Thank you for the new stable release Tomasz

Thank you Tomasz Grysztar, you really deserve your title of the 'Assembly Artist'

I have one question: why was 'fas file signature' changed from 0x1B736166
to 0x1A736166? How can it be that 'A' comes after 'B', and what that signature means?(sorry, those were three question)

Because PECVT.EXE crashed on this signature I had to rebuild it to work with that new one. If that signature changes with every new build of fasm, then
strap89 has to change his source code to take that in consideration

best regards

Please read this and the following posts: http://board.flatassembler.net/topic.php?p=92185#92185

Thanks

Some years ago, I decided to switch from TASM to FASM. I don't regret it.
Thank you Tomasz for your work.

Well done. Found that "formt ... as extension" missed in pdf or it me who missed? Also haven't noticed discussed here (on forum) large address aware flag for PE (for x64 its automatically set by default).

It's documented in the beginning of section 2.4

ok, I see. Just haven't seen it in self-explanatory example of "format" in PE file format section. Then it can be useful to add to example for PE description: <format... as "exe">. Some redundant information but representative indeed.

Hello,
thank you for the new version, but I think the added new mnemonics "rdmsrq", "wrmsrq", "sysexitq" and "sysretq" are not right: All gives the Opcode "48 0F 48" and not e.g. "48 0F 32" for rdmsrq etc.
Gruss
Helle

also could you explain rex prefixes are now available to all unary operations as -q suffix? even if operation is not supposed to get it?

Well, I guess this thread wasn't really a full success: http://board.flatassembler.net/topic.php?t=9967
I really hoped for such reports to come earlier. :>

Earlier? I read about this instructions in the Whatsnew for version 1.68 (Jun 13, 2009) and I have only this week-end time for tests - what´s your problem?
Thanks
Helle

The point of 1.68 was to come after the tests and fixes in 1.69.x development line, as a confirmed "stable" release, I waited until there were no new bug reports coming for some time, and taken iit as a good sign - my bad luck, it wasn't reliable enough.

As for RDMSRQ and family, it was available already quite a long time ago: http://board.flatassembler.net/topic.php?t=5539
Funny, if it was this buggy for all this time and no-one (including me) noticed.

I have no idea what those instructions are about, so sorry that I couldn't test it.

Oh sorry, I don´t known this (old!) post and in Intel-/AMD-papers or in the Programmer's Manual for 1.67 I don´t found this mnemonics. I have greatest respect for your work and I´m not a grumbler !
Please, don´t beat me, but in the last hour I made tests with the CRC32-Instruction and I got no right results with an 64-Bit-Register as destination.
Your example in the manual "crc32 rax,qword [rbx]" gives the opcode "F2 0F 38 F1 03". No REX-Prefix and the same as "crc32 eax,dword [rbx]". Please, take a look...
Thanks
Helle

Thank you for your reports, no need to feel guilty. You are helping to make fasm a better tool. I was only a little mad at myself that I did let such an obvious mistakes into supposed-to-be-stable release. Anyway, I'm going to make a 1.68 SP1 release in a few days.

SP1?! Hehe, fasm is now an MS product!

Well, that my way of saying, that I may release some bugfixes for 1.68 (1.68.1 perhaps), while at the same time already working on 1.69.0 (new and highly experimental development line).

Sound like FASM/we have a great future around the corner. One man who did it all. Time for some relaxation, so take your time Mr. Grysztar and don't forget to fully enjoy your summer vacation. You got 2 days and you too revolution just to Get Fresh, AGAIN for 2010 ! Grab your girl and go sailing, but don't get into any trouble

I uploaded the 1.68 package with a few small fixes and optimizations back-ported from 1.69.0 (hope I don't break anything this time ).