Re: [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

To: Andy Lutomirski <luto@xxxxxxxxxxxxxx>
Subject: Re: [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables
From: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Date: Sat, 21 Jul 2018 14:06:52 -0700
Cc: Joerg Roedel <jroedel@xxxxxxx>, Andrew Lutomirski <luto@xxxxxxxxxx>, Joerg Roedel <joro@xxxxxxxxxx>, Thomas Gleixner <tglx@xxxxxxxxxxxxx>, Ingo Molnar <mingo@xxxxxxxxxx>, Peter Anvin <hpa@xxxxxxxxx>, "the arch/x86 maintainers" <x86@xxxxxxxxxx>, Linux Kernel Mailing List <linux-kernel@xxxxxxxxxxxxxxx>, linux-mm <linux-mm@xxxxxxxxx>, Dave Hansen <dave.hansen@xxxxxxxxx>, Josh Poimboeuf <jpoimboe@xxxxxxxxxx>, Jürgen Groß <jgross@xxxxxxxx>, Peter Zijlstra <peterz@xxxxxxxxxxxxx>, Borislav Petkov <bp@xxxxxxxxx>, Jiri Kosina <jkosina@xxxxxxx>, Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx>, Brian Gerst <brgerst@xxxxxxxxx>, David Laight <David.Laight@xxxxxxxxxx>, Denys Vlasenko <dvlasenk@xxxxxxxxxx>, Eduardo Valentin <eduval@xxxxxxxxxx>, Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>, Will Deacon <will.deacon@xxxxxxx>, "Liguori, Anthony" <aliguori@xxxxxxxxxx>, Daniel Gruss <daniel.gruss@xxxxxxxxxxxxxx>, Hugh Dickins <hughd@xxxxxxxxxx>, Kees Cook <keescook@xxxxxxxxxx>, Andrea Arcangeli <aarcange@xxxxxxxxxx>, Waiman Long <llong@xxxxxxxxxx>, Pavel Machek <pavel@xxxxxx>, "David H . Gutteridge" <dhgutteridge@xxxxxxxxxxxx>, Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>, Alexander Shishkin <alexander.shishkin@xxxxxxxxxxxxxxx>, Jiri Olsa <jolsa@xxxxxxxxxx>, Namhyung Kim <namhyung@xxxxxxxxxx>
In-reply-to: <D89602E9-E620-4AF0-822C-206D7F0BA071@amacapital.net>
References: <1532103744-31902-1-git-send-email-joro@8bytes.org> <1532103744-31902-2-git-send-email-joro@8bytes.org> <CALCETrXJX8tPVgD=Ce41534uneAAobm-HyjeGwVYgJDJ_+-bDw@mail.gmail.com> <20180720174846.GF18541@8bytes.org> <CALCETrUj4cLpOKUbJUfLqKJFkjAgeraE=ORQ-e-bKU+AHda0=Q@mail.gmail.com> <20180720213700.gh6d2qd2ck6nt4ax@suse.de> <D89602E9-E620-4AF0-822C-206D7F0BA071@amacapital.net>

On Fri, Jul 20, 2018 at 3:20 PM Andy Lutomirski <luto@xxxxxxxxxxxxxx> wrote:
> Thanks for digging!  The problem was presumably that vmalloc_fault() will IRET and re-enable NMIs on the way out.
>  But we’ve supported page faults on user memory in NMI handlers on 32-bit and 64-bit for quite a while, and it’s fine now.
>
> I would remove the warning, re-test, and revert the other patch.

Agreed. I don't think we have any issues with page faults during NMI
any more.  Afaik the kprobe people depend on it.

That said, 64-bit mode has that scary PV-op case
(arch_flush_lazy_mmu_mode). Being PV mode, I can't find it in myself
to worry about it, I'm assuming it's ok.

                Linus

References:
- [PATCH 0/3] PTI for x86-32 Fixes and Updates
  - From: Joerg Roedel
- [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables
  - From: Joerg Roedel
- Re: [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables
  - From: Andy Lutomirski
- Re: [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables
  - From: Joerg Roedel
- Re: [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables
  - From: Andy Lutomirski
- Re: [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables
  - From: Joerg Roedel
- Re: [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables
  - From: Andy Lutomirski

Prev by Date: Re: [Bug 200105] High paging activity as soon as the swap is touched (with steps and code to reproduce it)
Next by Date: Re: [PATCH] mm: thp: remove use_zero_page sysfs knob
Previous by thread: Re: [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables
Next by thread: Re: [PATCH 1/3] perf/core: Make sure the ring-buffer is mapped in all page-tables
Index(es):
- Date
- Thread

[Index of Archives] [Linux ARM Kernel] [Linux ARM] [Linux Omap] [Fedora ARM] [IETF Annouce] [Bugtraq] [Linux OMAP] [Linux MIPS] [eCos] [Asterisk Internet PBX] [Linux API]