> On Jun 6, 2017, at 2:05 PM, [email protected] wrote: > > On June 6, 2017 1:49:34 PM PDT, Andrew Fish <[email protected]> wrote: >> >>> On Jun 6, 2017, at 12:41 PM, H. Peter Anvin <[email protected]> wrote: >>> >>> On 05/22/17 19:08, Fan, Jeff wrote: >>>> >>>> diff --git a/UefiCpuPkg/Library/MpInitLib/X64/MpFuncs.nasm >> b/UefiCpuPkg/Library/MpInitLib/X64/MpFuncs.nasm >>>> index fa54d01..0b14a53 100644 >>>> --- a/UefiCpuPkg/Library/MpInitLib/X64/MpFuncs.nasm >>>> +++ b/UefiCpuPkg/Library/MpInitLib/X64/MpFuncs.nasm >>>> @@ -1,5 +1,5 @@ >>>> >> ;------------------------------------------------------------------------------ >> ; -; Copyright (c) 2015 - 2016, Intel Corporation. All rights >> reserved.<BR> >>>> +; Copyright (c) 2015 - 2017, Intel Corporation. All rights >>>> +reserved.<BR> >>>> ; This program and the accompanying materials ; are licensed and >> made available under the terms and conditions of the BSD License ; >> which accompanies this distribution. The full text of the license may >> be found at @@ -201,7 +201,7 @@ CProcedureInvoke: >>>> push rbp >>>> mov rbp, rsp >>>> >>>> - mov rax, ASM_PFX(InitializeFloatingPointUnits) >>>> + mov rax, qword [esi + >> InitializeFloatingPointUnitsAddress] >> >> Does nasm remove the need for the ASM_PFX() macro? That macro hides if >> C is decorating with a _ prefix. >> >> Also given it is a #define (equ) why do we use camel case vs. all caps? >> >> Thanks, >> >> Andrew Fish >> >>>> sub rsp, 20h >>>> call rax ; Call assembly function to >> initialize FPU per UEFI spec >>>> add rsp, 20h >>> >>> FYI, the qword specifier is unnecessary since you are already >> specifying >>> rax. >>> >>> However, why not simply drop the use of rax entirely and do: >>> >>> call [esi + InitializeFloatingPointUnitsAddress] >>> >>> (Also: is this *really* supposed to be esi and not rsi? The former >>> means a 32-bit address.) >>> >>> -hpa >>> _______________________________________________ >>> edk2-devel mailing list >>> [email protected] >>> https://lists.01.org/mailman/listinfo/edk2-devel > > The NASM command-line option --prefix _ should do exactly that.
How does it know when to prefix the _? That could break dead stripping with the Xcode linker. L is used to imply local symbol, don't dead strip. Thanks, Andrew Fish > -- > Sent from my Android device with K-9 Mail. Please excuse my brevity. _______________________________________________ edk2-devel mailing list [email protected] https://lists.01.org/mailman/listinfo/edk2-devel
