Do not use pthread_getcpuclockid().
Using pthread_getcpuclockid() is fine for checking if a thread exists, but the problem is that this function may not exist. Using pthread_detach in this case results in the same functionality.
This should fix ticket #2712.
7295: reader-videoguard: Add support for Indovision (09C1). Patch by sat_hunter @ sb.
7296: Avoid filling the log with CACHEEX messages.
7297: Tryfix setting ifsc value
Implemented 3 retries instead of just returning error right away
7298: Seca card reporting:
Many providers use mediaguard to protect their content.
Its misleading to report all mediaguard cards as Canal+ NL
Keep in mind that a mediaguard V10.0 card is actually a nagra card tunneling seca!
7299: reader-videoguard2.c : Add mode detection patch #2706 for ins7e11 , tnx to lintendo.
7300: fixed a segfault
7301: config: Fix detection of OSX_VER.
7302: Implement almost full IPv6 support. The original patch was created by manio and discussed in ticket #2694. I've updated the patch to apply to the latest master and added some cleanups.
7303: build: Compile oscam.c like regular source file.
This removes special case in building of oscam.c. Before this oscam.c was build along with linking the final binary. This was rather strange and served no purpose at all.
7304: reader-seca: Use sensitive logging to hide SA.
7305: Use the correct operator.
It is certain that in this case instead of =+ the operator should be +=.
gcc accepts both but clang correctly warns about using =+.
7306: build: Do not enable HAVE_DVBAPI on OS X.
7307: pthread_self() result should not be casted to unsigned int.
7308: module-stat: Fix warning.
Seen when compiling on OSX with make CC=clang:
module-stat.c:357:47: warning: conversion specifies type 'unsigned short' but the argument has type 'uint32_t' (aka 'unsigned int') [-Wformat]
fprintf(file, "%s,%d,%04hX,%06X,%04hX,%04hX,%d,%d,%ld,%d,%02hX\n",
7309: Videoguard: some cards reply len=0x00 for not supported ins
7310: Videoguard: Improvements for ticket #2706, ins7e11 handling
7311: Mixed Up the setting of IFSD with IFSC -> so reverting 7297 and 7293 + fix of wrong operand use
7312: config: Add -O / --detect-osx-sdk-version option to help text.
7313: config: Update OS X SDK detection to handle 10.7+ versions.
In 10.7 the SDK location have moved. Thanks to stefansp for providing the new paths.
7314: dvbapi: Sent data as one packet.
Some time ago I've sent a patch for generic dvbapi under linux (boxtype=pc). It is working fine but I think there is need for one important enhancement. Currently it is sending each CA_SET_PID and CA_SET_DESCR as two separate packets which are: the request integer and the data. I think it will be much better to sent it in one packet.
At first it is not elegant to sent it this way, besides the UDP packet order is not the think that we should rely on.
About client implementation: as far as I know there is currently two clients using the api: vdr-plugin-dvbapi and ostryck MuMuDVB fork (maybe tvheadend will also use it). I contacted ostryck and he said there will be no problem with this small api change. Obviously also no problem on vdr-plugin-dvbapi side.
Patch by manio
7315: fixed a segfault
7316: Fix cs_inet_addr() to work correctly with IPv4 addresses when IPv6 is enabled.
Patch by manio.
oscam-1.20-svn.7425-ppc-tuxbox-linux-dbox2
Changeset [7425]:
Different approach to solving httpttl ending slash problem.
Changeset [7424]:
Fix use of possibly uninitialised values in readers.
ATR_GetHistoricalBytes() can return without initializing length parameter so valgrind correctly detects this for each reader:
Conditional jump or move depends on uninitialised value(s)
at 0x80A8A30: conax_card_init (reader-conax.c:117) by 0x80A6306: reader_get_cardsystem (reader-common.c:210) by 0x80A640B: reader_reset (reader-common.c:264) by 0x80563A9: work_thread (oscam.c:4051) by 0x4060AE9: start_thread (in /lib/libpthread-2.15.so) by 0x415B69D: clone (in /lib/libc-2.15.so)
Changeset [7423]:
Force valgrind to stop complaining about possibly uninitialized values.
The valgrind is wrong here but annoying. The problem is that it doesn't know what ioctl call does and assumes the structure is left not initialized.
Conditional jump or move depends on uninitialised value(s)
at 0x81062D7: IO_Serial_SetBitrate (io_serial.c:203) by 0x810642A: IO_Serial_SetParams (io_serial.c:233) by 0x810741F: IO_Serial_InitPnP (io_serial.c:747) by 0x81014EB: Phoenix_Init (ifd_phoenix.c:59) by 0x80FC78E: ICC_Async_Device_Init (icc_async.c:185) by 0x80A65AF: reader_device_init (reader-common.c:313) by 0x80D7CDC: reader_init (oscam-reader.c:844) by 0x805640B: work_thread (oscam.c:4065) by 0x4060AE9: start_thread (in /lib/libpthread-2.15.so) by 0x415B69D: clone (in /lib/libc-2.15.so)
Conditional jump or move depends on uninitialised value(s)
at 0x81062F3: IO_Serial_SetBitrate (io_serial.c:205) by 0x810642A: IO_Serial_SetParams (io_serial.c:233) by 0x810741F: IO_Serial_InitPnP (io_serial.c:747) by 0x81014EB: Phoenix_Init (ifd_phoenix.c:59) by 0x80FC78E: ICC_Async_Device_Init (icc_async.c:185) by 0x80A65AF: reader_device_init (reader-common.c:313) by 0x80D7CDC: reader_init (oscam-reader.c:844) by 0x805640B: work_thread (oscam.c:4065) by 0x4060AE9: start_thread (in /lib/libpthread-2.15.so) by 0x415B69D: clone (in /lib/libc-2.15.so)
Changeset [7422]:
Adding ending slash fixup for emmdir - for consistency with other configuration parameters pointing to folders
Changeset [7421]:
REVERT: cfg: Remove workaround for "httptpl" with missing ending slash.
This one reverts r7396 because my analysis was wrong.
All other variables that hold paths are working fine when slash (/) is missing at the end. Restore the needed fixup.
Changeset [7420]: Some clean up in message ouput
Changeset [7419]: Smal correction mac compile utils
Changeset [7418]:
Libusb will be static by default on all linux. (like before but with correct paths) You can override this by adding -DSTATIC_LIBUSB=0 behind the cmake command. For mac it will always compile with static libusb
Changeset [7417]: Updated 'aulow' parameter in documentation.
Changeset [7425]:
Different approach to solving httpttl ending slash problem.
Changeset [7424]:
Fix use of possibly uninitialised values in readers.
ATR_GetHistoricalBytes() can return without initializing length parameter so valgrind correctly detects this for each reader:
Conditional jump or move depends on uninitialised value(s)
at 0x80A8A30: conax_card_init (reader-conax.c:117) by 0x80A6306: reader_get_cardsystem (reader-common.c:210) by 0x80A640B: reader_reset (reader-common.c:264) by 0x80563A9: work_thread (oscam.c:4051) by 0x4060AE9: start_thread (in /lib/libpthread-2.15.so) by 0x415B69D: clone (in /lib/libc-2.15.so)
Changeset [7423]:
Force valgrind to stop complaining about possibly uninitialized values.
The valgrind is wrong here but annoying. The problem is that it doesn't know what ioctl call does and assumes the structure is left not initialized.
Conditional jump or move depends on uninitialised value(s)
at 0x81062D7: IO_Serial_SetBitrate (io_serial.c:203) by 0x810642A: IO_Serial_SetParams (io_serial.c:233) by 0x810741F: IO_Serial_InitPnP (io_serial.c:747) by 0x81014EB: Phoenix_Init (ifd_phoenix.c:59) by 0x80FC78E: ICC_Async_Device_Init (icc_async.c:185) by 0x80A65AF: reader_device_init (reader-common.c:313) by 0x80D7CDC: reader_init (oscam-reader.c:844) by 0x805640B: work_thread (oscam.c:4065) by 0x4060AE9: start_thread (in /lib/libpthread-2.15.so) by 0x415B69D: clone (in /lib/libc-2.15.so)
Conditional jump or move depends on uninitialised value(s)
at 0x81062F3: IO_Serial_SetBitrate (io_serial.c:205) by 0x810642A: IO_Serial_SetParams (io_serial.c:233) by 0x810741F: IO_Serial_InitPnP (io_serial.c:747) by 0x81014EB: Phoenix_Init (ifd_phoenix.c:59) by 0x80FC78E: ICC_Async_Device_Init (icc_async.c:185) by 0x80A65AF: reader_device_init (reader-common.c:313) by 0x80D7CDC: reader_init (oscam-reader.c:844) by 0x805640B: work_thread (oscam.c:4065) by 0x4060AE9: start_thread (in /lib/libpthread-2.15.so) by 0x415B69D: clone (in /lib/libc-2.15.so)
Changeset [7422]:
Adding ending slash fixup for emmdir - for consistency with other configuration parameters pointing to folders
Changeset [7421]:
REVERT: cfg: Remove workaround for "httptpl" with missing ending slash.
This one reverts r7396 because my analysis was wrong.
All other variables that hold paths are working fine when slash (/) is missing at the end. Restore the needed fixup.
Changeset [7420]: Some clean up in message ouput
Changeset [7419]: Smal correction mac compile utils
Changeset [7418]:
Libusb will be static by default on all linux. (like before but with correct paths) You can override this by adding -DSTATIC_LIBUSB=0 behind the cmake command. For mac it will always compile with static libusb
Changeset [7417]: Updated 'aulow' parameter in documentation.
oscam-1.20-svn.7440-ppc-tuxbox-linux-dbox2
Changes:
7426: cacheex: small stability enhancement
cacheex: debuglevel 1024 (ECM logging) now logs oscam+csp-hashs
7427: Added the possibility to compile libusb dynamic when LIBUSBDIR is used
7428: module-constcw: Do not kill the whole oscam if init fails
7429: module-pandora: Do not kill OSCam for stupid reasons.
The functions that are patched with this are all allowed to returing failure. Instead they were killing the whole OSCam, which is crazy.
7430: cfg: Free config values allocated from the parser
7431: webif: Do not return empty page when password authentication fails.
Lets not be antisocial and say to the poor user at least that the access was denied.
7432: webif: Do not display caid and srvid when they are set to internal values
7433: cfg/account: Convert simple variables.
This increases struct s_auth by 30 bytes, but we'll get some of that back once config parser start supporting boolean values.
Also dyndns, usr and pwd fields when converted to char * would save at least 180 (60+60+60-8+ bytes per struct s_account.
7434: cfg/accounts: Change account->dyndns to char *
This saves 60 bytes for each account in most cases.
7435: Add account_set_defaults() and use it where needed.
This fixes at least one bug in WebIf (the new accounts were added with cccignorereshare = 0 when they should be using -1).
7436: cfg: Prevent crash when freeing account values
7437: cacheex: fixed full running queue
7438: cfg: Fix build error when webif is disabled.
Fix an error introduced in the previous commit
7439: Add reader_set_defaults() and use it where it is needed.
This fixes couple of bugs in WebIf - the new readers were added with defaults that are not set correctly.
7440: cscrypt: keep valgrind happy
Changes:
7426: cacheex: small stability enhancement
cacheex: debuglevel 1024 (ECM logging) now logs oscam+csp-hashs
7427: Added the possibility to compile libusb dynamic when LIBUSBDIR is used
7428: module-constcw: Do not kill the whole oscam if init fails
7429: module-pandora: Do not kill OSCam for stupid reasons.
The functions that are patched with this are all allowed to returing failure. Instead they were killing the whole OSCam, which is crazy.
7430: cfg: Free config values allocated from the parser
7431: webif: Do not return empty page when password authentication fails.
Lets not be antisocial and say to the poor user at least that the access was denied.
7432: webif: Do not display caid and srvid when they are set to internal values
7433: cfg/account: Convert simple variables.
This increases struct s_auth by 30 bytes, but we'll get some of that back once config parser start supporting boolean values.
Also dyndns, usr and pwd fields when converted to char * would save at least 180 (60+60+60-8+ bytes per struct s_account.
7434: cfg/accounts: Change account->dyndns to char *
This saves 60 bytes for each account in most cases.
7435: Add account_set_defaults() and use it where needed.
This fixes at least one bug in WebIf (the new accounts were added with cccignorereshare = 0 when they should be using -1).
7436: cfg: Prevent crash when freeing account values
7437: cacheex: fixed full running queue
7438: cfg: Fix build error when webif is disabled.
Fix an error introduced in the previous commit
7439: Add reader_set_defaults() and use it where it is needed.
This fixes couple of bugs in WebIf - the new readers were added with defaults that are not set correctly.
7440: cscrypt: keep valgrind happy
oscam-1.20-svn.7460-ppc-tuxbox-linux-dbox2 .rar
Changes:
7441: cfg: Add config_list_gc_values()
7442: cfg/accounts: Change pwd type from char[64] to char *.
This change saves on average 54 bytes per struct s_account if we assume that all accounts have 8 chracter passwords. Since the accounts usually have shorter passwords the savings are even bigger.
7443: cacheex: camd35 another fix cacheex: increased queue warning to 2000
7444: cfg: Fix saving of keepalive value in accounts.
This fixes regression introduced in r7433
7445: Fallback to IPv6 if setting up IPv6 listener fails.
Patch by manio
7446: camd33: Silence some annoying warnings.
Thats what I get for not compiling the disabled module after changing it. This patch fixes:
module-camd33.c: In function 'camd33_auth_client':
module-camd33.c:89:5: warning: pointer targets in passing argument 1 of 'streq' differ in signedness [-Wpointer-sign]
In file included from globals.h:1763:0,
from module-camd33.c:1:
global-functions.h:417:12: note: expected 'const char *' but argument is of type 'uchar *'
module-camd33.c:89:5: warning: pointer targets in passing argument 1 of 'streq' differ in signedness [-Wpointer-sign]
In file included from globals.h:1763:0,
from module-camd33.c:1:
global-functions.h:417:12: note: expected 'const char *' but argument is of type 'uchar *'
No binary changes.
7447: cacheex: another fix for the queue problem
7448: cacheex: reconnect timeout now valid cacheex: cacheex-stats with ip
7449: fixed last commit when compiling without cacheex
7450: cacheex: fixed a camd35 issue
7451: cacheex: fixed high load
7452: cacheex: fixed wrong cccam cacheex stats nodeid
7453: cacheex: another try-fix
7454: cacheex/internal locking management: new locking mechanismn to avoid deadlocks on lock/destroy
7455: added ecmheaderwhilist patch. Thx to blaky!
7456: cacheex: added cw diagnostics to log (d1, use grep "Different CWs" on your log)
7457: Change some comments to accomodate a script that checks config tables.
I'm using a script to detect if certain variable is declared as config parser is expecting and the patched lines generate false positives. The script is pretty dumb but fixing it is a lot harder than just changing the comments a bit.
This commit contains no code changes.
7458: cfg/account: Reorder account functions.
No code changes, only code movement.
7459: cfg/account: Convert the rest of account config parsing.
7460: cfg: Split oscam-config.c to oscam-config-{account,global,reader}.c
No code changes, just code movement.
Changes:
7441: cfg: Add config_list_gc_values()
7442: cfg/accounts: Change pwd type from char[64] to char *.
This change saves on average 54 bytes per struct s_account if we assume that all accounts have 8 chracter passwords. Since the accounts usually have shorter passwords the savings are even bigger.
7443: cacheex: camd35 another fix cacheex: increased queue warning to 2000
7444: cfg: Fix saving of keepalive value in accounts.
This fixes regression introduced in r7433
7445: Fallback to IPv6 if setting up IPv6 listener fails.
Patch by manio
7446: camd33: Silence some annoying warnings.
Thats what I get for not compiling the disabled module after changing it. This patch fixes:
module-camd33.c: In function 'camd33_auth_client':
module-camd33.c:89:5: warning: pointer targets in passing argument 1 of 'streq' differ in signedness [-Wpointer-sign]
In file included from globals.h:1763:0,
from module-camd33.c:1:
global-functions.h:417:12: note: expected 'const char *' but argument is of type 'uchar *'
module-camd33.c:89:5: warning: pointer targets in passing argument 1 of 'streq' differ in signedness [-Wpointer-sign]
In file included from globals.h:1763:0,
from module-camd33.c:1:
global-functions.h:417:12: note: expected 'const char *' but argument is of type 'uchar *'
No binary changes.
7447: cacheex: another fix for the queue problem
7448: cacheex: reconnect timeout now valid cacheex: cacheex-stats with ip
7449: fixed last commit when compiling without cacheex
7450: cacheex: fixed a camd35 issue
7451: cacheex: fixed high load
7452: cacheex: fixed wrong cccam cacheex stats nodeid
7453: cacheex: another try-fix
7454: cacheex/internal locking management: new locking mechanismn to avoid deadlocks on lock/destroy
7455: added ecmheaderwhilist patch. Thx to blaky!
7456: cacheex: added cw diagnostics to log (d1, use grep "Different CWs" on your log)
7457: Change some comments to accomodate a script that checks config tables.
I'm using a script to detect if certain variable is declared as config parser is expecting and the patched lines generate false positives. The script is pretty dumb but fixing it is a lot harder than just changing the comments a bit.
This commit contains no code changes.
7458: cfg/account: Reorder account functions.
No code changes, only code movement.
7459: cfg/account: Convert the rest of account config parsing.
7460: cfg: Split oscam-config.c to oscam-config-{account,global,reader}.c
No code changes, just code movement.
oscam-1.20-svn.7507-ppc-tuxbox-linux-dbox2
Changes:
7461: Fix some bug.
7462: build: Show linker version in info screen.
7463: build: Strip white space from info screen.
7464: build: Do not pass unneeded flags to linker.
The linker have no use of the defines of compiler warning settings.
7465: build: Add more targets to PHONY.
These targets do not produce output file with the name of the target (or they must be run always) so they must be listed in PHONY targets.
7466: build: Rename LIBDIR to BUILD_DIR and improve clean and distclean targets.
7467: build: Stop using archives and allow parallel build.
This change simplifies the build and allows it to finally support parallel build (for example: make -j 4) and generation of proper dependency information (implemented in next patch).
7468: build: Generate and use proper dependencies for source files.
This should make using make clean a rare operation. Now we use the compiler to generate dependancy information for each source file. This means that only the correct files would be rebuild if some header or source file changed.
7469: build: Allow parallel build.
7470: cacheex: added cacheex-stats, logging errors
cacheex: increased garbage collector dispose time
cacheex: different cw error is logged and any further blocked
7471: build: Do not assume the shell is smart.
7472: cfg/global: Clear http_dyndns values before setting them.
Without this, once set http_dyndns value can not be removed in WebIf.
7473: Fix crash when saving ecmwhitelist values in readers.
7474: cacheex: fixed a cache flooding bug
7475: fixed a warning
7476: cfg/reader: Fix default cool_timeout_init/cool_timeout_after_init values.
The default values are mentioned in the wiki.
7477: Rename mode in struct s_reader to azbox_mode.
mode is pretty generic and since the variable is used only in azbox code its better to be called azbox_code.
No binary changes.
7478: module-webif.c : Fix a segmentation fault.
7479: module-webif.c: fix another segmentation fault.
oscam-config-account.c: checking invalid value should be top in the priority to prevent bad usages of resources.
7480: Added new parameter 'ecmheaderwhitelist' to documentation.
7481: anticasc: Move init_ac() to module-anticasc.c No code changes.
7482: config: Use proper temporary file when building config.mak
Without this fix, using parallel build results in a race because couple of config.sh scripts are run in parallel and the temporary file can be deleted before being checked which leads to errors.
7483: build: Save binary with debugging info in Distribution/oscam-XXXX.debug
Before to get a binary with debugging info you have to rebuild Oscam with DEBUG=1 parameter set to make because by default the outputed binary was stripped.
This created additional work when debugging and developing because you have to remember to run different command for building and also it was making it harder for our users when there was crash and we needed a backtrace.
This commit eases the pain. Now along with the stripped binary that is saved in Distribution/ we keep unstripped binary with the same name and .debug prefix. That way if there is problem that needs debugging we can just ask the user to use oscam-XXX.debug binary instead of asking for rebuilds.
7484: build: Use CC_OPTS, CC_WARN and CFLAGS when building list_smargo.
Thanks to hebusfoobar @ sb forum for the patch.
7485: Stop using -fno-strict-aliasing compiler parameter.
7486: Fix format string warnings introduced in commit r7485 on 64-bit arch.
7487: webif: Fix "fake" warning.
The fixed warning is fake, the compiler is not smart enough to realise that if we were getting to the point where it warns the variable would surely be initialized.
7488: webif: Fix lots of format string errors.
7489: change cooldown algorithm to close ticket 2341
7490: Oops forgot to delete something in oscam-chk.c see rev 7489
7491: Fix for ticket 2646: webif - status page - lb value is not shown
7492: Commit r7489 incidently turned on WITH_SSL by default. Disable it.
7493: webif: Fix a bug in fail2ban.
7494: webif: Fix deletion of fail2ban addresses.
7495: newcamd/webif: Move client id resolving function.
7496: newcamd: Add WiCard? client id ( 0x5743 )
7497: newcamd: Add tsdecrypt client id ( 0x7878 )
7498: Save a little bit of memory per client by storing ncd_client_id as binary.
7499: Prevent defined but not used warning.
7500: Fix regression introduced in r7495 (cccam version was not shown).
7501: Touch WebIF is now a part (addon) of OSCam. Use http :// host
ort/touch/ URL to access. Adds less than 10K to the binary and can be compiled without if desired.
7502: Update streamboard urls.
7503: webif: Remove unneeded ifdefs.
7504: Move irdeto guessing code to module-ird-guess.c
7505: Merge oscam-types.h into globals.h and remove unneeded defines.
7506: log: fix buffer overflow after a duplicate log line is skipped.
This fixes the bug on sh4-stapi that all log lines are skipped after the first duplicate line.
7507: webif/touch: Add missing build config bits.
Changes:
7461: Fix some bug.
7462: build: Show linker version in info screen.
7463: build: Strip white space from info screen.
7464: build: Do not pass unneeded flags to linker.
The linker have no use of the defines of compiler warning settings.
7465: build: Add more targets to PHONY.
These targets do not produce output file with the name of the target (or they must be run always) so they must be listed in PHONY targets.
7466: build: Rename LIBDIR to BUILD_DIR and improve clean and distclean targets.
7467: build: Stop using archives and allow parallel build.
This change simplifies the build and allows it to finally support parallel build (for example: make -j 4) and generation of proper dependency information (implemented in next patch).
7468: build: Generate and use proper dependencies for source files.
This should make using make clean a rare operation. Now we use the compiler to generate dependancy information for each source file. This means that only the correct files would be rebuild if some header or source file changed.
7469: build: Allow parallel build.
7470: cacheex: added cacheex-stats, logging errors
cacheex: increased garbage collector dispose time
cacheex: different cw error is logged and any further blocked
7471: build: Do not assume the shell is smart.
7472: cfg/global: Clear http_dyndns values before setting them.
Without this, once set http_dyndns value can not be removed in WebIf.
7473: Fix crash when saving ecmwhitelist values in readers.
7474: cacheex: fixed a cache flooding bug
7475: fixed a warning
7476: cfg/reader: Fix default cool_timeout_init/cool_timeout_after_init values.
The default values are mentioned in the wiki.
7477: Rename mode in struct s_reader to azbox_mode.
mode is pretty generic and since the variable is used only in azbox code its better to be called azbox_code.
No binary changes.
7478: module-webif.c : Fix a segmentation fault.
7479: module-webif.c: fix another segmentation fault.
oscam-config-account.c: checking invalid value should be top in the priority to prevent bad usages of resources.
7480: Added new parameter 'ecmheaderwhitelist' to documentation.
7481: anticasc: Move init_ac() to module-anticasc.c No code changes.
7482: config: Use proper temporary file when building config.mak
Without this fix, using parallel build results in a race because couple of config.sh scripts are run in parallel and the temporary file can be deleted before being checked which leads to errors.
7483: build: Save binary with debugging info in Distribution/oscam-XXXX.debug
Before to get a binary with debugging info you have to rebuild Oscam with DEBUG=1 parameter set to make because by default the outputed binary was stripped.
This created additional work when debugging and developing because you have to remember to run different command for building and also it was making it harder for our users when there was crash and we needed a backtrace.
This commit eases the pain. Now along with the stripped binary that is saved in Distribution/ we keep unstripped binary with the same name and .debug prefix. That way if there is problem that needs debugging we can just ask the user to use oscam-XXX.debug binary instead of asking for rebuilds.
7484: build: Use CC_OPTS, CC_WARN and CFLAGS when building list_smargo.
Thanks to hebusfoobar @ sb forum for the patch.
7485: Stop using -fno-strict-aliasing compiler parameter.
7486: Fix format string warnings introduced in commit r7485 on 64-bit arch.
7487: webif: Fix "fake" warning.
The fixed warning is fake, the compiler is not smart enough to realise that if we were getting to the point where it warns the variable would surely be initialized.
7488: webif: Fix lots of format string errors.
7489: change cooldown algorithm to close ticket 2341
7490: Oops forgot to delete something in oscam-chk.c see rev 7489
7491: Fix for ticket 2646: webif - status page - lb value is not shown
7492: Commit r7489 incidently turned on WITH_SSL by default. Disable it.
7493: webif: Fix a bug in fail2ban.
7494: webif: Fix deletion of fail2ban addresses.
7495: newcamd/webif: Move client id resolving function.
7496: newcamd: Add WiCard? client id ( 0x5743 )
7497: newcamd: Add tsdecrypt client id ( 0x7878 )
7498: Save a little bit of memory per client by storing ncd_client_id as binary.
7499: Prevent defined but not used warning.
7500: Fix regression introduced in r7495 (cccam version was not shown).
7501: Touch WebIF is now a part (addon) of OSCam. Use http :// host
ort/touch/ URL to access. Adds less than 10K to the binary and can be compiled without if desired.7502: Update streamboard urls.
7503: webif: Remove unneeded ifdefs.
7504: Move irdeto guessing code to module-ird-guess.c
7505: Merge oscam-types.h into globals.h and remove unneeded defines.
7506: log: fix buffer overflow after a duplicate log line is skipped.
This fixes the bug on sh4-stapi that all log lines are skipped after the first duplicate line.
7507: webif/touch: Add missing build config bits.
oscam-1.20-svn.7532-ppc-tuxbox-linux-dbox2
Changes:
7508: Added config of reader cooldown to webif
7509: Reworked the cooldown function (it was totally broken!)
7510: Fix Touch WebIF showing empty pages when OSCam compiled WITH_DEBUG disabled. In this case Debug Level block in System section of Touch WebIF will not be shown
7511: Rename cc_maxhop to cc_maxhops in struct s_reader.
This is in preparation for reader config update and also the variable is named closer to the configuration name.
This patch results in no binary changes.
7512: cfg: Rename DEF_OPT_{U}INT to DEF_OPT_{U}INT32.
The new names better reflect the expected variable type and allows adding OPT_{U}INT{8,16,64} when they are needed.
This patch results in no binary changes.
7513: cfg: Add support for OPT_{U}INT8 types.
7514: cfg: Change int variable types to the ones before config parser.
The variable types that were {u}int8_t were changed to {u}int32_t when conversion to table based config parser was made because the parser lacked support for smaller types.
Now that the support is added, lets return the old types and stop wasting memory. A lot of the variables really should be changed to "bool" but that can be made later.
7515: devtools: Introduce devtools.
7516: devtools: Add check_config_tables.sh script.
7517: Add README.config file which contains config.sh help.
7518: Some speed optimalisation in ratelimit/cooldown functionality. Cleaned a possible buffer overflow.
7519: cacheex: fixed a duplicate remote note request
7520: lb: new function "auto timeout": Enabled: if a request has not been answered in time, a timeout is invoked earlier based on lb stats avg values + percent+min time. Disabled: timeout like before from config.
This feature is also used for calling fallback readers!
7521: cfg: Fix minor build warning seen when CS_ANTICASC is disabled.
7522: Prevent warning about variable when WITH_LB is disabled.
7523: webif: http hide type parameter for hiding clients of these types from status page. For example httphidetype=sh hides both server+http process from the main status page
7524: ipv6: Remove unneeded ifdefs. Cleanup the code dealing with IPv4/IPv6, less ifdefs is always better.
7525: cacheex: Move cacheex code in module-cacheex.c
It is a lot of code (close to 600 lines), so putting it in one place might be a good idea.
7526: build: FreeBSD and OS X do not agree on the param to get linker version.
7527: cacheex: Introduce cacheex_init() function to simplify some ifdefs.
7528: cacheex: Introduce cacheex_update_hash() to remove some ifdefs.
7529: cacheex: Simplify ifdefs.
Chopping blocks with ifdefs makes my head hurt Lets simplify the code flow a little bit by duplicating two lines.
This patch produces no binary changes.
7530: cacheex: Introduce cacheex_reader() and cleanup ifdefs.
7531: build: Prevent users from setting BUILD_DIR
7532: build: Add qboxhd target in Makefile.extra
Changes:
7508: Added config of reader cooldown to webif
7509: Reworked the cooldown function (it was totally broken!)
7510: Fix Touch WebIF showing empty pages when OSCam compiled WITH_DEBUG disabled. In this case Debug Level block in System section of Touch WebIF will not be shown
7511: Rename cc_maxhop to cc_maxhops in struct s_reader.
This is in preparation for reader config update and also the variable is named closer to the configuration name.
This patch results in no binary changes.
7512: cfg: Rename DEF_OPT_{U}INT to DEF_OPT_{U}INT32.
The new names better reflect the expected variable type and allows adding OPT_{U}INT{8,16,64} when they are needed.
This patch results in no binary changes.
7513: cfg: Add support for OPT_{U}INT8 types.
7514: cfg: Change int variable types to the ones before config parser.
The variable types that were {u}int8_t were changed to {u}int32_t when conversion to table based config parser was made because the parser lacked support for smaller types.
Now that the support is added, lets return the old types and stop wasting memory. A lot of the variables really should be changed to "bool" but that can be made later.
7515: devtools: Introduce devtools.
7516: devtools: Add check_config_tables.sh script.
7517: Add README.config file which contains config.sh help.
7518: Some speed optimalisation in ratelimit/cooldown functionality. Cleaned a possible buffer overflow.
7519: cacheex: fixed a duplicate remote note request
7520: lb: new function "auto timeout": Enabled: if a request has not been answered in time, a timeout is invoked earlier based on lb stats avg values + percent+min time. Disabled: timeout like before from config.
This feature is also used for calling fallback readers!
7521: cfg: Fix minor build warning seen when CS_ANTICASC is disabled.
7522: Prevent warning about variable when WITH_LB is disabled.
7523: webif: http hide type parameter for hiding clients of these types from status page. For example httphidetype=sh hides both server+http process from the main status page
7524: ipv6: Remove unneeded ifdefs. Cleanup the code dealing with IPv4/IPv6, less ifdefs is always better.
7525: cacheex: Move cacheex code in module-cacheex.c
It is a lot of code (close to 600 lines), so putting it in one place might be a good idea.
7526: build: FreeBSD and OS X do not agree on the param to get linker version.
7527: cacheex: Introduce cacheex_init() function to simplify some ifdefs.
7528: cacheex: Introduce cacheex_update_hash() to remove some ifdefs.
7529: cacheex: Simplify ifdefs.
Chopping blocks with ifdefs makes my head hurt Lets simplify the code flow a little bit by duplicating two lines.
This patch produces no binary changes.
7530: cacheex: Introduce cacheex_reader() and cleanup ifdefs.
7531: build: Prevent users from setting BUILD_DIR
7532: build: Add qboxhd target in Makefile.extra
oscam-1.20-svn.7698-ppc-tuxbox-linux-dbox2
Changes:
7533: lcd: Create module-lcd.h file.
7534: Move led code in module-led.{c,h} and add LEDSUPPORT config.
That cleans up a lot of code. The changes were compile tested on x86, dockstar (arm) and qboxhd (sh4) with LEDSUPPORT enabled and disabled.
7535: led: Try to guess the correct machine type in order to use the LEDs.
After this patch the LED code for NSLU2, WRT350NV2 and DOCKSTAR is exactly the same, the ifdefs are gone.
7536: webif: Fix webif behaviour when opened through apache's mod_proxy.
If the oscam-webif is invoked via apache mod_proxy the uriok check failed strncmp was invoked only for the last path.
Patch by fu_man_tschu
7537: cfg: Fix saving of empty variables in the config files.
Bug spotted by nx111
7538: Just cosmetics, hide dvbapi info in normal log
7539: ipv6: Fix udp connections to IPv4 hosts when v6 support is enabled.
Patch by manio with small improvements from me.
7540: Init udp_sa_len, not that we are not using sizeof(udp_sa).
.
.
.
7695: debug: Add more debugging to try to pinpoint the problem in ticket #2850.
7696: webif: Move error message to D_TRACE (1) level.
This moves "WebIf: Error opening file descriptor using fdopen()" error message to TRACE level.
This message can be shown when a client tries to talk to SSL server without using SSL which is hardly something we have to bother the default logs by printing errors.
7697: Try to fix OK and ERROR definitions mess.
We have shot ourselves in the foot by using OK and ERROR defines.
In csctapi have:
OK 0
ERROR 1
but card systems are using the exact opposite:
OK 1
ERROR 0
On top of that CYGWIN have its own set of OK and ERROR defines.
This tries to fix ticket #2850 and cure a regression accidently introduced in r7688 by moving some defines around.
After this OK and ERROR in csctapi are defined in each C file that uses them so it is hard for them to escape outside of the file, also OK and ERROR are undefined in case they are already defined.
7698: webif: Fix build error when WITH_SSL is enabled.
Changes:
7533: lcd: Create module-lcd.h file.
7534: Move led code in module-led.{c,h} and add LEDSUPPORT config.
That cleans up a lot of code. The changes were compile tested on x86, dockstar (arm) and qboxhd (sh4) with LEDSUPPORT enabled and disabled.
7535: led: Try to guess the correct machine type in order to use the LEDs.
After this patch the LED code for NSLU2, WRT350NV2 and DOCKSTAR is exactly the same, the ifdefs are gone.
7536: webif: Fix webif behaviour when opened through apache's mod_proxy.
If the oscam-webif is invoked via apache mod_proxy the uriok check failed strncmp was invoked only for the last path.
Patch by fu_man_tschu
7537: cfg: Fix saving of empty variables in the config files.
Bug spotted by nx111
7538: Just cosmetics, hide dvbapi info in normal log
7539: ipv6: Fix udp connections to IPv4 hosts when v6 support is enabled.
Patch by manio with small improvements from me.
7540: Init udp_sa_len, not that we are not using sizeof(udp_sa).
.
.
.
7695: debug: Add more debugging to try to pinpoint the problem in ticket #2850.
7696: webif: Move error message to D_TRACE (1) level.
This moves "WebIf: Error opening file descriptor using fdopen()" error message to TRACE level.
This message can be shown when a client tries to talk to SSL server without using SSL which is hardly something we have to bother the default logs by printing errors.
7697: Try to fix OK and ERROR definitions mess.
We have shot ourselves in the foot by using OK and ERROR defines.
In csctapi have:
OK 0
ERROR 1
but card systems are using the exact opposite:
OK 1
ERROR 0
On top of that CYGWIN have its own set of OK and ERROR defines.
This tries to fix ticket #2850 and cure a regression accidently introduced in r7688 by moving some defines around.
After this OK and ERROR in csctapi are defined in each C file that uses them so it is hard for them to escape outside of the file, also OK and ERROR are undefined in case they are already defined.
7698: webif: Fix build error when WITH_SSL is enabled.
oscam-1.20-svn.7739-ppc-tuxbox-linux-dbox2.
Changes:
7699: Additional fix after r7696 (move error message to debug level).
7700: No one seems to test and report patches, so I'll have to commit it to the trunk to get reports!
- On the fly ATR decoding (tested ok on DREAMBOX PPC MIPSEL)
- Some changes to ioctl commands to internal readers
- Added EAGAIN construction similar as serial_read to io_serial Write en Write Timeout
7701: Tryfix for ATR on the fly -> tnx Gorgone for reporting!
7702: Stupid OK / ERROR returning makes no sense, before it was always OK.
Some Card ATR do a " softfail" but do work.
For now we return the stupid always OK again.
7703: Annoying EMM Filterpids messages now only visible with DVB_API loglevel or better!
7704: Tryfix ticket 2857 I presume this box (not PPC) needs same fix as PPC
Norbi20: Thanks for taking the effort to attaching a valid log to this ticket!
7705: Made the ATR parsing a bit more human readable and fixed the softfails Now ATR parser returns ERROR on ATRs that a realy invalid. For softfailing ATR it returns ATR_MALFORMED Correct ATR will return OK
Discovered Sci_FastReset is used nowhere in oscam so why is it still in the source?
7706: Tryfix ticket 2861
7707: Added Sci_FastReset to icc_async.c -> now ins7e11 can be used? test it!
Made Sci ATR checking a seperate function so we can recycle it
De-remarked Sci_FastReset and put it into functional use for the first time since rorothetroll added it 2 years ago in commit 2555.
7708: ifd_sci.c: Fix bug in on the fly parsing of ATR
7709: ifd_sci.c: Move ioctl out of on-the-fly parsing to be able to reuse the function for other readers
7710: smartreader: Fix usage of possibly uninitialized variable.
7711: Partly revert changeset 7708
In case of readerror while fetching the ATR char the actual readed ATR chars might be less than atrlength that is predicted
7712: csctapi: take it easy with the dtr/cts toggling
7713: Tryfix CDS NL 0100:00006a seca2 / nagra over sharereader protocol
You should add the CHID the log mentions to the specific reader(s) that answer the request ok
Perhaps in a later revision it will be added automatical
7714: Tryfix 2 does this fix ticket 2862, if not attach a detailed log to this ticket
7715: csctapi: Ported mp35 reader to new function pointer style API
7716: io_serial: I see no reason to mess with reset here
7717: Fix for ticket 2862 logging CDS FIX also on other cardsystems
7718: csctapi: Finish mp35 conversion to new API.
This removes unused ifd_mp35.h file, fixes includes in ifd_mp35.h (no reader should include icc_async.h), renames some functions and makes them static and fixes some log messages.
7719: csctapi: Port pcsc reader to new style API.
This conversion is only compile tested because I don't have PCSC reader. Please test this commit and report back in the dev forum or trac.
7720: Tryfix 4: Complete new approach of cds nl seca2 nagra problem
7721: Fixed co-existance issue of embedded Touch WebIF with httptpl/STATUS.tpl external template file
7722: csctapi/ifd_pcsc: Fix return codes to allow the reader to initialize correctly.
7723: pcsc: Add better debugging for card_write API.
7724: pcsc: Another fix for ticket #2864.
7725: pcsc: Yet another fix for ticket #2864.
7726: pcsc: Continuing fight with ticket #2864.
7727: icc_async: Return wrongly removed variable setting.
7728: Fix for ticket 2863...
Comments in source of io_serial were complete wrong, there was never done a reset.
Even no need to so thats why it was removed but this code did something else so it broke the readerfunction!
7729: Tryfix 5 Optimized the cds nl nagra/mediaguard3 versus seca2 issues
- DVBAPI quick&dirty fix to prevent CCcam from blacklisting sids was only working on cccamreaders hop 1 -> moved it to module-cccam so it should be hop X proof now.
- The code to determine the type of ecm request was not sound, moved the check to the already existing ecm blocker in reader-seca. Reader type is set right away and is always correct since its based on atr!
7730: Bugfix for last commit due to matching_reader not sending real ecm in case of dvbapi
7731: Another fix... forgot to remove my debug line in final version
7732: Another stupid bug squashed
7733: Case closed, CDS NL cant be fixed over X hop cccam!
7734: Optimize file_exists() function.
Going through all of the stdio by using fopen(3) just to check if file exists is not optimal. It is far easier (and faster) to use access(2).
7735: webif: Show picon image only when the picon exists.
It turns out that showing empty image in the place of non existing picon annoys users (who would've known With this changeset the code checks whether picon can be found and only if it exists then shows it, otherwise channel name is shown instead.
This is another fix for ticket #2672.
7736: Revert "webif: Return 1x1 gif image when an image is not found."
This reverts commit r7542. The problem was fixed better in r7735.
7737: webif: Return correct http error code when image was not found.
7738: Working CDS NL fix cccam tested over 2 hops... (who said it cant be done...)
7739: DVBAPI get rid of annoying log entries!
If you want to inspect them use loglevel -d128
Changes:
7699: Additional fix after r7696 (move error message to debug level).
7700: No one seems to test and report patches, so I'll have to commit it to the trunk to get reports!
- On the fly ATR decoding (tested ok on DREAMBOX PPC MIPSEL)
- Some changes to ioctl commands to internal readers
- Added EAGAIN construction similar as serial_read to io_serial Write en Write Timeout
7701: Tryfix for ATR on the fly -> tnx Gorgone for reporting!
7702: Stupid OK / ERROR returning makes no sense, before it was always OK.
Some Card ATR do a " softfail" but do work.
For now we return the stupid always OK again.
7703: Annoying EMM Filterpids messages now only visible with DVB_API loglevel or better!
7704: Tryfix ticket 2857 I presume this box (not PPC) needs same fix as PPC
Norbi20: Thanks for taking the effort to attaching a valid log to this ticket!
7705: Made the ATR parsing a bit more human readable and fixed the softfails Now ATR parser returns ERROR on ATRs that a realy invalid. For softfailing ATR it returns ATR_MALFORMED Correct ATR will return OK
Discovered Sci_FastReset is used nowhere in oscam so why is it still in the source?
7706: Tryfix ticket 2861
7707: Added Sci_FastReset to icc_async.c -> now ins7e11 can be used? test it!
Made Sci ATR checking a seperate function so we can recycle it
De-remarked Sci_FastReset and put it into functional use for the first time since rorothetroll added it 2 years ago in commit 2555.
7708: ifd_sci.c: Fix bug in on the fly parsing of ATR
7709: ifd_sci.c: Move ioctl out of on-the-fly parsing to be able to reuse the function for other readers
7710: smartreader: Fix usage of possibly uninitialized variable.
7711: Partly revert changeset 7708
In case of readerror while fetching the ATR char the actual readed ATR chars might be less than atrlength that is predicted
7712: csctapi: take it easy with the dtr/cts toggling
7713: Tryfix CDS NL 0100:00006a seca2 / nagra over sharereader protocol
You should add the CHID the log mentions to the specific reader(s) that answer the request ok
Perhaps in a later revision it will be added automatical
7714: Tryfix 2 does this fix ticket 2862, if not attach a detailed log to this ticket
7715: csctapi: Ported mp35 reader to new function pointer style API
7716: io_serial: I see no reason to mess with reset here
7717: Fix for ticket 2862 logging CDS FIX also on other cardsystems
7718: csctapi: Finish mp35 conversion to new API.
This removes unused ifd_mp35.h file, fixes includes in ifd_mp35.h (no reader should include icc_async.h), renames some functions and makes them static and fixes some log messages.
7719: csctapi: Port pcsc reader to new style API.
This conversion is only compile tested because I don't have PCSC reader. Please test this commit and report back in the dev forum or trac.
7720: Tryfix 4: Complete new approach of cds nl seca2 nagra problem
7721: Fixed co-existance issue of embedded Touch WebIF with httptpl/STATUS.tpl external template file
7722: csctapi/ifd_pcsc: Fix return codes to allow the reader to initialize correctly.
7723: pcsc: Add better debugging for card_write API.
7724: pcsc: Another fix for ticket #2864.
7725: pcsc: Yet another fix for ticket #2864.
7726: pcsc: Continuing fight with ticket #2864.
7727: icc_async: Return wrongly removed variable setting.
7728: Fix for ticket 2863...
Comments in source of io_serial were complete wrong, there was never done a reset.
Even no need to so thats why it was removed but this code did something else so it broke the readerfunction!
7729: Tryfix 5 Optimized the cds nl nagra/mediaguard3 versus seca2 issues
- DVBAPI quick&dirty fix to prevent CCcam from blacklisting sids was only working on cccamreaders hop 1 -> moved it to module-cccam so it should be hop X proof now.
- The code to determine the type of ecm request was not sound, moved the check to the already existing ecm blocker in reader-seca. Reader type is set right away and is always correct since its based on atr!
7730: Bugfix for last commit due to matching_reader not sending real ecm in case of dvbapi
7731: Another fix... forgot to remove my debug line in final version
7732: Another stupid bug squashed
7733: Case closed, CDS NL cant be fixed over X hop cccam!
7734: Optimize file_exists() function.
Going through all of the stdio by using fopen(3) just to check if file exists is not optimal. It is far easier (and faster) to use access(2).
7735: webif: Show picon image only when the picon exists.
It turns out that showing empty image in the place of non existing picon annoys users (who would've known With this changeset the code checks whether picon can be found and only if it exists then shows it, otherwise channel name is shown instead.
This is another fix for ticket #2672.
7736: Revert "webif: Return 1x1 gif image when an image is not found."
This reverts commit r7542. The problem was fixed better in r7735.
7737: webif: Return correct http error code when image was not found.
7738: Working CDS NL fix cccam tested over 2 hops... (who said it cant be done...)
7739: DVBAPI get rid of annoying log entries!
If you want to inspect them use loglevel -d128
oscam-1.20-svn.7777-ppc-tuxbox-linux-dbox2
Changes:
7740: Softfail malformed card should be initialized also
7741: Only fail if ATR_initFromArray returns ERROR
Some ATRs are malformed and do softfail, but actually do work so dont fail them
ATR_OK is returned if ATR is 100% OK
7742: Fix for misleading message ATR cardspeed and actual clock of readerdevice.
Take notice: they dont have to match to clock the card to the correct frequency!
Get information about the readerdevice that is driving the card to make sure you clock it correct!
7743: ATR on the fly fully interpreted... hope you all learn as much from it as I did
7744: csctapi: Fix couple of warnings seen on 32bit compile.
7745: Small fix for ATR on the fly and ATR_InitFromArry:
T14 protocol also doesnt use TCK byte.
Tnx to Gorgone for reporting!
7746: pcsc: Fix stack overwrite.
We trying to write 64 bytes into ATR_MAX_SIZE (31) bytes sized array.
7747: pcsc: Init card context. Try fix for ticket #2868.
7748: pcsc: Add debug message with card handle.
7749: Revert r7747, it is wrong to initialize hCard that way.
7750: csctapi: mp35 reader init fixes
7751: Fix for nonworking PCSC readers!
Tnx to Sulge and The_Raven for their good debugging and test work
7752: pcsc: Fix oscam crash on PCSC reader init.
Before the conversion to new cardreader API when PCSC reader was initialized the function just exited. After the conversion extra ATR parsing functions were called and this caused crashes.
7753: webif: Fix oscam crash caused by commit r7735.
Commit r7735 introduced checks if picons exist but that opened another bug that allowed tpl_getFilePathInSubdir() to be called with NULL parameter as path. Which of course causes segfault.
7754: config: Put an extra new line between services when writing oscam.services.
The extra empty line doesn't hurt and improves readability.
7755: Fix for inverse card conventions
7756: Fixed: Guardtime was display several times, only first was correct and valid
7757: Tryfix: inverse convention doesnt apply to ATR!
7758: Tryfix for crashing 7556 (?)
7759: added lb betatunnel patch. Thx to TNe-Two !
7760: stat: Rename shadowed variable.
This fixes warning about shadowed variable seen in dm500 and coolbox builds:
module-stat.c: In function 'stat_get_best_reader':
module-stat.c:918: warning: declaration of 'time' shadows a global declaration
No binary changes.
7761: Fix the build that broke when WITH_LB is disabled.
Fixes ticket #2876.
/Guys test the build with your favourite feature diabled when you are adding new stuff, not everyone enables everything /
7762: csctapi: Remove ifd_stapi.h
This file contains no public functions, all of the functions are implemented in liboscam_stapi and are used only in ifd_stapi.c
No binary changes.
7763: csctapi/ifd_phoenix: Add include guard in the header file.
No binary changes.
7764: csctapi/smartreader: Rename csctapi/smartreader_types.h to csctapi/ifd_smartreader_types.h
No binary changes.
7765: csctapi: Move mc_global.h data in icc_async.h
The mc_global.h file contains only couple of defines. Just kill it.
No binary changes.
7766: csctapi: Remove unused multicam defines and rename the rest.
Most of the multicam defines were not used anywhere so I just remove it. The two that are used get MULTICAM_ prefix for easier grepping.
No binary changes.
7767: csctapi: Rename sci_XXXX.h files to ifd_sci_XXXX.h
These headers are used only in ifd_sci.c
No binary changes.
7768: Update for 'lb_auto_betatunnel_prefer_beta' documentation.
7769: Small typo.
7770: Fixed a bad if else construction in atr on the fly
7771: beta: Remove unneeded and wrongly committed debug code in r7759.
Patch by TNe-TWo posted in ticket #2886.
7772: Update for 'failban' parameter in documentation.
7773: Do not disable readers that return errors from ICC_Async_Get_Status().
Before moving PCSC to new card reader API only when sc8in1 reader was used, ICC_Async_GetStatus() can ERROR. That in turn probably led to the fact that the code in reader-coomn.c which should handle the ERROR to not work very well.
Remove the offending code because it causing more harm than good.
This fixes ticket #2879.
7774: csctapi: Move smartreader (libusb) reader to the new API.
7775: csctapi: It's not an error not having a card inserted
7776: Tryfix Sci atr parser on the fly was not correct for T1 cards.
Tnx soupalonion for delivering a log of T1 tivusat card in internal reader
Please test!
7777: sctapi: Prepare ifd_cool for new cardreader structure
Changes:
7740: Softfail malformed card should be initialized also
7741: Only fail if ATR_initFromArray returns ERROR
Some ATRs are malformed and do softfail, but actually do work so dont fail them
ATR_OK is returned if ATR is 100% OK
7742: Fix for misleading message ATR cardspeed and actual clock of readerdevice.
Take notice: they dont have to match to clock the card to the correct frequency!
Get information about the readerdevice that is driving the card to make sure you clock it correct!
7743: ATR on the fly fully interpreted... hope you all learn as much from it as I did
7744: csctapi: Fix couple of warnings seen on 32bit compile.
7745: Small fix for ATR on the fly and ATR_InitFromArry:
T14 protocol also doesnt use TCK byte.
Tnx to Gorgone for reporting!
7746: pcsc: Fix stack overwrite.
We trying to write 64 bytes into ATR_MAX_SIZE (31) bytes sized array.
7747: pcsc: Init card context. Try fix for ticket #2868.
7748: pcsc: Add debug message with card handle.
7749: Revert r7747, it is wrong to initialize hCard that way.
7750: csctapi: mp35 reader init fixes
7751: Fix for nonworking PCSC readers!
Tnx to Sulge and The_Raven for their good debugging and test work
7752: pcsc: Fix oscam crash on PCSC reader init.
Before the conversion to new cardreader API when PCSC reader was initialized the function just exited. After the conversion extra ATR parsing functions were called and this caused crashes.
7753: webif: Fix oscam crash caused by commit r7735.
Commit r7735 introduced checks if picons exist but that opened another bug that allowed tpl_getFilePathInSubdir() to be called with NULL parameter as path. Which of course causes segfault.
7754: config: Put an extra new line between services when writing oscam.services.
The extra empty line doesn't hurt and improves readability.
7755: Fix for inverse card conventions
7756: Fixed: Guardtime was display several times, only first was correct and valid
7757: Tryfix: inverse convention doesnt apply to ATR!
7758: Tryfix for crashing 7556 (?)
7759: added lb betatunnel patch. Thx to TNe-Two !
7760: stat: Rename shadowed variable.
This fixes warning about shadowed variable seen in dm500 and coolbox builds:
module-stat.c: In function 'stat_get_best_reader':
module-stat.c:918: warning: declaration of 'time' shadows a global declaration
No binary changes.
7761: Fix the build that broke when WITH_LB is disabled.
Fixes ticket #2876.
/Guys test the build with your favourite feature diabled when you are adding new stuff, not everyone enables everything /
7762: csctapi: Remove ifd_stapi.h
This file contains no public functions, all of the functions are implemented in liboscam_stapi and are used only in ifd_stapi.c
No binary changes.
7763: csctapi/ifd_phoenix: Add include guard in the header file.
No binary changes.
7764: csctapi/smartreader: Rename csctapi/smartreader_types.h to csctapi/ifd_smartreader_types.h
No binary changes.
7765: csctapi: Move mc_global.h data in icc_async.h
The mc_global.h file contains only couple of defines. Just kill it.
No binary changes.
7766: csctapi: Remove unused multicam defines and rename the rest.
Most of the multicam defines were not used anywhere so I just remove it. The two that are used get MULTICAM_ prefix for easier grepping.
No binary changes.
7767: csctapi: Rename sci_XXXX.h files to ifd_sci_XXXX.h
These headers are used only in ifd_sci.c
No binary changes.
7768: Update for 'lb_auto_betatunnel_prefer_beta' documentation.
7769: Small typo.
7770: Fixed a bad if else construction in atr on the fly
7771: beta: Remove unneeded and wrongly committed debug code in r7759.
Patch by TNe-TWo posted in ticket #2886.
7772: Update for 'failban' parameter in documentation.
7773: Do not disable readers that return errors from ICC_Async_Get_Status().
Before moving PCSC to new card reader API only when sc8in1 reader was used, ICC_Async_GetStatus() can ERROR. That in turn probably led to the fact that the code in reader-coomn.c which should handle the ERROR to not work very well.
Remove the offending code because it causing more harm than good.
This fixes ticket #2879.
7774: csctapi: Move smartreader (libusb) reader to the new API.
7775: csctapi: It's not an error not having a card inserted
7776: Tryfix Sci atr parser on the fly was not correct for T1 cards.
Tnx soupalonion for delivering a log of T1 tivusat card in internal reader
Please test!
7777: sctapi: Prepare ifd_cool for new cardreader structure
oscam-1.20-svn.7895-ppc-tuxbox-linux-dbox2
Changes:
7778: Add per user fakedelay option.
This option can be used to override global fakedelay option on a per user basis. You can set it to higher or lower value or when set to 0 fakedelay is disabled for the user. By default this option is set to -1 which means "take the value from global fakedelay option".
WebIf was not updated and can't be used to edit this option currently. Patches are welcome (keep in mind that anticascading can be disabled and then this option does not exist).
7779: Atr.c -> removed the useless inverse ATR compare first byte = 0x03, this byte is never 0x03 only 0x3F or 0x3B
ifd_sci.c -> some bugfixes
7780: Fix the broken R_MOUSE protocol and documented the source so this wont happen again
.
.
.
7893: ifd_smargo: Tweaked reset for supporting PTS handshake, changed init and read functions
7894: csctapi: A flush is a flush and not some magic waiting for bytes to appear on the serial port. Card init is now much faster.
7895: Removed the WWT change for overclocked T14 cards.
Got several reports this specific overclockfix breaks it
Changes:
7778: Add per user fakedelay option.
This option can be used to override global fakedelay option on a per user basis. You can set it to higher or lower value or when set to 0 fakedelay is disabled for the user. By default this option is set to -1 which means "take the value from global fakedelay option".
WebIf was not updated and can't be used to edit this option currently. Patches are welcome (keep in mind that anticascading can be disabled and then this option does not exist).
7779: Atr.c -> removed the useless inverse ATR compare first byte = 0x03, this byte is never 0x03 only 0x3F or 0x3B
ifd_sci.c -> some bugfixes
7780: Fix the broken R_MOUSE protocol and documented the source so this wont happen again
.
.
.
7893: ifd_smargo: Tweaked reset for supporting PTS handshake, changed init and read functions
7894: csctapi: A flush is a flush and not some magic waiting for bytes to appear on the serial port. Card init is now much faster.
7895: Removed the WWT change for overclocked T14 cards.
Got several reports this specific overclockfix breaks it
oscam-1.20-svn.7940-ppc-tuxbox-linux-dbox2
Changes:
7896: build: Do not disable DVBAPI forcefully when building on OS X.
Apparently some people are using OS X as cross compilation platform and disabling CONFIG_DVBAPI is not what is wanted in this case.
Patch by foobar.
7897: config: Do not assume unknown reader protocol equals "mouse".
Let the user know that he/she should fix its config instead of assuming. Pretty soon protocol type MOUSE may not even be compiled in the binary.
7898: csctapi: Move DB2COM reader functions in csctapi/ifd_db2com.c
Also introduce separate "db2com" reader.
7899: csctapi: Add set_RTS_DTS callback for cardreaders.
db2com reader uses its own rts/dts detection function.
7900: csctapi: Add DB2COM support in mouse_test reader
7901: csctapi: Convert phoenix/mouse reader to the new card reader API.
This is the last reader that needed conversion! Now the road to disabling cardreaders in oscam config is opened.
7902: csctapi: Re-introduce Set reader ready to Send, it's needed for some readers
7903: Tryfix for ticket 2946
7904: Tryfix for ticket 2950 (Duplicate reader)
7905: Remove hardcoded card reader names in reader_get_type_desc().
All card readers have ->desc field initialized.
7906: Move reader related functions out of oscam-simples.c into oscam-reader.c
7907: webif: Remove duplicated pcsc reader in readers dropdown.
7908: csctapi: Remove active field in struct s_cardreader.
Now that all readers are converted this field is no longer needed.
7909: csctapi/db2com: Move the rest of db2com code into ifd_db2com.c
7910: config: Remove temp file after config.mak was build.
7911: config: Special case WITH_CARDREADER.
WITH_CARDREADER should not be user visible option.
7912: List supported card readers in oscam help text.
7913: csctapi/smartreader: Move lock initialization in ->lock_init().
7914: dvbapi/azbox: Add missing include file.
7915: Quick & Dirty fix for broken cccam server introduced in r7912
7916: Temp fix for broken viaccess cards -> to be continued!
7917: Revert "List supported card readers in oscam help text."
This reverts commit r7912. The whole approach is wrong because module init functions needs the config files to be loaded to be able to initialize the modules.
Also commit r7915 which was "Quick & Dirty fix for broken cccam server introduced in r7912". r7912 broke more than cccam.
7918: config: Add card readers configuration.
The configuration allows enabling and disabling of individual card reader drivers.
This commit is just a preparation it'll start to work in the next few commits.
7919: config: Allow card reader enable/disable to work.
7920: build: Build only enabled card readers.
7921: csctapi: Create generic IO_Serial_{Transmit,Receive} functions based on Phoenix.
Lots of drivers were using Phoenix_{Transmit,Receive} functions which are a tiny wrappers around IO_Serial_{Write,Read}.
Move the code in io_serial.c. This allows us to break internal reader dependency on phoenix reader.
7922: sctapi: Create generic IO_Serial_GetStatus() based on Phoenix code.
7923: csctapi: Remove smargo dependancy on Phoenix reader.
7924: csctapi: Create generic IO_Serial_SetBaudrate() based on Phoenix code.
7925: Show AZBOX and COOLAPI status in oscam.version.
7926: Add card reader support status in oscam.version file.
7927: Improve oscam.version file.
- Remove "Max PID: unlimited" line (it seems useless).
- Improve oscam version fomatting.
7928: Nothing stops write_version() from working on CYGWIN.
7929: Move write_versionfile() in oscam.c
7930: Add -V command line parameter to show oscam version info.
7931: Fit oscam help text in 80 columns and improve wording and white spacing.
7932: Move command line parsing in parse_cmdline_params() function.
7933: Rename ->l fields in EMM and ECM structs.
l is a pretty poor name for a field. Lets use something that we can grep for (ecmlen and emmlen).
7934: config: Fix azbox and coolapi build when internal reader is disabled.
7935: config: Rename DB2COM reader to dbox2.
7936: newcamd: Add Opticum client id. Patch by manio posted in ticket #2961.
7937: Added new command line parameter '-V' to documentation.
7938: Updated card reader descriptions
7939: fixing timings for coolstream internal reader
7940: revert config which accidently got permitted in last commit, sorry
Changes:
7896: build: Do not disable DVBAPI forcefully when building on OS X.
Apparently some people are using OS X as cross compilation platform and disabling CONFIG_DVBAPI is not what is wanted in this case.
Patch by foobar.
7897: config: Do not assume unknown reader protocol equals "mouse".
Let the user know that he/she should fix its config instead of assuming. Pretty soon protocol type MOUSE may not even be compiled in the binary.
7898: csctapi: Move DB2COM reader functions in csctapi/ifd_db2com.c
Also introduce separate "db2com" reader.
7899: csctapi: Add set_RTS_DTS callback for cardreaders.
db2com reader uses its own rts/dts detection function.
7900: csctapi: Add DB2COM support in mouse_test reader
7901: csctapi: Convert phoenix/mouse reader to the new card reader API.
This is the last reader that needed conversion! Now the road to disabling cardreaders in oscam config is opened.
7902: csctapi: Re-introduce Set reader ready to Send, it's needed for some readers
7903: Tryfix for ticket 2946
7904: Tryfix for ticket 2950 (Duplicate reader)
7905: Remove hardcoded card reader names in reader_get_type_desc().
All card readers have ->desc field initialized.
7906: Move reader related functions out of oscam-simples.c into oscam-reader.c
7907: webif: Remove duplicated pcsc reader in readers dropdown.
7908: csctapi: Remove active field in struct s_cardreader.
Now that all readers are converted this field is no longer needed.
7909: csctapi/db2com: Move the rest of db2com code into ifd_db2com.c
7910: config: Remove temp file after config.mak was build.
7911: config: Special case WITH_CARDREADER.
WITH_CARDREADER should not be user visible option.
7912: List supported card readers in oscam help text.
7913: csctapi/smartreader: Move lock initialization in ->lock_init().
7914: dvbapi/azbox: Add missing include file.
7915: Quick & Dirty fix for broken cccam server introduced in r7912
7916: Temp fix for broken viaccess cards -> to be continued!
7917: Revert "List supported card readers in oscam help text."
This reverts commit r7912. The whole approach is wrong because module init functions needs the config files to be loaded to be able to initialize the modules.
Also commit r7915 which was "Quick & Dirty fix for broken cccam server introduced in r7912". r7912 broke more than cccam.
7918: config: Add card readers configuration.
The configuration allows enabling and disabling of individual card reader drivers.
This commit is just a preparation it'll start to work in the next few commits.
7919: config: Allow card reader enable/disable to work.
7920: build: Build only enabled card readers.
7921: csctapi: Create generic IO_Serial_{Transmit,Receive} functions based on Phoenix.
Lots of drivers were using Phoenix_{Transmit,Receive} functions which are a tiny wrappers around IO_Serial_{Write,Read}.
Move the code in io_serial.c. This allows us to break internal reader dependency on phoenix reader.
7922: sctapi: Create generic IO_Serial_GetStatus() based on Phoenix code.
7923: csctapi: Remove smargo dependancy on Phoenix reader.
7924: csctapi: Create generic IO_Serial_SetBaudrate() based on Phoenix code.
7925: Show AZBOX and COOLAPI status in oscam.version.
7926: Add card reader support status in oscam.version file.
7927: Improve oscam.version file.
- Remove "Max PID: unlimited" line (it seems useless).
- Improve oscam version fomatting.
7928: Nothing stops write_version() from working on CYGWIN.
7929: Move write_versionfile() in oscam.c
7930: Add -V command line parameter to show oscam version info.
7931: Fit oscam help text in 80 columns and improve wording and white spacing.
7932: Move command line parsing in parse_cmdline_params() function.
7933: Rename ->l fields in EMM and ECM structs.
l is a pretty poor name for a field. Lets use something that we can grep for (ecmlen and emmlen).
7934: config: Fix azbox and coolapi build when internal reader is disabled.
7935: config: Rename DB2COM reader to dbox2.
7936: newcamd: Add Opticum client id. Patch by manio posted in ticket #2961.
7937: Added new command line parameter '-V' to documentation.
7938: Updated card reader descriptions
7939: fixing timings for coolstream internal reader
7940: revert config which accidently got permitted in last commit, sorry
oscam-1.20-svn.8340-ppc-tuxbox-linux-dbox2.
Changes:
7941: Fix csp hash for betatunnel, it was blocking cache2 responses!
7942: Make it easier to parse help output and learn the default config dir.
7943: Removed the for display only used ecm->checksum calculation and replaced it by MD5 hash
7944: Add more log to set parity, could be usefull for tickets like 2974
7945: Add some more debug to DT08 phase of nagra
7946: removed double implementation of same check functions
restored old behaviour of old oscam version: returning NULL ecm if no cachehit
7947: Removed some obsolete checks in set parity
7948: Tryfix ticket 2992
7949: Fix: some mouse readers could not initialize viaccess cards anymore
7950: Experimental: Removed the caid / betatunnel check on cacheanswers. Since the cacheanswer is only valid for a very short time I think for cacheanswers we can rely on csp-hash and MD5 hash checksum of the ecm. If they match the ecm must be the same even if caid is different.
7951: csctapi: Fixed MP35 reader init failed after reboot
7952: Add support for Matrix Cam Air in DVBAPI.
7953: build: CONF_DIR should not be set when USE_MCA is requested.
7954: log: Remove private mca logging.
7955: Move MCA defines in better place.
7956: dvbapi-azbox: Introduce some defines.
7957: module-dvbapi-mca: Make it closer to dvbapi-azbox.
7958: module-dvbapi-mca.c closer to azbox
7959: Bring module-dvbapi-mca.c closer to module-dvbapi-azbox.c
7960: Tryfix ticket 2995
7961: fixed a timeout bug
7962: reverted oscam-reader part from my last commit due to problems
7963: Tryfix2 for ticket 2995
Skipped a whole class due to double increase -> fixed!
Tnx XabiX for reports and testing!
7964: fixed output: cache2 instead found when answer is splitted because of different request user groups
7965: fixed another timeout situation: When a client disconnects, other clients waiting on same requests could lose their request-link
7966: cccam: Prevent segfault when procesing EMMs. This fixes ticket #2992.
7967: Start emmfilter before ecm processing (might be usefull for unactivated cards)
Prio fix for oscam.dvbapi without p rules
7968: smartreader: Three coding conventions in a single file of less than 150 lines is a bit too much
7969: oscam.dvbapi i: line was broken -> fixed!
dvbapi using preferlocalreader option ignored i: lines in dvbapi -> fixed!
7970: According to the GNU Make manual, the flags are automatically parsed to recursive versions of make.
Hence, $(MAKEFLAGS) is obsolete in Makefile.extra
7971: Revert "According to the GNU Make manual, the flags are automatically parsed to"
This reverts commit r7970.
7972: Tryfix Kill_Thread
Some users report crashing oscam latest message in log was "WARNING: invalid client". After analysing of the sources it seems to me it was possible that kill_thread killed the current client and by doing that oscam would crash for sure.
7973: Revert "Tryfix Kill_Thread"
This reverts commit r7972. It can be implemented much simpler and cleaner.
7974: Reimplement r7972 cleanly.
This is a much simpler implementaton of commit r7972.
7975: Update for Smargo Smartreader+ in documentation.
7976: Reader recover for internal readerdevices
While testdriving a HD02 card I stumbled over an unresponsive internal reader. Only a powercycle of the box got the reader responsive again. While searching for the culprit I stumpled over the fact that FS 0 made the reader responsive again without the need to powercycle the box.
7977: Added example for Smartreader in smargo mode
7978: Partly reverted commit 7967 (prio fix for oscam.dvbapi without p rules)
.
.
.
ist mir diesmal ein bissi viel
.
.
.
8340: build: Use -pipe compiler parameter.
On my machine this speeds up the build ~5%.
Changes:
7941: Fix csp hash for betatunnel, it was blocking cache2 responses!
7942: Make it easier to parse help output and learn the default config dir.
7943: Removed the for display only used ecm->checksum calculation and replaced it by MD5 hash
7944: Add more log to set parity, could be usefull for tickets like 2974
7945: Add some more debug to DT08 phase of nagra
7946: removed double implementation of same check functions
restored old behaviour of old oscam version: returning NULL ecm if no cachehit
7947: Removed some obsolete checks in set parity
7948: Tryfix ticket 2992
7949: Fix: some mouse readers could not initialize viaccess cards anymore
7950: Experimental: Removed the caid / betatunnel check on cacheanswers. Since the cacheanswer is only valid for a very short time I think for cacheanswers we can rely on csp-hash and MD5 hash checksum of the ecm. If they match the ecm must be the same even if caid is different.
7951: csctapi: Fixed MP35 reader init failed after reboot
7952: Add support for Matrix Cam Air in DVBAPI.
7953: build: CONF_DIR should not be set when USE_MCA is requested.
7954: log: Remove private mca logging.
7955: Move MCA defines in better place.
7956: dvbapi-azbox: Introduce some defines.
7957: module-dvbapi-mca: Make it closer to dvbapi-azbox.
7958: module-dvbapi-mca.c closer to azbox
7959: Bring module-dvbapi-mca.c closer to module-dvbapi-azbox.c
7960: Tryfix ticket 2995
7961: fixed a timeout bug
7962: reverted oscam-reader part from my last commit due to problems
7963: Tryfix2 for ticket 2995
Skipped a whole class due to double increase -> fixed!
Tnx XabiX for reports and testing!
7964: fixed output: cache2 instead found when answer is splitted because of different request user groups
7965: fixed another timeout situation: When a client disconnects, other clients waiting on same requests could lose their request-link
7966: cccam: Prevent segfault when procesing EMMs. This fixes ticket #2992.
7967: Start emmfilter before ecm processing (might be usefull for unactivated cards)
Prio fix for oscam.dvbapi without p rules
7968: smartreader: Three coding conventions in a single file of less than 150 lines is a bit too much
7969: oscam.dvbapi i: line was broken -> fixed!
dvbapi using preferlocalreader option ignored i: lines in dvbapi -> fixed!
7970: According to the GNU Make manual, the flags are automatically parsed to recursive versions of make.
Hence, $(MAKEFLAGS) is obsolete in Makefile.extra
7971: Revert "According to the GNU Make manual, the flags are automatically parsed to"
This reverts commit r7970.
7972: Tryfix Kill_Thread
Some users report crashing oscam latest message in log was "WARNING: invalid client". After analysing of the sources it seems to me it was possible that kill_thread killed the current client and by doing that oscam would crash for sure.
7973: Revert "Tryfix Kill_Thread"
This reverts commit r7972. It can be implemented much simpler and cleaner.
7974: Reimplement r7972 cleanly.
This is a much simpler implementaton of commit r7972.
7975: Update for Smargo Smartreader+ in documentation.
7976: Reader recover for internal readerdevices
While testdriving a HD02 card I stumbled over an unresponsive internal reader. Only a powercycle of the box got the reader responsive again. While searching for the culprit I stumpled over the fact that FS 0 made the reader responsive again without the need to powercycle the box.
7977: Added example for Smartreader in smargo mode
7978: Partly reverted commit 7967 (prio fix for oscam.dvbapi without p rules)
.
.
.
ist mir diesmal ein bissi viel
.
.
.
8340: build: Use -pipe compiler parameter.
On my machine this speeds up the build ~5%.