@Garthower - try refreshing your browser.

Thanks, I downloaded version 1.67.3. It was the problem with beta-version Outpost Firewall 64 installed at my computer.

I have such question. Really in PE64 are used 32-bit relocs?

No, see that other thread. fasm is currently able to generate 32-bit relocs for object output (and then the linker should warn you that the code has to be put in the low 2 GB, or ensure it in some other way), for relocatable PE64 there would have to be some modifier put in source, but I did not do it, as I didn't felt it really necessary right now - the standard PE no longer use the relocatable labels anyway.

The new important release today - 1.67.17 (the last one this year? not necessarily... ). I fixed two bugs that I stumbled upon myself when playing with some macro recently.

The first fix is to make precedence of macro line maker operators consistent - now not only conversion-to-string operator has higher precedence than concatenation, but both those operations allows the symbol escaping to be cancelled before they finish. This allows this nice trick in definitions of structures:


The second fix was needed in the new logical expression parser (introduced in 1.66), which failed to process properly the EQ or EQTYPE comparisons of chains of symbols containing round brackets, like:

The problem with logical expressions is that round brackets may be also used to enclose the logical values. However I think I managed to fix these problems even for some insane cases.

The other news are that FASMD is now officially part of DOS package, even though some of its features are not yet finished (these are .ini file and file browsing, to be exact). It even has a separate small documentation.

The FASMW also got a bit improved thanks to the FASMD's development. In particular, the Alt+[1-9] shortcuts for switching between first nine files has been back-ported from FASMD to FASMW and the few editor bugs has been fixed. The Win32 package contains also some corrections in examples and in some of the macros - the "import" macro will now refuse to define imports if it is not correctly referred by "library" macro.

Hopefully some day someone will understand the fasm macros ! Thanks for the release, I wonder what fasm 2.0 holds?