Re: rx2660 + debian
- To: Anatoly Pugachev <matorola@gmail.com>
- Cc: Pedro Miguel Justo <pmsjt@texair.net>, Anton Borisov <anton.borisov@gmail.com>, John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>, debian-ia64 <debian-ia64@lists.debian.org>, Sergei Trofimovich <slyich@gmail.com>
- Subject: Re: rx2660 + debian
- From: Frank Scheiner <frank.scheiner@web.de>
- Date: Wed, 4 May 2022 10:11:51 +0200
- Message-id: <[🔎] 996c2425-6ea2-188f-039f-91bb78dca6e2@web.de>
- In-reply-to: <CADxRZqzyoyhNwKJ9spwmNtNJVyLtsF0dL6CuPLkEqmMQPR-cpA@mail.gmail.com>
- References: <CANHzssBhm=A2Q+npDqdwYsd9DKhqYyT_VVnYwQr=7WpPBNccqg@mail.gmail.com> <db15a1d2-69be-8a02-476f-d7c183c2f46e@web.de> <CANHzssC5sa4JpiW_ZrbW0yDWnxa1u9j56eVCETE-M5qMD6YTVA@mail.gmail.com> <0927961a-1c49-e57f-37db-fd62e5b19a6e@web.de> <6c888a13-4737-8a1b-023f-c683f12e083d@web.de> <E8BAF1B0-6C33-4E65-BCEC-B2D258D97898@texair.net> <c7c5702b-8123-b3fc-c0f5-c5e132393d60@physik.fu-berlin.de> <E0CC7F7A-01A2-4F36-A080-266B56830C74@texair.net> <549c036f-fbb3-f990-deb3-4b47a57d38b2@physik.fu-berlin.de> <3FBD5838-E30A-4825-AD47-657F065062AE@texair.net> <8CDCF0F2-C2C2-4EED-ADD0-9E3889938908@texair.net> <6fc4b6d1-5c84-1eef-d7fa-952d9a09ee3f@physik.fu-berlin.de> <529EF8D2-F76E-4EC4-8642-ED7B0175E888@texair.net> <fa5bc5de-09f3-c047-5eb3-0253ed6959f5@physik.fu-berlin.de> <b1d000e5-1698-7f24-4d78-eced6c98e57b@web.de> <1E227A5F-0C03-4615-B921-D2EBED3DB058@texair.net> <42b98a5f-7f14-6c11-55e8-737ebd3ed50c@web.de> <CADxRZqzyoyhNwKJ9spwmNtNJVyLtsF0dL6CuPLkEqmMQPR-cpA@mail.gmail.com>
On 27.04.22 14:45, Anatoly Pugachev wrote:
is it possible to fill a bug report at linux kernel bugzilla and post
full stack trace with the description
to: linux-ia64@vger.k.o
cc: linux-kernel@vger.k.o , linux-mm@kvack.org
and CC to the author of the patch as well? https://lwn.net/Articles/694470/
Sorry, but I was out of time last week and this week's already pretty
full. I'll look into this as soon as time allows, although I believe I
am not the original poster of these memcopy problems.
But I can add some additional info now: Last week I also looked at my
rx4640 with 4 x Madison processors and it looks like this machine (or
its processors) are affected as much as the Montvale machines/processors
- so no successful operation w/ and w/o `hardened_usercopy=off`.
So in short, you're (relatively) fine with Montecitos (both in rx2620
and rx2660) and `hardened_usercopy=off` and not affected visibly with
Tukwilas (in rx2800 i2). But Madisons (in rx4640) and Montvales (in
rx2660 and rx6600 and most likely also rx3600) are not good due to this
memcopy problem.
Cheers,
Frank
Reply to: