path: root/doc/latex/src/changelog.tex

%
% vim: ts=4 sw=4 et
%
\xchapter{changelog}{\textindexlc{NASM Version History}}

\xsection{cl-2.xx}{NASM 2 Series}

The NASM 2 series supports x86-64, and is the production version of NASM
since 2007.

\xsubsection{cl-2.15}{Version 2.15}

\begin{itemize}
    \item{The state of warnings can now be saved and restored via
        the \code{[WARNING PUSH]} and \code{[WARNING POP]} directives.
        See \nref{asmdir-warning}}.
\end{itemize}

\xsubsection{cl-2.14.03}{Version 2.14.03}

\begin{itemize}
    \item{Suppress nuisance "\code{label changed during code
        generation}" messages after a real error.}
\end{itemize}

\xsubsection{cl-2.14.02}{Version 2.14.02}

\begin{itemize}
    \item{Fix crash due to multiple errors or warnings during the code
        generation pass if a list file is specified.}
\end{itemize}

\xsubsection{cl-2.14.01}{Version 2.14.01}

\begin{itemize}
    \item{Create all system-defined macros defore processing command-line
        given preprocessing directives (\code{-p}, \code{-d}, \code{-u}, \code{--pragma},
        \code{--before}).}

    \item{If debugging is enabled, define a \code{\_\_DEBUG\_FORMAT\_\_} predefined
        macro. See \nref{dfmtm}.}

    \item{Fix an assert for the case in the \code{obj} format when a \code{SEG}
        operator refers to an \code{EXTERN} symbol declared further down in the
        code.}

    \item{Fix a corner case in the floating-point code where a binary, octal
        or hexadecimal floating-point having at least 32, 11, or 8 mantissa
        digits could produce slightly incorrect results under very specific
        conditions.}

    \item{Support \code{-MD} without a filename, for \code{gcc} compatibility.
        \code{-MF} can be used to set the dependencies output filename.
        See \nref{opt-MD}.}

    \item{Fix \code{-E} in combination with \code{-MD}. See \nref{opt-E}.}

    \item{Fix missing errors on redefined labels; would cause convergence
        failure instead which is very slow and not easy to debug.}

    \item{Duplicate definitions of the same label \emph{with the same value} is now
        explicitly permitted (2.14 would allow it in some circumstances.)}

    \item{Add the option \code{--no-line} to ignore \code{\%line} directives in the
        source. See \nref{opt-no-line} and \nref{line}.}
\end{itemize}

\xsubsection{cl-2.14}{Version 2.14}

\begin{itemize}
    \item{Changed \code{-I} option semantics by adding a trailing path
        separator unconditionally.}

    \item{Fixed null dereference in corrupted invalid single line macros.}

    \item{Fixed division by zero which may happen if source code is malformed.}

    \item{Fixed out of bound access in processing of malformed segment override.}

    \item{Fixed out of bound access in certain \code{EQU} parsing.}

    \item{Fixed buffer underflow in float parsing.}

    \item{Added \code{SGX} (Intel Software Guard Extensions) instructions.}

    \item{Added \code{+n} syntax for multiple contiguous registers.}

    \item{Fixed \code{subsections\_via\_symbols} for \code{macho} object format.}

    \item{Added the \code{--gprefix}, \code{--gpostfix}, \code{--lprefix}, and
        \code{--lpostfix} command line options, to allow command line base symbol
        renaming. See \nref{opt-pfix}.}

    \item{Allow label renaming to be specified by \code{\%pragma} in addition to
        from the command line. See \nref{mangling}.}

    \item{Supported generic \code{\%pragma} namespaces, \code{output} and \code{debug}.
        See \nref{gen-namespace}.}

    \item{Added the \code{--pragma} command line option to inject a \code{\%pragma}
        directive. See \nref{opt-pragma}.}

    \item{Added the \code{--before} command line option to accept preprocess
        statement before input. See \nref{opt-before}.}

    \item{Added \code{AVX512} \code{VBMI2} (Additional Bit Manipulation), \code{VNNI}
        (Vector Neural Network), \code{BITALG} (Bit Algorithm), and \code{GFNI} (Galois
        Field New Instruction) instructions.}

    \item{Added the \code{STATIC} directive for local symbols that should be
        renamed using global-symbol rules. See \nref{static}.}

    \item{Allow a symbol to be defined as \code{EXTERN} and then later
        overridden as \code{GLOBAL} or \code{COMMON}. Furthermore, a symbol
        declared \code{EXTERN} and then defined will be treated as \code{GLOBAL}.
        See \nref{extern}.}

    \item{The \code{GLOBAL} directive no longer is required to precede the
        definition of the symbol.}

    \item{Support \code{private\_extern} as \code{macho} specific extension to the
        \code{GLOBAL} directive. See \nref{macho-pext}.}

    \item{Updated \code{UD0} encoding to match with the specification}

    \item{Added the \code{--limit-X} command line option to set execution
        limits. See \nref{opt-limit}.}

    \item{Updated the \code{Codeview} version number to be aligned with \code{MASM}.}

    \item{Added the \code{--keep-all} command line option to preserve output
        files. See \nref{opt-keep-all}.}

    \item{Added the \code{--include} command line option, an alias to \code{-P}
        (\nref{opt-p}).}

    \item{Added the \code{--help} command line option as an alias to \code{-h}
        (\nref{syntax}).}

    \item{Added \code{-W}, \code{-D}, and \code{-Q} suffix aliases for \code{RET}
        instructions so the operand sizes of these instructions can be
        encoded without using \code{o16}, \code{o32} or \code{o64}.}
\end{itemize}

\xsubsection{cl-2.13.03}{Version 2.13.03}

\begin{itemize}
    \item{Added AVX and AVX512 \code{VAES*} and \code{VPCLMULQDQ} instructions.}
    \item{Fixed missing dwarf record in x32 ELF output format.}
\end{itemize}

\xsubsection{cl-2.13.02}{Version 2.13.02}

\begin{itemize}
    \item{Fix false positive in testing of numeric overflows.}

    \item{Fix generation of \code{PEXTRW} instruction.}

    \item{Fix \code{smartalign} package which could trigger an error during
        optimization if the alignment code expanded too much due to
        optimization of the previous code.}

    \item{Fix a case where negative value in \code{TIMES} directive causes
        panic instead of an error.}

    \item{Always finalize \code{.debug\_abbrev} section with a null in
        \code{dwarf} output format.}

    \item{Support \code{debug} flag in section attributes for \code{macho}
        output format. See \nref{machosect}.}

    \item{Support up to 16 characters in section names for \code{macho}
        output format.}

    \item{Fix missing update of global \code{BITS} setting if \code{SECTION}
        directive specified a bit size using output format-specific
        extensions (e.g. \code{USE32} for the \code{obj} output format.)}

    \item{Fix the incorrect generation of VEX-encoded instruction when static
        mode decorators are specified on scalar instructions, losing the
        decorators as they require EVEX encoding.}

    \item{Option \code{-MW} to quote dependency outputs according to Watcom
        Make conventions instead of POSIX Make conventions. See \nref{opt-MW}.}

    \item{The \code{obj} output format now contains embedded dependency file
        information, unless disabled with \code{\%pragma obj nodepend}. See
        \nref{objdepend}.}

    \item{Fix generation of dependency lists.}

    \item{Fix a number of null pointer reference and memory allocation errors.}

    \item{Always generate symbol-relative relocations for the \code{macho64}
        output format; at least some versions of the XCode/LLVM linker fails
        for section-relative relocations.}
\end{itemize}

\xsubsection{cl-2.13.01}{Version 2.13.01}

\begin{itemize}
    \item{Fix incorrect output for some types of \code{FAR} or \code{SEG}
        references in the \code{obj} output format, and possibly other 16-bit
        output formats.}

    \item{Fix the address in the list file for an instruction containing a
        \code{TIMES} directive.}

    \item{Fix error with \code{TIMES} used together with an instruction which
        can vary in size, e.g. \code{JMP}.}

    \item{Fix breakage on some uses of the \code{DZ} pseudo-op.}
\end{itemize}

\xsubsection{cl-2.13}{Version 2.13}

\begin{itemize}
    \item{Support the official forms of the \code{UD0} and \code{UD1} instructions.}

    \item{Allow self-segment-relative expressions in immediates and
        displacements, even when combined with an external or otherwise
        out-of-segment special symbol, e.g.:
\begin{lstlisting}
extern foo
mov eax,[foo - $ + ebx] ; Now legal
\end{lstlisting}}

    \item{Handle a 64-bit origin in NDISASM.}

    \item{NASM can now generate sparse output files for relevant output
        formats, if the underlying operating system supports them.}

    \item{The \code{macho} object format now supports the
        \code{subsections\_via\_symbols} and \code{no\_dead\_strip}
        directives, see \nref{macho-ssvs}.}

    \item{The \code{macho} object format now supports the \code{no\_dead\_strip},
        \code{live\_support} and \code{strip\_static\_syms} section flags,
        see \nref{machosect}.}

    \item{The \code{macho} object format now supports the \code{dwarf} debugging
        format, as required by newer toolchains.}

    \item{All warnings can now be suppressed if desired; warnings not
        otherwise part of any warning class are now considered its own
        warning class called \code{other} (e.g. \code{-w-other}). Furthermore,
        warning-as-error can now be controlled on a per warning class
        basis, using the syntax \code{-w+error=}\emph{warning-class} and its
        equivalent for all other warning control options. See \nref{opt-w}
        for the command-line options and warning classes and
        \nref{asmdir-warning} for the \code{[WARNING]} directive.}

    \item{Fix a number of bugs related to AVX-512 decorators.}

    \item{Significant improvements to building NASM with Microsoft Visual
        Studio via \code{Mkfiles/msvc.mak}. It is now possible to build the
        full Windows installer binary as long as the necessary
        prerequisites are installed; see \code{Mkfiles/README}}

    \item{To build NASM with custom modifications (table changes) or from the
        git tree now requires Perl 5.8 at the very minimum, quite possibly
        a higher version (Perl 5.24.1 tested.) There is no requirement to
        have Perl on your system at all if all you want to do is build
        unmodified NASM from source archives.}

    \item{Fix the \code{\{z\}} decorator on AVX-512 \code{VMOVDQ*}
        instructions.}

    \item{Add new warnings for certain dangerous constructs which never ought
        to have been allowed. In particular, the \code{RESB} family of
        instructions should have been taking a critical expression all
        along.}

    \item{Fix the EVEX (AVX-512) versions of the \code{VPBROADCAST}, \code{VPEXTR},
        and \code{VPINSR} instructions.}

    \item{Support contracted forms of additional instructions. As a general
        rule, if an instruction has a non-destructive source immediately
        after a destination register that isn't used as an input, NASM
        supports omitting that source register, using the destination
        register as that value. This among other things makes it easier to
        convert SSE code to the equivalent AVX code:
\begin{lstlisting}
addps xmm1,xmm0         ; SSE instruction
vaddps ymm1,ymm1,ymm0   ; AVX official long form
vaddps ymm1,ymm0        ; AVX contracted form
\end{lstlisting}}

    \item{Fix Codeview malformed compiler version record.}

    \item{Add the \code{CLWB} and \code{PCOMMIT} instructions. Note that the
        \code{PCOMMIT} instruction has been deprecated and will never be
        included in a shipping product; it is included for completeness
        only.}

    \item{Add the \code{\%pragma} preprocessor directive for
        soft-error directives.}

    \item{Add the \code{RDPID} instruction.}
\end{itemize}

\xsubsection{cl-2.12.02}{Version 2.12.02}

\begin{itemize}
    \item{Fix preprocessor errors, especially \code{\%error} and
        \code{\%warning}, inside \code{\%if} statements.}

    \item{Fix relative relocations in 32-bit Mach-O.}

    \item{More Codeview debug format fixes.}

    \item{If the MASM \code{PTR} keyword is encountered, issue a warning. This is
        much more likely to indicate a MASM-ism encountered in NASM than it
        is a valid label. This warning can be suppressed with \code{-w-ptr},
        the \code{[warning]} directive (see \nref{opt-w}) or by the macro
        definition \code{\%idefine ptr \$\%} (see \nref{selfref}).}

    \item{When an error or a warning comes from the expansion of a multi-line
        macro, display the file and line numbers for the expanded macros.
        Macros defined with \code{.nolist} do not get displayed.}

    \item{Add macros \code{ilog2fw()} and \code{ilog2cw()} to the \code{ifunc} macro
        package. See \nref{ilog2}.}
\end{itemize}

\xsubsection{cl-2.12.01}{Version 2.12.01}

\begin{itemize}
    \item{Portability fixes for some platforms.}
    \item{Fix error when not specifying a list file.}
    \item{Correct the handling of macro-local labels in the Codeview debugging format.}
    \item{Add \code{CLZERO}, \code{MONITORX} and \code{MWAITX} instructions.}
\end{itemize}

\xsubsection{cl-2.12}{Version 2.12}

\begin{itemize}
    \item{Major fixes to the \code{macho} backend (\nref{machofmt}); earlier versions
        would produce invalid symbols and relocations on a regular basis.}

    \item{Support for thread-local storage in Mach-O.}

    \item{Support for arbitrary sections in Mach-O.}

    \item{Fix wrong negative size treated as a big positive value passed into
        backend causing NASM to crash.}

    \item{Fix handling of zero-extending unsigned relocations, we have been printing
        wrong message and forgot to assign segment with predefined value before
        passing it into output format.}

    \item{Fix potential write of oversized (with size greater than allowed in
        output format) relative relocations.}

    \item{Portability fixes for building NASM with the LLVM compiler.}

    \item{Add support of Codeview version 8 (\code{cv8}) debug format for
        \code{win32} and \code{win64} formats in the \code{COFF} backend,
        see \nref{codeview}.}

    \item{Allow 64-bit outputs in 16/32-bit only backends. Unsigned 64-bit
        relocations are zero-extended from 32-bits with a warning
        (suppressible via \code{-w-zext-reloc}); signed 64-bit relocations are
        an error.}

    \item{Line numbers in list files now correspond to the lines in the source
        files, instead of simply being sequential.}

    \item{There is now an official 64-bit (x64 a.k.a. x86-64) build for Windows.}
\end{itemize}

\xsubsection{cl-2.11.09}{Version 2.11.09}

\begin{itemize}
    \item{Fix potential stack overwrite in \code{macho32} backend.}

    \item{Fix relocation records in \code{macho64} backend.}

    \item{Fix symbol lookup computation in \code{macho64} backend.}

    \item{Adjust \code{.symtab} and \code{.rela.text} sections alignments to 8 bytes
        in \code{elf64} backed.}

    \item{Fix section length computation in \code{bin} backend which leaded in incorrect
        relocation records.}
\end{itemize}

\xsubsection{cl-2.11.08}{Version 2.11.08}

\begin{itemize}
    \item{Fix section length computation in \code{bin} backend which leaded in incorrect
        relocation records.}

    \item{Add a warning for numeric preprocessor definitions passed via command
        line which might have unexpected results otherwise.}

    \item{Add ability to specify a module name record in \code{rdoff} linker with
        \code{-mn} option.}

    \item{Increase label length capacity up to 256 bytes in \code{rdoff} backend for
        FreePascal sake, which tends to generate very long labels for procedures.}

    \item{Fix segmentation failure when rip addressing is used in \code{macho64} backend.}

    \item{Fix access on out of memory when handling strings with a single
        grave. We have sixed similar problem in previous release but not
        all cases were covered.}

    \item{Fix NULL dereference in disassembled on \code{BND} instruction.}
\end{itemize}

\xsubsection{cl-2.11.07}{Version 2.11.07}

\begin{itemize}
    \item{Fix 256 bit \code{VMOVNTPS} instruction.}

    \item{Fix \code{-MD} option handling, which was rather broken in previous
        release changing command line api.}

    \item{Fix access to unitialized space when handling strings with
        a single grave.}

    \item{Fix nil dereference in handling memory reference parsing.}
\end{itemize}

\xsubsection{cl-2.11.06}{Version 2.11.06}

\begin{itemize}
    \item{Update AVX512 instructions based on the Extension Reference (319433-021 Sept 2014).}
    \item{Fix the behavior of \code{-MF} and \code{-MD} options (Bugzilla 3392280)}
    \item{Updated Win32 Makefile to fix issue with build}
\end{itemize}

\xsubsection{cl-2.11.05}{Version 2.11.05}

\begin{itemize}
    \item{Add \code{--v} as an alias for \code{-v} (see \nref{opt-v}), for
        command-line compatibility with Yasm.}
    \item{Fix a bug introduced in 2.11.03 whereby certain instructions would
        contain multiple REX prefixes, and thus be corrupt.}
\end{itemize}

\xsubsection{cl-2.11.04}{Version 2.11.04}

\begin{itemize}
    \item{Removed an invalid error checking code. Sometimes a memref only with
        a displacement can also set an evex flag. For example:
\begin{lstlisting}
vmovdqu32 [0xabcd]{k1}, zmm0
\end{lstlisting}}

    \item{Fixed a bug in disassembler that EVEX.L'L vector length was not matched
        when EVEX.b was set because it was simply considered as EVEC.RC.
        Separated EVEX.L'L case from EVEX.RC which is ignored in matching.}
\end{itemize}

\xsubsection{cl-2.11.03}{Version 2.11.03}

\begin{itemize}
    \item{Fix a bug there REX prefixes were missing on instructions
        inside a \code{TIMES} statement.}
\end{itemize}

\xsubsection{cl-2.11.02}{Version 2.11.02}

\begin{itemize}
    \item{Add the \code{XSAVEC}, \code{XSAVES} and \code{XRSTORS} family instructions.}
    \item{Add the \code{CLFLUSHOPT} instruction.}
\end{itemize}

\xsubsection{cl-2.11.01}{Version 2.11.01}

\begin{itemize}
    \item{Allow instructions which implicitly uses \code{XMM0} (\code{VBLENDVPD},
        \code{VBLENDVPS}, \code{PBLENDVB} and \code{SHA256RNDS2}) to be specified
        without an explicit \code{xmm0} on the assembly line. In other words,
        the following two lines produce the same output:
\begin{lstlisting}
vblendvpd xmm2,xmm1,xmm0    ; Last operand is fixed xmm0
vblendvpd xmm2,xmm1         ; Implicit xmm0 omitted
\end{lstlisting}}

    \item{In the ELF backends, don't crash the assembler if \code{section align}
        is specified without a value.}
\end{itemize}

\xsubsection{cl-2.11}{Version 2.11}

\begin{itemize}
    \item{Add support for the Intel AVX-512 instruction set:}

    \item{16 new, 512-bit SIMD registers. Total 32 \code{(ZMM0 \textasciitilde ZMM31)}}

    \item{8 new opmask registers \code{(K0 \textasciitilde K7)}.
        One of 7 registers \code{(K1 \textasciitilde K7)} can
        be used as an opmask for conditional execution.}

    \item{A new EVEX encoding prefix. EVEX is based on VEX and provides more
        capabilities: opmasks, broadcasting, embedded rounding and compressed
        displacements.
\begin{lstlisting}
; opmask: conditional vector operation
; using opmask k1 and {z} is for zero-masking
    VDIVPD zmm0{k1}{z}, zmm1, zmm3

; broadcasting: load single-precision float and
; replicate it 16 times. 32 * 16 = 512
    VDIVPS zmm4, zmm5, [rbx]{1to16}

; embedded rounding: round toward zero. note that
: it is used as if a separate operand. it comes after
; the last SIMD operand
    VCVTSI2SD xmm6, xmm7, {rz-sae}, rax
\end{lstlisting}}

    \item{Add support for \code{ZWORD} (512 bits), \code{DZ} and \code{RESZ}.}

    \item{Add support for the MPX and SHA instruction sets.}

    \item{Better handling of section redefinition.}

    \item{Generate manpages when running \code{'make dist'}.}

    \item{Handle all token chains in mmacro params range.}

    \item{Support split [base,index] effective address:
\begin{lstlisting}
; eax=base, ecx=index, 4=scale, 8=disp
mov eax,[eax+8,ecx*4]
\end{lstlisting}
        This is expected to be most useful for the MPX instructions.}

    \item{Support \code{BND} prefix for branch instructions (for MPX).}

    \item{The \code{DEFAULT} directive can now take \code{BND} and \code{NOBND}
        options to indicate whether all relevant branches should be getting
        \code{BND} prefixes. This is expected to be the normal for use in MPX
        code.}

    \item{Add \code{{evex}}, \code{{vex3}} and \code{{vex2}} instruction prefixes to
        have NASM encode the corresponding instruction, if possible, with an EVEX,
        3-byte VEX, or 2-byte VEX prefix, respectively.}

    \item{Support for section names longer than 8 bytes in Win32/Win64 COFF.}

    \item{The \code{NOSPLIT} directive by itself no longer forces a single
        register to become an index register, unless it has an explicit
        multiplier.
\begin{lstlisting}
mov eax,[nosplit eax]   ; eax as base register
mov eax,[nosplit eax*1] ; eax as index register
\end{lstlisting}}
\end{itemize}

\xsubsection{cl-2.10.09}{Version 2.10.09}

\begin{itemize}
    \item{Pregenerate man pages.}
\end{itemize}

\xsubsection{cl-2.10.08}{Version 2.10.08}

\begin{itemize}
    \item{Fix \code{VMOVNTDQA}, \code{MOVNTDQA} and \code{MOVLPD} instructions.}
    \item{Fix collision for \code{VGATHERQPS}, \code{VPGATHERQD} instructions.}
    \item{Fix \code{VPMOVSXBQ}, \code{VGATHERQPD}, \code{VSPLLW} instructions.}
    \item{Add a bunch of AMD TBM instructions.}
    \item{Fix potential stack overwrite in numbers conversion.}
    \item{Allow byte size in \code{PREFETCHTx} instructions.}
    \item{Make manual pages up to date.}
    \item{Make \code{F3} and \code{F2} SSE prefixes to override \code{66}.}
    \item{Support of AMD SVM instructions in 32 bit mode.}
    \item{Fix near offsets code generation for \code{JMP}, \code{CALL} instrictions in long mode.}
    \item{Fix preprocessor parse regression when id is expanding to a whitespace.}
\end{itemize}

\xsubsection{cl-2.10.07}{Version 2.10.07}

\begin{itemize}
    \item{Fix line continuation parsing being broken in previous version.}
\end{itemize}

\xsubsection{cl-2.10.06}{Version 2.10.06}

\begin{itemize}
    \item{Always quote the dependency source names when using the automatic
        dependency generation options.}
    \item{If no dependency target name is specified via the \code{-MT} or
        \code{-MQ} options, quote the default output name.}
    \item{Fix assembly of shift operations in \code{CPU 8086} mode.}
    \item{Fix incorrect generation of explicit immediate byte for shift by 1
        under certain circumstances.}
    \item{Fix assembly of the \code{VPCMPGTQ} instruction.}
    \item{Fix RIP-relative relocations in the \code{macho64} backend.}
\end{itemize}

\xsubsection{cl-2.10.05}{Version 2.10.05}

\begin{itemize}
    \item{Add the \code{CLAC} and \code{STAC} instructions.}
\end{itemize}

\xsubsection{cl-2.10.04}{Version 2.10.04}

\begin{itemize}
    \item{Add back the inadvertently deleted 256-bit version of the \code{VORPD} instruction.}
    \item{Correct disassembly of instructions starting with byte \code{82} hex.}
    \item{Fix corner cases in token pasting, for example:
\begin{lstlisting}
%define N 1e%++%+ 5
dd N, 1e+5
\end{lstlisting}}
\end{itemize}

\xsubsection{cl-2.10.03}{Version 2.10.03}

\begin{itemize}
    \item{Correct the assembly of the instruction:
\begin{lstlisting}
XRELEASE MOV [absolute],AL
\end{lstlisting}
    Previous versions would incorrectly generate \code{F3 A2} for this
    instruction and issue a warning; correct behavior is to emit \code{F3 88 05}.}
\end{itemize}

\xsubsection{cl-2.10.02}{Version 2.10.02}

\begin{itemize}
    \item{Add the \code{ifunc} macro package with integer functions, currently
        only integer logarithms. See \nref{pkgifunc}.}
    \item{Add the \code{RDSEED}, \code{ADCX} and \code{ADOX} instructions.}
\end{itemize}

\xsubsection{cl-2.10.01}{Version 2.10.01}

\begin{itemize}
    \item{Add missing VPMOVMSKB instruction with reg32, ymmreg operands.}
\end{itemize}

\xsubsection{cl-2.10}{Version 2.10}

\begin{itemize}
    \item{When optimization is enabled, \code{mov r64,imm} now optimizes to the
        shortest form possible between:
\begin{lstlisting}
mov r32,imm32   ;  5 bytes
mov r64,imm32   ;  7 bytes
mov r64,imm64   ; 10 bytes
\end{lstlisting}
        To force a specific form, use the \code{STRICT} keyword, see \nref{strict}.}
    \item{Add support for the Intel AVX2 instruction set.}
    \item{Add support for Bit Manipulation Instructions 1 and 2.}
    \item{Add support for Intel Transactional Synchronization Extensions (TSX).}
    \item{Add support for x32 ELF (32-bit ELF with the CPU in 64-bit mode.) See \nref{elffmt}.}
    \item{Add support for bigendian UTF-16 and UTF-32. See \nref{unicode}.}
\end{itemize}

\xsubsection{cl-2.09.10}{Version 2.09.10}

\begin{itemize}
    \item{Fix up NSIS script to protect uninstaller against registry keys
        absence or corruption. It brings in a few additional questions
        to a user during deinstallation procedure but still it is better
        than unpredictable file removal.}
\end{itemize}

\xsubsection{cl-2.09.09}{Version 2.09.09}

\begin{itemize}
    \item{Fix initialization of section attributes of \code{bin} output format.}
    \item{Fix \code{mach64} output format bug that crashes NASM due to NULL symbols.}
\end{itemize}

\xsubsection{cl-2.09.08}{Version 2.09.08}

\begin{itemize}
    \item{Fix \code{\_\_OUTPUT\_FORMAT\_\_} assignment when output driver alias
        is used. For example when \code{-f elf} is used \code{\_\_OUTPUT\_FORMAT\_\_}
        must be set to \code{elf}, if \code{-f elf32} is used \code{\_\_OUTPUT\_FORMAT\_\_}
        must be assigned accordingly, i.e. to \code{elf32}. The rule applies to
        all output driver aliases. See \nref{ofmtm}.}
\end{itemize}

\xsubsection{cl-2.09.07}{Version 2.09.07}

\begin{itemize}
    \item{Fix attempts to close same file several times when \code{-a} option is used.}
    \item{Fixes for VEXTRACTF128, VMASKMOVPS encoding.}
\end{itemize}

\xsubsection{cl-2.09.06}{Version 2.09.06}

\begin{itemize}
    \item{Fix missed section attribute initialization in \code{bin} output target.}
\end{itemize}

\xsubsection{cl-2.09.05}{Version 2.09.05}

\begin{itemize}
    \item{Fix arguments encoding for VPEXTRW instruction.}
    \item{Remove invalid form of VPEXTRW instruction.}
    \item{Add \code{VLDDQU} as alias for \code{VLDQQU} to match specification.}
\end{itemize}

\xsubsection{cl-2.09.04}{Version 2.09.04}

\begin{itemize}
    \item{Fix incorrect labels offset for VEX intructions.}
    \item{Eliminate bogus warning on implicit operand size override.}
    \item{\code{\%if} term could not handle 64 bit numbers.}
    \item{The COFF backend was limiting relocations number to 16 bits even if
        in real there were a way more relocations.}
\end{itemize}

\xsubsection{cl-2.09.03}{Version 2.09.03}

\begin{itemize}
    \item{Print \code{\%macro} name inside \code{\%rep} blocks on error.}
    \item{Fix preprocessor expansion behaviour. It happened sometime
        too early and sometime simply wrong. Move behaviour back to
        the origins (down to NASM 2.05.01).}
    \item{Fix unitialized data dereference on OMF output format.}
    \item{Issue warning on unterminated \code{\%\{} construct.}
    \item{Fix for documentation typo.}
\end{itemize}

\xsubsection{cl-2.09.02}{Version 2.09.02}

\begin{itemize}
    \item{Fix reversed tokens when \code{\%deftok} produces more than one output token.}
    \item{Fix segmentation fault on disassembling some VEX instructions.}
    \item{Missing \code{\%endif} did not always cause error.}
    \item{Fix typo in documentation.}
    \item{Compound context local preprocessor single line macro identifiers
        were not expanded early enough and as result lead to unresolved symbols.}
\end{itemize}

\xsubsection{cl-2.09.01}{Version 2.09.01}

\begin{itemize}
    \item{Fix NULL dereference on missed \code{\%deftok} second parameter.}
    \item{Fix NULL dereference on invalid \code{\%substr} parameters.}
\end{itemize}

\xsubsection{cl-2.09}{Version 2.09}

\begin{itemize}
    \item{Fixed assignment the magnitude of \code{\%rep} counter.
        It is limited to 62 bits now.}
    \item{Fixed NULL dereference if argument of \code{\%strlen} resolves
        to whitespace. For example if nonexistent macro parameter is used.}
    \item{\code{\%ifenv}, \code{\%elifenv}, \code{\%ifnenv}, and
        \code{\%elifnenv} directives introduced. See \nref{ifenv}.}
    \item{Fixed NULL dereference if environment variable is missed.}
    \item{Updates of new AVX v7 Intel instructions.}
    \item{\code{PUSH imm32} is now officially documented.}
    \item{Fix for encoding the LFS, LGS and LSS in 64-bit mode.}
    \item{Fixes for compatibility with OpenWatcom compiler and DOS 8.3 file format limitation.}
    \item{Macros parameters range expansion introduced. See \nref{mlmacrange}.}
    \item{Backward compatibility on expanging of local sigle macros restored.}
    \item{8 bit relocations for \code{elf} and \code{bin} output formats are introduced.}
    \item{Short intersegment jumps are permitted now.}
    \item{An alignment more than 64 bytes are allowed for \code{win32},
        \code{win64} output formats.}
    \item{\code{SECTALIGN} directive introduced. See \nref{sectalign}.}
    \item{\code{nojmp} option introduced in \code{smartalign} package. See
        \nref{pkgsmartalign}.}
    \item{Short aliases \code{win}, \code{elf} and \code{macho} for output formats are
        introduced. Each stands for \code{win32}, \code{elf32} and \code{macho32}
        accordingly.}
    \item{Faster handling of missing directives implemented.}
    \item{Various small improvements in documentation.}
    \item{No hang anymore if unable to open malloc.log file.}
    \item{The environments without vsnprintf function are able to build nasm again.}
    \item{AMD LWP instructions updated.}
    \item{Tighten EA checks. We warn a user if there overflow in EA addressing.}
    \item{Make \code{-Ox} the default optimization level. For the legacy
        behavior, specify \code{-O0} explicitly. See \nref{opt-O}.}
    \item{Environment variables read with \code{\%!} or tested with \code{\%ifenv}
        can now contain non-identifier characters if surrounded by quotes.
        See \nref{getenv}.}
    \item{Add a new standard macro package \code{\%use fp} for floating-point
        convenience macros. See \nref{pkgfp}.}
\end{itemize}

\xsubsection{cl-2.08.02}{Version 2.08.02}

\begin{itemize}
    \item{Fix crash under certain circumstances when using
        the \code{\%+} operator.}
\end{itemize}

\xsubsection{cl-2.08.01}{Version 2.08.01}

\begin{itemize}
    \item{Fix the \code{\%use} statement, which was broken in 2.08.}
\end{itemize}

\xsubsection{cl-2.08}{Version 2.08}

\begin{itemize}
    \item{A number of enhancements/fixes in macros area.}
    \item{Support for converting strings to tokens. See \nref{deftok}.}
    \item{Fuzzy operand size logic introduced.}
    \item{Fix COFF stack overrun on too long export identifiers.}
    \item{Fix Macho-O alignment bug.}
    \item{Fix crashes with \code{-fwin32} on file with many exports.}
    \item{Fix stack overrun for too long [DEBUG id].}
    \item{Fix incorrect sbyte usage in IMUL (hit only if optimization flag passed).}
    \item{Append ending token for \code{.stabs} records in the ELF output format.}
    \item{New NSIS script which uses ModernUI and MultiUser approach.}
    \item{Visual Studio 2008 NASM integration (rules file).}
    \item{Warn a user if a constant is too long (and as result will be stripped).}
    \item{The obsoleted pre-XOP AMD SSE5 instruction set which was never actualized was removed.}
    \item{Fix stack overrun on too long error file name passed from the command line.}
    \item{Bind symbols to the .text section by default (ie in case if
        \code{SECTION} directive was omitted) in the ELF output format.}
    \item{Fix sync points array index wrapping.}
    \item{A few fixes for FMA4 and XOP instruction templates.}
    \item{Add AMD Lightweight Profiling (LWP) instructions.}
    \item{Fix the offset for \code{\%arg} in 64-bit mode.}
    \item{An undefined local macro (\code{\%\$}) no longer matches a global macro
            with the same name.}
        \item{Fix NULL dereference on too long local labels.}
\end{itemize}

\xsubsection{cl-2.07}{Version 2.07}

\begin{itemize}
    \item{NASM is now under the 2-clause BSD license. See \nref{license}.}
    \item{Fix the section type for the \code{.strtab} section in the \code{elf64}
        output format.}
    \item{Fix the handling of \code{COMMON} directives in the \code{obj} output format.}
    \item{New \code{ith} and \code{srec} output formats; these are variants of the
        \code{bin} output format which output Intel hex and Motorola S-records,
        respectively. See \nref{ithfmt} and \nref{srecfmt}.}
    \item{\code{rdf2ihx} replaced with an enhanced \code{rdf2bin}, which can output
        binary, COM, Intel hex or Motorola S-records.}
    \item{The Windows installer now puts the NASM directory first in the
        \code{PATH} of the ``NASM Shell''.}
    \item{Revert the early expansion behavior of \code{\%+} to
        pre-2.06 behavior: \code{\%+} is only expanded late.}
    \item{Yet another Mach-O alignment fix.}
    \item{Don't delete the list file on errors. Also, include error and
        warning information in the list file.}
    \item{Support for 64-bit Mach-O output, see \nref{machofmt}.}
    \item{Fix assert failure on certain operations that involve strings with
        high-bit bytes.}
\end{itemize}

\xsubsection{cl-2.06}{Version 2.06}

\begin{itemize}
    \item{This release is dedicated to the memory of Charles A. Crayne, long
        time NASM developer as well as moderator of \code{comp.lang.asm.x86} and
        author of the book \emph{Serious Assembler}. We miss you, Chuck.}
    \item{Support for indirect macro expansion (\code{\%[...]}). See \nref{indmacro}.}
    \item{\code{\%pop} can now take an argument, see \nref{pushpop}.}
    \item{The argument to \code{\%use} is no longer macro-expanded. Use
        \code{\%[...]} if macro expansion is desired.}
    \item{Support for thread-local storage in ELF32 and ELF64. See \nref{elftls}.}
    \item{Fix crash on \code{\%ifmacro} without an argument.}
    \item{Correct the arguments to the \code{POPCNT} instruction.}
    \item{Fix section alignment in the Mach-O format.}
    \item{Update AVX support to version 5 of the Intel specification.}
    \item{Fix the handling of accesses to context-local macros from higher
        levels in the context stack.}
    \item{Treat \code{WAIT} as a prefix rather than as an instruction, thereby
        allowing constructs like \code{O16 FSAVE} to work correctly.}
    \item{Support for structures with a non-zero base offset. See \nref{struc}.}
    \item{Correctly handle preprocessor token concatenation (see \nref{concat})
        involving floating-point numbers.}
    \item{The \code{PINSR} series of instructions have been corrected and rationalized.}
    \item{Removed AMD SSE5, replaced with the new XOP/FMA4/CVT16 (rev 3.03) spec.}
    \item{The ELF backends no longer automatically generate a \code{.comment} section.}
    \item{Add additional "well-known" ELF sections with default attributes. See \nref{elfsect}.}
\end{itemize}

\xsubsection{cl-2.05.01}{Version 2.05.01}

\begin{itemize}
    \item{Fix the \code{-w}/\code{-W} option parsing, which was broken in NASM 2.05.}
\end{itemize}

\xsubsection{cl-2.05}{Version 2.05}

\begin{itemize}
    \item{Fix redundant REX.W prefix on \code{JMP reg64}.}
    \item{Make the behaviour of \code{-O0} match NASM 0.98 legacy behavior.
        See \nref{opt-O}.}
    \item{\code{-w-user} can be used to suppress the output of \code{\%warning} directives.
        See \nref{opt-w}.}
    \item{Fix bug where \code{ALIGN} would issue a full alignment datum instead of
        zero bytes.}
    \item{Fix offsets in list files.}
    \item{Fix \code{\%include} inside multi-line macros or loops.}
    \item{Fix error where NASM would generate a spurious warning on valid
        optimizations of immediate values.}
    \item{Fix arguments to a number of the \code{CVT} SSE instructions.}
    \item{Fix RIP-relative offsets when the instruction carries an immediate.}
    \item{Massive overhaul of the ELF64 backend for spec compliance.}
    \item{Fix the Geode \code{PFRCPV} and \code{PFRSQRTV} instruction.}
    \item{Fix the SSE 4.2 \code{CRC32} instruction.}
\end{itemize}


\xsubsection{cl-2.04}{Version 2.04}

\begin{itemize}
    \item{Sanitize macro handing in the \code{\%error} directive.}
    \item{New \code{\%warning} directive to issue user-controlled warnings.}
    \item{\code{\%error} directives are now deferred to the final assembly phase.}
    \item{New \code{\%fatal} directive to immediately terminate assembly.}
    \item{New \code{\%strcat} directive to join quoted strings together.}
    \item{New \code{\%use} macro directive to support standard macro directives. See \nref{use}.}
    \item{Excess default parameters to \code{\%macro} now issues a warning by default.
        See \nref{mlmacro}.}
    \item{Fix \code{\%ifn} and \code{\%elifn}.}
    \item{Fix nested \code{\%else} clauses.}
    \item{Correct the handling of nested \code{\%rep}s.}
    \item{New \code{\%unmacro} directive to undeclare a multi-line macro.
        See \nref{unmacro}.}
    \item{Builtin macro \code{\_\_PASS\_\_} which expands to the current assembly pass.
        See \nref{passdef}.}
    \item{\code{\_\_utf16\_\_} and \code{\_\_utf32\_\_} operators to generate
        UTF-16 and UTF-32 strings. See \nref{unicode}.}
    \item{Fix bug in case-insensitive matching when compiled on platforms that
        don't use the \code{configure} script. Of the official release binaries,
        that only affected the OS/2 binary.}
    \item{Support for x87 packed BCD constants. See \nref{bcdconst}.}
    \item{Correct the \code{LTR} and \code{SLDT} instructions in 64-bit mode.}
    \item{Fix unnecessary REX.W prefix on indirect jumps in 64-bit mode.}
    \item{Add AVX versions of the AES instructions (\code{VAES}...).}
    \item{Fix the 256-bit FMA instructions.}
    \item{Add 256-bit AVX stores per the latest AVX spec.}
    \item{VIA XCRYPT instructions can now be written either with or without
        \code{REP}, apparently different versions of the VIA spec wrote them
        differently.}
    \item{Add missing 64-bit \code{MOVNTI} instruction.}
    \item{Fix the operand size of \code{VMREAD} and \code{VMWRITE}.}
    \item{Numerous bug fixes, especially to the AES, AVX and VTX instructions.}
    \item{The optimizer now always runs until it converges. It also runs even
        when disabled, but doesn't optimize. This allows most forward references
        to be resolved properly.}
    \item{\code{\%push} no longer needs a context identifier; omitting the context
        identifier results in an anonymous context.}
\end{itemize}

\xsubsection{cl-2.03.01}{Version 2.03.01}

\begin{itemize}
    \item{Fix buffer overflow in the listing module.}
    \item{Fix the handling of hexadecimal escape codes in `...` strings.}
    \item{The Postscript/PDF documentation has been reformatted.}
    \item{The \code{-F} option now implies \code{-g}.}
\end{itemize}

\xsubsection{cl-2.03}{Version 2.03}

\begin{itemize}
    \item{Add support for Intel AVX, CLMUL and FMA instructions,
        including YMM registers.}
    \item{\code{dy}, \code{resy} and \code{yword} for 32-byte operands.}
    \item{Fix some SSE5 instructions.}
    \item{Intel \code{INVEPT}, \code{INVVPID} and \code{MOVBE} instructions.}
    \item{Fix checking for critical expressions when the optimizer is enabled.}
    \item{Support the DWARF debugging format for ELF targets.}
    \item{Fix optimizations of signed bytes.}
    \item{Fix operation on bigendian machines.}
    \item{Fix buffer overflow in the preprocessor.}
    \item{\code{SAFESEH} support for Win32, \code{IMAGEREL} for Win64 (SEH).}
    \item{\code{\%?} and \code{\%??} to refer to the name of a macro itself.
        In particular, \code{\%idefine keyword \$\%?} can be used to make
        a keyword "disappear".}
    \item{New options for dependency generation: \code{-MD}, \code{-MF},
        \code{-MP}, \code{-MT}, \code{-MQ}.}
    \item{New preprocessor directives \code{\%pathsearch} and \code{\%depend};
        \code{INCBIN} reimplemented as a macro.}
    \item{\code{\%include} now resolves macros in a sane manner.}
    \item{\code{\%substr} can now be used to get other than one-character substrings.}
    \item{New type of character/string constants, using backquotes (\code{`...`}),
        which support C-style escape sequences.}
    \item{\code{\%defstr} and \code{\%idefstr} to stringize macro definitions
        before creation.}
    \item{Fix forward references used in \code{EQU} statements.}
\end{itemize}

\xsubsection{cl-2.02}{Version 2.02}

\begin{itemize}
    \item{Additional fixes for MMX operands with explicit \code{qword}, as well as
        (hopefully) SSE operands with \code{oword}.}
    \item{Fix handling of truncated strings with \code{DO}.}
    \item{Fix segfaults due to memory overwrites when floating-point constants were used.}
    \item{Fix segfaults due to missing include files.}
    \item{Fix OpenWatcom Makefiles for DOS and OS/2.}
    \item{Add autogenerated instruction list back into the documentation.}
    \item{ELF: Fix segfault when generating stabs, and no symbols have been defined.}
    \item{ELF: Experimental support for DWARF debugging information.}
    \item{New compile date and time standard macros.}
    \item{\code{\%ifnum} now returns true for negative numbers.}
    \item{New \code{\%iftoken} test for a single token.}
    \item{New \code{\%ifempty} test for empty expansion.}
    \item{Add support for the \code{XSAVE} instruction group.}
    \item{Makefile for Netware/gcc.}
    \item{Fix issue with some warnings getting emitted way too many times.}
    \item{Autogenerated instruction list added to the documentation.}
\end{itemize}

\xsubsection{cl-2.01}{Version 2.01}

\begin{itemize}
    \item{Fix the handling of MMX registers with explicit \code{qword} tags on
        memory (broken in 2.00 due to 64-bit changes.)}
    \item{Fix the PREFETCH instructions.}
    \item{Fix the documentation.}
    \item{Fix debugging info when using \code{-f elf}
        (backwards compatibility alias for \code{-f elf32}).}
    \item{Man pages for rdoff tools (from the Debian project.)}
    \item{ELF: handle large numbers of sections.}
    \item{Fix corrupt output when the optimizer runs out of passes.}
\end{itemize}

\xsubsection{cl-2.00}{Version 2.00}

\begin{itemize}
    \item{Added c99 data-type compliance.}
    \item{Added general x86-64 support.}
    \item{Added win64 (x86-64 COFF) output format.}
    \item{Added \code{\_\_BITS\_\_} standard macro.}
    \item{Renamed the \code{elf} output format to \code{elf32} for clarity.}
    \item{Added \code{elf64} and \code{macho} (MacOS X) output formats.}
    \item{Added Numeric constants in \code{dq} directive.}
    \item{Added \code{oword}, \code{do} and \code{reso} pseudo operands.}
    \item{Allow underscores in numbers.}
    \item{Added 8-, 16- and 128-bit floating-point formats.}
    \item{Added binary, octal and hexadecimal floating-point.}
    \item{Correct the generation of floating-point constants.}
    \item{Added floating-point option control.}
    \item{Added Infinity and NaN floating point support.}
    \item{Added ELF Symbol Visibility support.}
    \item{Added setting OSABI value in ELF header directive.}
    \item{Added Generate Makefile Dependencies option.}
    \item{Added Unlimited Optimization Passes option.}
    \item{Added \code{\%ifn} and \code{\%elifn} support.}
    \item{Added Logical Negation Operator.}
    \item{Enhanced Stack Relative Preprocessor Directives.}
    \item{Enhanced ELF Debug Formats.}
    \item{Enhanced Send Errors to a File option.}
    \item{Added SSSE3, SSE4.1, SSE4.2, SSE5 support.}
    \item{Added a large number of additional instructions.}
    \item{Significant performance improvements.}
    \item{\code{-w+warning} and \code{-w-warning} can now be written
        as -Wwarning and -Wno-warning, respectively. See \nref{opt-w}.}
    \item{Add \code{-w+error} to treat warnings as errors. See \nref{opt-w}.}
    \item{Add \code{-w+all} and \code{-w-all} to enable or disable all suppressible
        warnings. See \nref{opt-w}.}
\end{itemize}

\xsection{cl-0.98.xx}{NASM 0.98 Series}

The 0.98 series was the production versions of NASM from 1999 to 2007.

\xsubsection{cl-0.98.39}{Version 0.98.39}

\begin{itemize}
    \item{fix buffer overflow}
    \item{fix outas86's \code{.bss} handling}
    \item{"make spotless" no longer deletes config.h.in.}
    \item{\code{\%(el)if(n)idn} insensitivity to string quotes difference (\#809300).}
    \item{(nasm.c)\code{\_\_OUTPUT\_FORMAT\_\_} changed to string value instead of symbol.}
\end{itemize}

\xsubsection{cl-0.98.38}{Version 0.98.38}

\begin{itemize}
    \item{Add Makefile for 16-bit DOS binaries under OpenWatcom, and modify
        \code{mkdep.pl} to be able to generate completely pathless dependencies, as
        required by OpenWatcom wmake (it supports path searches, but not
        explicit paths.)}
    \item{Fix the \code{STR} instruction.}
    \item{Fix the ELF output format, which was broken under certain
        circumstances due to the addition of stabs support.}
    \item{Quick-fix Borland format debug-info for \code{-f obj}}
    \item{Fix for \code{\%rep} with no arguments (\#560568)}
    \item{Fix concatenation of preprocessor function call (\#794686)}
    \item{Fix long label causes coredump (\#677841)}
    \item{Use autoheader as well as autoconf to keep configure from generating
        ridiculously long command lines.}
    \item{Make sure that all of the formats which support debugging output
        actually will suppress debugging output when \code{-g} not specified.}
\end{itemize}

\xsubsection{cl-0.98.37}{Version 0.98.37}

\begin{itemize}
    \item{Paths given in \code{-I} switch searched for \code{incbin}-ed as
        well as \code{\%include}-ed files.}
    \item{Added stabs debugging for the ELF output format, patch from
        Martin Wawro.}
    \item{Fix \code{output/outbin.c} to allow origin > 80000000h.}
    \item{Make \code{-U} switch work.}
    \item{Fix the use of relative offsets with explicit prefixes, e.g. 
        \code{a32 loop foo}.}
    \item{Remove \code{backslash()}.}
    \item{Fix the \code{SMSW} and \code{SLDT} instructions.}
    \item{\code{-O2} and \code{-O3} are no longer aliases for \code{-O10} and \code{-O15}.
        If you mean the latter, please say so! :)}
\end{itemize}

\xsubsection{cl-0.98.36}{Version 0.98.36}

\begin{itemize}
    \item{Update rdoff - librarian/archiver - common rec - docs!}
    \item{Fix signed/unsigned problems.}
    \item{Fix \code{JMP FAR label} and \code{CALL FAR label}.}
    \item{Add new multisection support - map files - fix align bug}
    \item{Fix sysexit, movhps/movlps reg,reg bugs in insns.dat}
    \item{\code{Q} or \code{O} suffixes indicate octal}
    \item{Support Prescott new instructions (PNI).}
    \item{Cyrix \code{XSTORE} instruction.}
\end{itemize}


\xsubsection{cl-0.98.35}{Version 0.98.35}

\begin{itemize}
    \item{Fix build failure on 16-bit DOS (Makefile.bc3 workaround for compiler bug.)}
    \item{Fix dependencies and compiler warnings.}
    \item{Add "const" in a number of places.}
    \item{Add -X option to specify error reporting format (use -Xvc to
        integrate with Microsoft Visual Studio.)}
    \item{Minor changes for code legibility.}
    \item{Drop use of tmpnam() in rdoff (security fix.)}
\end{itemize}

\xsubsection{cl-0.98.34}{Version 0.98.34}

\begin{itemize}
    \item{Correct additional address-size vs. operand-size confusions.}
    \item{Generate dependencies for all Makefiles automatically.}
    \item{Add support for unimplemented (but theoretically available)
        registers such as tr0 and cr5. Segment registers 6 and 7 are called
        segr6 and segr7 for the operations which they can be represented.}
    \item{Correct some disassembler bugs related to redundant address-size prefixes.
        Some work still remains in this area.}
    \item{Correctly generate an error for things like "SEG eax".}
    \item{Add the JMPE instruction, enabled by "CPU IA64".}
    \item{Correct compilation on newer gcc/glibc platforms.}
    \item{Issue an error on things like "jmp far eax".}
\end{itemize}

\xsubsection{cl-0.98.33}{Version 0.98.33}

\begin{itemize}
    \item{New \_\_NASM\_PATCHLEVEL\_\_ and \_\_NASM\_VERSION\_ID\_\_ standard macros
        to round out the version-query macros. version.pl now understands
        X.YYplWW or X.YY.ZZplWW as a version number, equivalent to
        X.YY.ZZ.WW (or X.YY.0.WW, as appropriate).}
    \item{New keyword "strict" to disable the optimization of specific operands.}
    \item{Fix the handing of size overrides with JMP instructions
        (instructions such as "jmp dword foo".)}
    \item{Fix the handling of "ABSOLUTE label", where "label" points into a
        relocatable segment.}
    \item{Fix OBJ output format with lots of externs.}
    \item{More documentation updates.}
    \item{Add -Ov option to get verbose information about optimizations.}
    \item{Undo a braindead change which broke \code{\%elif} directives.}
    \item{Makefile updates.}
\end{itemize}

\xsubsection{cl-0.98.32}{Version 0.98.32}

\begin{itemize}
    \item{Fix NASM crashing when \code{\%macro} directives were left unterminated.}
    \item{Lots of documentation updates.}
    \item{Complete rewrite of the PostScript/PDF documentation generator.}
    \item{The MS Visual C++ Makefile was updated and corrected.}
    \item{Recognize .rodata as a standard section name in ELF.}
    \item{Fix some obsolete Perl4-isms in Perl scripts.}
    \item{Fix configure.in to work with autoconf 2.5x.}
    \item{Fix a couple of "make cleaner" misses.}
    \item{Make the normal "./configure \&\& make" work with Cygwin.}
\end{itemize}

\xsubsection{cl-0.98.31}{Version 0.98.31}

\begin{itemize}
    \item{Correctly build in a separate object directory again.}
    \item{Derive all references to the version number from the version file.}
    \item{New standard macros \code{\_\_NASM\_SUBMINOR\_\_} and
        \code{\_\_NASM\_VER\_\_} macros.}
    \item{Lots of Makefile updates and bug fixes.}
    \item{New \code{\%ifmacro} directive to test for multiline macros.}
    \item{Documentation updates.}
    \item{Fixes for 16-bit OBJ format output.}
    \item{Changed the NASM environment variable to NASMENV.}
\end{itemize}

\xsubsection{cl-0.98.30}{Version 0.98.30}

\begin{itemize}
    \item{Changed doc files a lot: completely removed old READMExx and
        Wishlist files, incorporating all information in CHANGES and TODO.}
    \item{I waited a long time to rename zoutieee.c to (original) outieee.c}
    \item{moved all output modules to output/ subdirectory.}
    \item{Added 'make strip' target to strip debug info from nasm \& ndisasm.}
    \item{Added INSTALL file with installation instructions.}
    \item{Added -v option description to nasm man.}
    \item{Added dist makefile target to produce source distributions.}
    \item{16-bit support for ELF output format (GNU extension, but useful.)}
\end{itemize}


\xsubsection{cl-0.98.28}{Version 0.98.28}

\begin{itemize}
    \item{Fastcooked this for Debian's Woody release:
        Frank applied the \code{INCBIN} bug patch to 0.98.25alt
        and called it 0.98.28 to not confuse poor little apt-get.}
\end{itemize}

\xsubsection{cl-0.98.26}{Version 0.98.26}

\begin{itemize}
    \item{Reorganised files even better from 0.98.25alt}
\end{itemize}


\xsubsection{cl-0.98.25alt}{Version 0.98.25alt}

\begin{itemize}
    \item{Prettified the source tree. Moved files to more reasonable places.}
    \item{Added findleak.pl script to misc/ directory.}
    \item{Attempted to fix doc.}
\end{itemize}

\xsubsection{cl-0.98.25}{Version 0.98.25}

\begin{itemize}
    \item{Line continuation character \code{\\}.}
    \item{Docs inadvertantly reverted - "dos packaging".}
\end{itemize}


\xsubsection{cl-0.98.24p1}{Version 0.98.24p1}

\begin{itemize}
    \item{FIXME: Someone, document this please.}
\end{itemize}

\xsubsection{cl-0.98.24}{Version 0.98.24}

\begin{itemize}
    \item{Documentation - Ndisasm doc added to Nasm.doc.}
\end{itemize}

\xsubsection{cl-0.98.23}{Version 0.98.23}

\begin{itemize}
    \item{Attempted to remove rdoff version1}
    \item{Lino Mastrodomenico's patches to preproc.c (\%\$\$ bug?).}
\end{itemize}


\xsubsection{cl-0.98.22}{Version 0.98.22}

\begin{itemize}
    \item{Update rdoff2 - attempt to remove v1.}
\end{itemize}

\xsubsection{cl-0.98.21}{Version 0.98.21}

\begin{itemize}
    \item{Optimization fixes.}
\end{itemize}

\xsubsection{cl-0.98.20}{Version 0.98.20}

\begin{itemize}
    \item{Optimization fixes.}
\end{itemize}

\xsubsection{cl-0.98.19}{Version 0.98.19}

\begin{itemize}
    \item{H. J. Lu's patch back out.}
\end{itemize}

\xsubsection{cl-0.98.18}{Version 0.98.18}

\begin{itemize}
    \item{Added ".rdata" to "-f win32".}
\end{itemize}

\xsubsection{cl-0.98.17}{Version 0.98.17}

\begin{itemize}
    \item{H. J. Lu's "bogus elf" patch. (Red Hat problem?)}
\end{itemize}


\xsubsection{cl-0.98.16}{Version 0.98.16}

\begin{itemize}
    \item{Fix whitespace before \code{[section \dots]} bug.}
\end{itemize}

\xsubsection{cl-0.98.15}{Version 0.98.15}

\begin{itemize}
    \item{Rdoff changes (?).}
    \item{Fix fixes to memory leaks.}
\end{itemize}

\xsubsection{cl-0.98.14}{Version 0.98.14}

\begin{itemize}
    \item{Fix memory leaks.}
\end{itemize}

\xsubsection{cl-0.98.13}{Version 0.98.13}

\begin{itemize}
    \item{There was no 0.98.13}
\end{itemize}


\xsubsection{cl-0.98.12}{Version 0.98.12}

\begin{itemize}
    \item{Update optimization (new function of \code{-O1}).}
    \item{Changes to test/bintest.asm (?).}
\end{itemize}

\xsubsection{cl-0.98.11}{Version 0.98.11}

\begin{itemize}
    \item{Optimization changes.}
    \item{Ndisasm fixed.}
\end{itemize}

\xsubsection{cl-0.98.10}{Version 0.98.10}

\begin{itemize}
    \item{There was no 0.98.10}
\end{itemize}

\xsubsection{cl-0.98.09}{Version 0.98.09}

\begin{itemize}
    \item{Add multiple sections support to \code{-f bin}.}
    \item{Changed \code{GLOBAL\_TEMP\_BASE} in outelf.c from 6 to 15.}
    \item{Add \code{-v} as an alias to the \code{-r} switch.}
    \item{Remove \code{\#ifdef} from Tasm compatibility options.}
    \item{Remove redundant size-overrides on \code{mov ds, ex}, etc.}
    \item{Fixes to SSE2, other insns.dat (?).}
    \item{Enable uppercase "I" and "P" switches.}
    \item{Case insinsitive \code{seg} and \code{wrt}.}
    \item{Update install.sh (?).}
    \item{Allocate tokens in blocks.}
    \item{Improve "invalid effective address" messages.}
\end{itemize}

\xsubsection{cl-0.98.08}{Version 0.98.08}

\begin{itemize}
    \item{Add \code{\%strlen} and \code{\%substr} macro operators.}
    \item{Fixed broken c16.mac.}
    \item{Unterminated string error reported.}
    \item{Fixed bugs as per 0.98bf}
\end{itemize}

\xsubsection{cl-0.98.09b}{Version 0.98.09b with John Coffman patches released 28-Oct-2001}

Changes from 0.98.07 release to 98.09b as of 28-Oct-2001

\begin{itemize}
    \item{More closely compatible with 0.98 when \code{-O0} is implied
        or specified. Not strictly identical, since backward
        branches in range of short offsets are recognized, and signed
        byte values with no explicit size specification will be
        assembled as a single byte.}
    \item{More forgiving with the PUSH instruction. 0.98 requires
        a size to be specified always. 0.98.09b will imply the size
        from the current BITS setting (16 or 32).}
    \item{Changed definition of the optimization flag
        \begin{itemize}
            \item{\code{-O0}: strict two-pass assembly, JMP and Jcc are
                handled more like 0.98, except that back ward JMPs are
                short, if possible.}
            \item{\code{-O1}: strict two-pass assembly, but forward
                branches are assembled with code guaranteed
                to reach; may produce larger code than
                \code{-O0}, but will produce successful assembly
                more often if branch offset sizes are not
                specified.}
            \item{\code{-O2}: multi-pass optimization, minimize branch
                offsets; also will minimize signed immediate bytes,
                overriding size specification.}
            \item{\code{-O3}: like \code{-O2}, but more passes taken,
                if needed}
        \end{itemize}}
\end{itemize}

\xsubsection{cl-0.98.07}{Version 0.98.07 released 01/28/01}

\begin{itemize}
    \item{Added Stepane Denis' SSE2 instructions to a \emph{working}
        version of the code - some earlier versions were based on
        broken code - sorry 'bout that. version "0.98.07"}
    \item{Cosmetic modifications to nasm.c, nasm.h, AUTHORS, MODIFIED}
\end{itemize}

\xsubsection{cl-0.98.06f}{Version 0.98.06f released 01/18/01}

\begin{itemize}
    \item{Add "metalbrain"s jecxz bug fix in insns.dat}
    \item{Alter nasmdoc.src to match - version "0.98.06f"}
\end{itemize}

\xsubsection{cl-0.98.06e}{Version 0.98.06e released 01/09/01}

\begin{itemize}
    \item{Removed the "outforms.h" file - it appears to be
        someone's old backup of "outform.h". version "0.98.06e"}
    \item{fbk - finally added the fix for the "multiple \code{\%include} bug",
        known since 7/27/99 - reported originally (?) and sent to
        us by Austin Lunnen - he reports that John Fine had a fix
        within the day. Here it is...}
    \item{Nelson Rush resigns from the group. Big thanks to Nelson for
        his leadership and enthusiasm in getting these changes
        incorporated into Nasm!}
    \item{fbk - \code{[list +]}, \code{[list -]} directives - ineptly implemented,
        should be re-written or removed, perhaps.}
    \item{Brian Raiter / fbk - "elfso bug" fix - applied to aoutb format
        as well - testing might be desirable...}
    \item{James Seter - -postfix, -prefix command line switches.}
    \item{Yuri Zaporozhets - rdoff utility changes.}
\end{itemize}

\xsubsection{cl-0.98p1}{Version 0.98p1}

\begin{itemize}
    \item{GAS-like palign (Panos Minos)}
    \item{FIXME: Someone, fill this in with details}
\end{itemize}

\xsubsection{cl-0.98bf}{Version 0.98bf (bug-fixed)}

\begin{itemize}
    \item{Fixed - elf and aoutb bug - shared libraries}
    \item{multiple \code{\%include} bug in "-f obj"}
    \item{jcxz, jecxz bug}
    \item{unrecognized option bug in ndisasm}
\end{itemize}

\xsubsection{cl-0.98.03}{Version 0.98.03 with John Coffman's changes released 27-Jul-2000}

\begin{itemize}
    \item{Added signed byte optimizations for the 0x81/0x83 class
        of instructions: ADC, ADD, AND, CMP, OR, SBB, SUB, XOR:
        when used as 'ADD reg16,imm' or 'ADD reg32,imm.'  Also
        optimization of signed byte form of 'PUSH imm' and 'IMUL
        reg,imm'/'IMUL reg,reg,imm.'  No size specification is needed.}
    \item{Added multi-pass JMP and Jcc offset optimization. Offsets
        on forward references will preferentially use the short form,
        without the need to code a specific size (short or near) for
        the branch. Added instructions for 'Jcc label' to use the
        form 'Jnotcc \$+3/JMP label', in cases where a short offset
        is out of bounds. If compiling for a 386 or higher CPU, then
        the 386 form of Jcc will be used instead.

        This feature is controlled by a new command-line switch: "O",
        (upper case letter O). "-O0" reverts the assembler to no
        extra optimization passes, "-O1" allows up to 5 extra passes,
        and "-O2"(default), allows up to 10 extra optimization passes.}
    \item{Added a new directive:  'cpu XXX', where XXX is any of:
        8086, 186, 286, 386, 486, 586, pentium, 686, PPro, P2, P3 or
        Katmai. All are case insensitive. All instructions will
        be selected only if they apply to the selected cpu or lower.
        Corrected a couple of bugs in cpu-dependence in 'insns.dat'.}
    \item{Added to 'standard.mac', the "use16" and "use32" forms of
        the "bits 16/32" directive. This is nothing new, just conforms
        to a lot of other assemblers. (minor)}
    \item{Changed label allocation from 320/32 (10000 labels @ 200K+)
        to 32/37 (1000 labels); makes running under DOS much easier.
        Since additional label space is allocated dynamically, this
        should have no effect on large programs with lots of labels.
        The 37 is a prime, believed to be better for hashing. (minor)}
\end{itemize}

\xsubsection{cl-0.98.03}{Version 0.98.03}

"Integrated patchfile 0.98-0.98.01. I call this version 0.98.03 for
historical reasons: 0.98.02 was trashed." --John Coffman
<johninsd@san.rr.com>, 27-Jul-2000

\begin{itemize}
    \item{Kendall Bennett's SciTech MGL changes}
    \item{Note that you must define "TASM\_COMPAT" at compile-time
        to get the Tasm Ideal Mode compatibility.}
    \item{All changes can be compiled in and out using the TASM\_COMPAT macros,
        and when compiled without TASM\_COMPAT defined we get the exact same
        binary as the unmodified 0.98 sources.}
    \item{standard.mac, macros.c: Added macros to ignore TASM directives before
        first include}
    \item{nasm.h: Added extern declaration for tasm\_compatible\_mode}
    \item{nasm.c: Added global variable tasm\_compatible\_mode}
    \item{Added command line switch for TASM compatible mode (-t)}
    \item{Changed version command line to reflect when compiled with TASM additions}
    \item{Added response file processing to allow all arguments on a single
        line (response file is @resp rather than -@resp for NASM format).}
    \item{labels.c: Changes islocal() macro to support TASM style @@local labels.}
    \item{Added islocalchar() macro to support TASM style @@local labels.}
    \item{parser.c: Added support for TASM style memory references (ie: mov
        [DWORD eax],10 rather than the NASM style mov DWORD [eax],10).}
    \item{preproc.c: Added new directives, \code{\%arg}, \code{\%local},
        \code{\%stacksize} to directives table}
    \item{Added support for TASM style directives without a leading \% symbol.}
    \item{Integrated a block of changes from Andrew Zabolotny <bit@eltech.ru>:}
    \item{A new keyword \code{\%xdefine} and its case-insensitive counterpart
        \code{\%ixdefine}. They work almost the same way as \code{\%define} and
        \code{\%idefine} but expand the definition immediately, not on the invocation.
        Something like a cross between \code{\%define} and \code{\%assign}. The "x"
        suffix stands for "eXpand", so "xdefine" can be deciphered as "expand-and-define".
        Thus you can do things like this:
\begin{lstlisting}
%assign ofs     0

%macro  arg     1
    %xdefine %1 dword [esp+ofs]
    %assign ofs ofs+4
%endmacro
\end{lstlisting}}
    \item{Changed the place where the expansion of \%\$name macros are expanded.
        Now they are converted into ..@ctxnum.name form when detokenizing, so
        there are no quirks as before when using \%\$name arguments to macros,
        in macros etc. For example:
\begin{lstlisting}
%macro  abc     1
    %define %1 hello
%endm
    abc     %$here
%$here
\end{lstlisting}
        Now last line will be expanded into "hello" as expected. This also allows
        for lots of goodies, a good example are extended "proc" macros included
        in this archive.}

    \item{Added a check for "cstk" in smacro\_defined() before calling get\_ctx() -
        this allows for things like:
\begin{lstlisting}
%ifdef %$abc
%endif
\end{lstlisting}
        work without warnings even in no context.}
    \item{Added a check for "cstk" in \%if*ctx and \%elif*ctx directives -
        this allows to use \code{\%ifctx} without excessive warnings. If there is
        no active context, \code{\%ifctx} goes through "false" branch.}
    \item{Removed "user error: " prefix with \code{\%error} directive: it just clobbers the
        output and has absolutely no functionality. Besides, this allows to write
        macros that does not differ from built-in functions in any way.}
    \item{Added expansion of string that is output by \code{\%error} directive. Now you
        can do things like:
\begin{lstlisting}
%define hello(x) Hello, x!

%define %$name andy
%error "hello(%$name)"
\end{lstlisting}
        Same happened with \code{\%include} directive.}
    \item{Now all directives that expect an identifier will try to expand and
        concatenate everything without whitespaces in between before usage.
        For example, with "unfixed" nasm the commands
\begin{lstlisting}
%define %$abc hello
%define __%$abc goodbye
__%$abc
\end{lstlisting}
        would produce "incorrect" output: last line will expand to
\begin{lstlisting}
hello goodbyehello
\end{lstlisting}
        Not quite what you expected, eh? :-) The answer is that preprocessor
        treats the \code{\%define} construct as if it would be
\begin{lstlisting}
%define __ %$abc goodbye
\end{lstlisting}
        (note the white space between \_\_ and \%\$abc). After my "fix" it
        will "correctly" expand into
\begin{lstlisting}
goodbye
\end{lstlisting}
        as expected. Note that I use quotes around words "correct", "incorrect"
        etc because this is rather a feature not a bug; however current behaviour
        is more logical (and allows more advanced macro usage :-).

        Same change was applied to: \code{\%push},\code{\%macro},\code{\%imacro},
        \code{\%define},\code{\%idefine},\code{\%xdefine},\code{\%ixdefine},
        \code{\%assign},\code{\%iassign},\code{\%undef}}
    \item{A new directive \code{[WARNING {+|-}warning-id]} have been added. It works only
        if the assembly phase is enabled (i.e. it doesn't work with nasm -e).}
    \item{A new warning type: macro-selfref. By default this warning is disabled;
        when enabled NASM warns when a macro self-references itself; for example
        the following source:
\begin{lstlisting}
[WARNING macro-selfref]
%macro          push    1-*
    %rep    %0
        push    %1
        %rotate 1
    %endrep
%endmacro

        push    eax,ebx,ecx
\end{lstlisting}
        will produce a warning, but if we remove the first line we won't see it
        anymore (which is The Right Thing To Do \texttrademark IMHO since C
        preprocessor eats such constructs without warnings at all).}
    \item{Added a "error" routine to preprocessor which always will set ERR\_PASS1
        bit in severity\_code. This removes annoying repeated errors on first
        and second passes from preprocessor.}
    \item{Added the \%+ operator in single-line macros for concatenating two
        identifiers. Usage example:
\begin{lstlisting}
%define _myfunc _otherfunc
%define cextern(x) _ %+ x
cextern (myfunc)
\end{lstlisting}
        After first expansion, third line will become "\_myfunc". After this
        expansion is performed again so it becomes "\_otherunc".}
    \item{Now if preprocessor is in a non-emitting state, no warning or error
        will be emitted. Example:
\begin{lstlisting}
%if 1
    mov eax,ebx
%else
    put anything you want between these two brackets,
    even macro-parameter references %1 or local
    labels %$zz or macro-local labels %%zz - no
    warning will be emitted.
%endif
\end{lstlisting}}
    \item{Context-local variables on expansion as a last resort are looked up
        in outer contexts. For example, the following piece:
\begin{lstlisting}
%push   outer
%define %$a [esp]

    %push   inner
    %$a
    %pop
%pop
\end{lstlisting}
        will expand correctly the fourth line to [esp]; if we'll define another
        \%\$a inside the "inner" context, it will take precedence over outer
        definition. However, this modification has been applied only to
        expand\_smacro and not to smacro\_define: as a consequence expansion
        looks in outer contexts, but \code{\%ifdef} won't look in outer contexts.

        This behaviour is needed because we don't want nested contexts to
        act on already defined local macros. Example:
\begin{lstlisting}
%define %$arg1  [esp+4]
test    eax,eax
if      nz
    mov     eax,%$arg1
endif
\end{lstlisting}
        In this example the "if" mmacro enters into the "if" context, so \%\$arg1
        is not valid anymore inside "if". Of course it could be worked around
        by using explicitely \%\$\$arg1 but this is ugly IMHO.}
    \item{Fixed memory leak in \code{\%undef}. The origline wasn't freed before exiting on success.}
    \item{Fixed trap in preprocessor when line expanded to empty set of tokens.
        This happens, for example, in the following case:
\begin{lstlisting}
#define SOMETHING
SOMETHING
\end{lstlisting}}
\end{itemize}

\xsubsection{cl-0.98}{Version 0.98}

All changes since NASM 0.98p3 have been produced by H. Peter Anvin <hpa@zytor.com>.

\begin{itemize}
    \item{The documentation comment delimiter is \code{\textbackslash \#} not \code{\#}.}
    \item{Allow EQU definitions to refer to external labels; reported by Pedro Gimeno.}
    \item{Re-enable support for RDOFF v1; reported by Pedro Gimeno.}
    \item{Updated License file per OK from Simon and Julian.}
\end{itemize}

\xsubsection{cl-0.98p9}{Version 0.98p9}

\begin{itemize}
    \item{Update documentation (although the instruction set reference will
        have to wait; I don't want to hold up the 0.98 release for it.)}
    \item{Verified that the NASM implementation of the PEXTRW and PMOVMSKB
        instructions is correct. The encoding differs from what the Intel
        manuals document, but the Pentium III behaviour matches NASM, not
        the Intel manuals.}
    \item{Fix handling of implicit sizes in PSHUFW and PINSRW, reported by
        Stefan Hoffmeister.}
    \item{Resurrect the -s option, which was removed when changing the
        diagnostic output to stdout.}
\end{itemize}

\xsubsection{cl-0.98p8}{Version 0.98p8}

\begin{itemize}
    \item{Fix for "DB" when NASM is running on a bigendian machine.}
    \item{Invoke insns.pl once for each output script, making Makefile.in
        legal for "make -j".}
    \item{Improve the Unix configure-based makefiles to make package
        creation easier.}
    \item{Included an RPM .spec file for building RPM (RedHat Package Manager)
        packages on Linux or Unix systems.}
    \item{Fix Makefile dependency problems.}
    \item{Change src/rdsrc.pl to include sectioning information in info
        output; required for install-info to work.}
    \item{Updated the RDOFF distribution to version 2 from Jules; minor
        massaging to make it compile in my environment.}
    \item{Split doc files that can be built by anyone with a Perl interpreter off
        into a separate archive.}
    \item{"Dress rehearsal" release!}
\end{itemize}

\xsubsection{cl-0.98p7}{Version 0.98p7}

\begin{itemize}
    \item{Fixed opcodes with a third byte-sized immediate argument to not
        complain if given "byte" on the immediate.}
    \item{Allow \code{\%undef} to remove single-line macros with arguments. This
        matches the behaviour of \#undef in the C preprocessor.}
    \item{Allow -d, -u, -i and -p to be specified as -D, -U, -I and -P for
        compatibility with most C compilers and preprocessors. This allows
        Makefile options to be shared between cc and nasm, for example.}
    \item{Minor cleanups.}
    \item{Went through the list of Katmai instructions and hopefully fixed the
        (rather few) mistakes in it.}
    \item{(Hopefully) fixed a number of disassembler bugs related to ambiguous
        instructions (disambiguated by -p) and SSE instructions with REP.}
    \item{Fix for bug reported by Mark Junger: "call dword 0x12345678" should
        work and may add an OSP (affected CALL, JMP, Jcc).}
    \item{Fix for environments when "stderr" isn't a compile-time constant.}
\end{itemize}

\xsubsection{cl-0.98p6}{Version 0.98p6}

\begin{itemize}
    \item{Took officially over coordination of the 0.98 release; so drop
        the p3.x notation. Skipped p4 and p5 to avoid confusion with John
        Fine's J4 and J5 releases.}
    \item{Update the documentation; however, it still doesn't include
        documentation for the various new instructions. I somehow wonder if
        it makes sense to have an instruction set reference in the assembler
        manual when Intel et al have PDF versions of their manuals online.}
    \item{Recognize "idt" or "centaur" for the -p option to ndisasm.}
    \item{Changed error messages back to stderr where they belong, but add an
        -E option to redirect them elsewhere (the DOS shell cannot redirect
        stderr.)}
    \item{-M option to generate Makefile dependencies (based on code from Alex Verstak.)}
    \item{\code{\%undef} preprocessor directive, and -u option, that undefines a
        single-line macro.}
    \item{OS/2 Makefile (Mkfiles/Makefile.os2) for Borland under OS/2; from
        Chuck Crayne.}
    \item{Various minor bugfixes (reported by):
        - Dangling \code{\%s} in preproc.c (Martin Junker)}
    \item{THERE ARE KNOWN BUGS IN SSE AND THE OTHER KATMAI INSTRUCTIONS. I am
        on a trip and didn't bring the Katmai instruction reference, so I
        can't work on them right now.}
    \item{Updated the License file per agreement with Simon and Jules to
        include a GPL distribution clause.}
\end{itemize}

\xsubsection{cl-0.98p3.7}{Version 0.98p3.7}

\begin{itemize}
    \item{(Hopefully) fixed the canned Makefiles to include the outrdf2 and
        zoutieee modules.}
    \item{Renamed changes.asm to changed.asm.}
\end{itemize}

\xsubsection{cl-0.98p3.6}{Version 0.98p3.6}

\begin{itemize}
    \item{Fixed a bunch of instructions that were added in 0.98p3.5 which had
        memory operands, and the address-size prefix was missing from the
        instruction pattern.}
\end{itemize}

\xsubsection{cl-0.98p3.5}{Version 0.98p3.5}

\begin{itemize}
    \item{Merged in changes from John S. Fine's 0.98-J5 release. John's based
        0.98-J5 on my 0.98p3.3 release; this merges the changes.}
    \item{Expanded the instructions flag field to a long so we can fit more
        flags; mark SSE (KNI) and AMD or Katmai-specific instructions as
        such.}
    \item{Fix the "PRIV" flag on a bunch of instructions, and create new
        "PROT" flag for protected-mode-only instructions (orthogonal to if
        the instruction is privileged!) and new "SMM" flag for SMM-only
        instructions.}
    \item{Added AMD-only SYSCALL and SYSRET instructions.}
    \item{Make SSE actually work, and add new Katmai MMX instructions.}
    \item{Added a -p (preferred vendor) option to ndisasm so that it can
        distinguish e.g. Cyrix opcodes also used in SSE. For example:
\begin{lstlisting}
ndisasm -p cyrix aliased.bin
00000000  670F514310        paddsiw mm0,[ebx+0x10]
00000005  670F514320        paddsiw mm0,[ebx+0x20]
ndisasm -p intel aliased.bin
00000000  670F514310        sqrtps xmm0,[ebx+0x10]
00000005  670F514320        sqrtps xmm0,[ebx+0x20]
\end{lstlisting}}
    \item{Added a bunch of Cyrix-specific instructions.}
\end{itemize}

\xsubsection{cl-0.98p3.4}{Version 0.98p3.4}

\begin{itemize}
    \item{Made at least an attempt to modify all the additional Makefiles (in
        the Mkfiles directory). I can't test it, but this was the best I
        could do.}
    \item{DOS DJGPP+"Opus Make" Makefile from John S. Fine.}
    \item{changes.asm changes from John S. Fine.}
\end{itemize}

\xsubsection{cl-0.98p3.3}{Version 0.98p3.3}

\begin{itemize}
    \item{Patch from Conan Brink to allow nesting of \code{\%rep} directives.}
    \item{If we're going to allow INT01 as an alias for INT1/ICEBP (one of
        Jules 0.98p3 changes), then we should allow INT03 as an alias for INT3
        as well.}
    \item{Updated changes.asm to include the latest changes.}
    \item{Tried to clean up the <CR>s that had snuck in from a DOS/Windows
        environment into my Unix environment, and try to make sure than
        DOS/Windows users get them back.}
    \item{We would silently generate broken tools if insns.dat wasn't sorted
        properly. Change insns.pl so that the order doesn't matter.}
    \item{Fix bug in insns.pl (introduced by me) which would cause conditional
        instructions to have an extra "cc" in disassembly, e.g. "jnz"
        disassembled as "jccnz".}
\end{itemize}

\xsubsection{cl-0.98p3.2}{Version 0.98p3.2}

\begin{itemize}
    \item{Merged in John S. Fine's changes from his 0.98-J4 prerelease; see
        http://www.csoft.net/cz/johnfine/}
    \item{Changed previous "spotless" Makefile target (appropriate for distribution)
        to "distclean", and added "cleaner" target which is same as "clean"
        except deletes files generated by Perl scripts; "spotless" is union.}
    \item{Removed BASIC programs from distribution. Get a Perl interpreter
        instead (see below.)}
    \item{Calling this "pre-release 3.2" rather than "p3-hpa2" because of
        John's contributions.}
    \item{Actually link in the IEEE output format (zoutieee.c); fix a bunch of
        compiler warnings in that file. Note I don't know what IEEE output
        is supposed to look like, so these changes were made "blind".}
\end{itemize}

\xsubsection{cl-0.98p3-hpa}{Version 0.98p3-hpa}

\begin{itemize}
    \item{Merged nasm098p3.zip with nasm-0.97.tar.gz to create a fully
        buildable version for Unix systems (Makefile.in updates, etc.)}
    \item{Changed insns.pl to create the instruction tables in nasm.h and
        names.c, so that a new instruction can be added by adding it *only*
        to insns.dat.}
    \item{Added the following new instructions: SYSENTER, SYSEXIT, FXSAVE,
        FXRSTOR, UD1, UD2 (the latter two are two opcodes that Intel
        guarantee will never be used; one of them is documented as UD2 in
        Intel documentation, the other one just as "Undefined Opcode" --
        calling it UD1 seemed to make sense.)}
    \item{MAX\_SYMBOL was defined to be 9, but LOADALL286 and LOADALL386 are 10
        characters long. Now MAX\_SYMBOL is derived from insns.dat.}
    \item{A note on the BASIC programs included: forget them. insns.bas is
        already out of date. Get yourself a Perl interpreter for your
        platform of choice at \href{http://www.cpan.org/ports/index.html}
        {http://www.cpan.org/ports/index.html}.}
\end{itemize}

\xsubsection{cl-0.98p3}{Version 0.98 pre-release 3}

\begin{itemize}
    \item{added response file support, improved command line handling,
        new layout help screen}
    \item{fixed limit checking bug, 'OUT byte nn, reg' bug, and
        a couple of rdoff related bugs, updated Wishlist; 0.98 Prerelease 3.}
\end{itemize}

\xsubsection{cl-0.98p2}{Version 0.98 pre-release 2}

\begin{itemize}
    \item{fixed bug in outcoff.c to do with truncating section names longer
        than 8 characters, referencing beyond end of string; 0.98 pre-release 2}
\end{itemize}

\xsubsection{cl-0.98p1}{Version 0.98 pre-release 1}

\begin{itemize}
    \item{Fixed a bug whereby STRUC didn't work at all in RDF.}
    \item{Fixed a problem with group specification in PUBDEFs in OBJ.}
    \item{Improved ease of adding new output formats. Contribution due to Fox Cutter.}
    \item{Fixed a bug in relocations in the `bin' format: was showing up when
        a relocatable reference crossed an 8192-byte boundary in any output
        section.}
    \item{Fixed a bug in local labels: local-label lookups were inconsistent
        between passes one and two if an EQU occurred between the definition
        of a global label and the subsequent use of a local label local to
        that global.}
    \item{Fixed a seg-fault in the preprocessor (again) which happened when
        you use a blank line as the first line of a multi-line macro
        definition and then defined a label on the same line as a call to
        that macro.}
    \item{Fixed a stale-pointer bug in the handling of the NASM environment
        variable. Thanks to Thomas McWilliams.}
    \item{ELF had a hard limit on the number of sections which caused
        segfaults when transgressed. Fixed.}
    \item{Added ability for ndisasm to read from stdin by using `-' as the
        filename.}
    \item{ndisasm wasn't outputting the TO keyword. Fixed.}
    \item{Fixed error cascade on bogus expression in \code{\%if} - an error in
        evaluation was causing the entire \code{\%if} to be discarded, thus creating
        trouble later when the \code{\%else} or \code{\%endif} was encountered.}
    \item{Forward reference tracking was instruction-granular not operand-
        granular, which was causing 286-specific code to be generated
        needlessly on code of the form `shr word [forwardref],1'. Thanks to
        Jim Hague for sending a patch.}
    \item{All messages now appear on stdout, as sending them to stderr serves
        no useful purpose other than to make redirection difficult.}
    \item{Fixed the problem with EQUs pointing to an external symbol - this
        now generates an error message.}
    \item{Allowed multiple size prefixes to an operand, of which only the first
        is taken into account.}
    \item{Incorporated John Fine's changes, including fixes of a large number
        of preprocessor bugs, some small problems in OBJ, and a reworking of
        label handling to define labels before their line is assembled, rather
        than after.}
    \item{Reformatted a lot of the source code to be more readable. Included
        'coding.txt' as a guideline for how to format code for contributors.}
    \item{Stopped nested \code{\%reps} causing a panic - they now cause a slightly more
        friendly error message instead.}
    \item{Fixed floating point constant problems (patch by Pedro Gimeno)}
    \item{Fixed the return value of insn\_size() not being checked for -1, indicating
        an error.}
    \item{Incorporated 3Dnow! instructions.}
    \item{Fixed the 'mov eax, eax + ebx' bug.}
    \item{Fixed the GLOBAL EQU bug in ELF. Released developers release 3.}
    \item{Incorporated John Fine's command line parsing changes}
    \item{Incorporated David Lindauer's OMF debug support}
    \item{Made changes for LCC 4.0 support (\code{\_\_NASM\_CDecl\_\_}, removed register size
        specification warning when sizes agree).}
\end{itemize}

\xsection{cl-0.9x}{NASM 0.9 Series}

Revisions before 0.98.

\xsubsection{cl-0.97}{Version 0.97 released December 1997}

\begin{itemize}
    \item{This was entirely a bug-fix release to 0.96, which seems to have got
        cursed. Silly me.}
    \item{Fixed stupid mistake in OBJ which caused `MOV EAX,<constant>' to
        fail. Caused by an error in the `MOV EAX,<segment>' support.}
    \item{ndisasm hung at EOF when compiled with lcc on Linux because lcc on
        Linux somehow breaks feof(). ndisasm now does not rely on feof().}
    \item{A heading in the documentation was missing due to a markup error in
        the indexing. Fixed.}
    \item{Fixed failure to update all pointers on realloc() within extended-
        operand code in parser.c. Was causing wrong behaviour and seg faults
        on lines such as `dd 0.0,0.0,0.0,0.0,...'}
    \item{Fixed a subtle preprocessor bug whereby invoking one multi-line
        macro on the first line of the expansion of another, when the second
        had been invoked with a label defined before it, didn't expand the
        inner macro.}
    \item{Added internal.doc back in to the distribution archives - it was
        missing in 0.96 *blush*}
    \item{Fixed bug causing 0.96 to be unable to assemble its own test files,
        specifically objtest.asm. *blush again*}
    \item{Fixed seg-faults and bogus error messages caused by mismatching
        \code{\%rep} and \code{\%endrep} within multi-line macro definitions.}
    \item{Fixed a problem with buffer overrun in OBJ, which was causing
        corruption at ends of long PUBDEF records.}
    \item{Separated DOS archives into main-program and documentation to reduce
        download size.}
\end{itemize}

\xsubsection{cl-0.96}{Version 0.96 released November 1997}

\begin{itemize}
    \item{Fixed a bug whereby, if `nasm sourcefile' would cause a filename
        collision warning and put output into `nasm.out', then `nasm
        sourcefile -o outputfile' still gave the warning even though the
        `-o' was honoured.
        Fixed name pollution under Digital UNIX: one of its header files
        defined R\_SP, which broke the enum in nasm.h.}
    \item{Fixed minor instruction table problems: FUCOM and FUCOMP didn't have
        two-operand forms; NDISASM didn't recognise the longer register
        forms of PUSH and POP (eg FF F3 for PUSH BX); TEST mem,imm32 was
        flagged as undocumented; the 32-bit forms of CMOV had 16-bit operand
        size prefixes; `AAD imm' and `AAM imm' are no longer flagged as
        undocumented because the Intel Architecture reference documents
        them.}
    \item{Fixed a problem with the local-label mechanism, whereby strange
        types of symbol (EQUs, auto-defined OBJ segment base symbols)
        interfered with the `previous global label' value and screwed up
        local labels.}
    \item{Fixed a bug whereby the stub preprocessor didn't communicate with
        the listing file generator, so that the -a and -l options in
        conjunction would produce a useless listing file.}
    \item{Merged `os2' object file format back into `obj', after discovering
        that `obj' \emph{also} shouldn't have a link pass separator in a module
        containing a non-trivial MODEND. Flat segments are now declared
        using the FLAT attribute. `os2' is no longer a valid object format
        name: use `obj'.}
    \item{Removed the fixed-size temporary storage in the evaluator. Very very
        long expressions (like `mov ax,1+1+1+1+...' for two hundred 1s or
        so) should now no longer crash NASM.}
    \item{Fixed a bug involving segfaults on disassembly of MMX instructions,
        by changing the meaning of one of the operand-type flags in nasm.h.
        This may cause other apparently unrelated MMX problems; it needs to
        be tested thoroughly.}
    \item{Fixed some buffer overrun problems with large OBJ output files.
        Thanks to DJ Delorie for the bug report and fix.}
    \item{Made preprocess-only mode actually listen to the \code{\%line} markers as it
        prints them, so that it can report errors more sanely.}
    \item{Re-designed the evaluator to keep more sensible track of expressions
        involving forward references: can now cope with previously-nightmare
        situations such as:
\begin{lstlisting}
mov ax,foo | bar
foo equ 1
bar equ 2
\end{lstlisting}}
    \item{Added the ALIGN and ALIGNB standard macros.}
    \item{Added PIC support in ELF: use of WRT to obtain the four extra
        relocation types needed.}
    \item{Added the ability for output file formats to define their own
        extensions to the GLOBAL, COMMON and EXTERN directives.}
    \item{Implemented common-variable alignment, and global-symbol type and
        size declarations, in ELF.}
    \item{Implemented NEAR and FAR keywords for common variables, plus
        far-common element size specification, in OBJ.}
    \item{Added a feature whereby EXTERNs and COMMONs in OBJ can be given a
        default WRT specification (either a segment or a group).}
    \item{Transformed the Unix NASM archive into an auto-configuring package.}
    \item{Added a sanity-check for people applying SEG to things which are
        already segment bases: this previously went unnoticed by the SEG
        processing and caused OBJ-driver panics later.}
    \item{Added the ability, in OBJ format, to deal with `MOV EAX,<segment>'
        type references: OBJ doesn't directly support dword-size segment
        base fixups, but as long as the low two bytes of the constant term
        are zero, a word-size fixup can be generated instead and it will
        work.}
    \item{Added the ability to specify sections' alignment requirements in
        Win32 object files and pure binary files.}
    \item{Added preprocess-time expression evaluation: the \code{\%assign} (and
        \code{\%iassign}) directive and the bare \code{\%if} (and \code{\%elif}) conditional.
        Added relational operators to the evaluator, for use only in \code{\%if}
        constructs: the standard relationals = < > <= >= <> (and C-like
        synonyms == and !=) plus low-precedence logical operators \&\&,
        \textasciicircum \textasciicircum and ||.}
    \item{Added a preprocessor repeat construct: \code{\%rep}, \code{\%exitrep},
        \code{\%endrep}.}
    \item{Added the \_\_FILE\_\_ and \_\_LINE\_\_ standard macros.}
    \item{Added a sanity check for number constants being greater than
        0xFFFFFFFF. The warning can be disabled.}
    \item{Added the \%0 token whereby a variadic multi-line macro can tell how
        many parameters it's been given in a specific invocation.}
    \item{Added \code{\%rotate}, allowing multi-line macro parameters to be cycled.}
    \item{Added the `*' option for the maximum parameter count on multi-line
        macros, allowing them to take arbitrarily many parameters.}
    \item{Added the ability for the user-level forms of EXTERN, GLOBAL and
        COMMON to take more than one argument.}
    \item{Added the IMPORT and EXPORT directives in OBJ format, to deal with
        Windows DLLs.}
    \item{Added some more preprocessor \code{\%if} constructs: \code{\%ifidn},
        \code{\%ifidni} (exact textual identity), and \code{\%ifid},
        \code{\%ifnum}, \code{\%ifstr} (token type testing).}
    \item{Added the ability to distinguish SHL AX,1 (the 8086 version) from
        SHL AX,BYTE 1 (the 286-and-upwards version whose constant happens to
        be 1).}
    \item{Added NetBSD/FreeBSD/OpenBSD's variant of a.out format, complete
        with PIC shared library features.}
    \item{Changed NASM's idiosyncratic handling of FCLEX, FDISI, FENI, FINIT,
        FSAVE, FSTCW, FSTENV, and FSTSW to bring it into line with the
        otherwise accepted standard. The previous behaviour, though it was a
        deliberate feature, was a deliberate feature based on a
        misunderstanding. Apologies for the inconvenience.}
    \item{Improved the flexibility of ABSOLUTE: you can now give it an
        expression rather than being restricted to a constant, and it can
        take relocatable arguments as well.}
    \item{Added the ability for a variable to be declared as EXTERN multiple
        times, and the subsequent definitions are just ignored.}
    \item{We now allow instruction prefixes (CS, DS, LOCK, REPZ etc) to be
        alone on a line (without a following instruction).}
    \item{Improved sanity checks on whether the arguments to EXTERN, GLOBAL
        and COMMON are valid identifiers.}
    \item{Added misc/exebin.mac to allow direct generation of .EXE files by
        hacking up an EXE header using DB and DW; also added test/binexe.asm
        to demonstrate the use of this. Thanks to Yann Guidon for
        contributing the EXE header code.}
    \item{ndisasm forgot to check whether the input file had been successfully
        opened. Now it does. Doh!}
    \item{Added the Cyrix extensions to the MMX instruction set.}
    \item{Added a hinting mechanism to allow [EAX+EBX] and [EBX+EAX] to be
        assembled differently. This is important since [ESI+EBP] and
        [EBP+ESI] have different default base segment registers.}
    \item{Added support for the PharLap OMF extension for 4096-byte segment
        alignment.}
\end{itemize}

\xsubsection{cl-0.95}{Version 0.95 released July 1997}

\begin{itemize}
    \item{Fixed yet another ELF bug. This one manifested if the user relied on
        the default segment, and attempted to define global symbols without
        first explicitly declaring the target segment.}
    \item{Added makefiles (for NASM and the RDF tools) to build Win32 console
        apps under Symantec C++. Donated by Mark Junker.}
    \item{Added `macros.bas' and `insns.bas', QBasic versions of the Perl
        scripts that convert `standard.mac' to `macros.c' and convert
        `insns.dat' to `insnsa.c' and `insnsd.c'. Also thanks to Mark
        Junker.}
    \item{Changed the diassembled forms of the conditional instructions so
        that JB is now emitted as JC, and other similar changes. Suggested
        list by Ulrich Doewich.}
    \item{Added `@' to the list of valid characters to begin an identifier with.}
    \item{Documentary changes, notably the addition of the `Common Problems'
        section in nasm.doc.}
    \item{Fixed a bug relating to 32-bit PC-relative fixups in OBJ.}
    \item{Fixed a bug in perm\_copy() in labels.c which was causing exceptions
        in cleanup\_labels() on some systems.}
    \item{Positivity sanity check in TIMES argument changed from a warning to
        an error following a further complaint.}
    \item{Changed the acceptable limits on byte and word operands to allow
        things like `\textasciitilde 10111001b' to work.}
    \item{Fixed a major problem in the preprocessor which caused seg-faults if
        macro definitions contained blank lines or comment-only lines.}
    \item{Fixed inadequate error checking on the commas separating the
        arguments to `db', `dw' etc.}
    \item{Fixed a crippling bug in the handling of macros with operand counts
        defined with a `+' modifier.}
    \item{Fixed a bug whereby object file formats which stored the input file
        name in the output file (such as OBJ and COFF) weren't doing so
        correctly when the output file name was specified on the command
        line.}
    \item{Removed [INC] and [INCLUDE] support for good, since they were
        obsolete anyway.}
    \item{Fixed a bug in OBJ which caused all fixups to be output in 16-bit
        (old-format) FIXUPP records, rather than putting the 32-bit ones in
        FIXUPP32 (new-format) records.}
    \item{Added, tentatively, OS/2 object file support (as a minor variant
        on OBJ).}
    \item{Updates to Fox Cutter's Borland C makefile, Makefile.bc2.}
    \item{Removed a spurious second fclose() on the output file.}
    \item{Added the `-s' command line option to redirect all messages which
        would go to stderr (errors, help text) to stdout instead.}
    \item{Added the `-w' command line option to selectively suppress some
        classes of assembly warning messages.}
    \item{Added the `-p' pre-include and `-d' pre-define command-line options.}
    \item{Added an include file search path: the `-i' command line option.}
    \item{Fixed a silly little preprocessor bug whereby starting a line with a
        `\%!' environment-variable reference caused an `unknown directive'
        error.}
    \item{Added the long-awaited listing file support: the `-l' command line
        option.}
    \item{Fixed a problem with OBJ format whereby, in the absence of any
        explicit segment definition, non-global symbols declared in the
        implicit default segment generated spurious EXTDEF records in the
        output.}
    \item{Added the NASM environment variable.}
    \item{From this version forward, Win32 console-mode binaries will be
        included in the DOS distribution in addition to the 16-bit binaries.
        Added Makefile.vc for this purpose.}
    \item{Added `return 0;' to test/objlink.c to prevent compiler warnings.}
    \item{Added the \_\_NASM\_MAJOR\_\_ and \_\_NASM\_MINOR\_\_ standard defines.}
    \item{Added an alternative memory-reference syntax in which prefixing an
        operand with `\&' is equivalent to enclosing it in square brackets,
        at the request of Fox Cutter.}
    \item{Errors in pass two now cause the program to return a non-zero error
        code, which they didn't before.}
    \item{Fixed the single-line macro cycle detection, which didn't work at
        all on macros with no parameters (caused an infinite loop). Also
        changed the behaviour of single-line macro cycle detection to work
        like cpp, so that macros like `extrn' as given in the documentation
        can be implemented.}
    \item{Fixed the implementation of WRT, which was too restrictive in that
        you couldn't do `mov ax,[di+abc wrt dgroup]' because (di+abc) wasn't
        a relocatable reference.}
\end{itemize}

\xsubsection{cl-0.94}{Version 0.94 released April 1997}

\begin{itemize}
    \item{Major item: added the macro processor.}
    \item{Added undocumented instructions SMI, IBTS, XBTS and LOADALL286. Also
        reorganised CMPXCHG instruction into early-486 and Pentium forms.
        Thanks to Thobias Jones for the information.}
    \item{Fixed two more stupid bugs in ELF, which were causing `ld' to
        continue to seg-fault in a lot of non-trivial cases.}
    \item{Fixed a seg-fault in the label manager.}
    \item{Stopped FBLD and FBSTP from \emph{requiring} the TWORD keyword, which is
        the only option for BCD loads/stores in any case.}
    \item{Ensured FLDCW, FSTCW and FSTSW can cope with the WORD keyword, if
        anyone bothers to provide it. Previously they complained unless no
        keyword at all was present.}
    \item{Some forms of FDIV/FDIVR and FSUB/FSUBR were still inverted: a
        vestige of a bug that I thought had been fixed in 0.92. This was
        fixed, hopefully for good this time...}
    \item{Another minor phase error (insofar as a phase error can \emph{ever} be
        minor) fixed, this one occurring in code of the form
\begin{lstlisting}
rol ax,forward_reference
forward_reference equ 1
\end{lstlisting}}
    \item{The number supplied to TIMES is now sanity-checked for positivity,
        and also may be greater than 64K (which previously didn't work on
        16-bit systems).}
    \item{Added Watcom C makefiles, and misc/pmw.bat, donated by Dominik Behr.}
    \item{Added the INCBIN pseudo-opcode.}
    \item{Due to the advent of the preprocessor, the [INCLUDE] and [INC]
        directives have become obsolete. They are still supported in this
        version, with a warning, but won't be in the next.}
    \item{Fixed a bug in OBJ format, which caused incorrect object records to
        be output when absolute labels were made global.}
    \item{Updates to RDOFF subdirectory, and changes to outrdf.c.}
\end{itemize}

\xsubsection{cl-0.93}{Version 0.93 released January 1997}

This release went out in a great hurry after semi-crippling bugs
were found in 0.92.

\begin{itemize}
    \item{Really \emph{did} fix the stack overflows this time. *blush*}
    \item{Had problems with EA instruction sizes changing between passes, when
        an offset contained a forward reference and so 4 bytes were
        allocated for the offset in pass one; by pass two the symbol had
        been defined and happened to be a small absolute value, so only 1
        byte got allocated, causing instruction size mismatch between passes
        and hence incorrect address calculations. Fixed.}
    \item{Stupid bug in the revised ELF section generation fixed (associated
        string-table section for .symtab was hard-coded as 7, even when this
        didn't fit with the real section table). Was causing `ld' to
        seg-fault under Linux.}
    \item{Included a new Borland C makefile, Makefile.bc2, donated by Fox
        Cutter <lmb@comtch.iea.com>.}
\end{itemize}

\xsubsection{cl-0.92}{Version 0.92 released January 1997}

\begin{itemize}
    \item{The FDIVP/FDIVRP and FSUBP/FSUBRP pairs had been inverted: this was
        fixed. This also affected the LCC driver.}
    \item{Fixed a bug regarding 32-bit effective addresses of the form
        \code{[other\_register+ESP]}.}
    \item{Documentary changes, notably documentation of the fact that Borland
        Win32 compilers use `obj' rather than `win32' object format.}
    \item{Fixed the COMENT record in OBJ files, which was formatted
        incorrectly.}
    \item{Fixed a bug causing segfaults in large RDF files.}
    \item{OBJ format now strips initial periods from segment and group
        definitions, in order to avoid complications with the local label
        syntax.}
    \item{Fixed a bug in disassembling far calls and jumps in NDISASM.}
    \item{Added support for user-defined sections in COFF and ELF files.}
    \item{Compiled the DOS binaries with a sensible amount of stack, to
        prevent stack overflows on any arithmetic expression containing
        parentheses.}
    \item{Fixed a bug in handling of files that do not terminate in a newline.}
\end{itemize}

\xsubsection{cl-0.91}{Version 0.91 released November 1996}

\begin{itemize}
    \item{Loads of bug fixes.}
    \item{Support for RDF added.}
    \item{Support for DBG debugging format added.}
    \item{Support for 32-bit extensions to Microsoft OBJ format added.}
    \item{Revised for Borland C: some variable names changed, makefile added.}
    \item{LCC support revised to actually work.}
    \item{JMP/CALL NEAR/FAR notation added.}
    \item{`a16', `o16', `a32' and `o32' prefixes added.}
    \item{Range checking on short jumps implemented.}
    \item{MMX instruction support added.}
    \item{Negative floating point constant support added.}
    \item{Memory handling improved to bypass 64K barrier under DOS.}
    \item{\code{\$} prefix to force treatment of reserved words as identifiers added.}
    \item{Default-size mechanism for object formats added.}
    \item{Compile-time configurability added.}
    \item{\code{\#}, \code{\@}, \code{\textasciitilde} and \code{?} are now valid characters in labels.}
    \item{\code{-e} and \code{-k} options in NDISASM added.}
\end{itemize}

\xsubsection{cl-0.90} Version 0.90 released October 1996

First release version. First support for object file output. Other
changes from previous version (0.3x) too numerous to document.