[
https://issues.apache.org/jira/browse/THRIFT-1400?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Carlos Carrasco updated THRIFT-1400:
------------------------------------
Attachment: fix-field-value-buf.diff
> Ruby native extension aborts with __stack_chk_fail in OSX
> ---------------------------------------------------------
>
> Key: THRIFT-1400
> URL: https://issues.apache.org/jira/browse/THRIFT-1400
> Project: Thrift
> Issue Type: Bug
> Components: Ruby - Library
> Affects Versions: 0.7, 0.8
> Environment: Mac OSX 10.7.2
> Xcode 4.2
> MRI 1.9.3-rc1 built with clang, 64 bit
> Reporter: Carlos Carrasco
> Attachments: fix-field-value-buf.diff
>
>
> get_field_value and set_field_value in struct.c apparently allocate one less
> byte than is required when building the field name, which I guess it works
> most of the time everywhere since there is nothing else in the stack, but
> OSX+clang appear to be serious about it.
> My patch follows the rationale that RSTRING_LEN does not include the 0-ended
> length, and apparently in Ruby 1.9(.3?) RSTRING_PTR is not zero-terminated,
> so I add one extra byte to the char[] and fix the passed buffer length to be
> actually the string length plus the extra byte for the zero.
> Stack trace:
> #0 0x00007fff9504bce2 in __pthread_kill ()
> #1 0x00007fff93d617d2 in pthread_kill ()
> #2 0x00007fff93d52b4a in __abort ()
> #3 0x00007fff93d4f070 in __stack_chk_fail ()
> #4 0x000000010077a417 in set_field_value (obj=<value temporarily
> unavailable, due to optimizations>, field_name=<value temporarily
> unavailable, due to optimizations>, value=<value temporarily unavailable, due
> to optimizations>) at struct.c:426
> #5 0x0000000100779141 in rb_thrift_struct_read (self=4306092880,
> protocol=4306073480) at struct.c:592
> #6 0x0000000100779cb9 in read_anything (protocol=4306073480, ttype=<value
> temporarily unavailable, due to optimizations>, field_info=<value temporarily
> unavailable, due to optimizations>) at struct.c:469
> #7 0x0000000100779a5a in read_anything (protocol=4306073480, ttype=<value
> temporarily unavailable, due to optimizations>, field_info=4304975600) at
> struct.c:522
> #8 0x0000000100779132 in rb_thrift_struct_read (self=4306096120,
> protocol=4306073480) at struct.c:592
> #9 0x0000000100779cb9 in read_anything (protocol=4306073480, ttype=<value
> temporarily unavailable, due to optimizations>, field_info=<value temporarily
> unavailable, due to optimizations>) at struct.c:469
> #10 0x0000000100779132 in rb_thrift_struct_read (self=4306097000,
> protocol=4306073480) at struct.c:592
> #11 0x000000010013e7c5 in vm_call_cfunc [inlined] () at
> /Users/ccm/.rvm/src/ruby-1.9.3-rc1/vm_insnhelper.c:404
> #12 0x000000010013e7c5 in vm_call_method (th=0x100301b10, cfp=0x1004ffb30,
> num=1, blockptr=0x0, flag=<value temporarily unavailable, due to
> optimizations>, id=<value temporarily unavailable, due to optimizations>,
> me=<value temporarily unavailable, due to optimizations>, recv=<value
> temporarily unavailable, due to optimizations>) at vm_insnhelper.c:534
> #13 0x000000010012f5ee in vm_exec_core (th=0x100301b10, initial=<value
> temporarily unavailable, due to optimizations>) at vm.inc:1015
> #14 0x0000000100138c94 in vm_exec (th=0x100301b10) at vm.c:1220
> #15 0x00000001001395ab in rb_iseq_eval_main (iseqval=4304251720) at vm.c:1461
> #16 0x000000010003701f in ruby_exec_internal (n=<value temporarily
> unavailable, due to optimizations>) at eval.c:204
> #17 0x0000000100036f67 in ruby_run_node (n=0x1008dab48) at eval.c:251
> #18 0x0000000100000edf in main (argc=2, argv=0x7fff5fbff838) at main.c:38
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
