| Age | Commit message (Collapse) | Author |
|---|
|
This is sound because the liveness data structure only stores liveness
info at basic block boundaries, and the rewrites that happen in
TargetSpecific:check_and_rewrite/2 preserves all existing definitions
and uses, and all new liveness intervals, belonging to newly introduced
temporaries, are always local to a basic block, and thus do not show up
in the liveout or livein sets for the basic block.
|
|
ra_partitioned significantly speeds up register allocation of larger
functions without affecting allocation quality negatively. This is the
final change needed to make o1 suitable for compiling really large
functions without choking.
|
|
The division into an initial pass that may introduce temps, and
following passes that must not forces us to make the same heuristic
decision during each of these passes. Thus, the splitting heuristic
can't be based on the number of temporaries -- at least without
excluding temporaries above SpillLimit.
|
|
If temps introduced by hipe_regalloc_prepass end up above SpillLimit,
the register allocators will not spill them. This constraint is
unnecessarily limiting the allocators and might theoretically lead to
unallocatable programs (more temps above SpillLimit alive at a time than
there are physical registers).
|
|
|
|
These will not only be useful for hipe_regalloc_prepass, but also, after
the introduction of a mk_move/2 (or similar) callback, for the purpose
of range splitting.
Since the substitution needed to case over all the instructions, a new
module, hipe_ppc_subst, was introduced to the ppc backend.
|
|
These will not only be useful for hipe_regalloc_prepass, but also, after
the introduction of a mk_move/2 (or similar) callback, for the purpose
of range splitting.
Since the substitution needed to case over all the instructions, a new
module, hipe_sparc_subst, was introduced to the sparc backend.
|
|
These will not only be useful for hipe_regalloc_prepass, but also, after
the introduction of a mk_move/2 (or similar) callback, for the purpose
of range splitting.
Since the substitution needed to case over all the instructions, a new
module, hipe_arm_subst, was introduced to the arm backend.
|
|
These will not only be useful for hipe_regalloc_prepass, but also, after
the introduction of a mk_move/2 (or similar) callback, for the purpose
of range splitting.
Since the substitution needed to case over all the instructions, a new
module, hipe_x86_subst, was introduced to the x86 backend.
Due to differences in the 'jtab' field of a #jmp_switch{} between x86
and amd64, it regrettably needed to be duplicated to hipe_amd64_subst.
|
|
Now that all backends do register allocation on a CFG directly and
define the defun_to_cfg/1 callback as the identity function, it can be
removed.
|
|
As the just_as_good_as assertion was loosened with the `NowRegs >=
CheckRegs` check, it no longer verified that hipe_regalloc_prepass had
not incorrectly labeled a temp as unallocatable. We add that behaviour
back.
|
|
|
|
This is due to the improvements in hipe_temp_map, removing the need for
duplicated logic in the backends.
|
|
|
|
|
|
|
|
|
|
|
|
hipe_regalloc_prepass speeds up register allocation by spilling any temp
that is live over a call (which clobbers all register).
In order to detect these, a new function was added to the target
interface; defines_all_alloc/1, that takes an instruction and returns a
boolean.
|
|
* sverker/hipe-performance-o1/PR-1154:
hipe_sparc: Minimise CFG<->linear conversions
hipe_ppc: Minimise CFG<->linear conversions
hipe_arm: Minimise CFG<->linear conversions
hipe_x86: Use lea instead of move+add
hipe_arm: Improve peephole optimiser
hipe_arm: Be resilient to crappy RTL
hipe_ppc: Be resilient to crappy RTL
hipe_sparc: Be resilient to crappy RTL
hipe: Reuse liveness info for spillmin
hipe_x86: Minimise CFG<->linear conversions
hipe: Fix o0 and o1
hipe: Add o0 and o1 to tests
hipe_rtl_binary:get_word_integer/4: Handle imms
hipe_x86: Be resilient to crappy RTL
hipe_x86: LSRA for SSE2
|
|
|
|
* sverker/hipe-sparc-19/PR-1148:
Eliminate catch-all clause from two functions
Increase the time limit used by the test suite
|
|
|
|
* bjorn/eliminate-sys_pre_expand/OTP-13856:
Remove sys_pre_expand
xref: Eliminate use of sys_pre_expand
debugger: Eliminate use of sys_pre_expand
compiler: Eliminate use of sys_pre_expand
shell_SUITE: Eliminate references to sys_pre_expand in comments
qlc_pt: Simplify code because of updated erl_expand_records
Teach erl_expand_records to translate module-less calls
erl_internal: Add add_predefined_functions/1
cover: Remove always stale comment
|
|
* maint:
Fix documentation of an option of erl_tidy
|
|
* fenollp/syntax_tools/erl_tidy_option/PR-1151:
Fix documentation of an option of erl_tidy
|
|
* maint:
doc: Correct errors introduced by Editorial changes
|
|
* hasse/doc/fix_editorial_changes:
doc: Correct errors introduced by Editorial changes
|
|
* rickard/default-dirty-schedulers/OTP-13860:
Enable dirty scheduler support as default
|
|
|
|
* rickard/remove-old-purge-strategy/OTP-13844:
Avoid selective receive in code-purger process
Remove old purge strategy
Conflicts:
erts/emulator/beam/beam_bif_load.c
erts/emulator/beam/global.h
|
|
* maint:
Fix reclaim of literal areas
Conflicts:
erts/emulator/beam/global.h
|
|
* rickard/new-purge-strategy/OTP-13833:
Fix reclaim of literal areas
Conflicts:
erts/emulator/beam/beam_bif_load.c
|
|
The previous commits have made sys_pre_expand superfluous. Since
sys_pre_expand is undocumented and unsupported it can be removed in
a major release without prior deprecation.
Also remove code in erl_parse that handles abstract code that has
passed through sys_pre_expand.
We considered deprecating sys_pre_expand just in case, but decided
against it for the following reasons:
- Anyone brave and knowledgeable enough to use sys_pre_expand should
be able to cope with sys_pre_expand being removed.
- If we kept it, but didn't test it anywhere in OTP, it could
potentially stop working. So we would probably have to add some test
cases.
|
|
|
|
|
|
sys_pre_expand previously did a lot more work, for example,
translating records and funs, but now is merely a grab bag
of small transformations. Move those transformations to
v3_core.
|
|
Make it clear that we test that the shell is not dependent on
the compiler application being in the path.
|
|
erl_expand_records now adds "erlang:" to calls to BIFs, so we
can simplify the code.
|
|
As the next step in eliminating sys_pre_expand, teach
erl_expand_records to handle calls without explicit module name. If
such call refer to a BIF or imported function, add an explicit module
name.
That means that any subsequent pass will know that a call without
a module name is always to a local function defined in the module.
|
|
The sys_pre_expand module used to do a lot more (translate records and
funs, for example), but now it does very little. Most of the code is
an identify transformation of the abstract format.
The identity transformation part of the code must be maintained and
kept correct when new forms are added to the abstract format. That
adds to the maintance burden. It also adds (slightly) to compilation
times. Therefore, we want to eliminate sys_pre_expand, moving all of
its (non-identity) transformations to better places.
As a preliminary first step, move the code that adds the pre-defined
functions (such as module_info/0) to a new function in erl_internal.
|
|
The comment in the catch-all clause in munge_expr/2 will always be
stale. There is no way it can be kept up to date.
For example, 'eof' is listed, but 'eof' is a top-level form and is
actually handled in the catch-all clause for munge/4.
|
|
* maint:
beam_lib: Correct type for the abstract code chunk
|
|
|
|
The abstract_code chunk can also contain {eof,_}, {error,_}, and
{warning,_} in addition to the abstract forms that make up the Erlang
syntax proper.
|
|
Fix some older errors as well.
|
|
* maint:
observer: Fix error msg loop in table viewer
|
|
* dgud/observer/table-viewer-bug/ERL-237/OTP-13845:
observer: Fix error msg loop in table viewer
|
|
When observing mnesia table on remote node and connection to that node
is lost, a loop of error messages occur if mnesia is not running on
the observer node. ERL-237
|
|
* maint:
dialyzer: Increase time limit of suites
dialyzer: Remove a check that always fails
dialyzer: Optimize an opaque type case
|
