| Summary: | Error occur when clicking on Contacts icon. | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [Build System, Metadata & Runtime] OE-Core | Reporter: | Liang Wang <liangliang.wang> | ||||
| Component: | core | Assignee: | Edwin Zhai <edwin.zhai> | ||||
| Status: | VERIFIED FIXED | QA Contact: | |||||
| Severity: | minor | ||||||
| Priority: | Medium | CC: | dongxiao.xu, jiajun.xu, liang.li2, liangliang.wang, meta.mr.watcher, meta.watcher, yi.zhao | ||||
| Version: | 1.0 | ||||||
| Target Milestone: | 1.0 M4 | ||||||
| Hardware: | Other | ||||||
| OS: | mips | ||||||
| Whiteboard: | |||||||
| OS type for building Yocto: | --- | Type of Regression: | --- | ||||
| Verified: | Documentation change: | --- | |||||
| Attachments: |
|
||||||
|
Description
Liang Wang
2011-01-17 19:13:07 UTC
I checked the Jan 14 changesets. Contacts used e-addressbook-factory to create DB file, but our db didn't support fast mutexes and caused failure: "(e-addressbook-factory:346): libebookbackend-WARNING **: libdb error: architecture lacks fast mutexes: applications cannot be threaded (e-addressbook-factory:346): libebookbackend-WARNING **: db_env_open failed with Invalid argument" This bug occur when HAVE_MUTEX_THREADS not defined, but application try to use db_env_open with *_ENV_THREAD flag. This issue of db only exists on mips, while x86/arm are okay. x86 is generic, and arm has "--with-mutex=ARM/gcc-assembly". Do we need something similar for mips? CC db's owner dongxiao. BTW, I also checked latest master (6cb8fd6def4912e4aa76330649ba42a9ed2694fd), this issue got fixed. (maybe after mark hatle upgrading db @ e77207) So mark it as fixed. Dongxiao can investigate it in future if needed. I compared the db's log.do_configure from 4.2.52 to 5.1.19: === 4.2.52 === checking for mutexes... UNIX/fcntl configure: WARNING: NO FAST MUTEXES FOUND FOR THIS COMPILER/ARCHITECTURE. === 5.1.19 === checking for mutexes... POSIX/pthreads/library checking pthread.h usability... yes checking pthread.h presence... yes checking for pthread.h... yes checking for main in -lpthread... yes checking for atomic operations... no checking for library containing sched_yield... none required checking for library containing fdatasync... none required checking for library containing getaddrinfo... none required checking for library containing hstrerror... none required checking for main in -lm... yes checking for main in -lnsl... yes checking for main in -lsocket... no So maybe db failed to detect the pthread library? We can revisit it if this bug occur again in future. This issue does *not* occur on sato/sdk qemumips image. Image location: http://autobuilder.yoctoproject.org/nightly/20110305-4/machines/qemu/qemumips/ commit c192fefd868be53c20dfd8bc44e3e815cf53a10a |