Age | Commit message (Collapse) | Author |
|
beam_bool would crash when attempting to optimize BEAM code similar
to this code:
bif '=:=' Reg1 SomeValue => y(0)
bif '=:=' Reg2 {atom,true} => x(2)
bif '=:=' Reg3 {atom,true} => x(3)
bif 'or' x(2) x(3) => x(2)
is_eq_exact Fail x(2) {atom,true}
The problem is that the first instruction that assigns a value to a Y
register. beam_bool:ssa_assign/2 will not accept a Y register
argument.
We could change ssa_assign/2 to accept a Y register, but that would
only cause the entire optimization to be rejected later because the Y
register is alive in the code that follows. Therefore, a better
solution is to modify extend_block/3 so that the instruction that
assign to Y registers are not added to the block. That is, the
optimizer will only operate on the following code:
bif '=:=' Reg2 {atom,true} => x(2)
bif '=:=' Reg3 {atom,true} => x(3)
bif 'or' x(2) x(3) => x(2)
is_eq_exact Fail x(2) {atom,true}
Usually the optimization will succeed, rewriting the four instructions
to a select_val instruction.
Assembly code such as the above can be produced by code similar to:
Y = Something == SomethingElse,
case Y of
Condition; OtherCondition ->
. . .
end,
. . .,
Y.
Reported-by: http://bugs.erlang.org/browse/ERL-143
Reported-by: José Valim
|
|
Rewrite code such as:
X = not_a_fun,
X()
to:
error({badfun,not_a_fun})
Also generate a warning.
|
|
'callback' and 'optional_callbacks' are no longer wild attributes.
|
|
* bjorn/compiler/core-erlang-fixes:
Slightly optimize core_pp
v3_core: Don't depend on sys_core_fold for cleaning up
|
|
Dialyzer relies heavily on the assumption that the type of a literal
that is used as a pattern is the type of any value that can match that
pattern. For maps, that is not true, and it was causing bad analysis
results. A new help function dialyzer_utils:refold_pattern/1 identifies
maps in literal patterns, and unfolds and labels them, allowing them to
be properly analysed.
|
|
|
|
v3_core would generate unsafe code for the following example:
f() ->
{ok={error,E}} = foo(),
E.
Internally, the code would look similar to:
f() ->
Var = foo(),
error({badmatch,Var}),
E.
That is, there would remain a reference to an unbound variable.
Normally, sys_core_fold would remove the reference to 'E', but if
if optimization was disabled the compiler would crash.
|
|
Use case in compile.erl is cryptographical so use
crypto:strong_rand_bytes/1 instead.
Use case in test suite is not cryptographical so use
other test instead.
|
|
* bjorn/compiler/remove-timestamps/OTP-13504:
Remove timestamps from BEAM files
|
|
As long as anyone can remember, the compilation time has been included
in BEAM files (and can be retrieved using Mod:module_info(compile)).
The timestamp has caused problems for anyone using tools such as 'cmp'
to compare BEAM files or for package managers:
http://erlang.org/pipermail/erlang-questions/2016-April/088717.html
Rarely has the timestamp been of any use. Yes, sometimes the timestamp
could help to figuring out which version of a module was used, but
nowadays a better way is to use Mod:module_info(md5).
To get rid of this problem, remove the timestamp from BEAM files in
OTP 19. Don't add an option to include timestamps.
Utilities that depend on the timestamp will need to be modified.
For example:
http://erlang.org/pipermail/erlang-questions/2016-April/088730.html
Instead of using the compilation time, the MD5 for the BEAM code can
be used. Example:
1> c:module_info(md5).
<<79,26,188,243,168,60,58,45,34,69,19,222,138,190,214,118>>
2> beam_lib:md5(code:which(c)).
{ok,{c,<<79,26,188,243,168,60,58,45,34,69,19,222,138,190,214,118>>}}
3>
|
|
|
|
This will speed up test cases that print all annotations.
|
|
|
|
If we pretty print to a file and read it back in, we expect to
get the same cerl data structures back.
|
|
Annotations would not be accepted for all constructs.
|
|
Make sure that we don't convert a map pattern to a map expression.
|
|
|
|
|
|
* bjorn/compiler/misc-opt:
v3_kernel: Construct literal lists properly
Use the register map in %live in beam_utils:is_killed_block/2
Teach beam_utils to check liveness for put_map instructions
beam_peep: Help out beam_jump
|
|
Use cerl:make_list/1 instead of a home-made make_list/1 to ensure that
literal lists are constructed as literals. In a future release, we
would like to forbid in the loader construction of literal lists using
instructions like:
put_list {atom,a} [] Dst
The proper way is:
move {literal,[a]} {x,0}
Also update the comment about "put_list Const [] Dst" in ops.tab.
|
|
In 1f0ae04d374, a complete register map was introduced in the %live
instructions thar are added by beam_utils:live_opt/1.
Use the register map to improve beam_utils:is_killed_block/2.
|
|
* henrik/update-copyrightyear:
update copyright-year
|
|
|
|
beam_jump fails to optimize the following:
jump 2
label 1
label 2
Since this situation is rare, instead of complicating beam_jump,
add the optimization to beam_peep. It will always succeed, since
adjacent labels have been coalesced.
|
|
Remove the unreachable instructions after a 'raise' instruction
(e.g. a 'jump' or 'deallocate', 'return') to decrease code size.
|
|
compile:forms/1,2 will crash when the current working directory has
been deleted. Fix that problem, and while we are at it, also stop
including {source,""} in module_info() when no source code file is
given.
Reported-at: http://bugs.erlang.org/browse/ERL-113
Reported-by: Adam Lindberg
|
|
Slightly speed up 'erlc' by pre-loading the modules used
by the compiler. Write a test case to ensure that the correct
set of modules are loaded.
|
|
|
|
Consider this code:
%% Start of block
get_tuple_element Tuple 0 Element
get_map_elements Fail Map [Key => Dest]
.
.
.
move Element UltimateDest
%% End of block
Fail:
%% Code that uses Element.
beam_block (more precisely, otp_tuple_element/1) would
incorrectly transform the code to this:
%% Start of block
get_map_elements Fail Map [Key => Dest]
.
.
.
get_tuple_element Tuple 0 UltimateDest
%% End of block
Fail:
%% Code that uses Element.
That is, the code at label Fail would use register Element,
which is either uninitalized or contains the wrong value.
We could fix this problem by always keeping label information
at hand when optimizing blocks so that we could check the code
at the failure label for get_map_elements. That would require
changes to beam_block and beam_utils. We might consider doing
that in the future if it turns out be worth it.
For now, I have decided that I want to keep the simplicity of blocks
(allowing them to be optimized without keeping label information).
That could be achieved by not including get_map_elements in
blocks. Another way, which I have chosen, is to only allow
get_map_elements as the first instruction in the block.
For background on the bug: c288ab8 introduced the beam_reorder pass
and 5f431276 introduced opt_tuple_element() in beam_block.
|
|
The expression in a bit string comprehension is limited to a
literal bit string expression. That is, the following code
is legal:
<< <<X>> || X <- List >>
but not this code:
<< foo(X) || X <- List >>
The limitation is annoying. For one thing, tools that transform
the abstract format must be careful not to produce code such as:
<< begin
%% Some instrumentation code.
<<X>>
end || X <- List >>
One reason for the limitation could be that we'll get
reduce/reduce conflicts if we try to allow an arbitrary
expression in a bit string comprehension:
binary_comprehension -> '<<' expr '||' lc_exprs '>>' :
{bc,?anno('$1'),'$2','$4'}.
Unfortunately, there does not seem to be an easy way to work
around that problem. The best we can do is to allow 'expr_max'
expressions (as in the binary syntax):
binary_comprehension -> '<<' expr_max '||' lc_exprs '>>' :
{bc,?anno('$1'),'$2','$4'}.
That will work, but functions calls must be enclosed in
parentheses:
<< (foo(X)) || X <- List >>
|
|
Binary matching can be confusing. For example:
1> <<-1>> = <<-1>>.
** exception error: no match of right hand side value <<"ÿ">>
2>
When constructing binaries, the value will be masked to fit in
the binary segment. But no such masking happens when matching
binaries.
One solution that we considered was to do the same masking when
matching. We have rejected that solution for several reasons:
* Masking in construction is highly controversial and by some
people considered a bad design decision.
* While masking of unsigned numbers can be understood, masking of
signed numbers it not easy to understand.
* Then there is the question of backward compatibility. Adding
masking to matching would mean that clauses that did not match
earlier would start to match. That means that code that has
never been tested will be executed. Code that has not been
tested will usually not work.
Therefore, we have decided to warn for binary patterns that cannot
possibly match.
While we are it, we will also warn for the following example where
size for a binary segment is invalid:
bad_size(Bin) ->
BadSize = bad_size,
<<42:BadSize>> = Bin.
That example would crash the HiPE compiler because the BEAM compiler
would generate a bs_get_integer2 instruction with an invalid size
field. We can avoid that crash if sys_core_fold not only warns for bad
binary pattern, but also removes the clauses that will not match.
Reported-by: http://bugs.erlang.org/browse/ERL-44
Reported-by: Kostis Sagonas
|
|
We will need them when we start to produce warnings for patterns
that can't match.
|
|
As a preparation for checking binary patterns we will add
var_list/2 that will work as pattern_list/2 but is guaranteed
not to throw an exception. That way, we will only have to use
try...catch for the few remaining calls to pattern_list/2.
|
|
Save work the *extremely* common case that the guard is
a literal.
|
|
|
|
* maint:
Eliminate crash because of unsafe delaying of sub-binary creation
|
|
The following code would fail to compile:
decode(<<Code/integer, Bin/binary>>) ->
<<C1/integer, B1/binary>> = Bin,
case C1 of
X when X =:= 1 orelse X =:= 2 ->
Bin2 = <<>>;
_ ->
Bin2 = B1
end,
case Code of
1 -> decode(Bin2);
_ -> Bin2
end.
The error message would be:
t: function decode/1+28:
Internal consistency check failed - please report this bug.
Instruction: return
Error: {match_context,{x,0}}:
The beam_bsm pass would delay the creation of a sub-binary when it was
unsafe to do so. The culprit was the btb_follow_branch/3 function that
for performance reasons cached labels that had already been checked.
The problem was the safety of a label also depends on the contents
of the registers. Therefore, the key for caching needs to be both
the label and the register contents.
Reported-by: José Valim
|
|
Internally in the v3_core pass, an #imatch{} record represents
a match expression:
Pattern = Expression
If Pattern is a single, unbound variable, #imatch{} will be
rewritten to #iset{}; otherwise it will be rewritten to #icase{}.
To determine how #imatch{} should be translated, the pattern is
processed using upattern/3. The return value from upattern/3 is thrown
away (after having been used for determing how the #imatch{} record
should be translated).
That means that every pattern in an #imatch{} is processed twice,
which is wasteful.
We can easily avoid the double processing of patterns by
introducing a new helper function that determines whether the
pattern is a new variable.
|
|
When manipulating Core Erlang trees it may be useful to perform some
operation when a node is visited, before inspecting children nodes. The
definition of cerl_tree:mapfold/3 does not allow that, as it applies the
given function only after all the recursive calls on the children nodes
have been completed.
This patch adds a new argument to mapfold: a function that is applied when
a node is first entered.
As an example of its use, consider the case where one wants to move a
'call' node earlier, by adding 'let' node and replacing the 'call' node
with the defined variable. The name of that variable must be specified
before one traverses the inner tree (especially if such replacements can be
nested).
|
|
|
|
|
|
Literal maps could cause dialyzer to crash when pretty printing the results.
Reported-by: Chris McGrath <[email protected]>
|
|
* maint:
Fix crash when attempting to update a fun as if it were a map
|
|
The following example would cause an internal consistency
failure in the compiler:
f() -> ok.
update() -> (fun f/0)#{u => 42}.
The reason is that internally, v3_core will (incorrectly)
rewrite update/0 to code similar to this:
update() ->
if
is_map(fun f/0) ->
maps:update(u, 42, fun f/0)
end.
Since funs are not allowed to be created in guards, incorrect and
unsafe code would be generated.
It is easy to fix the bug. There already is a is_valid_map_src/1
function in v3_core that tests whether the argument for the map update
operation can possibly be a valid map. A fun is represented as a
variable with a special name in Core Erlang, so it would not be
recognized as unsafe. All we'll need to do to fix the bug is to look
closer at variables to ensure they don't represent funs. That will
ensure that the code is rewritten in the correct way:
update() ->
error({badmap,fun f/0})
end.
Reported-by: Thomas Arts
|
|
* maint:
Eliminate crash in v3_codegen
|
|
The following code would crash v3_codegen:
order(From) ->
catch
if
From#{[] => sufficient} ->
saint
end.
Before explaining the crash, first some background on the stack
frame and the Y registers.
Certain instructions, most notably the 'call' instructions, clobber
all X registers. Before any such instruction, all X registers that
have values that will be used after the call must be saved to Y
registers (i.e. to the stack frame). adjust_stack/4 will be called
when X registers must be saved.
There is also another situation when X registers must be saved, namely
within a 'catch' if we are about to execute any instruction that may
cause an exception. Examples of such instructions are some guard BIFs
(such as length/1) and construction of binaries or maps. Within a
'catch', X registers must be be saved because if an exception is
thrown and catched all X registers will be destroyed. The same
adjust_stack/4 function will be called for those instructions, but
only if they occur within a 'catch'.
There is actually one more complication. If there is code in
a guard within a catch, the X registers should not be saved, because
the code in a guard never clobbers any X registers that were alive
before the guard code was entered. v3_codegen is written with the
implicit assumption that code in guards never cause anything
to be saved to Y registers.
The code for building maps and binaries would incorrectly save X
registers within a guard inside a 'catch'.
For construction of binaries, that would mean that a useless but
harmelss 'move' instruction was generated.
But for construction of maps, the saving of the Y register would not
be harmless. There would be a crash when attempting to merge #sr{}
records. #sr{} records keeps track of the contents of X and Y
registers. When two separate code paths are joined (e.g. at the end of
'case' statement), the register descriptors must be reconciled.
Basically, the register descriptors for both paths must be identical.
The #sr{} record for one path must not claim that {y,0} contains
a certain value, while another path claims that {y,0} is dead.
Thus, the crash occurs in sr_merge/2 when failing to reconcile the
Y registers.
To fix this bug this bug we will introduce a new function called
maybe_adjust_stack/5. It will save X registers on the stack only
if the code is inside a catch but not inside a guard. We will
change all existing code to use this new function when appropriate.
Reported-by: Thomas Arts
|
|
* maint:
beam_bool: Fix unsafe optimization
|
|
beam_bool would make the following code unsafe (which would be
reported by beam_validator):
scotland(Echo) ->
found(case Echo of
Echo when true; Echo, Echo, Echo ->
Echo;
echo ->
[]
end,
Echo = placed).
found(_, _) -> million.
Basically, beam_bool would see that the 'case' would always return
the value of Echo. Thus:
scotland(Echo) ->
found(Echo, Echo = placed).
The only problem is that beam_bool would also remove a 'move'
instruction that would save Echo to the stack. Here is the
assembly code for part of the function:
{allocate_zero,1,1}.
{move,{x,0},{y,0}}. %% Save Echo on stack.
{bif,'=:=',{f,7},[{x,0},{atom,true}],{x,1}}.
{bif,'=:=',{f,7},[{x,0},{atom,true}],{x,2}}.
{bif,'=:=',{f,7},[{x,0},{atom,true}],{x,3}}.
{bif,'and',{f,7},[{x,2},{x,3}],{x,2}}.
{bif,'and',{f,7},[{x,1},{x,2}],{x,1}}.
{jump,{f,8}}.
{label,7}.
{move,{atom,false},{x,1}}.
{label,8}.
{bif,'or',{f,6},[{atom,true},{x,1}],{x,1}}.
{test,is_eq_exact,{f,6},[{x,1},{atom,true}]}. %% Jump never taken.
{jump,{f,5}}.
{label,6}.
{test,is_eq_exact,{f,9},[{x,0},{atom,echo}]}.
{move,nil,{x,0}}.
{jump,{f,5}}.
{label,9}.
{test_heap,3,0}.
{put_tuple,2,{x,0}}.
{put,{atom,case_clause}}.
{put,{y,0}}.
{line,[{location,"t.erl",5}]}.
{call_ext,1,{extfunc,erlang,error,1}}.
{jump,{f,5}}.
{label,5}.
{test,is_eq_exact,{f,12},[{atom,placed},{y,0}]}.
beam_bool would see that the is_eq_exact test at label 8 would
always succeed. It could therefore remove most of the code before
the jump to label 5. Unfortunately it also removed the essential
move of Echo to the stack:
{allocate_zero,1,1}.
%% Instruction incorrectly removed: {move,{x,0},{y,0}}.
{jump,{f,5}}.
{label,5}.
{test,is_eq_exact,{f,12},[{atom,placed},{y,0}]}.
The root cause of the problem is that the 'move' instruction is
included in the block of 'bif' instructions before label 8.
Normally the 'move' instruction would not have been discarded,
but because the left operand to the 'or' BIF is 'true', the
entire block with 'bif' instructions are dropped.
As far as I can see, there is no gain by including 'move'
instructions in the first place. There is no way that better
code will be produced. In fact, the entire optimization can
be given up if 'move' instructions are found in the block.
Thus we can fix this bug by never including any 'move' instructions
in the block of 'bif' instructions. We can also remove all the
code that deals with 'move' instructions within blocks.
Reported-by: Thomas Arts
|
|
In most cases, we don't have to seed the random number generator,
as the rand:uniform/1 takes care about that itself.
|
|
The 'random' module is used to pad the end of a block with random
bytes. The appropriate function to use in this case
crypto:rand_bytes/1.
|