From dmytro.a.kovalov at globallogic.com  Fri Nov  4 18:56:19 2022
From: dmytro.a.kovalov at globallogic.com (Dmytro Kovalov)
Date: Fri, 4 Nov 2022 19:56:19 +0200
Subject: Fwd: Fwd: cipher-gcm-armv7-neon.S issue
In-Reply-To: <791cc2b6-f9c6-706d-07f1-3dbbe5709bd2@iki.fi>
References: <CAGOFUGMuUo5GdoYXVEUR170hTJiakqzuUomCZLQs7yLeKfXsyQ@mail.gmail.com>
 <CAGOFUGOJritHOfBRj=FE_oT56EGjE_oRkviiWfdVAe+Qy-fZFQ@mail.gmail.com>
 <791cc2b6-f9c6-706d-07f1-3dbbe5709bd2@iki.fi>
Message-ID: <CAGOFUGP43P3dqu5ogJcuVP4ObPCU8D9j9W=2anCjsFUL+aDN3Q@mail.gmail.com>

---------- Forwarded message ---------
From: Jussi Kivilinna <jussi.kivilinna at iki.fi>
Date: Fri, Nov 4, 2022 at 7:50 PM
Subject: Re: Fwd: cipher-gcm-armv7-neon.S issue
To: Dmytro Kovalov <dmytro.a.kovalov at globallogic.com>


Hello,

Please send questions about libgcrypt to "gcrypt-devel at gnupg.org"

-Jussi

On 4.11.2022 19.33, Dmytro Kovalov wrote:
>
> Hello Jussi,
>
> Little update.
> I found the root cause of thecipher-gcm-armv7-neon.S
(sha512-armv7-neon.S) issue .
> It was an extra '$' symbol.
>
> This source describe diffs between AT&T and INTEL dialect
>
https://imada.sdu.dk/~kslarsen/dm546/Material/IntelnATT.htm#:~:text=In%20Intel%20syntax%20the%20first,in%20this%20situation%20is%20obvious
<https://imada.sdu.dk/~kslarsen/dm546/Material/IntelnATT.htm#:~:text=In%20Intel%20syntax%20the%20first,in%20this%20situation%20is%20obvious>
<
https://imada.sdu.dk/~kslarsen/dm546/Material/IntelnATT.htm#:~:text=In%20Intel%20syntax%20the%20first,in%20this%20situation%20is%20obvious
<https://imada.sdu.dk/~kslarsen/dm546/Material/IntelnATT.htm#:~:text=In%20Intel%20syntax%20the%20first,in%20this%20situation%20is%20obvious>
>.
> and the AT&T syntax is different from either Intel or libgcrypt asm
source code.
> I mean from all rules described on source there only *Prefixes* rule
partly uses by libgcrypt code (two '%' and $ are from AT&T )
> The others (*Memory Operands, **Direction of Operands, **Suffixes***)
haven't applied to  libgcrypt code.
>
> Could you please comment difference between AT&T Intel and gcc approaches
?
> I am a bit confused , but can confirm that at least clang build
(sha512-neon-arm) works now after removal of '%' '$' on the arm machine.
>
> Best Regards,
> Dmytro
> ---------- Forwarded message ---------
> From: *Dmytro Kovalov* <dmytro.a.kovalov at globallogic.com <mailto:
dmytro.a.kovalov at globallogic.com>>
> Date: Fri, Nov 4, 2022 at 5:50 PM
> Subject: cipher-gcm-armv7-neon.S issue
> To: Jussi Kivilinna <jussi.kivilinna at iki.fi <mailto:jussi.kivilinna at iki.fi
>>
>
>
> Hello Jussi,
>
> Just found an issue: the cipher-gcm-armv7-neon.S failed on clang compile
stage.
> The fix is just disable NEON on version 1.8.6 no such issue because of
file absence.
>
> Regards,
> Dmytro.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.gnupg.org/pipermail/gcrypt-devel/attachments/20221104/b3fb9ee8/attachment.html>