Date: Tue, 23 Oct 2018 00:00:39 +0200 From: Hannes Hauswedell <h2+lists2018@fsfe.org> To: freebsd-current@freebsd.org Subject: Re: Current status of Ryzen (non-)support Message-ID: <6845b53c-63db-7a76-cddb-2ca1b8408e1b@hauswedell.net> In-Reply-To: <20181021173327.GJ5335@kib.kiev.ua> References: <632587eb-2c27-2e7f-2571-25c06025cd80@fsfe.org> <20181021173327.GJ5335@kib.kiev.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --8x5LMPkBHahSYc4UliLd6P2QVCpF5GcuV Content-Type: multipart/mixed; boundary="LDrIDj0YwWjWeXysQ05ONSkk1imU4W9Me"; protected-headers="v1" From: Hannes Hauswedell <h2+lists2018@fsfe.org> To: freebsd-current@freebsd.org Message-ID: <6845b53c-63db-7a76-cddb-2ca1b8408e1b@hauswedell.net> Subject: Re: Current status of Ryzen (non-)support References: <632587eb-2c27-2e7f-2571-25c06025cd80@fsfe.org> <20181021173327.GJ5335@kib.kiev.ua> In-Reply-To: <20181021173327.GJ5335@kib.kiev.ua> --LDrIDj0YwWjWeXysQ05ONSkk1imU4W9Me Content-Type: text/plain; charset=utf-8 Content-Language: en-GB Content-Transfer-Encoding: quoted-printable On 21.10.18 19:33, Konstantin Belousov wrote: > On Sun, Oct 21, 2018 at 06:09:50PM +0200, Hannes Hauswedell wrote: >> Hi everyone, >> >> I wanted to ask what the current status of Ryzen support is and/or >> whether any new changes are planned. >> >> My situation: >> * second generation Ryzen: 2600X >> * running -CURRENT >> * I have done the things described here: >> https://lists.freebsd.org/pipermail/freebsd-current/2018-June/069799.h= tml > This was based on the public errata document for the first gen ryzen. > I am not aware of the erratas ('revision guides' in AMD terminology) > for the second gen ryzens. >=20 > For now you can only get fixes for microcode from your BIOS updates. >=20 Thanks for the replies everyone! I ran stress-ng --vm 12 --vm-bytes 12G --verify -v and could reproducibly produce memory issues on FreeBSD, but not on Linux= =2E After updating the BIOS and deactivating swap on FreeBSD they seem to be gone now, I am still investigating whether turning swap back on brings them back which would hint at a possible disk problem or whether the BIOS update changed this behaviour. Disk initially appeared more plausible, but further tests suggest against it (it's a fairly new SSD, multiple ZFS scrubs show now errors, there are no smart errors or other errors reported by the kernel in regard to the disk). Best regards, Hannes --=20 pgp-key: https://hannes.hauswedell.net/hannes_hauswedell_public_key.asc fingerprint: EF56 F66A A683 3E32 CE05 3F55 6F9F 2E30 ECB3 7977 --LDrIDj0YwWjWeXysQ05ONSkk1imU4W9Me-- --8x5LMPkBHahSYc4UliLd6P2QVCpF5GcuV Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- iHUEARYIAB0WIQTvVvZqpoM+Ms4FP1Vvny4w7LN5dwUCW85IkAAKCRBvny4w7LN5 dzX/AP9X1o39KyRRuXhQBtEbQwhpQciafOyd2rCcy+ouZRZOUgD/dy1ctvyxi+wV uMvKRVnKyvqfHIwj76iE5qRnE/uXGQY= =hMi1 -----END PGP SIGNATURE----- --8x5LMPkBHahSYc4UliLd6P2QVCpF5GcuV--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?6845b53c-63db-7a76-cddb-2ca1b8408e1b>