Re: NULL (was Re: cvs commit: parrot interpreter.c stacks.c)
Nicholas Clark wrote: On Sat, Nov 09, 2002 at 11:09:39AM -, [EMAIL PROTECTED] wrote: -interpreter->ctx.int_reg_base->prev = NULL; It's not likely to be a portability problem, as I seem to remember that the C FAQ is only able to mention one very obscure platform where NULL was not all bits zero in memory. Yep C-FAQ 5.17 lists these machines (Prime50, Eclipse MV, CDC 180, old HP 3000) To NULL or not to NULL, this is the question. I believe that a bigger assumption (not sure if we're making it; didn't spot it) is that float 0.0 and double 0.0 are written as all bits zero in memory. NRegs were memset to 0, Parrot_clear_n() sets them to 0.0. But we seem to have a real problem WRT JIT and floats. Some perl6 tests still break, when run through -j. I could imagine different float formats for 80/64 bit floats is the reason. Nicholas Clark leo
NULL (was Re: cvs commit: parrot interpreter.c stacks.c)
On Sat, Nov 09, 2002 at 11:09:39AM -, [EMAIL PROTECTED] wrote: > -/* Initialize the integer register chunk */ > -interpreter->ctx.int_reg_base->used = 0; > +/* Initialize the register chunks */ >interpreter->ctx.int_reg_base->free = FRAMES_PER_INT_REG_CHUNK; > -interpreter->ctx.int_reg_base->next = NULL; > -interpreter->ctx.int_reg_base->prev = NULL; By doing this you are assuming that the in memory representation of a NULL pointer is all bits NULL. This isn't guaranteed. (What *is* guaranteed by the standard is that a NULL pointer casts to 0, and that all other pointers cast to non-zero) It's not likely to be a portability problem, as I seem to remember that the C FAQ is only able to mention one very obscure platform where NULL was not all bits zero in memory. (Sorry, not online at the moment, so no URL) However, I have a niggling thought that someone telling me that some compiler product in memory leak debug mode (possibly a Microsoft compiler) was able to have NULL pointers *not* stored as all bits zero in memory, and so could fault confusion between 0 and NULL I believe that a bigger assumption (not sure if we're making it; didn't spot it) is that float 0.0 and double 0.0 are written as all bits zero in memory. Nicholas Clark -- Brainfuck better than perl? http://www.perl.org/advocacy/spoofathon/
[CVS ci] (was: Should memory be washed?)
Dan Sugalski wrote:
At 8:58 PM +0100 11/6/02, Leopold Toetsch wrote:
If we want this, then lets have Parrot_{re,}allocate{,zeroed}.
The allocate_string variants are ok with unzeroed mem already.
Which was my thought here. Things that care can ask for zeroed memory,
which they may get anyway. (Or we may keep a big pool of zeroed memory
around as it's cheaper to clean large blocks or something)
This patch implements the first step of mem alloc cleanup:
- NEW mem_sys_allocate_zeroed (calloc)
- mem_sys_allocate uses malloc now
- NEW Parrot_allocate_zeroed
- Parrot_allocate uses malloc in res_lea now
implying the following rules:
Buffer headers and such are guaranteed to be zeroed.
Mem allocated via Parrot_allocate may or may not be zeroed, depending on
allocator scheme.
The _allocate_string variants do never guarantee to deliver zeroed memory.
Next step is going through files and look, where allocating zeroed
memory looks cheaper and use mem_sys_allocate_zeroed.
leo
mem leaks test summary
After removing IntQueue this is the state of memory leaks[1] (i386/linux): Failed Test Status Wstat Total Fail Failed List of failed --- t/op/hacks.t 21 50,00% 1 t/op/rx.t237 30,43% 8-9, 12-14, 21-22 9 subtests skipped. Failed 2/35 test scripts, 94.29% okay. 8/494 subtests failed, 98.38% okay. - hacks.t uses PIO_open/PIO_new which doesn't keep track of IO objects - rx.t is broken WRT bitmaps and intstack [1] a test passes, when output matches and when it doesn't show WARNINGs/ERRORs (s. testyamd script) leo
[CVS ci] removed (was: removiing classes/intqueue.pmc)
Leopold Toetsch wrote: If no one hollers, I'll remove the IntQueue class. - only used in some tests AFAIK - functionality may be written in terms of intlist push/pop - has a weird syntax for queue/dequeue - is broken WRT memory management and clone IntQueue is history. Please run 'make realclean; perl Configure.pl; make -s' to get pmc class enums reordered. leo
