Unified Extensible Firmware Interface (UEFI) Example

[johnsa]

You need to use at least .686 to allow the simd instructions, at least it's always been this way in jwasm/uasm. .686 should then work for both uasm and masm.

I've changed MEMALIGN to ALIGNADDR .. makes sense and it shouldn't conflict with anything.

Updated the packages on the site.

[aw27]

I'm happy to remove the case sensitive version for this or rename it.. we can take a vote.
Personally I'd opt to rename it so that the case can be freely applied.

I don't know if this makes sense, but the assembler could simply supersede any internal macro whenever there is a user defined macro with the same name. This will prevent the need for another switch which people tend to forget (and UASM already has a large number of them to remember).

[johnsa]

Agreed, for now I'm going to leave it renamed so there is no immediate conflict.
I think in future I will add a flag to the definitions of the macro-lib marking them as "internal", so that we can override them with any user defined ones.

[hutch--]

 

> Actually MASM is taught in college and high-school and Masm32 is not allowed there.

You may be surprised just how many UNI courses use the MASM32 SDK and over how many years. Where MASM is taught at universities, it is usually DOS 16 bit and others use Kip Irvine's book and examples.

[aw27]

You may be surprised just how many UNI courses use the MASM32 SDK and over how many years. Where MASM is taught at universities, it is usually DOS 16 bit and others use Kip Irvine's book and examples.

Bad luck, when I was doing online tutoring not a single one requested Masm32. For x86/x64 ASM, all of them required Irvine for 16, 32 and 64-bit.

[hutch--]

 

Bigger fool them, commerce dictates BUYING the software. 

[HSE]

Bad luck, when I was doing online tutoring not a single one requested Masm32.

Then: MASM32 is very easy. 

[aw27]

I suspect teachers were receiving a fat commission for doing classes based on the Irvine's book. 

[TimoVJL]

 

[LiaoMi]

You need to use at least .686 to allow the simd instructions, at least it's always been this way in jwasm/uasm. .686 should then work for both uasm and masm.

I've changed MEMALIGN to ALIGNADDR .. makes sense and it shouldn't conflict with anything.

Updated the packages on the site.

Hi johnsa,

I updated the cache, but the links are old 

[johnsa]

I've only updated the packages not the site, just waiting for confirmation that initial testing was ok.

[jj2007]

John,
Where is the latest UAsm64.exe? The one with the memalign problem is 20 Oct 20:50.

[TimoVJL]

 

[jj2007]

UAsm 2.50 compiled with Clang 9 and using msvcrt.dll, so it's so slow.

Now it works, but it's indeed not that fast:
1750 ms  Uasm64msvcrt, ML 6.15 & 10.0
1500 ms  Uasm64
1150 ms  AsmC

[TimoVJL]

clang 9 & libc

Code Select Expand

5.382s        asmc.exe 2.30.23
7.831s        uasm64libc.exe 2.50
8.393s        uasm64vc2019.exe 2.50
8.830s        uasm64.exe 2.49
9.235s        uasm64poc9.exe 2.50
10.124s       uasm32libc.exe 2.50

EDIT: fix WIN64