Thanks for the report! As it happens, I think I fixed this a few minutes ago:
http://git.racket-lang.org/plt/commit/3b032893c1b9f74b04d0803ecd0f20c1d101dd11 Just to make sure before trying a new version, you might confirm that the bug goes away if you disable the JIT (e.g., with "-j"). At Wed, 29 Dec 2010 02:27:56 +0100, Jakub Piotr Cłapa wrote: > Hello, > > I found out a very peculiar for for/list combo that segfaults both the > CLI racket and DrRacket (Intel OS X) on the current nightly build. > > When I run the code (see below) it only works for empty input. When I > pass at least one line it segfaults. The (invalid) use of multiple > values is critical as is actually calling eof-object? on the result of > peek-char. > > My other version does not crash (just reports the invalidity of multiple > values in this context). It was compiled somewhere after 5.0.99.4 > (probably just after the websocket-client SSL merge but I am not sure). > > Below is the misbehaving code, some test run outputs and a backtrace. > > > #lang racket > (let ([in (current-input-port)]) > (for/list ([p (in-producer (λ () (file-position in)) #f)] > [l (in-lines in)] > #:when (eof-object? (peek-char in))) > (values l p))) > > > Expected output: > > > [...@jpc WK2009] echo "1" | racket- for-test.rkt > context expected 1 value, received 2 values: "1" 0 > > === context === > /Users/jpc/Projects/regaz/WK2009/for-test.rkt:3:2: for-loop > /Users/jpc/Projects/regaz/WK2009/for-test.rkt: [running body] > > > Got: > > > [...@jpc WK2009] echo "1" | racket for-test.rkt > Seg fault (internal error) at 0x6 > > > A quick backtrace I was able to get: > > > Program received signal EXC_BAD_ACCESS, Could not access memory. > Reason: KERN_PROTECTION_FAILURE at address: 0x00000006 > 0x0015c8fb in check_cycles_fast () > (gdb) bt > #0 0x0015c8fb in check_cycles_fast () > #1 0x0015c9d7 in check_cycles_fast () > #2 0x00167924 in print_to_string () > #3 0x00167e9c in print_to_port () > #4 0x00167f1c in scheme_internal_print () > #5 0x00156270 in sch_default_global_port_print_handler () > #6 0x000520a4 in scheme_do_eval () > #7 0x0014f8e1 in sch_default_print_handler () > #8 0x001593ed in display_write () > #9 0x000520a4 in scheme_do_eval () > #10 0x00086444 in scheme_default_print_handler () > #11 0x000520a4 in scheme_do_eval () > #12 0x0007953c in for_each () > #13 0x00756ead in ?? () > #14 0x0005158c in scheme_do_eval () > #15 0x000fae1e in body_one_expr () > #16 0x0005259c in scheme_do_eval () > #17 0x0007b1b8 in scheme_finish_apply_for_prompt () > #18 0x0007b357 in scheme_apply_for_prompt () > #19 0x00085130 in call_with_prompt () > #20 0x000520a4 in scheme_do_eval () > #21 0x0007b572 in do_call_with_prompt () > #22 0x000fb160 in scheme_module_run_finish () > #23 0x002ff3a3 in ?? () > #24 0x000e8af6 in scheme_module_run_start () > #25 0x0010107a in eval_module_body () > #26 0x001011f9 in do_start_module () > #27 0x00106f81 in start_module () > #28 0x0010affc in parse_requires () > #29 0x0010c59c in do_namespace_require () > #30 0x000520a4 in scheme_do_eval () > #31 0x00074e44 in apply_k () > #32 0x00074536 in scheme_top_level_do_worker () > #33 0x00074778 in scheme_top_level_do () > #34 0x000035f0 in finish_cmd_line_run () > #35 0x0000458a in main_after_stack () > #36 0x0000c0a3 in scheme_main_stack_setup () > #37 0x00002b46 in main () > > > -- > regards, > Jakub Piotr Cłapa > _________________________________________________ > For list-related administrative tasks: > http://lists.racket-lang.org/listinfo/dev _________________________________________________ For list-related administrative tasks: http://lists.racket-lang.org/listinfo/dev
