From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 6DA2546DE3; Tue, 26 Aug 2025 14:02:34 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id EF5DE402A3; Tue, 26 Aug 2025 14:02:33 +0200 (CEST) Received: from mail-qv1-f41.google.com (mail-qv1-f41.google.com [209.85.219.41]) by mails.dpdk.org (Postfix) with ESMTP id BD9754021E for ; Tue, 26 Aug 2025 14:02:32 +0200 (CEST) Received: by mail-qv1-f41.google.com with SMTP id 6a1803df08f44-70d9eb2e976so34286256d6.0 for ; Tue, 26 Aug 2025 05:02:32 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=uetpeshawar-edu-pk.20230601.gappssmtp.com; s=20230601; t=1756209752; x=1756814552; darn=dpdk.org; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=JPlPwVdyJt1Fpa3Z3muUPBz0KFKV79bDKqM54CmGLMY=; b=zZ//lDkdLJEqbIxlkVF78SU6T8qUykSDxILH9fDhNzMHcX2BrvwqLwRqg8+HNb1cj7 3M+q4PPJpPB/u+xE0tWwSqWWRanULZS6uIRAXMf1d02g1tEZY5No4LENPBuKCqeyMliV EnGQeG03ezjU0rcnJD7u/j+hxe3tHioYeD4+9LD3m3OSGF73JjFBizfT3jjpkpMdt6Qh Ei5MoGa5TLgC/c8TVS5AqJArPPgyY9YVvMkPfD/vus06xNpEUrEslUCs4xKYTS0E8eRd AtgxpDRUHuD6Ovz7OuiswfNwd2zZuD+HbXUu/3xpCsYqkOqD8tVrVu4zum2KnihgIfJh Fypg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1756209752; x=1756814552; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=JPlPwVdyJt1Fpa3Z3muUPBz0KFKV79bDKqM54CmGLMY=; b=EssSgDTqPWk5IBm5hj2wKDU0fwxy1BFSIVgclNeSnF7QFDD21gvq58Fvz6jl9zgFam iy3nQxDwAxRVZpBL4r8kd0Aga3zYP+iyMAfEKpQeh9ZOmFJTw4q42SOov6OOXVv0WDOy yho/V1Vpb7ndHIs+8dhIYvJs1kGVlM6rP5sL/SpIYL2o+8o+hexqlK4zw5hQQMczzJbg SX5suzqFZszyLsjddGtVKoOZXunV2nZF8tZoHuL3xbHvZmW/5a5ydwawfsP+yRfb1+rO nZtR0sUYgX+BWS6s7q49j0ZKbk6tpmoMBL//XSXnb+kpwdS9iUHvO4t6iZd/3z17hq3D 3dLg== X-Forwarded-Encrypted: i=1; AJvYcCVpQZxetNFN/q3qh2apjvMXWST5sw9VW/7Q0/c5DZEDbNoPwyRx0UHgSY5rhgUXHmVqokI=@dpdk.org X-Gm-Message-State: AOJu0Yx0PeSxejxnWQLk+PVKhwr1AzE+b9utqi4ErFI7qn/DP9nErnxJ VNZU742pz7dGtMJzT8398G4e/anpvRK0H09+tc5QFPikil0mxXNzj1ni834S7Y0JFBJUqAZFASc 5rPJ2RKjHAeuyQMgWWJS1QQqtiFcswOb5R5nhFBreKA== X-Gm-Gg: ASbGncs1BtbYm0CBq075y1wE41ibttC4d3MWDqrgLzOkj7aCExOdYHDahp0B7B6yzKA gH8DtsgjFM0q2ar0skLJKpHMYcV2XVd4yz2sxaLMklrqzZ5I2XBR9WWjbKOO0JvHo7sP+1yXtqm vPaU9N/aCRNzqC5U55YTXGQYS9RAB+PSRe2D4CBRt4mX2DSK2LRgTPAJk2kfqv2chnBM/ubJ1m3 uW6QoJ9h9KjReE6Mm4sgEPBVRj0JdsurQMOpfXRqd9ZOJBB5AcZ/mEOnsqcbBL8nTIeIYPBdBSA 8MSyx1eqCw== X-Google-Smtp-Source: AGHT+IFp1IYhkxkm3E4JXS1H4Qll8yxEqYoZKVIt20qq5mBpQuAKXrtONOG+bTm4ntHGQu2uT6SG6BercDl65FJm8Z4= X-Received: by 2002:ad4:5cc9:0:b0:707:71d9:d6da with SMTP id 6a1803df08f44-70d9722a7abmr184684006d6.10.1756209739161; Tue, 26 Aug 2025 05:02:19 -0700 (PDT) MIME-Version: 1.0 References: <20250825123548.1177090-1-ciara.loftus@intel.com> <20250825123548.1177090-4-ciara.loftus@intel.com> In-Reply-To: From: Khadem Ullah <14pwcse1224@uetpeshawar.edu.pk> Date: Tue, 26 Aug 2025 17:02:07 +0500 X-Gm-Features: Ac12FXyZyc-LFqTCkmhABhGq3NU7BgexgmmvpNdsogzUVp23qhCgH8wHUq493oU Message-ID: Subject: Re: [PATCH 3/4] net/ice: remove SSE vector path To: Bruce Richardson Cc: Ciara Loftus , dev@dpdk.org Content-Type: multipart/alternative; boundary="000000000000a7e844063d437176" X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org --000000000000a7e844063d437176 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Tue, Aug 26, 2025 at 3:50=E2=80=AFPM Bruce Richardson wrote: > On Tue, Aug 26, 2025 at 02:29:28PM +0500, Khadem Ullah wrote: > > Hi Bruce, > > Thanks for the feedback! > > No problem. > BTW: Please don't top-post in replying - it's best practice to put > the reply below the text you are replying to. Thanks. > > Ohh, I got it :) I was trimming the quotes when replying but in top-post format, I will always try to avoid a top-post reply in future! > AVX2 was first available in systems starting in 2013, (and AMD systems > since 2015), so at this point it's been around a long time. The SSE code > paths in the drivers will only be used by systems which do not have AVX2 = on > them - which should be relatively rare, I hope, at this point. There are = no > features in the SSE driver that are not available in the AVX2 one, so, I'= m > not aware of any reason why one would need to use the SSE code path in a > deployment of DPDK Yes, I think all features in SSE do already exist in AVX2 paths. > > Even without this patch, there will be no features added to the SSE code > paths in the drivers. Any new additions would just be to the AVX2 and > AVX-512 code paths. Even for systems without AVX2, if the SSE path is > removed the driver will fall-back to the scalar paths, which have far mor= e > features available in them than the SSE codepaths, which were simplified > for > performance reasons. > > /Bruce > Thanks for the update. I could not exactly get the meaning of fall-back to the scalar path. Does that mean that the driver automatically switches to the scalar path ? which is slower but includes all the necessary features that were simplified in the AVX2 path. I believe AVX2 provides an average performance much better for small frame or packet size (about 14 Gbps). Please forget the previous email, some mess up with practicing not doing top-posts. Hope this will be fine :) Regards, Khadem --000000000000a7e844063d437176 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable


On T= ue, Aug 26, 2025 at 3:50=E2=80=AFPM Bruce Richardson <bruce.richardson@intel.com> wrote:
On Tue, Aug 26, 2025 = at 02:29:28PM +0500, Khadem Ullah wrote:
>=C2=A0 =C2=A0 Hi Bruce,
>=C2=A0 =C2=A0 Thanks for the feedback!

No problem.
BTW: Please don't top-post in replying - it's best practice to put<= br> the reply below the text you are replying to. Thanks.

<= div>
Ohh, I got it :)=C2=A0=C2=A0
I was trimming the quot= es when replying but in top-post format,=C2=A0
I will always try t= o avoid a top-post reply in future!=C2=A0
=C2=A0
AVX2 was first available in systems starti= ng in 2013, (and AMD systems
since 2015), so at this point it's been around a long time. The SSE cod= e
paths in the drivers will only be used by systems which do not have AVX2 on=
them - which should be relatively rare, I hope, at this point. There are no=
features in the SSE driver that are not available in the AVX2 one, so, I= 9;m
not aware of any reason why one would need to use the SSE code path in a deployment of DPDK
=C2=A0
Yes, I think all features in= SSE do already exist in AVX2 paths.=C2=A0 =C2=A0

Even without this patch, there will be no features added to the SSE code paths in the drivers. Any new additions would just be to the AVX2 and
AVX-512 code paths. Even for systems without AVX2, if the SSE path is
removed the driver will fall-back to the scalar paths, which have far more<= br> features available in them than the SSE codepaths, which were simplified fo= r
performance reasons.

/Bruce
Thanks for the update. I could not e= xactly get the meaning of fall-back to the scalar path.=C2=A0
Does that = mean=C2=A0that the driver automatically switches to= the scalar path ?
which is slower but includes all the necessary featur= es that were simplified in the AVX2 path.=C2=A0=C2=A0
I be= lieve AVX2 provides an average performance much better for small frame or p= acket size (about 14 Gbps).=C2=A0

Please forget the previ= ous email, some mess up with practicing not doing top-posts.=C2=A0
Hope = this will be fine :)
Regards,
Khadem=C2=A0


=C2=A0
=
--000000000000a7e844063d437176--