mirror of
https://github.com/Mercury-Language/mercury.git
synced 2026-04-22 04:43:53 +00:00
083d376e6598628362ee91c2da170febd83590f4
19 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Mark Brown
|
d465fa53cb |
Update the COPYING.LIB file and references to it.
Discussion of these changes can be found on the Mercury developers
mailing list archives from June 2018.
COPYING.LIB:
Add a special linking exception to the LGPL.
*:
Update references to COPYING.LIB.
Clean up some minor errors that have accumulated in copyright
messages.
|
||
|
Zoltan Somogyi
|
53b573692a |
Convert C code to use // style comments.
runtime/*.[ch]:
trace/*.[chyl]:
As above. In some places, improve comments, e.g. by expanding contractions
such as "we've". Add #ifndef guards against double inclusion around
the trace/*.h files that did not already have them.
tools/*:
Make the corresponding changes in shell scripts that generate .[ch] files
in the runtime.
tests/*:
Conform to a slight change in the text of a message.
|
||
|
Zoltan Somogyi
|
67326f16e4 |
Fix style issues in the runtime.
Move all .h and .c files to four-space indentation without tabs, if they weren't there already. Use the same vim line for all .h and .c files. Align all backslashes at the ends of lines in macro definitions. Align close comment signs. In some places, fix inconsistent indentation. Fix a bunch of comments. Add XXXs to a few of them. |
||
|
Julien Fischer
|
787f8b2c6d |
Fix spelling and grammer in runtime comments.
runtime/*.[ch]:
As above.
|
||
|
Julien Fischer
|
f522a3dbaf |
Define atomic ops in high-level .par grades.
runtime/mercury_atomic_ops.c: Fix a mismatch between the macros protecting the declarations and the definitions of the atomic ops. |
||
|
Paul Bone
|
a9f82d004b |
On some systems the CPU's time stamp counter (TSC) cannot reliabily be
used. Mercury's ThreadScope support will now use gettimeofday() by
default, but use of the TSC may be enabled.
Note that in Linux, gettimeofday() does not always make a system call.
runtime/mercury_threadscope.[ch]:
Add support for measuring time with gettimeofday().
Use gettimeofday() to measure time by default.
runtime/mercury_atomic_ops.[ch]
Add a new function MR_tsc_is_sensible(), It returns true if the TSC can
(as far as the RTS can detect) be used.
Fix trailing whitespace.
runtime/mercury_wrapper.c:
Add a new runtime option --threadscope-use-tsc.
When specified this option allows threadscope to use the CPU's TSC to
measure time.
doc/userguide.texi:
Document the --threadscope-use-tsc option. This documentation is
commented out.
|
||
|
Julien Fischer
|
8af00f7a2a |
Avoid using the __GNUC__ macro in the runtime as a test for the presence of
Branches: main, 11.07 Avoid using the __GNUC__ macro in the runtime as a test for the presence of gcc, since clang also defines that macro. Since clang doesn't support all of the GNU C extensions, we can't actually use __GNUC__ without also checking whether we are actually using clang. runtime/mercury_conf_param.h: Add three new macros, MR_CLANG, MR_GNUC and MR_MSVC that are defined only when the C compiler is clang, gcc, or Visual C respectively. (In particular, MR_GNUC will _not_ be defined when the C compiler is clang.) runtime/mercury.c: runtime/mercury.h: runtime/mercury_atomic_ops.c: runtime/mercury_atomic_ops.h runtime/mercury_bitmap.h: runtime/mercury_float.h: runtime/mercury_getopt.c: runtime/mercury_goto.h: runtime/mercury_heap.h: runtime/mercury_std.h: Replace uses of the __GNUC__ and __clang__ macros with the above. runtime/mercury_regs.h: As above, also #include mercury_conf_param.h directly since this file is #included by some of the tests in the configure script. |
||
|
Peter Wang
|
7e26b55e74 |
Implement a new form of memory profiling, which tells the user what memory
Branches: main
Implement a new form of memory profiling, which tells the user what memory
is being retained during a program run. This is done by allocating an extra
word before each cell, which is used to "attribute" the cell to an
allocation site. The attribution, or "allocation id", is an address to an
MR_AllocSiteInfo structure generated by the Mercury compiler, giving the
procedure, filename and line number of the allocation, and the type
constructor and arity of the cell that it allocates.
The user must manually instrument the program with calls to
`benchmarking.report_memory_attribution', which forces a GC and summarises
the live objects on the heap using the attributions. The mprof tool is
extended with a new mode to parse and present that data.
Objects which are unattributed (e.g. by hand-written C code which hasn't
been updated) are still accounted for, but show up in profiles as "unknown".
Currently this profiling mode only works in conjunction with the Boehm
garbage collector, though in principle it can work with any memory allocator
for which we can access a list of the live objects. Since term size
profiling relies on the same technique of using an extra word per memory
cell, the two profiling modes are incompatible.
The output from `mprof -s' looks like this:
------ [1] some label ------
cells words cumul procedure / type (location)
14150 38872 total
* 1949/ 13.8% 4872/ 12.5% 12.5% <predicate `parser.parse_rest/7' mode 0>
975/ 6.9% 1950/ 5.0% list.list/1 (parser.m:502)
487/ 3.4% 1948/ 5.0% term.term/1 (parser.m:501)
487/ 3.4% 974/ 2.5% term.const/0 (parser.m:501)
* 1424/ 10.1% 4272/ 11.0% 23.5% <predicate `parser.parse_simple_term_2/6' mode 0>
708/ 5.0% 2832/ 7.3% term.term/1 (parser.m:643)
708/ 5.0% 1416/ 3.6% term.const/0 (parser.m:643)
...
boehm_gc/alloc.c:
boehm_gc/include/gc.h:
boehm_gc/misc.c:
boehm_gc/reclaim.c:
Add a callback function to be called for every live object after a GC.
Add a function to write out the GC_size_map array.
compiler/layout.m:
Define the alloc_site_info type which is equivalent to the
MR_AllocSiteInfo C structure.
Add alloc_site_array as a kind of "layout" array.
compiler/llds.m:
Add allocation sites to `cfile' structure.
Replace TypeMsg argument (which was also for profiling) on `incr_hp'
instructions by an allocation site identifier.
Add a new foreign_proc_component for allocation site ids.
compiler/code_info.m:
compiler/global_data.m:
compiler/proc_gen.m:
Keep the set of allocation sites in the code_info and global_data
structures.
compiler/unify_gen.m:
Add allocation sites to LLDS allocation instructions.
compiler/layout_out.m:
compiler/llds_out_file.m:
compiler/llds_out_instr.m:
Output MR_AllocSiteInfo arrays in generated C files.
Output code to register the MR_AllocSiteInfo array with the Mercury
runtime.
Output allocation site ids for memory allocation instructions.
compiler/llds_out_util.m:
Add allocation sites to llds_out_info.
compiler/pragma_c_gen.m:
compiler/ml_foreign_proc_gen.m:
Generate a macro MR_ALLOC_ID which resolves to an allocation site
structure, for every foreign_proc whose C code contains the string
"MR_ALLOC_ID". This is to be used by hand-written C code which
allocates memory.
MR_PROC_LABELs are retained for backwards compatibility. Though
they were introduced for profiling, they seem to have been co-opted
for printf-debugging since then.
compiler/ml_global_data.m:
Add allocation site structures to the MLDS global data.
compiler/mlds.m:
compiler/ml_unify_gen.m:
Add allocation site id to `new_object' instruction.
compiler/mlds_to_c.m:
Output allocation site arrays and allocation ids in high-level C code.
Output a call to register the allocation site array with the Mercury
runtime.
Delete an unused predicate.
compiler/exprn_aux.m:
compiler/jumpopt.m:
compiler/livemap.m:
compiler/mercury_compile_llds_back_end.m:
compiler/middle_rec.m:
compiler/ml_accurate_gc.m:
compiler/ml_elim_nested.m:
compiler/ml_optimize.m:
compiler/ml_util.m:
compiler/mlds_to_cs.m:
compiler/mlds_to_gcc.m:
compiler/mlds_to_il.m:
compiler/mlds_to_java.m:
compiler/mlds_to_managed.m:
compiler/opt_debug.m:
compiler/opt_util.m:
compiler/use_local_vars.m:
compiler/var_locn.m:
Conform to changes.
compiler/pickle.m:
compiler/prog_event.m:
compiler/timestamp.m:
Conform to changes in memory allocation macros.
library/benchmarking.m:
Add the `report_memory_attribution' instrumentation predicates.
Conform to changes to MR_memprof_record.
library/array.m:
library/bit_buffer.m:
library/bitmap.m:
library/construct.m:
library/deconstruct.m:
library/dir.m:
library/io.m:
library/mutvar.m:
library/store.m:
library/string.m:
library/thread.semaphore.m:
library/version_array.m:
Use attributed memory allocation throughout the standard library so
that objects don't show up in the memory profile as "unknown".
Replace MR_PROC_LABEL by MR_ALLOC_ID.
mdbcomp/program_representation.m:
mdbcomp/rtti_access.m:
Replace MR_PROC_LABEL by MR_ALLOC_ID.
profiler/Mercury.options:
profiler/globals.m:
profiler/mercury_profile.m:
profiler/options.m:
profiler/output.m:
profiler/snapshots.m:
Add a new mode to `mprof' to parse and present the data from
`Prof.Snapshots' files.
Add options for the new profiling mode.
profiler/process_file.m:
Fix a typo.
runtime/mercury_conf_param.h:
#define MR_MPROF_PROFILE_MEMORY_ATTRIBUTION if memory profiling
is enabled and we are using Boehm GC.
runtime/mercury.h:
Make MR_new_object take an allocation id argument.
Conform to changes in memory allocation macros.
runtime/mercury_memory.c:
runtime/mercury_memory.h:
runtime/mercury_types.h:
Define MR_AllocSiteInfo.
Add memory allocation functions and macros which take into the
account the additional word necessary for the new profiling mode.
These should be used in preferences to the raw memory allocation
functions wherever possible so that objects do not show up in the
profile as "unknown".
Add analogues of realloc/free which take into account the offset
introduced by the attribution word.
Add function versions of the MR_new_object macros, which can't be
written in standard C. They are only used when necessary.
Add built-in allocation site ids, to be used in the runtime and
other hand-written code when context-specific ids are unavailable.
runtime/mercury_heap.h:
Make MR_tag_offset_incr_hp_msg and MR_tag_offset_incr_hp_atomic_msg
allocate an extra word when memory attribution is desired, and store
the allocation id there.
Similarly for MR_create{1,2,3}_msg.
Replace proclabel arguments in allocation macros by alloc_id
arguments.
Replace MR_hp_alloc_atomic by MR_hp_alloc_atomic_msg. It was only
used for boxing floats.
Conform to change to MR_new_object macro.
runtime/mercury_bootstrap.h:
Delete obsolete macro hp_alloc_atomic.
runtime/mercury_heap_profile.c:
runtime/mercury_heap_profile.h:
Add the code to summarise the live objects on the Boehm GC heap and
writes out the data to `Prof.Snapshots', for display by mprof.
Don't store the procedure name in MR_memprof_record: the procedure
address is enough and faster to compare.
runtime/mercury_prof.c:
Finish and close the `Prof.Snapshots' file when the program
terminates.
Conform to changes in MR_memprof_record.
runtime/mercury_misc.h:
Add a macro to expand to the name of the allocation sites array
in LLDS grades.
runtime/mercury_bitmap.c:
runtime/mercury_bitmap.h:
Pass allocation id through bitmap allocation functions.
Delete unused function MR_string_to_bitmap.
runtime/mercury_string.h:
Add MR_make_aligned_string_copy_msg.
Make string allocation macros take allocation id arguments.
runtime/mercury.c:
runtime/mercury_array_macros.h:
runtime/mercury_context.c:
runtime/mercury_deconstruct.c:
runtime/mercury_deconstruct_macros.h:
runtime/mercury_dlist.c:
runtime/mercury_engine.c:
runtime/mercury_float.h:
runtime/mercury_hash_table.c:
runtime/mercury_ho_call.c:
runtime/mercury_label.c:
runtime/mercury_prof_mem.c:
runtime/mercury_stacks.c:
runtime/mercury_stm.c:
runtime/mercury_string.c:
runtime/mercury_thread.c:
runtime/mercury_trace_base.c:
runtime/mercury_trail.c:
runtime/mercury_type_desc.c:
runtime/mercury_type_info.c:
runtime/mercury_wsdeque.c:
Use attributed memory allocation throughout the runtime so that
objects don't show up in the profile as "unknown".
runtime/mercury_memory_zones.c:
Attribute memory zones to the Mercury runtime.
runtime/mercury_tabling.c:
runtime/mercury_tabling.h:
Use attributed memory allocation macros for tabling structures.
Delete unused MR_table_realloc_* and MR_table_copy_bytes macros.
runtime/mercury_deep_copy_body.h:
Try to retain the original attribution word when copying values.
runtime/mercury_ml_expand_body.h:
Conform to changes in memory allocation macros.
runtime/mercury_tags.h:
Replace proclabel arguments by alloc_id arguments in allocation macros.
runtime/mercury_wrapper.c:
If memory attribution is enabled, tell Boehm GC that pointers may be
displaced by an extra word.
trace/mercury_trace.c:
trace/mercury_trace_tables.c:
Conform to changes in memory allocation macros.
extras/net/tcp.m:
extras/solver_types/library/any_array.m:
extras/trailed_update/tr_array.m:
Conform to changes in memory allocation macros.
doc/user_guide.texi:
Document the new profiling mode.
doc/reference_manual.texi:
Update a commented out example.
|
||
|
Paul Bone
|
edc230406e |
Fix a number of errors and warnings in the runtime picked up by GCC 4.x in
parallel and threadscope grades.
We had been using types with the wrong signedness well calling atomic operations.
GCC 4.x also picked up an error where #elif was used instead of #else.
While testing these changes on a 32bit system more bugs where found on the i386
architecture and on AMD brand processors.
runtime/mercury_atomic_ops.h:
runtime/mercury_atomic_ops.c:
Add unsigned variants of the following atomic operations:
increment,
add,
add_and_fetch,
dec_and_is_zero,
Add a signed variant for compare and swap.
Rename the MR_atomic_dec_<type>_and_is_zero operation to move the type to
the end of the name.
Use volatile storage in the MR_Stats structure.
A 32bit machine cannot do atomic operations on 64bit values and MR_Stats
must use 64bit values. Therefore 64bit values in the MR_Stats structure
are now protected by a lock on 32bit machines.
runtime/mercury_atomic_ops.h:
Fix a typeo in the i386 version of MR_atomic_dec_and_is_zero_uint().
runtime/mercury_atomic_ops.c:
AMD CPUs do not conform to Intel's specification for being able to
extract the CPU clock speed from the brand string. When we cannot
determine the CPU's clock speed then we write out threadscope
timestamps in raw clock cycles rather than nanoseconds.
On i386 machines the ebx register is used to implement PIC code,
however the CPUID instruction uses it to output information. Save
this register on C's stack while we issue CPUID and retrieve the
result in ebx.
We now pass native machine sized values to the inline assembler code
that implements RDTSC and RDTSCP.
Fix commenting style in some places.
runtime/mercury_atomic_ops.c:
Fix some incorrect C preprocessor code for conditional compilation.
runtime/mercury_grade.h:
Increment binary compatibility number. This should have been done in a
prior change when the MR_runnext macro changed which broke binary
compatibility in the parallel low-level C grades.
runtime/mercury_context.h:
In MR_SyncTerm_Struct use an unsigned value for the number of conjuncts
remaining before the conjunction is complete.
runtime/mercury_threadscope.c:
Record raw cpu clock ticks rather than milliseconds when we don't
know the processor's clock speed.
runtime/mercury_context.c:
runtime/mercury_wsdeque.h:
runtime/mercury_wsdeque.c:
Conform to changes in mercury_atomic_ops.h
|
||
|
Paul Bone
|
6b2bc6a66a |
When an engine steals a spark and executes it using the context it is
currently holding it did not allocate a new context ID. A user looking at
this behaviour from threadscope would see thread 27 (for instance) finish, and
then immediately begin executing again. Therefore we now allocates a new
context ID when a context is reused making the context look different from
threadscope's point of view. New context IDs are already allocated to
contexts that are allocated from the free context lists.
runtime/mercury_context.c:
As above.
The next context id variable is now accessed atomically rather than being
protected by the free context list lock.
runtime/mercury_atomic_ops.h:
runtime/mercury_atomic_ops.c:
Implement a new atomic operation, MR_atomic_add_and_fetch_int, this is
used to allocate context ids.
Reimplement MR_atomic_add_int in terms of MR_atomic_add_and_fetch_int when
handwritten assembler support is not available.
runtime/mercury_atomic_ops.c:
Re-order atomic operations to match the order in the header file.
runtime/mercury_atomic_ops.h:
Place the definition of the MR_ATOMIC_PAUSE macro before the other atomic
operations since MR_atomic_add_and_fetch_int depends on it. This also
conforms with the coding standard.
runtime/mercury_threadscope.h:
Make the Context ID type a MR_Integer to match the argument size on the
available atomic operations.
|
||
|
Paul Bone
|
83a6f14708 |
Create a threadscope grade component.
Threadscope grades are enabled by using the grade component 'threadscope'.
They are supported only with low-lavel C parallel grades. Support for
threadscope in high level C grades is intended in the future but does not work
now.
runtime/mercury_conf_param.h:
Create the MR_THREADSCOPE macro that is defined if the grade is a
threadscope grade.
Define MR_PROFILE_FOR_PARALLEL_EXECUTION if MR_THREADSCOPE is defined.
Emit an error if MR_LL_PARALLEL_CONJ is defined before it is implied by
MR_THREADSAFE and ! MR_HIGHLEVEL_CODE
runtime/mercury_grade.h
Update the grade symbol for the threadscope grade component.
runtime/mercury_atomic_ops.c:
runtime/mercury_atomic_ops.h:
runtime/mercury_context.c:
runtime/mercury_context.h:
runtime/mercury_engine.c:
runtime/mercury_engine.h:
runtime/mercury_thread.c:
runtime/mercury_threadscope.c:
runtime/mercury_threadscope.h:
runtime/mercury_wrapper.c:
Now that MR_PROFILE_FOR_IMPLICIT_PARALLELISM is implied by MR_THREADSAFE we
don't need to test for MR_THREADSAFE when we test for
MR_PROFILE_FOR_IMPLICIT_PARALLELISM. The same is true for
MR_LL_PARALLEL_CONJ which is implied by MR_THREADSAFE &&
!MR_HIGHLEVEL_CODE.
Replace some occurances of MR_PROFILE_FOR_IMPLICIT_PARALLELISM with
MR_THREADSCOPE where the conditionally compiled code is used to support
threadscope profiling.
scripts/init_grade_options.sh-subr:
scripts/canonical_grade.sh-subr:
scripts/parse_grade_options.sh-subr:
scripts/final_grade_options.sh-subr:
scripts/mgnuc.in:
compiler/handle_options.m:
compiler/options.m:
compiler/compile_target_code.m:
configure.in:
Add support for the new grade component.
Pass -DMR_THREADSCOPE to the C compiler when using a threadscope grade.
Add assertions to ensure that the 'threadscope' grade component is used
only with the 'par' grade component.
doc/user_guide.texi:
Added commented-out documentation for the threadscope greate component.
Adjusted documentation of the --profile-parallel-execution runtime option
to describe the correct prerequisite compile time options.
Added my name to the authors list.
runtime/mercury_context.c:
Corrected grammar and prose in comments in the MR_do_join_and_continue code.
|
||
|
Paul Bone
|
5cfd73644a |
Implement work stealing.
This patch is heavily based on earlier, uncommitted work by Peter Wang. It
has been updated so that it applies against the current version of the source.
A number of other changes have been made. Peter's original ChangeLog
follows:
Implement work stealing for parallel conjunctions. This builds on an
older patch which introduced work-stealing deques to the runtime but
didn't perform work stealing.
Previously when we came across a parallel conjunct, we would place a spark
into either the _global spark queue_ or the _local spark stack_ of the
Mercury context. A spark on the global spark queue may be picked up for
parallel execution by an idle Mercury engine, whereas a spark on a local
spark stack is confined to execution in the context that originated it.
The problem is that we have to decide, ahead of time, where to put a
spark. Ideally, we should have just enough sparks in the global queue to
keep the available Mercury engines busy, and leave the rest of the sparks
to execute in their original contexts since that is more efficient. But
we can't predict the future so have to make do with guesses using simple
heuristics. A bad decision, once made, cannot be reversed. An engine may
sit idle due to an empty global spark queue, even while there are sparks
available in some local spark stacks.
In the work stealing scheme, sparks are always placed into each context's
_local spark deque_. Idle engines actively try to steal sparks from
random spark deques. We don't need to make irreversible and potentially
suboptimal decisions about where to put sparks. Making a spark available
for parallel execution is cheap and happens by default because of the
work-stealing deques; putting a spark on a global queue implies
synchronisation with other threads. The downside is that idle engines
need to expend more time and effort to find the work from multiple places
instead of just one place.
Practically, the new scheme seems to work as well as the old scheme and
vice versa, except that the old scheme often required
`--max-context-per-threads' to be set "correctly" to get good results.
Only tested on x86-64, which has a relatively constrained memory model.
My modifications include:
The difference between 'shared' and 'private' synchronisation terms has
been removed. All sync terms are assumed to be shared and thread-safe
operations are used everywhere. This allows us to remove complicated code
used when a private synchronisation term became shared. This may change
the performance of thread stealing, in particular it may become slower due
to the assumption that all sync terms are shared and therefore atomic
operations must always be used when decrementing their count field.
I've re-factored MR_do_join_and_continue, It is now much simpler as the
conditional code in it enumerates the possible cases clearly.
This change bootchecks and successfully runs the test suite in asm_fast.gc
asm_fast.gc.par hlc.gc and hlc.par, no other grades where tested. I have not
yet tested performance.
runtime/mercury_context.c:
runtime/mercury_context.h:
Keep pointers to all spark deques in a flat array, so we have access
to them for stealing.
Added functions to manage the global array of spark deques.
Modify MR_do_run_next, it now attempts to steal work from other context's
spark queues. Threads sleeping on the condition variable in
MR_do_run_next now use a timed wait so they can wakeup and try to steal
sparks.
Re-factored MR_do_join_and_continue.
MR_num_idle_engines is used by atomic operations, it has been made an
MR_Integer so that it's size matches the expectations of the atomic
operations we have defined.
Modified the MR_SyncTerm and MR_Spark structures. Sparks now point to
their sync terms. The perant stack pointer has been moved into the
SyncTerm structure. The MR_st_is_shared field in the MR_SyncTerm
structure has been removed.
runtime/mercury_atomic_ops.c:
runtime/mercury_atomic_ops.h:
Implement a new atomic operation: decrement integer and is zero. On the
x86/x86_64 one can't atomically decrement an integer and fetch the result
in a single instruction, a loop with a 'compare and exchange' instruction
is necessary. However since we only want to test if the value has become
zero after the decrement we can use the processor's flags. This can be
done in two instructions, but more importantly a loop is not required and
only one instruction is atomic.
runtime/mercury_wrapper.c:
runtime/mercury_wrapper.h:
Added runtime tunable options for work stealing. These control the number
of attempts an idle engine will make when looking for work, and the
duration to sleep after failing to find any work.
runtime/mercury_thread.c:
runtime/mercury_thread.h:
Added MR_COND_TIMED_WAIT, which waits on condition variables like
MR_COND_WAIT except that it may time out.
runtime/mercury_wsdeque.h:
runtime/mercury_wsdeque.c:
MR_wsdeque_pop_bottom now uses it's second argument to return the code
address to jump to rather the whole spark.
runtime/mercury_conf.h.in:
configure.in:
Test for sched_yield()
Change the synchronisation term structure.
doc/user_guide.texi:
Add commented out documentation for two new tunable parameters,
`--worksteal-max-attempts' and `--worksteal-sleep-msecs'.
Implementors may want to experiment with different values but end
users shouldn't need to know about them.
|
||
|
Paul Bone
|
92afa23af5 |
Support for threadscope profiling of the parallel runtime.
This change adds support for threadscope profiling of the parallel runtime in
low level C grades. It can be enabled by compiling _all_ code with the
MR_PROFILE_PARALLEL_EXECUTION_SUPPORT C macro defined. The runtime, libraries
and applications must all have this flag defined as it alters the MercuryEngine
and MR_Context structures.
See Don Jones Jr, Simon Marlow, Satnam Singh - Parallel Performance Tuning for
Haskell.
This change also includes:
Smarter thread pinning (the primordial thread is pinned to the thread that
it is currently running on).
The addition of callbacks from the Boehm GC to notify the runtime of
stop the world garbage collections.
Implement some userspace spin loops and conditions. These are cheaper than
their POSIX equivalents, do not support sleeping, and are signal handler
safe.
boehm_gc/alloc.h:
boehm_gc/alloc.c:
Declare and define the new callback functions.
boehm_gc/alloc.c:
Call the start and stop collect callbacks when we start and stop a
stop-the-world collection.
Correct how we record the time spent collecting, it now includes
collections that stop prematurely.
boehm_gc/pthread_stop_world.c:
Call the pause and resume thread callbacks in each thread where the GC
arranges for that thread to be stopped during a stop-the-world collection.
runtime/mercury_threadscope.c:
runtime/mercury_threadscope.h:
New files implementing the threadscope support.
runtime/mercury_atomic_ops.c:
runtime/mercury_atomic_ops.h:
Rename MR_configure_profiling_timers to MR_do_cpu_feature_detection.
Add a new function MR_read_cpu_tsc() to read the TSC register from the CPU,
this simply abstracts the static MR_rdtsc function.
runtime/mercury_atomic_ops.h:
Modify the C inline assembler to ensure we tell the C compiler that the
value in the register mapped to the 'old' parameter is also an output from
the instructions. That is, the C compiler must not depend on the value of
'old' being the same before and after the instruction is executed. This
has never been a problem in practice though.
Implement some cheap userspace mutual exclusion locks and condition
variables. These will be faster than pthread's mutexes when critical
sections are short and threads are pinned to separate CPUs.
runtime/mercury_context.c:
runtime/mercury_context.h:
Add a new function for pinning the primordial thread. If the OS supports
sched_getcpu we use it to determine which CPU the primordial thread should
use. No other thread will be pinned to this CPU.
Add a numeric id field to each context, this id is uniquely assigned and
identifies each context for threadscope.
MR_schedule_context posts the 'context runnable' threadscope event.
MR_do_runnext has been modified to destroy engines differently, it ensures
they cleanup properly so that their threadscope events are flushed properly
and then calls pthread_exit(0)
MR_do_runnext posts events for threadscope.
MR_do_join_and_continue posts events for threadscope.
runtime/mercury_engine.h:
Add new fields to the MercuryEngine structure including a buffer of
threadscope events, a clock offset (used to synchronize the TSC clocks) and
a unique identifier for the engine,
runtime/mercury_engine.c:
Call MR_threadscope_setup_engine() and MR_threadscope_finalize_engine for
newly created and about-to-be-destroyed engines.
When the main context finishes on a thread that's not the primordial thread
post a 'context is yielding' message before re-scheduling the context on
the primordial thread.
runtime/mercury_thread.c:
Added an XXX comment about a potential problem, it's only relevant for
programs using thread.spawn.
Added calls to the TSC synchronisation code used for threadscope profiling.
It appears that this is not necessary on modern x86 machines, it has been
commented out.
Post a threadscope event when we create a new context.
Don't call pthread_exit in MR_destroy_thread, we now do this in
MR_do_runnext so that we can unlock the runqueue mutex after cleaning up.
runtime/mercury_wrapper.c:
Conform to changes in mercury_atomic_ops.[ch]
Post an event immediately before calling main to mark the beginning of the
program in the threadscope profile.
Post a "context finished" event at the end of the program.
Wait until all engines have exited before cleaning up global data, this is
important for finishing writing the threadscope data file.
configure.in:
runtime/mercury_conf.h.in:
Test for the sched_getcpu C function and utmpx.h header file, these are
used for thread pinning.
runtime/Mmakefile:
Include the mercury_threadscope.[hc] files in the list of runtime headers
and sources respectively.
|
||
|
Paul Bone
|
3e75c9bd61 |
Print the CPU's clock speed in the profiling data for the parallelism runtime.
While the CPU's clock speed is not always the speed that the CPU operates at,
it is the number of TSC units per second when the CPU supports the constant TSC
feature (see /proc/cpuinfo / CPUID). This change set is used in preparation for
upcoming work on exporting ThreadScope profiling data, see "Parallel
Performance Tuning for Haskell" - Don Jones Jr, Simon Marlow and Satnam Singh.
runtime/mercury_atomic_ops.c:
Retrieve the CPU frequency from the CPUID instruction and store it in
MR_cpu_cycles_per_sec
runtime/mercury_atomic_ops.h:
Export MR_cpu_cycles_per_sec
runtime/mercury_context.c:
Print the number of cycles per second in the parallel profiling data file
if we where able to detect it.
|
||
|
Paul Bone
|
db9a526d6b |
Corrections in response to review comments on recent parallel runtime changes.
Made thread pinning off by default, the operating system should handle this
unless we have a good reason to.
configure.in:
Removed newly added declaration checking code.
doc/user_guide.texi:
Documentation corrections.
Adjusted the --pin-threads runtime option default.
runtime/mercury_atomic_ops.c
runtime/mercury_atomic_ops.h
Use __x86_64__ instead of __amd64__
Altered comments at the beginning of sections of the file to better
describe the contents of that section.
Placed comments at the end of long conditional compilation blocks that
match the condition at the beginning of the block.
runtime/mercury_conf.h.in:
Added editor hint for vim at the top of the file.
Remove newly added declarations section.
runtime/mercury_context.c:
Adjusted default behaviour of --pin-threads
Fixed some style issues.
runtime/mercury_context.h:
Fixed grammatical error.
runtime/mercury_wrapper.c:
Fixed grammatical error.
Fixed a missing break statement in a switch statement.
|
||
|
Paul Bone
|
4f1bfc2ebc |
Parallel runtime profiling improvements.
Improve the profiling of the parallel runtime code in two main ways: + Record data for more events. + Record high-precision timing data on x86 machines via the TSC where access to the TSC is available. Access to the TSC is available via two machine instructions. RDTSC - read TSC. and RDTSCP - read TSC and processor ID. We prefer the latter as a process migrated between two calls to RDTSC may cause an incorrect time duration to be calculated (since TSC counts are seldom synchronized). We fall back to RDTSC when RDTSCP is not available and gracefully record no timing information when neither is available. Availability is detected via the CPUID instruction, see MR_configure_profiling_timers(). runtime/mercury_context.c: runtime/mercury_context.h: Runtime profiling changes as above. runtime/mercury_atomic_ops.c: runtime/mercury_atomic_ops.h: Add runtime profiling timing code. Add new add and subtract atomic operations. runtime/mercury_wrapper.c: Call the new MR_configure_profiling_timers() procedure to detect the CPU and configure access to the TSC. Mmakefile: runtime/Mmakefile: 'mmake tags' at the top level now builds the tags file for the runtime directory. The tags target in the runtime directory is now marked as PHONY so it is generated even if it already exists. |
||
|
Paul Bone
|
4d41cf6c23 |
Rename the runtime granularity control macros, variables and predicates.
Estimated hours taken: 3 Branches: main Rename the runtime granularity control macros, variables and predicates. Names of the runtime granularity control macros, variables and predicates are now more descriptive and more consistent. An alternative runtime granularity control predicate and macro is now available, it considers the number of contexts and all sparks whereas the original predicate and macro considers only the number of contexts and sparks on the global queue. A new predicate has been added to determine the number of worker threads that the mercury runtime is configured to use. library/par_builtin.m: Renamed predicates. Conform to changes in runtime/mercury_thread.h Added the new predicates. Removed some old foreign procedure attributes. Addressed an XXX comment left by Zoltan. runtime/mercury_context.c: runtime/mercury_context.h: Rename existing runtime granularity control variables and macros. Add new runtime granularity control variable and macro. runtime/mercury_wrapper.c: runtime/mercury_wrapper.h: Export MR_num_threads variable. Make this variable an MR_Unsigned. runtime/mercury_atomic_ops.c: runtime/mercury_atomic_ops.h: Introduce new atomic increment and decrement instructions. These are used to count the number of local sparks created which is done outside of a critical section. library/Mmakefile: Rebuild the par_builtin module when either runtime/mercury_context.h or runtime/mercury_thread.h change. compiler/granularity.m: Conform to changes in runtime/mercury_context.h |
||
|
Julien Fischer
|
cd849f451a |
Avoid problems with some versions of gcc when compiling the runtime atomic_ops
Estimated hours taken: 0.2 Branches: main Avoid problems with some versions of gcc when compiling the runtime atomic_ops module at -O0 in non low-level .par grades. runtime/mercury_atomic_ops.[hc]: Implement MR_compare_and_swap_word using the builtin gcc compare and swap if it is available. (In particular prefer this to the handcoded assembler versions on x86 and x86_64.) Only include the code in these modules if MR_LL_PARALLEL_CONJ is defined. This helps to avoid problems with gcc in grades where atomic operations are not used. |
||
|
Peter Wang
|
cb8459d517 |
Make the parallel conjunction execution mechanism more efficient.
Branches: main Make the parallel conjunction execution mechanism more efficient. 1. Don't allocate sync terms on the heap. Sync terms are now allocated in the stack frame of the procedure call which originates a parallel conjunction. 2. Don't allocate individual sparks on the heap. Sparks are now stored in preallocated, growing arrays using an algorithm that doesn't use locks. 3. Don't have one mutex per sync term. Just use one mutex to protect concurrent accesses to all sync terms (it's is rarely needed anyway). This makes sync terms smaller and saves initialising a mutex for each parallel conjunction encountered. 4. We don't bother to acquire the global sync term lock if we know a parallel conjunction couldn't be executing in parallel. In a highly parallel program, the majority of parallel conjunctions will be executed sequentially so protecting the sync terms from concurrent accesses is unnecessary. par_fib(39) is ~8.4 times faster (user time) on my laptop (Linux 2.6, x86_64), which is ~3.5 as slow as sequential execution. configure.in: Update the configuration for a changed MR_SyncTerm structure. compiler/llds.m: Make the fork instruction take a second argument, which is the base stack slot of the sync term. Rename it to fork_new_child to match the macro name in the runtime. compiler/par_conj_gen.m: Change the generated code for parallel conjunctions to allocate sync terms on the stack and to pass the sync term to fork_new_child. compiler/dupelim.m: compiler/dupproc.m: compiler/exprn_aux.m: compiler/global_data.m: compiler/jumpopt.m: compiler/livemap.m: compiler/llds_out.m: compiler/llds_to_x86_64.m: compiler/middle_rec.m: compiler/opt_debug.m: compiler/opt_util.m: compiler/reassign.m: compiler/use_local_vars.m: Conform to the change in the fork instruction. compiler/liveness.m: compiler/proc_gen.m: Disable use of the parallel conjunction operator in the compiler as older versions of the compiler will generate code incompatible with the new runtime. runtime/mercury_context.c: runtime/mercury_context.h: Remove the next pointer field from MR_Spark as it's no longer needed. Remove the mutex from MR_SyncTerm. Add a field to record if a spark belonging to the sync term was scheduled globally, i.e. if the parallel conjunction might be executed in parallel. Define MR_SparkDeque and MR_SparkArray. Use MR_SparkDeques to hold per-context sparks and global sparks. Change the abstract machine instructions MR_init_sync_term, MR_fork_new_child, MR_join_and_continue as per the main change log. Use a preprocessor macro MR_LL_PARALLEL_CONJ as a shorthand for !MR_HIGHLEVEL_CODE && MR_THREAD_SAFE. Take the opportunity to clean things up a bit. runtime/mercury_wsdeque.c: runtime/mercury_wsdeque.h: New files containing an implementation of work-stealing deques. We don't do work stealing yet but we use the underlying data structure. runtime/mercury_atomic_ops.c: runtime/mercury_atomic_ops.h: New files to contain atomic operations. Currently it just contains compare-and-swap for gcc/x86_64, gcc/x86 and gcc-4.1. runtime/Mmakefile: Add the new files. runtime/mercury_engine.h: runtime/mercury_mm_own_stacks.c: runtime/mercury_wrapper.c: Conform to runtime changes. runtime/mercury_conf_param.h: Update an outdated comment. |