aboutsummaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)Author
2011-08-08ic: Support parallel makeBjörn Gustavsson
2011-08-08orber: Support parallel makeBjörn Gustavsson
2011-08-08.gitignore: Ignore IDL-GENERATEDBjörn Gustavsson
One input file that generate multiple output file are tricky to handle in make. One way is to use a stand-in to represent all the other generated files.
2011-08-08public_key: Support parallel makeBjörn Gustavsson
2011-08-08ssh: Support parallel makeBjörn Gustavsson
2011-08-08os_mon: Support parallel makeBjörn Gustavsson
2011-08-08diameter: Support parallel makeBjörn Gustavsson
2011-08-08snmp: Support parallel makeBjörn Gustavsson
2011-08-08megaco: Support parallel makeBjörn Gustavsson
2011-08-08megaco/src/flex/Makefile.in: Support parallel makeBjörn Gustavsson
Make sure that directories are created before they are used. I tried to use order-only prerequisites to create the directories, but run into two problems: First, order-only prerequisites are only implemented in Make 3.80 and later. Second, on a computer running Solaris/Intel 2.8 (with Make 3.80), order-only prerequisites seemed to work like ordinary prerequisites, causing targets to be re-built if the timestamp for the directory changed. Therefore, using a shell command to run mkdir seems to be the more portable solution.
2011-08-08*/c_src/Makefile*: Support parallel makeBjörn Gustavsson
Create directories first, not within implicit rules. If two 'install' instances runs at the same time attempting to create a directory, one of them may fail with an "File exists" error. I tried to use order-only prerequisites to create the directories, but run into two problems: First, order-only prerequisites are only implemented in Make 3.80 and later. Second, on a computer running Solaris/Intel 2.8 (with Make 3.80), order-only prerequisites seemed to work like ordinary prerequisites, causing targets to be re-built if the timestamp for the directory changed. Therefore, using a shell command to run mkdir seems to be the more portable solution.
2011-08-08eunit: Support parallel makeBjörn Gustavsson
2011-08-08gs: Support parallel makeBjörn Gustavsson
2011-08-08common_test Makefile: Support parallel makeBjörn Gustavsson
Add dependecies to force the ct_line parse transform module to be built before all other modules.
2011-08-08erts/emulator/Makefile.in: Support parallel makeBjörn Gustavsson
2011-08-08erts: Fix dependency generationBjörn Gustavsson
Apparantly, an invocation like "make generate depend" will run the 'generate' and 'depend' targets in parallel, even if the Makefile contains a .NOTPARALLEL target. In this case, it will result in incorrect dependencies since not all files have been generated when the dependencies are computed. Split up the single 'make' invocation to two invocations, to force the targets to be run sequentially.
2011-08-08erts/Makefile.in: Suppress parallel makeBjörn Gustavsson
2011-08-08Top-level Makefile: Suppress parallel makeBjörn Gustavsson
2011-08-08lib/Makefile: Only build the runtime_tools application onceBjörn Gustavsson
2011-08-08lib/Makefile: Remove support for OSE deltaBjörn Gustavsson
2011-08-08jinterface: Use otp_subdir.mk instead of homebrewed solutionBjörn Gustavsson
There once was a reason to have a "Makefile.otp" makefile, but it doesn't apply any longer. Rename it to "Makefile" so that the standard otp_subdir.mk file can be used for recursion into sub directories.
2011-08-08erts/emulator/Makefile.in: Remove clearmake supportBjörn Gustavsson
2011-08-08otp_subdir.mk: Remove support for clearmakeBjörn Gustavsson
2011-08-08Makefile.in: Remove the obsolete bootstrap targetsBjörn Gustavsson
2011-08-05Merge branch 'bjorn/erts/loader-related-bs-bugs/OTP-9284' into devBjörn Gustavsson
* bjorn/erts/loader-related-bs-bugs/OTP-9284: Fix binary construction with huge literal sizes beam_load.c: Add overflow check of tag values beam_makeops: Add some sanity checks Fix construction of <<0:((1 bsl 32)-1)>>
2011-08-03Fix binary construction with huge literal sizesBjörn Gustavsson
Constructing binaries using the bit syntax with literals sizes that would not fit in an Uint will either cause an emulator crash or the loading to be aborted. Use the new TAG_o tag introduced in the previous commit to make sure that the attempt to create huge binary literals will generate a system_limit exception at run-time.
2011-08-03beam_load.c: Add overflow check of tag valuesBjörn Gustavsson
The handling of large values for other tags than TAG_i (integer) is buggy. Any tag value equal to or greater than 2^40 (5 bytes) will abort loading. Tag values fitting in 5 bytes will be truncated to 4 bytes values. Those bugs cause real problems because the bs_init2/6 and bs_init_bits/6 instructions unfortunately use TAG_u to encode literal sizes (using TAG_i would have been a better choice, but it is too late to change that now). Any binary size that cannot fit in an Uint should cause a system_limit exception at run-time, but instead the buggy handling will either cause an emulator crash (for values in the range 2^32 to 2^40-1) or abort loading. In this commit, implement overflow checking of tag values as a preparation for fixing the binary construction instructions. If any tag value cannot fit in an Uint (except for TAG_i), change the tag to the special TAG_o overflow tag.
2011-08-03beam_makeops: Add some sanity checksBjörn Gustavsson
We want to make sure that a tag/type name is not defined more than once and that we don't define too many primitive tags. Primitive tags must be named with lowercase letters (or they will be confused with variable names in transformations in the ops.tab file).
2011-08-03Fix construction of <<0:((1 bsl 32)-1)>>Björn Gustavsson
Attempting to construct <<0:((1 bsl 32)-1)>>, the largest bitstring allowed in a 32 bit emulator, would cause an emulator crash because of integer overflow. Fix the problem by using an Uint64 to avoid integer overflow. Do not attempt to handle construction of <<0:((1 bsl 64)-1>> in a 64-bit emulator, because that will certainly cause the emulator to terminate anyway because of insufficient memory.
2011-07-29Merge branch 'sverker/allocator-aoff/OTP-9424' into devSverker Eriksson
* sverker/allocator-aoff/OTP-9424: New allocator: Address order first fit (aoff)
2011-07-27Merge branch 'sverker/fun_SUITE-refc_dist-gcfix' into devSverker Eriksson
* sverker/fun_SUITE-refc_dist-gcfix: Fix test case fun_SUITE:refc_dist
2011-07-27Merge branch 'sverker/epmd-vxworks-select-bug/OTP-9427' into devSverker Eriksson
* sverker/epmd-vxworks-select-bug/OTP-9427: Fix epmd crash on vxworks caused by faulty argument to select
2011-07-27Fix epmd crash on vxworks caused by faulty argument to selectSverker Eriksson
The first argument g->max_conn to select() in epmd_srv.c seems to be wrong as it excludes the listening socket(s). Instead we keep track of the highest fd seen. However I still don't understand why select() started to fail for vxworks with R14B03 when the feature of more than one listening socket was introduced. The default is still 1 listening socket (num_sockets=1), which would make the first argument to select() unchanged.
2011-07-26Merge branch 'maint-r14' into devMicael Karlberg
Conflicts: lib/snmp/doc/src/notes.xml lib/snmp/src/agent/snmpa_mpd.erl lib/snmp/src/app/snmp.appup.src lib/snmp/vsn.mk
2011-07-26Merge branch 'sverker/testcase/OTP-9422' into devSverker Eriksson
* sverker/testcase/OTP-9422: Test case for OTP-9422
2011-07-26Merge branch 'sverker/testcase/OTP-9423' into devSverker Eriksson
* sverker/testcase/OTP-9423: Add test case for ETS bug OTP-9423
2011-07-26Merge branch 'sverker/sbmbc-fixup' into devSverker Eriksson
* sverker/sbmbc-fixup: Small fixups for rickard/sbmbc/OTP-9339
2011-07-26Fix test case fun_SUITE:refc_distSverker Eriksson
It failed sometimes depending on GC invocation.
2011-07-22Merge branch 'bmk/snmp/snmp421_integration' into maint-r14Erlang/OTP
* bmk/snmp/snmp421_integration: Fixed install directory typo for man3. Peter Lemenkov. Hans Ulrich Niedermann. The snmp config tool could not handle (manager) audit trail config because the option seqno was not handled. OTP-9354 [agent] The SNMP ACM cache was not properly updated when changes where made to the VACM security-to-group, access and view-tree-family tables. OTP-9367 Added marker id's in the snmpc(command) to allow linking top specific compiler options. Updated release notes accordingly. OTP-9437 [manager] There was no way to specify transport domain. The transport domains was assumed to be IPv4 (transportDomainUdpIpv4). This has now been changed so that it can also be IPv6 (transportDomainUdpIpv6). To facilitate this, the transport domain, tdomain, is now a (new) valid option when registering (snmpm#register_agent) a new agent (and updating (snmpm#update_agent_info) agent info). This also mean that the transport behaviour has changed. Fixed appup file. OTP-9305 [compiler] Make documentation for the snmpc escript regarding the the new option for warnings as errors ("--wae" in this case). Tuncer Ayaz. OTP-9437 [compiler] Added the option warnings_as_errors which specifies whether warnings should be treated as errors. Tuncer Ayaz. OTP-9437 Fixed unintended build to ebin dir of snmpc escript. Cosmetics indenting... Removed "crap" that somehow had gotten into the file (escape sequence). Fixed install of MIB compiler escript. Was incorrectly installed both in bin end ebin. Fixed (uncommented) test cases. Improved (transport) domain error reporting. Fixed release notes and API documentation for tdomain.xs transport domain config option named tdomain. Removed debug printouts. transport domain error detection. [manager] No way to specify transport domain. Transport domains was assumed to be IPv4. This has now been changed so that it can also be IPv6. To facilitate this, the transportDomain is now a (new) valid option. This also mean that the transport behaviour has changed. OTP-9305 Updated version. [agent] Did not handle transport domains properly in some cases. OTP-9400 Utility make target for cleanig the repo... ...
2011-07-22Fixed install directory typo for man3.Micael Karlberg
Peter Lemenkov Hans Ulrich Niedermann OTP-9442 Merge branch 'bmk/snmp/install_dir_typo_man3/OTP-9442' into bmk/snmp/snmp421_integration Conflicts: lib/snmp/doc/src/notes.xml
2011-07-22Fixed install directory typo for man3.Micael Karlberg
Peter Lemenkov. Hans Ulrich Niedermann. OTP-9442
2011-07-22Merge branch 'raimo/io_lib_fread/OTP-9439' into devRaimo Niskanen
* raimo/io_lib_fread/OTP-9439: Code cleanup, unduplicate test for whitespace EOF before first field is only ok for first character in io:fread At end of input ask for more also for for literal format characters
2011-07-22Test case for OTP-9422Sverker Eriksson
2011-07-22[agent] The SNMP ACM cache was not properly updated whenMicael Karlberg
changes where made to the VACM security-to-group, access and view-tree-family tables. OTP-9367 Merge branch 'bmk/snmp/agent/acm_cache_cleanup/OTP-9367' into bmk/snmp/snmp421_integration Conflicts: lib/snmp/doc/src/notes.xml lib/snmp/src/app/snmp.appup.src
2011-07-21Code cleanup, unduplicate test for whitespaceRaimo Niskanen
2011-07-21The snmp config tool could not handle (manager) audit trail configMicael Karlberg
because the option seqno was not handled. OTP-9354 Merge branch 'bmk/snmp/audit_trail_config_bug/OTP-9354' into bmk/snmp/snmp421_integration Conflicts: lib/snmp/doc/src/notes.xml lib/snmp/src/app/snmp.appup.src
2011-07-21The snmp config tool could not handle (manager) audit trail configMicael Karlberg
because the option seqno was not handled. OTP-9354
2011-07-21EOF before first field is only ok for first character in io:freadRaimo Niskanen
2011-07-21At end of input ask for more also for for literal format charactersRaimo Niskanen
2011-07-21[agent] The SNMP ACM cache was not properly updated whenMicael Karlberg
changes where made to the VACM security-to-group, access and view-tree-family tables. OTP-9367