gh-mirror/mikepaul-LuaJIT
1
0
Fork 0
mirror of https://github.com/LuaJIT/LuaJIT.git synced 2025-02-07 23:24:09 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge branch 'v2.1' into luajit-s390x

Browse Source
This commit is contained in:
Aditya Bisht 2022-08-29 17:19:27 +05:30 committed by GitHub
commit 3c878599b1
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
29 changed files with 162 additions and 113 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
Makefile
View File

@ -33,7 +33,8 @@ DPREFIX= $(DESTDIR)$(PREFIX)
INSTALL_BIN= $(DPREFIX)/bin
INSTALL_LIB= $(DPREFIX)/$(MULTILIB)
INSTALL_SHARE= $(DPREFIX)/share
INSTALL_INC= $(DPREFIX)/include/luajit-$(MAJVER).$(MINVER)
INSTALL_DEFINC= $(DPREFIX)/include/luajit-$(MAJVER).$(MINVER)
INSTALL_INC= $(INSTALL_DEFINC)
INSTALL_LJLIBD= $(INSTALL_SHARE)/luajit-$(VERSION)
INSTALL_JITLIB= $(INSTALL_LJLIBD)/jit
@ -78,6 +79,9 @@ UNINSTALL= $(RM)
LDCONFIG= ldconfig -n 2>/dev/null
SED_PC= sed -e "s|^prefix=.*|prefix=$(PREFIX)|" \
-e "s|^multilib=.*|multilib=$(MULTILIB)|"
ifneq ($(INSTALL_DEFINC),$(INSTALL_INC))
SED_PC+= -e "s|^includedir=.*|includedir=$(INSTALL_INC)|"
endif
FILE_T= luajit
FILE_A= libluajit.a

14
doc/ext_buffer.html
View File

@ -86,7 +86,7 @@ overhead. In conjunction with the FFI library, they allow zero-copy
operations.
</p>
<p>
The string buffer libary also includes a high-performance
The string buffer library also includes a high-performance
<a href="serialize">serializer</a> for Lua objects.
</p>
@ -434,8 +434,8 @@ unsupported object types, circular references or deeply nested tables.
<h3 id="buffer_decode"><tt>obj = buffer.decode(str)<br>
obj = buf:decode()</tt></h3>
<p>
The stand-alone function de-serializes (decodes) the string
<tt>str</tt>, the buffer method de-serializes one object from the
The stand-alone function deserializes (decodes) the string
<tt>str</tt>, the buffer method deserializes one object from the
buffer. Both return a Lua object <tt>obj</tt>.
</p>
<p>
@ -449,7 +449,7 @@ data after decoding a single top-level object. The buffer method leaves
any left-over data in the buffer.
</p>
<p>
Attempting to de-serialize an FFI type will throw an error, if the FFI
Attempting to deserialize an FFI type will throw an error, if the FFI
library is not built-in or has not been loaded, yet.
</p>
@ -462,7 +462,7 @@ the following members (all optional):
<li>
<tt>dict</tt> is a Lua table holding a <b>dictionary of strings</b> that
commonly occur as table keys of objects you are serializing. These keys
are compactly encoded as indexes during serialization. A well chosen
are compactly encoded as indexes during serialization. A well-chosen
dictionary saves space and improves serialization performance.
</li>
<li>
@ -473,7 +473,7 @@ for the table objects you are serializing.
<p>
<tt>dict</tt> needs to be an array of strings and <tt>metatable</tt> needs
to be an array of tables. Both starting at index 1 and without holes (no
<tt>nil</tt> inbetween). The tables are anchored in the buffer object and
<tt>nil</tt> in between). The tables are anchored in the buffer object and
internally modified into a two-way index (don't do this yourself, just pass
a plain array). The tables must not be modified after they have been passed
to <tt>buffer.new()</tt>.
@ -624,7 +624,7 @@ errors are best caught with an outer wrapper for larger parts of code.
There's not much one can do after that, anyway.
</p>
<p>
OTOH you may want to catch some errors individually. Buffer methods need
OTOH, you may want to catch some errors individually. Buffer methods need
to receive the buffer object as the first argument. The Lua colon-syntax
<tt>obj:method()</tt> does that implicitly. But to wrap a method with
<tt>pcall()</tt>, the arguments need to be passed like this:

4
doc/ext_c_api.html
View File

@ -107,7 +107,7 @@ Turn the whole JIT compiler on or off or flush the whole cache of compiled code.
This sets the mode for the function at the stack index <tt>idx</tt> or
the parent of the calling function (<tt>idx = 0</tt>). It either
enables JIT compilation for a function, disables it and flushes any
already compiled code or only flushes already compiled code. This
already compiled code, or only flushes already compiled code. This
applies recursively to all sub-functions of the function with
<tt>LUAJIT_MODE_ALLFUNC</tt> or only to the sub-functions with
<tt>LUAJIT_MODE_ALLSUBFUNC</tt>.
@ -126,7 +126,7 @@ traces which link to it.
This mode defines a wrapper function for calls to C functions. If
called with <tt>LUAJIT_MODE_ON</tt>, the stack index at <tt>idx</tt>
must be a <tt>lightuserdata</tt> object holding a pointer to the wrapper
function. From now on all C functions are called through the wrapper
function. From now on, all C functions are called through the wrapper
function. If called with <tt>LUAJIT_MODE_OFF</tt> this mode is turned
off and all C functions are directly called.
</p>

14
doc/ext_ffi.html
View File

@ -157,7 +157,7 @@ call the binding function. Phew!
<h2 id="cdata">Motivating Example: Using C Data Structures</h2>
<p>
The FFI library allows you to create and access C&nbsp;data
structures. Of course the main use for this is for interfacing with
structures. Of course, the main use for this is for interfacing with
C&nbsp;functions. But they can be used stand-alone, too.
</p>
<p>
@ -169,7 +169,7 @@ implemented with a big table holding lots of tiny tables. This imposes
both a substantial memory overhead as well as a performance overhead.
</p>
<p>
Here's a sketch of a library that operates on color images plus a
Here's a sketch of a library that operates on color images, plus a
simple benchmark. First, the plain Lua version:
</p>
<pre class="code">
@ -184,7 +184,7 @@ local function image_ramp_green(n)
return img
end
local function image_to_grey(img, n)
local function image_to_gray(img, n)
for i=1,n do
local y = floor(0.3*img[i].red + 0.59*img[i].green + 0.11*img[i].blue)
img[i].red = y; img[i].green = y; img[i].blue = y
@ -194,14 +194,14 @@ end
local N = 400*400
local img = image_ramp_green(N)
for i=1,1000 do
image_to_grey(img, N)
image_to_gray(img, N)
end
</pre>
<p>
This creates a table with 160.000 pixels, each of which is a table
holding four number values in the range of 0-255. First an image with
holding four number values in the range of 0-255. First, an image with
a green ramp is created (1D for simplicity), then the image is
converted to greyscale 1000 times. Yes, that's silly, but I was in
converted to grayscale 1000 times. Yes, that's silly, but I was in
need of a simple example ...
</p>
<p>
@ -308,7 +308,7 @@ be more compact and faster. This is certainly true (by a factor of
~1.7x). Switching to a struct-of-arrays would help, too.
</p>
<p style="font-size: 8pt;">
However the resulting code would be less idiomatic and rather
However, the resulting code would be less idiomatic and rather
error-prone. And it still doesn't get even close to the performance of
the FFI version of the code. Also, high-level data structures cannot
be easily passed to other C&nbsp;functions, especially I/O functions,

10
doc/ext_ffi_api.html
View File

@ -121,7 +121,7 @@ separated by semicolons. The trailing semicolon for a single
declaration may be omitted.
</p>
<p>
Please note that external symbols are only <em>declared</em>, but they
Please note, that external symbols are only <em>declared</em>, but they
are <em>not bound</em> to any specific address, yet. Binding is
achieved with C&nbsp;library namespaces (see below).
</p>
@ -209,7 +209,7 @@ parse the cdecl only once and get its ctype with
<tt>ffi.typeof()</tt>. Then use the ctype as a constructor repeatedly.
</p>
<p style="font-size: 8pt;">
Please note that an anonymous <tt>struct</tt> declaration implicitly
Please note, that an anonymous <tt>struct</tt> declaration implicitly
creates a new and distinguished ctype every time you use it for
<tt>ffi.new()</tt>. This is probably <b>not</b> what you want,
especially if you create more than one cdata object. Different anonymous
@ -256,12 +256,12 @@ afterwards. Neither the contents of the <tt>metatable</tt> nor the
contents of an <tt>__index</tt> table (if any) may be modified
afterwards. The associated metatable automatically applies to all uses
of this type, no matter how the objects are created or where they
originate from. Note that pre-defined operations on types have
originate from. Note that predefined operations on types have
precedence (e.g. declared field names cannot be overridden).
</p>
<p>
All standard Lua metamethods are implemented. These are called directly,
without shortcuts and on any mix of types. For binary operations, the
without shortcuts, and on any mix of types. For binary operations, the
left operand is checked first for a valid ctype metamethod. The
<tt>__gc</tt> metamethod only applies to <tt>struct</tt>/<tt>union</tt>
types and performs an implicit <a href="#ffi_gc"><tt>ffi.gc()</tt></a>
@ -492,7 +492,7 @@ have some extra methods:
<p>
Free the resources associated with a callback. The associated Lua
function is unanchored and may be garbage collected. The callback
function pointer is no longer valid and must not be called anymore
function pointer is no longer valid and must not be called again
(it may be reused by a subsequently created callback).
</p>

58
doc/ext_ffi_semantics.html
View File

@ -88,7 +88,7 @@ footprint. It's used by the <a href="ext_ffi_api.html">ffi.* library
functions</a> to declare C&nbsp;types or external symbols.
</p>
<p>
It's only purpose is to parse C&nbsp;declarations, as found e.g. in
Its only purpose is to parse C&nbsp;declarations, as found e.g. in
C&nbsp;header files. Although it does evaluate constant expressions,
it's <em>not</em> a C&nbsp;compiler. The body of <tt>inline</tt>
C&nbsp;function definitions is simply ignored.
@ -165,7 +165,7 @@ function declarations.</li>
</ul>
<p>
The following C&nbsp;types are pre-defined by the C&nbsp;parser (like
The following C&nbsp;types are predefined by the C&nbsp;parser (like
a <tt>typedef</tt>, except re-declarations will be ignored):
</p>
<ul>
@ -583,9 +583,9 @@ ffi.new("struct nested", {x=1,y={2,3}}) --> x = 1, y.a = 2, y.b = 3
<h2 id="cdata_ops">Operations on cdata Objects</h2>
<p>
All of the standard Lua operators can be applied to cdata objects or a
All standard Lua operators can be applied to cdata objects or a
mix of a cdata object and another Lua object. The following list shows
the pre-defined operations.
the predefined operations.
</p>
<p>
Reference types are dereferenced <em>before</em> performing each of
@ -593,7 +593,7 @@ the operations below &mdash; the operation is applied to the
C&nbsp;type pointed to by the reference.
</p>
<p>
The pre-defined operations are always tried first before deferring to a
The predefined operations are always tried first before deferring to a
metamethod or index table (if any) for the corresponding ctype (except
for <tt>__new</tt>). An error is raised if the metamethod lookup or
index table lookup fails.
@ -643,7 +643,7 @@ assigning to an index of a vector raises an error.</li>
</ul>
<p>
A ctype object can be indexed with a string key, too. The only
pre-defined operation is reading scoped constants of
predefined operation is reading scoped constants of
<tt>struct</tt>/<tt>union</tt> types. All other accesses defer
to the corresponding metamethods or index tables (if any).
</p>
@ -656,7 +656,7 @@ certain optimizations.
<p>
As a consequence, the <em>elements</em> of complex numbers and
vectors are immutable. But the elements of an aggregate holding these
types <em>may</em> be modified of course. I.e. you cannot assign to
types <em>may</em> be modified, of course. I.e. you cannot assign to
<tt>foo.c.im</tt>, but you can assign a (newly created) complex number
to <tt>foo.c</tt>.
</p>
@ -675,8 +675,8 @@ through unions is explicitly detected and allowed.
to <tt>ffi.new(ct, ...)</tt>, unless a <tt>__new</tt> metamethod is
defined. The <tt>__new</tt> metamethod is called with the ctype object
plus any other arguments passed to the constructor. Note that you have to
use <tt>ffi.new</tt> inside of it, since calling <tt>ct(...)</tt> would
cause infinite recursion.</li>
use <tt>ffi.new</tt> inside the metamethod, since calling <tt>ct(...)</tt>
would cause infinite recursion.</li>
<li><b>C&nbsp;function call</b>: a cdata function or cdata function
pointer can be called. The passed arguments are
@ -687,7 +687,7 @@ variable argument part of vararg C&nbsp;function use
C&nbsp;function is called and the return value (if any) is
<a href="#convert_tolua">converted to a Lua object</a>.<br>
On Windows/x86 systems, <tt>__stdcall</tt> functions are automatically
detected and a function declared as <tt>__cdecl</tt> (the default) is
detected, and a function declared as <tt>__cdecl</tt> (the default) is
silently fixed up after the first call.</li>
</ul>
@ -697,7 +697,7 @@ silently fixed up after the first call.</li>
<li><b>Pointer arithmetic</b>: a cdata pointer/array and a cdata
number or a Lua number can be added or subtracted. The number must be
on the right hand side for a subtraction. The result is a pointer of
on the right-hand side for a subtraction. The result is a pointer of
the same type with an address plus or minus the number value
multiplied by the element size in bytes. An error is raised if the
element size is undefined.</li>
@ -712,7 +712,7 @@ operators (<tt>+&nbsp;-&nbsp;*&nbsp;/&nbsp;%&nbsp;^</tt> and unary
minus) can be applied to two cdata numbers, or a cdata number and a
Lua number. If one of them is an <tt>uint64_t</tt>, the other side is
converted to an <tt>uint64_t</tt> and an unsigned arithmetic operation
is performed. Otherwise both sides are converted to an
is performed. Otherwise, both sides are converted to an
<tt>int64_t</tt> and a signed arithmetic operation is performed. The
result is a boxed 64&nbsp;bit cdata object.<br>
@ -759,7 +759,7 @@ which is compatible with any other pointer type.</li>
<li><b>64&nbsp;bit integer comparison</b>: two cdata numbers, or a
cdata number and a Lua number can be compared with each other. If one
of them is an <tt>uint64_t</tt>, the other side is converted to an
<tt>uint64_t</tt> and an unsigned comparison is performed. Otherwise
<tt>uint64_t</tt> and an unsigned comparison is performed. Otherwise,
both sides are converted to an <tt>int64_t</tt> and a signed
comparison is performed.<br>
@ -784,9 +784,9 @@ keys!</b>
A cdata object is treated like any other garbage-collected object and
is hashed and compared by its address for table indexing. Since
there's no interning for cdata value types, the same value may be
boxed in different cdata objects with different addresses. Thus
boxed in different cdata objects with different addresses. Thus,
<tt>t[1LL+1LL]</tt> and <tt>t[2LL]</tt> usually <b>do not</b> point to
the same hash slot and they certainly <b>do not</b> point to the same
the same hash slot, and they certainly <b>do not</b> point to the same
hash slot as <tt>t[2]</tt>.
</p>
<p>
@ -808,7 +808,7 @@ the resulting Lua number as a key when indexing tables.<br>
One obvious benefit: <tt>t[tonumber(2LL)]</tt> <b>does</b> point to
the same slot as <tt>t[2]</tt>.</li>
<li>Otherwise use either <tt>tostring()</tt> on 64&nbsp;bit integers
<li>Otherwise, use either <tt>tostring()</tt> on 64&nbsp;bit integers
or complex numbers or combine multiple fields of a cdata aggregate to
a Lua string (e.g. with
<a href="ext_ffi_api.html#ffi_string"><tt>ffi.string()</tt></a>). Then
@ -816,7 +816,7 @@ use the resulting Lua string as a key when indexing tables.</li>
<li>Create your own specialized hash table implementation using the
C&nbsp;types provided by the FFI library, just like you would in
C&nbsp;code. Ultimately this may give much better performance than the
C&nbsp;code. Ultimately, this may give much better performance than the
other alternatives or what a generic by-value hash table could
possibly provide.</li>
@ -882,7 +882,7 @@ garbage collector will automatically free the memory used by it (at
the end of the next GC cycle).
</p>
<p>
Please note that pointers themselves are cdata objects, however they
Please note, that pointers themselves are cdata objects, however they
are <b>not</b> followed by the garbage collector. So e.g. if you
assign a cdata array to a pointer, you must keep the cdata object
holding the array alive as long as the pointer is still in use:
@ -931,18 +931,18 @@ of the function pointer and the Lua function object (closure).
</p>
<p>
This can happen implicitly due to the usual conversions, e.g. when
passing a Lua function to a function pointer argument. Or you can use
passing a Lua function to a function pointer argument. Or, you can use
<tt>ffi.cast()</tt> to explicitly cast a Lua function to a
C&nbsp;function pointer.
</p>
<p>
Currently only certain C&nbsp;function types can be used as callback
Currently, only certain C&nbsp;function types can be used as callback
functions. Neither C&nbsp;vararg functions nor functions with
pass-by-value aggregate argument or result types are supported. There
are no restrictions for the kind of Lua functions that can be called
are no restrictions on the kind of Lua functions that can be called
from the callback &mdash; no checks for the proper number of arguments
are made. The return value of the Lua function will be converted to the
result type and an error will be thrown for invalid conversions.
result type, and an error will be thrown for invalid conversions.
</p>
<p>
It's allowed to throw errors across a callback invocation, but it's not
@ -1003,7 +1003,7 @@ convention cannot be automatically detected, unlike for
<tt>__stdcall</tt> calls <em>to</em> Windows functions.
</p>
<p>
For some use cases it's necessary to free up the resources or to
For some use cases, it's necessary to free up the resources or to
dynamically redirect callbacks. Use an explicit cast to a
C&nbsp;function pointer and keep the resulting cdata object. Then use
the <a href="ext_ffi_api.html#callback_free"><tt>cb:free()</tt></a>
@ -1056,7 +1056,7 @@ GUI application, which waits for user input most of the time, anyway.
</p>
<p>
For new designs <b>avoid push-style APIs</b>: a C&nbsp;function repeatedly
calling a callback for each result. Instead <b>use pull-style APIs</b>:
calling a callback for each result. Instead, <b>use pull-style APIs</b>:
call a C&nbsp;function repeatedly to get a new result. Calls from Lua
to C via the FFI are much faster than the other way round. Most well-designed
libraries already use pull-style APIs (read/write, get/put).
@ -1075,7 +1075,7 @@ function.
</p>
<p>
Indexing a C&nbsp;library namespace object with a symbol name (a Lua
string) automatically binds it to the library. First the symbol type
string) automatically binds it to the library. First, the symbol type
is resolved &mdash; it must have been declared with
<a href="ext_ffi_api.html#ffi_cdef"><tt>ffi.cdef</tt></a>. Then the
symbol address is resolved by searching for the symbol name in the
@ -1130,7 +1130,7 @@ Performance notice: the JIT compiler specializes to the identity of
namespace objects and to the strings used to index it. This
effectively turns function cdata objects into constants. It's not
useful and actually counter-productive to explicitly cache these
function objects, e.g. <tt>local strlen = ffi.C.strlen</tt>. OTOH it
function objects, e.g. <tt>local strlen = ffi.C.strlen</tt>. OTOH, it
<em>is</em> useful to cache the namespace itself, e.g. <tt>local C =
ffi.C</tt>.
</p>
@ -1155,14 +1155,14 @@ This behavior is inevitable, since the goal is to provide full
interoperability with C&nbsp;code. Adding extra safety measures, like
bounds checks, would be futile. There's no way to detect
misdeclarations of C&nbsp;functions, since shared libraries only
provide symbol names, but no type information. Likewise there's no way
provide symbol names, but no type information. Likewise, there's no way
to infer the valid range of indexes for a returned pointer.
</p>
<p>
Again: the FFI library is a low-level library. This implies it needs
to be used with care, but it's flexibility and performance often
outweigh this concern. If you're a C or C++ developer, it'll be easy
to apply your existing knowledge. OTOH writing code for the FFI
to apply your existing knowledge. OTOH, writing code for the FFI
library is not for the faint of heart and probably shouldn't be the
first exercise for someone with little experience in Lua, C or C++.
</p>
@ -1190,7 +1190,7 @@ currently incomplete:
<li>C&nbsp;declarations are not passed through a C&nbsp;pre-processor,
yet.</li>
<li>The C&nbsp;parser is able to evaluate most constant expressions
commonly found in C&nbsp;header files. However it doesn't handle the
commonly found in C&nbsp;header files. However, it doesn't handle the
full range of C&nbsp;expression semantics and may fail for some
obscure constructs.</li>
<li><tt>static const</tt> declarations only work for integer types

24
doc/ext_ffi_tutorial.html
View File

@ -85,7 +85,7 @@ of its functions:
local ffi = require("ffi")
</pre>
<p>
Please note this doesn't define an <tt>ffi</tt> variable in the table
Please note, this doesn't define an <tt>ffi</tt> variable in the table
of globals &mdash; you really need to use the local variable. The
<tt>require</tt> function ensures the library is only loaded once.
</p>
@ -194,7 +194,7 @@ don't need to declare them as such.
<span class="mark">&#9316;</span> The <tt>poll()</tt>
function takes a couple more arguments we're not going to use. You can
simply use <tt>nil</tt> to pass a <tt>NULL</tt> pointer and <tt>0</tt>
for the <tt>nfds</tt> parameter. Please note that the
for the <tt>nfds</tt> parameter. Please note, that the
number&nbsp;<tt>0</tt> <em>does not convert to a pointer value</em>,
unlike in C++. You really have to pass pointers to pointer arguments
and numbers to number arguments.
@ -291,12 +291,12 @@ Here's the step-by-step explanation:
<p>
<span class="mark">&#9312;</span> This defines some of the
C&nbsp;functions provided by zlib. For the sake of this example, some
type indirections have been reduced and it uses the pre-defined
type indirections have been reduced and it uses the predefined
fixed-size integer types, while still adhering to the zlib API/ABI.
</p>
<p>
<span class="mark">&#9313;</span> This loads the zlib shared
library. On POSIX systems it's named <tt>libz.so</tt> and usually
library. On POSIX systems, it's named <tt>libz.so</tt> and usually
comes pre-installed. Since <tt>ffi.load()</tt> automatically adds any
missing standard prefixes/suffixes, we can simply load the
<tt>"z"</tt> library. On Windows it's named <tt>zlib1.dll</tt> and
@ -324,7 +324,7 @@ actual length that was used.
<p>
In C you'd pass in the address of a local variable
(<tt>&amp;buflen</tt>). But since there's no address-of operator in
Lua, we'll just pass in a one-element array. Conveniently it can be
Lua, we'll just pass in a one-element array. Conveniently, it can be
initialized with the maximum buffer size in one step. Calling the
actual <tt>zlib.compress2</tt> function is then straightforward.
</p>
@ -348,7 +348,7 @@ for garbage collection and string interning.
<span class="mark">&#9317;</span> The <tt>uncompress</tt>
functions does the exact opposite of the <tt>compress</tt> function.
The compressed data doesn't include the size of the original string,
so this needs to be passed in. Otherwise no surprises here.
so this needs to be passed in. Otherwise, no surprises here.
</p>
<p>
<span class="mark">&#9318;</span> The code, that makes use
@ -382,7 +382,7 @@ Ok, so the <tt>ffi.*</tt> functions generally accept cdata objects
wherever you'd want to use a number. That's why we get a away with
passing <tt>n</tt> to <tt>ffi.string()</tt> above. But other Lua
library functions or modules don't know how to deal with this. So for
maximum portability one needs to use <tt>tonumber()</tt> on returned
maximum portability, one needs to use <tt>tonumber()</tt> on returned
<tt>long</tt> results before passing them on. Otherwise the
application might work on some systems, but would fail in a POSIX/x64
environment.
@ -454,7 +454,7 @@ the origin.
</p>
<p>
<span class="mark">&#9315;</span> If we run out of operators, we can
define named methods, too. Here the <tt>__index</tt> table defines an
define named methods, too. Here, the <tt>__index</tt> table defines an
<tt>area</tt> function. For custom indexing needs, one might want to
define <tt>__index</tt> and <tt>__newindex</tt> <em>functions</em> instead.
</p>
@ -468,13 +468,13 @@ be used e.g. to create an array of points. The metamethods automatically
apply to any and all uses of this type.
</p>
<p>
Please note that the association with a metatable is permanent and
Please note, that the association with a metatable is permanent and
<b>the metatable must not be modified afterwards!</b> Ditto for the
<tt>__index</tt> table.
</p>
<p>
<span class="mark">&#9317;</span> Here are some simple usage examples
for the point type and their expected results. The pre-defined
for the point type and their expected results. The predefined
operations (such as <tt>a.x</tt>) can be freely mixed with the newly
defined metamethods. Note that <tt>area</tt> is a method and must be
called with the Lua syntax for methods: <tt>a:area()</tt>, not
@ -483,7 +483,7 @@ called with the Lua syntax for methods: <tt>a:area()</tt>, not
<p>
The C&nbsp;type metamethod mechanism is most useful when used in
conjunction with C&nbsp;libraries that are written in an object-oriented
style. Creators return a pointer to a new instance and methods take an
style. Creators return a pointer to a new instance, and methods take an
instance pointer as the first argument. Sometimes you can just point
<tt>__index</tt> to the library namespace and <tt>__gc</tt> to the
destructor and you're done. But often enough you'll want to add
@ -569,7 +569,7 @@ end
</pre>
<p>
This turns them into indirect calls and generates bigger and slower
machine code. Instead you'll want to cache the namespace itself and
machine code. Instead, you'll want to cache the namespace itself and
rely on the JIT compiler to eliminate the lookups:
</p>
<pre class="code">

4
doc/ext_jit.html
View File

@ -154,7 +154,7 @@ Contains the target architecture name:
<h2 id="jit_opt"><tt>jit.opt.*</tt> &mdash; JIT compiler optimization control</h2>
<p>
This sub-module provides the backend for the <tt>-O</tt> command line
This submodule provides the backend for the <tt>-O</tt> command line
option.
</p>
<p>
@ -174,7 +174,7 @@ which was one of the ways to enable optimization.
<h2 id="jit_util"><tt>jit.util.*</tt> &mdash; JIT compiler introspection</h2>
<p>
This sub-module holds functions to introspect the bytecode, generated
This submodule holds functions to introspect the bytecode, generated
traces, the IR and the generated machine code. The functionality
provided by this module is still in flux and therefore undocumented.
</p>

8
doc/ext_profiler.html
View File

@ -158,7 +158,7 @@ To see how much time is spent in different VM states or
Combinations of <tt>v/z</tt> with <tt>f/F/l</tt> produce two-level
views, e.g. <tt>-jp=vf</tt> or <tt>-jp=fv</tt>. This shows the time
spent in a VM state or zone vs. hotspots. This can be used to answer
questions like "Which time consuming functions are only interpreted?" or
questions like "Which time-consuming functions are only interpreted?" or
"What's the garbage collector overhead for a specific function?".
</p>
<p>
@ -217,7 +217,7 @@ local profile = require("jit.profile")
This module can be used to implement your own higher-level profiler.
A typical profiling run starts the profiler, captures stack dumps in
the profiler callback, adds them to a hash table to aggregate the number
of samples, stops the profiler and then analyzes all of the captured
of samples, stops the profiler and then analyzes all captured
stack dumps. Other parameters can be sampled in the profiler callback,
too. But it's important not to spend too much time in the callback,
since this may skew the statistics.
@ -271,9 +271,9 @@ returns a string with a stack dump for the <tt>thread</tt> (coroutine),
formatted according to the <tt>fmt</tt> argument:
</p>
<ul>
<li><tt>p</tt> &mdash; Preserve the full path for module names. Otherwise
<li><tt>p</tt> &mdash; Preserve the full path for module names. Otherwise,
only the file name is used.</li>
<li><tt>f</tt> &mdash; Dump the function name if it can be derived. Otherwise
<li><tt>f</tt> &mdash; Dump the function name if it can be derived. Otherwise,
use module:line.</li>
<li><tt>F</tt> &mdash; Ditto, but dump module:name.</li>
<li><tt>l</tt> &mdash; Dump module:line.</li>

10
doc/extensions.html
View File

@ -88,7 +88,7 @@ or LuaJIT.
</p>
<p>
LuaJIT extends the standard Lua VM with new functionality and adds
several extension modules. Please note this page is only about
several extension modules. Please note, this page is only about
<em>functional</em> enhancements and not about performance enhancements,
such as the optimized VM, the faster interpreter or the JIT compiler.
</p>
@ -197,7 +197,7 @@ usage. See also the
</p>
<p>
The generated bytecode is portable and can be loaded on any architecture
that LuaJIT supports, independent of word size or endianess. However the
that LuaJIT supports, independent of word size or endianess. However, the
bytecode compatibility versions must match. Bytecode stays compatible
for dot releases (x.y.0 &rarr; x.y.1), but may change with major or
minor releases (2.0 &rarr; 2.1) or between any beta release. Foreign
@ -229,7 +229,7 @@ avoids managing backlinks, saves an allocation and the overhead of
incremental array/hash part growth.
</p>
<p>
Please note this function is meant for very specific situations. In most
Please note, this function is meant for very specific situations. In most
cases it's better to replace the (usually single) link with a new table
and let the GC do its work.
</p>
@ -239,7 +239,7 @@ and let the GC do its work.
LuaJIT uses a Tausworthe PRNG with period 2^223 to implement
<tt>math.random()</tt> and <tt>math.randomseed()</tt>. The quality of
the PRNG results is much superior compared to the standard Lua
implementation which uses the platform-specific ANSI rand().
implementation, which uses the platform-specific ANSI rand().
</p>
<p>
The PRNG generates the same sequences from the same seeds on all
@ -257,7 +257,7 @@ Important: Neither this nor any other PRNG based on the simplistic
<h3 id="io"><tt>io.*</tt> functions handle 64&nbsp;bit file offsets</h3>
<p>
The file I/O functions in the standard <tt>io.*</tt> library handle
64&nbsp;bit file offsets. In particular this means it's possible
64&nbsp;bit file offsets. In particular, this means it's possible
to open files larger than 2&nbsp;Gigabytes and to reposition or obtain
the current file position for offsets beyond 2&nbsp;GB
(<tt>fp:seek()</tt> method).

6
doc/faq.html
View File

@ -120,7 +120,7 @@ Direct3D version 10 or higher do not show this behavior anymore.
Consider testing your application with older versions, too.<br>
Similarly, the Borland/Delphi runtime modifies the FPU control word and
enables FP exceptions. Of course this violates the Windows ABI, too.
enables FP exceptions. Of course, this violates the Windows ABI, too.
Please check the Delphi docs for the Set8087CW method.</dd>
</dl>
@ -130,7 +130,7 @@ Please check the Delphi docs for the Set8087CW method.</dd>
ignored by compiled code. If your program is running in a tight loop
and never falls back to the interpreter, the debug hook never runs and
can't throw the "interrupted!" error.<br>
You have to press Ctrl-C twice to get stop your program. That's similar
You have to press Ctrl-C twice to stop your program. That's similar
to when it's stuck running inside a C function under the Lua interpreter.</dd>
</dl>
@ -150,7 +150,7 @@ so it doesn't rely on the key order. Or sort the table keys, if you must.</dd>
<dl id="sandbox">
<dt>Q: Can Lua code be safely sandboxed?</dt>
<dd>
Maybe for an extremly restricted subset of Lua and if you relentlessly
Maybe for an extremely restricted subset of Lua and if you relentlessly
scrutinize every single interface function you offer to the untrusted code.<br>
Although Lua provides some sandboxing functionality (<tt>setfenv()</tt>, hooks),

6
doc/install.html
View File

@ -328,7 +328,7 @@ The recommended way to fetch the latest version is to do a pull from
the git repository.
</p>
<p>
Alternatively download the latest source package of LuaJIT (pick the .tar.gz).
Alternatively, download the latest source package of LuaJIT (pick the .tar.gz).
Move it to a directory of your choice, open a terminal window and change
to this directory. Now unpack the archive and change to the newly created
directory (replace XX.YY.ZZ with the version you downloaded):
@ -697,7 +697,7 @@ allocator from your system (no support for this on 64&nbsp;bit architectures).</
of calling <tt>luaopen_base</tt> etc. directly.</li>
<li>To change or extend the list of standard libraries to load, copy
<tt>src/lib_init.c</tt> to your project and modify it accordingly.
Make sure the <tt>jit</tt> library is loaded or the JIT compiler
Make sure the <tt>jit</tt> library is loaded, or the JIT compiler
will not be activated.</li>
<li>The <tt>bit.*</tt> module for bitwise operations
is already built-in. There's no need to statically link
@ -716,7 +716,7 @@ in unspeakable ways.
There should be absolutely no need to patch <tt>luaconf.h</tt> or any
of the Makefiles. And please do not hand-pick files for your packages &mdash;
simply use whatever <tt>make install</tt> creates. There's a reason
for all of the files <em>and</em> directories it creates.
for all the files <em>and</em> directories it creates.
</p>
<p>
The build system uses GNU make and auto-detects most settings based on

5
doc/running.html
View File

@ -111,6 +111,7 @@ are accepted:
<li><tt>-t type</tt> &mdash; Set output file type (default: auto-detect from output name).</li>
<li><tt>-a arch</tt> &mdash; Override architecture for object files (default: native).</li>
<li><tt>-o os</tt> &mdash; Override OS for object files (default: native).</li>
<li><tt>-F name</tt> &mdash; Override filename (default: input filename).</li>
<li><tt>-e chunk</tt> &mdash; Use chunk string as input.</li>
<li><tt>-</tt> (a single minus sign) &mdash; Use stdin as input and/or stdout as output.</li>
</ul>
@ -184,7 +185,7 @@ written in Lua. They are mainly used for debugging the JIT compiler
itself. For a description of their options and output format, please
read the comment block at the start of their source.
They can be found in the <tt>lib</tt> directory of the source
distribution or installed under the <tt>jit</tt> directory. By default
distribution or installed under the <tt>jit</tt> directory. By default,
this is <tt>/usr/local/share/luajit-XX.YY.ZZ>/jit</tt> on POSIX
systems (replace XX.YY.ZZ by the installed version).
</p>
@ -216,7 +217,7 @@ to a specific value.
You can either use this option multiple times (like <tt>-Ocse
-O-dce -Ohotloop=10</tt>) or separate several settings with a comma
(like <tt>-O+cse,-dce,hotloop=10</tt>). The settings are applied from
left to right and later settings override earlier ones. You can freely
left to right, and later settings override earlier ones. You can freely
mix the three forms, but note that setting an optimization level
overrides all earlier flags.
</p>

2
doc/status.html
View File

@ -83,7 +83,7 @@ Known incompatibilities and issues in LuaJIT&nbsp;2.0:
<ul>
<li>
There are some differences in <b>implementation-defined</b> behavior.
These either have a good reason, are arbitrary design choices
These either have a good reason, are arbitrary design choices,
or are due to quirks in the VM. The latter cases may get fixed if a
demonstrable need is shown.
</li>

43
src/jit/bcsave.lua
View File

@ -34,6 +34,7 @@ Save LuaJIT bytecode: luajit -b[options] input output
-t type Set output file type (default: auto-detect from output name).
-a arch Override architecture for object files (default: native).
-o os Override OS for object files (default: native).
-F name Override filename (default: input filename).
-e chunk Use chunk string as input.
-- Stop handling options.
- Use stdin as input and/or stdout as output.
@ -50,10 +51,22 @@ local function check(ok, ...)
os.exit(1)
end
local function readfile(input)
local function readfile(ctx, input)
if type(input) == "function" then return input end
if input == "-" then input = nil end
return check(loadfile(input))
if ctx.filename then
local data
if input == "-" then
data = io.stdin:read("*a")
else
local fp = assert(io.open(input, "rb"))
data = assert(fp:read("*a"))
assert(fp:close())
end
return check(load(data, ctx.filename))
else
if input == "-" then input = nil end
return check(loadfile(input))
end
end
local function savefile(name, mode)
@ -457,18 +470,18 @@ typedef struct {
uint32_t value;
} mach_nlist;
typedef struct {
uint32_t strx;
int32_t strx;
uint8_t type, sect;
uint16_t desc;
uint64_t value;
} mach_nlist_64;
typedef struct
{
uint32_t magic, nfat_arch;
int32_t magic, nfat_arch;
} mach_fat_header;
typedef struct
{
uint32_t cputype, cpusubtype, offset, size, align;
int32_t cputype, cpusubtype, offset, size, align;
} mach_fat_arch;
typedef struct {
struct {
@ -502,6 +515,18 @@ typedef struct {
mach_nlist sym_entry;
uint8_t space[4096];
} mach_fat_obj;
typedef struct {
mach_fat_header fat;
mach_fat_arch fat_arch[2];
struct {
mach_header_64 hdr;
mach_segment_command_64 seg;
mach_section_64 sec;
mach_symtab_command sym;
} arch[2];
mach_nlist_64 sym_entry;
uint8_t space[4096];
} mach_fat_obj_64;
]]
local symname = '_'..LJBC_PREFIX..ctx.modname
local isfat, is64, align, mobj = false, false, 4, "mach_obj"
@ -510,7 +535,7 @@ typedef struct {
elseif ctx.arch == "arm" then
isfat, mobj = true, "mach_fat_obj"
elseif ctx.arch == "arm64" then
is64, align, isfat, mobj = true, 8, true, "mach_fat_obj"
is64, align, isfat, mobj = true, 8, true, "mach_fat_obj_64"
else
check(ctx.arch == "x86", "unsupported architecture for OSX")
end
@ -599,7 +624,7 @@ local function bclist(input, output, lineinfo)
end
local function bcsave(ctx, input, output)
local f = readfile(input)
local f = readfile(ctx, input)
local s = string.dump(f, ctx.strip)
local t = ctx.type
if not t then
@ -656,6 +681,8 @@ local function docmd(...)
ctx.arch = checkarg(tremove(arg, n), map_arch, "architecture")
elseif opt == "o" then
ctx.os = checkarg(tremove(arg, n), map_os, "OS name")
elseif opt == "F" then
ctx.filename = "@"..tremove(arg, n)
else
usage()
end

2
src/lib_buffer.c
View File

@ -323,6 +323,7 @@ LJLIB_CF(buffer_new)
setgcref(sbx->dict_str, obj2gco(dict_str));
setgcref(sbx->dict_mt, obj2gco(dict_mt));
if (sz > 0) lj_buf_need2((SBuf *)sbx, sz);
lj_gc_check(L);
return 1;
}
@ -339,6 +340,7 @@ LJLIB_CF(buffer_decode) LJLIB_REC(.)
GCstr *str = lj_lib_checkstrx(L, 1);
setnilV(L->top++);
lj_serialize_decode(L, L->top-1, str);
lj_gc_check(L);
return 1;
}

4
src/lib_ffi.c
View File

@ -639,7 +639,7 @@ LJLIB_CF(ffi_alignof) LJLIB_REC(ffi_xof FF_ffi_alignof)
CTState *cts = ctype_cts(L);
CTypeID id = ffi_checkctype(L, cts, NULL);
CTSize sz = 0;
CTInfo info = lj_ctype_info(cts, id, &sz);
CTInfo info = lj_ctype_info_raw(cts, id, &sz);
setintV(L->top-1, 1 << ctype_align(info));
return 1;
}
@ -770,7 +770,7 @@ LJLIB_CF(ffi_metatype)
CTypeID id = ffi_checkctype(L, cts, NULL);
GCtab *mt = lj_lib_checktab(L, 2);
GCtab *t = cts->miscmap;
CType *ct = ctype_get(cts, id); /* Only allow raw types. */
CType *ct = ctype_raw(cts, id);
TValue *tv;
GCcdata *cd;
if (!(ctype_isstruct(ct->info) || ctype_iscomplex(ct->info) ||

2
src/lib_package.c
View File

@ -57,7 +57,7 @@ static lua_CFunction ll_sym(lua_State *L, void *lib, const char *sym)
static const char *ll_bcsym(void *lib, const char *sym)
{
#if defined(RTLD_DEFAULT)
#if defined(RTLD_DEFAULT) && !defined(NO_RTLD_DEFAULT)
if (lib == NULL) lib = RTLD_DEFAULT;
#elif LJ_TARGET_OSX || LJ_TARGET_BSD
if (lib == NULL) lib = (void *)(intptr_t)-2;

2
src/lj_api.c
View File

@ -779,7 +779,7 @@ LUA_API void lua_concat(lua_State *L, int n)
L->top -= n;
break;
}
n -= (int)(L->top - top);
n -= (int)(L->top - (top - 2*LJ_FR2));
L->top = top+2;
lj_vm_call(L, top, 1+1);
L->top -= 1+LJ_FR2;

3
src/lj_asm_x86.h
View File

@ -485,7 +485,8 @@ static Reg asm_fuseload(ASMState *as, IRRef ref, RegSet allow)
asm_fusexref(as, ir->op1, xallow);
return RID_MRM;
}
} else if (ir->o == IR_VLOAD && !(LJ_GC64 && irt_isaddr(ir->t))) {
} else if (ir->o == IR_VLOAD && IR(ir->op1)->o == IR_AREF &&
!(LJ_GC64 && irt_isaddr(ir->t))) {
asm_fuseahuref(as, ir->op1, xallow);
as->mrm.ofs += 8 * ir->op2;
return RID_MRM;

2
src/lj_clib.c
View File

@ -25,7 +25,7 @@
#include <dlfcn.h>
#include <stdio.h>
#if defined(RTLD_DEFAULT)
#if defined(RTLD_DEFAULT) && !defined(NO_RTLD_DEFAULT)
#define CLIB_DEFHANDLE RTLD_DEFAULT
#elif LJ_TARGET_OSX || LJ_TARGET_BSD
#define CLIB_DEFHANDLE ((void *)(intptr_t)-2)

2
src/lj_cparse.c
View File

@ -468,7 +468,7 @@ static void cp_expr_sizeof(CPState *cp, CPValue *k, int wantsz)
} else {
cp_expr_unary(cp, k);
}
info = lj_ctype_info(cp->cts, k->id, &sz);
info = lj_ctype_info_raw(cp->cts, k->id, &sz);
if (wantsz) {
if (sz != CTSIZE_INVALID)
k->u32 = sz;

8
src/lj_ctype.c
View File

@ -333,6 +333,14 @@ CTInfo lj_ctype_info(CTState *cts, CTypeID id, CTSize *szp)
return qual;
}
/* Ditto, but follow a reference. */
CTInfo lj_ctype_info_raw(CTState *cts, CTypeID id, CTSize *szp)
{
CType *ct = ctype_get(cts, id);
if (ctype_isref(ct->info)) id = ctype_cid(ct->info);
return lj_ctype_info(cts, id, szp);
}
/* Get ctype metamethod. */
cTValue *lj_ctype_meta(CTState *cts, CTypeID id, MMS mm)
{

1
src/lj_ctype.h
View File

@ -468,6 +468,7 @@ LJ_FUNC CType *lj_ctype_rawref(CTState *cts, CTypeID id);
LJ_FUNC CTSize lj_ctype_size(CTState *cts, CTypeID id);
LJ_FUNC CTSize lj_ctype_vlsize(CTState *cts, CType *ct, CTSize nelem);
LJ_FUNC CTInfo lj_ctype_info(CTState *cts, CTypeID id, CTSize *szp);
LJ_FUNC CTInfo lj_ctype_info_raw(CTState *cts, CTypeID id, CTSize *szp);
LJ_FUNC cTValue *lj_ctype_meta(CTState *cts, CTypeID id, MMS mm);
LJ_FUNC GCstr *lj_ctype_repr(lua_State *L, CTypeID id, GCstr *name);
LJ_FUNC GCstr *lj_ctype_repr_int64(lua_State *L, uint64_t n, int isunsigned);

3
src/lj_gc.c
View File

@ -700,9 +700,12 @@ static size_t gc_onestep(lua_State *L)
}
case GCSfinalize:
if (gcref(g->gc.mmudata) != NULL) {
GCSize old = g->gc.total;
if (tvref(g->jit_base)) /* Don't call finalizers on trace. */
return LJ_MAX_MEM;
gc_finalize(L); /* Finalize one userdata object. */
if (old >= g->gc.total && g->gc.estimate > old - g->gc.total)
g->gc.estimate -= old - g->gc.total;
if (g->gc.estimate > GCFINALIZECOST)
g->gc.estimate -= GCFINALIZECOST;
return GCFINALIZECOST;

4
src/lj_record.c
View File

@ -1956,7 +1956,7 @@ static void rec_varg(jit_State *J, BCReg dst, ptrdiff_t nresults)
emitir(IRTGI(IR_EQ), fr,
lj_ir_kint(J, (int32_t)frame_ftsz(J->L->base-1)));
vbase = emitir(IRT(IR_SUB, IRT_IGC), REF_BASE, fr);
vbase = emitir(IRT(IR_ADD, IRT_PGC), vbase, lj_ir_kint(J, frofs-8));
vbase = emitir(IRT(IR_ADD, IRT_PGC), vbase, lj_ir_kint(J, frofs-8*(1+LJ_FR2)));
for (i = 0; i < nload; i++) {
IRType t = itype2irt(&J->L->base[i-1-LJ_FR2-nvararg]);
J->base[dst+i] = lj_record_vload(J, vbase, i, t);
@ -2660,6 +2660,8 @@ static const BCIns *rec_setup_root(jit_State *J)
J->bc_min = pc;
break;
case BC_ITERL:
if (bc_op(pc[-1]) == BC_JLOOP)
lj_trace_err(J, LJ_TRERR_LINNER);
lj_assertJ(bc_op(pc[-1]) == BC_ITERC, "no ITERC before ITERL");
J->maxslot = ra + bc_b(pc[-1]) - 1;
J->bc_extent = (MSize)(-bc_j(ins))*sizeof(BCIns);

2
src/vm_arm64.dasc
View File

@ -3988,6 +3988,7 @@ static void emit_asm_debug(BuildCtx *ctx)
"\t.align 3\n"
".LEFDE1:\n\n", (int)ctx->codesz - fcofs);
#endif
#if !LJ_NO_UNWIND
fprintf(ctx->fp, "\t.section .eh_frame,\"a\",%%progbits\n");
fprintf(ctx->fp,
".Lframe1:\n"
@ -4055,6 +4056,7 @@ static void emit_asm_debug(BuildCtx *ctx)
"\t.byte 0x94\n\t.uleb128 4\n" /* offset x20 */
"\t.align 3\n"
".LEFDE3:\n\n", (int)ctx->codesz - fcofs);
#endif
#endif
break;
#if !LJ_NO_UNWIND

13
src/vm_x64.dasc
View File

@ -359,9 +359,6 @@
|.macro sseconst_1, reg, tmp // Synthesize 1.0.
| sseconst_hi reg, tmp, 3ff00000
|.endmacro
|.macro sseconst_m1, reg, tmp // Synthesize -1.0.
| sseconst_hi reg, tmp, bff00000
|.endmacro
|.macro sseconst_2p52, reg, tmp // Synthesize 2^52.
| sseconst_hi reg, tmp, 43300000
|.endmacro
@ -2543,15 +2540,17 @@ static void build_subroutines(BuildCtx *ctx)
| addsd xmm1, xmm3 // (|x| + 2^52) - 2^52
| subsd xmm1, xmm3
| orpd xmm1, xmm2 // Merge sign bit back in.
| sseconst_1 xmm3, RD
| .if mode == 1 // ceil(x)?
| sseconst_m1 xmm2, RD // Must subtract -1 to preserve -0.
| cmpsd xmm0, xmm1, 6 // x > result?
| andpd xmm0, xmm3
| addsd xmm1, xmm0 // If yes, add 1.
| orpd xmm1, xmm2 // Merge sign bit back in (again).
| .else // floor(x)?
| sseconst_1 xmm2, RD
| cmpsd xmm0, xmm1, 1 // x < result?
| andpd xmm0, xmm3
| subsd xmm1, xmm0 // If yes, subtract 1.
| .endif
| andpd xmm0, xmm2
| subsd xmm1, xmm0 // If yes, subtract +-1.
|.endif
| movaps xmm0, xmm1
|1:

13
src/vm_x86.dasc
View File

@ -464,9 +464,6 @@
|.macro sseconst_1, reg, tmp // Synthesize 1.0.
| sseconst_hi reg, tmp, 3ff00000
|.endmacro
|.macro sseconst_m1, reg, tmp // Synthesize -1.0.
| sseconst_hi reg, tmp, bff00000
|.endmacro
|.macro sseconst_2p52, reg, tmp // Synthesize 2^52.
| sseconst_hi reg, tmp, 43300000
|.endmacro
@ -3004,15 +3001,17 @@ static void build_subroutines(BuildCtx *ctx)
| addsd xmm1, xmm3 // (|x| + 2^52) - 2^52
| subsd xmm1, xmm3
| orpd xmm1, xmm2 // Merge sign bit back in.
| sseconst_1 xmm3, RDa
| .if mode == 1 // ceil(x)?
| sseconst_m1 xmm2, RDa // Must subtract -1 to preserve -0.
| cmpsd xmm0, xmm1, 6 // x > result?
| andpd xmm0, xmm3
| addsd xmm1, xmm0 // If yes, add 1.
| orpd xmm1, xmm2 // Merge sign bit back in (again).
| .else // floor(x)?
| sseconst_1 xmm2, RDa
| cmpsd xmm0, xmm1, 1 // x < result?
| andpd xmm0, xmm3
| subsd xmm1, xmm0 // If yes, subtract 1.
| .endif
| andpd xmm0, xmm2
| subsd xmm1, xmm0 // If yes, subtract +-1.
|.endif
| movaps xmm0, xmm1
|1: