| Summary: | bitbake "NoProvider" message doesn't aid in resolving the problem | ||
|---|---|---|---|
| Product: | [Build System, Metadata & Runtime] OE-Core | Reporter: | Darren Hart <dvhart> |
| Component: | core | Assignee: | Richard Purdie <richard.purdie> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | enhancement | ||
| Priority: | Medium | CC: | jiajun.xu, josh, ke.yu, kevin.tian, meta.mr.watcher, meta.watcher, richard.purdie, sgw, song.liu |
| Version: | unspecified | ||
| Target Milestone: | 1.2 M3 | ||
| Hardware: | x86 | ||
| OS: | Multiple | ||
| Whiteboard: | (P1)(QA Testing Completed)(Done) | ||
| OS type for building Yocto: | --- | Type of Regression: | --- |
| Verified: | Documentation change: | --- | |
| Bug Depends on: | |||
| Bug Blocks: | 1543 | ||
This looks to be a non-trivial amount of work. I'd like to postpone it till after 1.0... Another example is when COMPATIBLE_MACHINE prevents a recipe from building it's not obvious that it's happened without spotting the line in the bb file. After http://git.yoctoproject.org/cgit.cgi/poky/commit/?id=a5463088fe57462bf681b648d21212808454fc2a the python stack trace is no longer printed. Since Paul already added patches which show information about COMPATIBLE_HOST and COMPATIBLE_MACHINE recipe incompatibility a long time ago, resolving this bug. Output is now: ERROR: Nothing PROVIDES 'valgrind' ERROR: valgrind was skipped: incompatible with host mips-poky-linux (which is from COMPATIBLE_HOST and handled the same as COMPATIBLE_MACHINE) The feature is tested workable in 1.2 M2. |
While working on a new linux kernel recipe, bitbake reported the following when my local.conf MACHINE setting wasn't listed in the COMPATIBLE_MACHINES of the new recipe: dvhart@doubt:build$ bitbake linux-linaro NOTE: Handling BitBake files: | (0876/0876) [100 %] Parsing of 876 .bb files complete (823 cached, 53 parsed). 1036 targets, 35 skipped, 0 masked, 0 errors. OE Build Configuration: BB_VERSION = "1.11.0" METADATA_BRANCH = "master-dev" METADATA_REVISION = "3f3499a0e18ffcc6171ca0d148940a92990098d4" TARGET_ARCH = "i586" TARGET_OS = "linux" MACHINE = "qemux86" DISTRO = "poky" DISTRO_VERSION = "0.9+snapshot-20101118" TARGET_FPU = "" ERROR: Nothing PROVIDES 'linux-linaro' Command execution failed: Traceback (most recent call last): File "/home/dvhart/source/poky.git/scripts/..//bitbake/lib/bb/command.py", line 88, in runAsyncCommand commandmethod(self.cmds_async, self, options) File "/home/dvhart/source/poky.git/scripts/..//bitbake/lib/bb/command.py", line 184, in buildTargets command.cooker.buildTargets(pkgs_to_build, task) File "/home/dvhart/source/poky.git/scripts/..//bitbake/lib/bb/cooker.py", line 756, in buildTargets taskdata.add_provider(localdata, self.status, k) File "/home/dvhart/source/poky.git/scripts/..//bitbake/lib/bb/taskdata.py", line 354, in add_provider self.add_provider_internal(cfgData, dataCache, item) File "/home/dvhart/source/poky.git/scripts/..//bitbake/lib/bb/taskdata.py", line 374, in add_provider_internal raise bb.providers.NoProvider(item) NoProvider: linux-linaro For sane user input (ie package-name typos and and bad configs like my case above) we should avoid dumping python traces to the console. Ideally it would say something more along the lines of: dvhart@doubt:build$ bitbake linux-linaro Parsing of 876 .bb files complete (823 cached, 53 parsed). 1036 targets, 35 skipped, 0 masked, 0 errors. OE Build Configuration: BB_VERSION = "1.11.0" METADATA_BRANCH = "master-dev" METADATA_REVISION = "3f3499a0e18ffcc6171ca0d148940a92990098d4" TARGET_ARCH = "i586" TARGET_OS = "linux" MACHINE = "qemux86" DISTRO = "poky" DISTRO_VERSION = "0.9+snapshot-20101118" TARGET_FPU = "" ERROR: Nothing PROVIDES 'linux-linaro' for MACHINE=qemux86-64 dvhart@doubt:build$