Date: Fri, 26 Oct 2018 18:15:59 +0300 From: Lev Serebryakov <lev@FreeBSD.org> To: "Pieper, Jeffrey E" <jeffrey.e.pieper@intel.com>, "freebsd-net@freebsd.org" <freebsd-net@freebsd.org> Subject: Re: What are conditions to spread traffic across different NIC queues? Message-ID: <49927120-c7d4-eb30-38ae-94ee31fbd4cb@FreeBSD.org> In-Reply-To: <08D0D8DC-FCA8-4B56-9F4C-F407F69EFDE6@intel.com> References: <cbcadec2-bf74-a1a9-2c66-97d9d409907f@FreeBSD.org> <370ccdb0-cb71-a4da-0a35-4e878eb6ba1c@FreeBSD.org> <08D0D8DC-FCA8-4B56-9F4C-F407F69EFDE6@intel.com>
next in thread | previous in thread | raw e-mail | index | archive | help
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --OZ41y2YXX30tagcM0iSSUTih6sKyOSMuA Content-Type: multipart/mixed; boundary="wHA7UhnXZ2SFoqNaAh1E0mKIkiE4Et0vn"; protected-headers="v1" From: Lev Serebryakov <lev@FreeBSD.org> Reply-To: lev@FreeBSD.org To: "Pieper, Jeffrey E" <jeffrey.e.pieper@intel.com>, "freebsd-net@freebsd.org" <freebsd-net@freebsd.org> Message-ID: <49927120-c7d4-eb30-38ae-94ee31fbd4cb@FreeBSD.org> Subject: Re: What are conditions to spread traffic across different NIC queues? References: <cbcadec2-bf74-a1a9-2c66-97d9d409907f@FreeBSD.org> <370ccdb0-cb71-a4da-0a35-4e878eb6ba1c@FreeBSD.org> <08D0D8DC-FCA8-4B56-9F4C-F407F69EFDE6@intel.com> In-Reply-To: <08D0D8DC-FCA8-4B56-9F4C-F407F69EFDE6@intel.com> --wHA7UhnXZ2SFoqNaAh1E0mKIkiE4Et0vn Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: quoted-printable On 26.10.2018 18:04, Pieper, Jeffrey E wrote: > So what exactly is your complaint? That a higher end chipset has more f= eatures than a low-end chipset?=20 Please, don't take my complain too serious, it was emotional reaction after 2 hours of googling, reading datasheets and investigation :-) I'm perfectly happy with Intel hardware, to be honest. And all newer chipsets support this feature even desktop ones (I210, for example, supports it according to datasheet). There is slight problem =E2=80=94 I can not pay something more and get b= etter in these days, as typically, nobody produce "same motherboard but with slightly better NICs for +$5/$10/$15". But it is not Intel problem, per se, but market problem. For example, MoBo in question is Intel D2500CC (yes, it is OLD, I know), and there was NO same MoBo with better NIC chips. Same with Supermicro, same with other =C2=ABembedded=C2=BB hardware manufacturers. = And don't ask me about desktop MoBos :-) --=20 // Lev Serebryakov --wHA7UhnXZ2SFoqNaAh1E0mKIkiE4Et0vn-- --OZ41y2YXX30tagcM0iSSUTih6sKyOSMuA Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- iQKTBAEBCgB9FiEE+W0coLX0MYtnSzMK6rA8WL/cR48FAlvTL69fFIAAAAAALgAo aXNzdWVyLWZwckBub3RhdGlvbnMub3BlbnBncC5maWZ0aGhvcnNlbWFuLm5ldEY5 NkQxQ0EwQjVGNDMxOEI2NzRCMzMwQUVBQjAzQzU4QkZEQzQ3OEYACgkQ6rA8WL/c R4/NiQ/9G+VdPVIksqx0RXzWRXG0lbxYq54K4acpLHaAOiyuS9k/xSN1BiZzA3qt Kij+1QLGltIa8Eiie5u2vXfZLuoH+OAcR3ZgMaQh+nvPZ9jHn0ChuwtLLSPiUrcK n03mR+v20Thme0uRt1NBlD27I4/5Wc4gAWxwXeYZZdOWQ1qnE0ChtfjtO1rZOTaT HVBCgf6+VsJsnkZlV/HBjKGumcWi87zMnXBoxZTrL8TJ+DSvhyst8FLOpc/UpPtF Ic8L6m12vsQ0MB2TXYyM/yajzyah6XXizhdNpCHdq96BmWtdTKdiz1gVz4BQH2AM g1+Ydw3XUD6Wqi/Rn0kHwY1ZjPEevIkUyKPLSnwQzemZDWqv5jVguE6crpYDdoRL Rff1uxBI5CQX4GP1/U9SX4R0e8pZ+qddWsPUBJTmPuDI4enEn0gb6xwerexQlOSm TtOIF2r7uHIOMzeKAXZ//7KDcDWJchr6zt8OjrSVtMVHbmgiUO1py++hbEh6rwlX dlpbyXc3gLO21R6GiCkdKybcBmsrzU56VJQFa/76iXqTTTOvvuV3Ivdf7hrefEbN S9SyK/v90rj/i+kyYVDGXH7+KjutDemMlxUjeeLUDjoNJ2+sEcRd3IcZLhJ3eMpT /dXui438X5l8SUmB6MTIi6AlMR8tsH4M8v2N3QHbO3Boxe2E6NY= =js2o -----END PGP SIGNATURE----- --OZ41y2YXX30tagcM0iSSUTih6sKyOSMuA--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?49927120-c7d4-eb30-38ae-94ee31fbd4cb>