Linux Kernel Selftest development
 help / color / mirror / Atom feed
From: "Arnd Bergmann" <arnd@arndb.de>
To: "Maciej W. Rozycki" <macro@orcam.me.uk>
Cc: "Andy Lutomirski" <luto@kernel.org>,
	"Thomas Weißschuh" <thomas.weissschuh@linutronix.de>,
	"Thomas Gleixner" <tglx@linutronix.de>,
	"Vincenzo Frascino" <vincenzo.frascino@arm.com>,
	"David S . Miller" <davem@davemloft.net>,
	"Andreas Larsson" <andreas@gaisler.com>,
	"Nick Alcock" <nick.alcock@oracle.com>,
	"John Stultz" <jstultz@google.com>,
	"Stephen Boyd" <sboyd@kernel.org>,
	"John Paul Adrian Glaubitz" <glaubitz@physik.fu-berlin.de>,
	shuah <shuah@kernel.org>,
	"Catalin Marinas" <catalin.marinas@arm.com>,
	"Will Deacon" <will@kernel.org>, "Theodore Ts'o" <tytso@mit.edu>,
	"Jason A . Donenfeld" <Jason@zx2c4.com>,
	"Russell King" <linux@armlinux.org.uk>,
	"Madhavan Srinivasan" <maddy@linux.ibm.com>,
	"Michael Ellerman" <mpe@ellerman.id.au>,
	"Nicholas Piggin" <npiggin@gmail.com>,
	"Christophe Leroy" <christophe.leroy@csgroup.eu>,
	"Huacai Chen" <chenhuacai@kernel.org>,
	"WANG Xuerui" <kernel@xen0n.name>,
	"Thomas Bogendoerfer" <tsbogend@alpha.franken.de>,
	"Heiko Carstens" <hca@linux.ibm.com>,
	"Vasily Gorbik" <gor@linux.ibm.com>,
	"Alexander Gordeev" <agordeev@linux.ibm.com>,
	"Christian Borntraeger" <borntraeger@linux.ibm.com>,
	"Sven Schnelle" <svens@linux.ibm.com>,
	"Shannon Nelson" <sln@onemain.com>,
	"Linux Kernel Mailing List" <linux-kernel@vger.kernel.org>,
	sparclinux@vger.kernel.org, linux-kselftest@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linuxppc-dev@lists.ozlabs.org, loongarch@lists.linux.dev,
	linux-mips@vger.kernel.org, linux-s390@vger.kernel.org,
	"Arnd Bergmann" <arnd@kernel.org>
Subject: Re: [PATCH v5 00/34] sparc64: vdso: Switch to the generic vDSO library
Date: Sun, 09 Nov 2025 17:08:27 +0100	[thread overview]
Message-ID: <7eb1a661-e5bf-45cb-a3cf-7337fb0c4173@app.fastmail.com> (raw)
In-Reply-To: <alpine.DEB.2.21.2511090221080.25436@angie.orcam.me.uk>

On Sun, Nov 9, 2025, at 04:23, Maciej W. Rozycki wrote:
> On Sat, 8 Nov 2025, Arnd Bergmann wrote:
>
>> On other architectures, I see that parisc (always aliasing) has stubbed
>> out the vdso functions, while mips/loongson has limited the page size
>> selection to never alias. A few other mips platforms can theoretically
>> enable both small pages and vdso, but my guess is that in practice
>> they don't use the vdso (mips32/ath79) or they use 16KB pages
>> (rm, dec, ip22) based on the defconfig settings.
>
>  Umm, I'd have to dive into the details (and I hardly have the resources 
> at hand), but quite a bunch of MIPS microarchitectures suffer from cache 
> aliases; some even have VIVT caches.

I was going with this list:

$ git grep define.cpu_has_dc_aliases arch/mips/
arch/mips/include/asm/cpu-features.h:#define cpu_has_dc_aliases (cpu_data[0].dcache.flags & MIPS_CACHE_ALIASES)
arch/mips/include/asm/mach-ath79/cpu-feature-overrides.h:#define cpu_has_dc_aliases     1
arch/mips/include/asm/mach-au1x00/cpu-feature-overrides.h:#define cpu_has_dc_aliases            0
arch/mips/include/asm/mach-bcm63xx/cpu-feature-overrides.h:#define cpu_has_dc_aliases           0
arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h:#define cpu_has_dc_aliases     0
arch/mips/include/asm/mach-dec/cpu-feature-overrides.h:#define cpu_has_dc_aliases               0
arch/mips/include/asm/mach-dec/cpu-feature-overrides.h:#define cpu_has_dc_aliases               (PAGE_SIZE < 0x4000)
arch/mips/include/asm/mach-ingenic/cpu-feature-overrides.h:#define cpu_has_dc_aliases   0
arch/mips/include/asm/mach-ip22/cpu-feature-overrides.h:#define cpu_has_dc_aliases      (PAGE_SIZE < 0x4000)
arch/mips/include/asm/mach-ip27/cpu-feature-overrides.h:#define cpu_has_dc_aliases              0
arch/mips/include/asm/mach-ip28/cpu-feature-overrides.h:#define cpu_has_dc_aliases      0 /* see probe_pcache() */
arch/mips/include/asm/mach-ip30/cpu-feature-overrides.h:#define cpu_has_dc_aliases              0
arch/mips/include/asm/mach-loongson2ef/cpu-feature-overrides.h:#define cpu_has_dc_aliases       (PAGE_SIZE < 0x4000)
arch/mips/include/asm/mach-loongson64/cpu-feature-overrides.h:#define cpu_has_dc_aliases        (PAGE_SIZE < 0x4000)
arch/mips/include/asm/mach-malta/cpu-feature-overrides.h:/* #define cpu_has_dc_aliases  ? */
arch/mips/include/asm/mach-malta/cpu-feature-overrides.h:/* #define cpu_has_dc_aliases  ? */
arch/mips/include/asm/mach-ralink/mt7621/cpu-feature-overrides.h:#define cpu_has_dc_aliases     0
arch/mips/include/asm/mach-rm/cpu-feature-overrides.h:#define cpu_has_dc_aliases        (PAGE_SIZE < 0x4000)
arch/mips/include/asm/mach-sibyte/cpu-feature-overrides.h:#define cpu_has_dc_aliases    0

which for many platforms seems to come up with a compile-time
constant value. I hadn't checked the exact conditions for
the runtime MIPS_CACHE_ALIASES flag, but I see that the kernel
has a fairly centralized detection function in probe_pcache()
https://elixir.bootlin.com/linux/v6.17.7/source/arch/mips/mm/c-r4k.c#L1251

which is used for the 'generic' platform and those that don't set
a compile-time constant (ath25, bcm47xx, bmips, cobalt, ip32, malta,
pic32, rc32434 and tx49xx).

> (see the figures at the bottom; uptime quoted for an idea of the rate, 
> though the system hasn't been heavily loaded).  It is possible with the 
> aid of S$, which is inclusive and PIPT.

I now found commit 0f02cfbc3d9e ("MIPS: VDSO: Match data page
cache colouring when D$ aliases"), which probably does everything
necessary to just make it work reliably on mips, and should
be portable to sparc as Andy suggested as well.

     Arnd

  reply	other threads:[~2025-11-09 16:09 UTC|newest]

Thread overview: 54+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-06 10:01 [PATCH v5 00/34] sparc64: vdso: Switch to the generic vDSO library Thomas Weißschuh
2025-11-06 10:01 ` [PATCH v5 01/34] selftests: vDSO: vdso_test_correctness: Handle different tv_usec types Thomas Weißschuh
2025-11-06 10:01 ` [PATCH v5 02/34] arm64: vDSO: getrandom: Explicitly include asm/alternative.h Thomas Weißschuh
2025-11-06 10:01 ` [PATCH v5 03/34] arm64: vDSO: gettimeofday: Explicitly include vdso/clocksource.h Thomas Weißschuh
2025-11-06 10:01 ` [PATCH v5 04/34] arm64: vDSO: compat_gettimeofday: Add explicit includes Thomas Weißschuh
2025-11-06 10:01 ` [PATCH v5 05/34] ARM: vdso: gettimeofday: " Thomas Weißschuh
2025-11-06 10:01 ` [PATCH v5 06/34] powerpc/vdso/gettimeofday: Explicitly include vdso/time32.h Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 07/34] powerpc/vdso: Explicitly include asm/cputable.h and asm/feature-fixups.h Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 08/34] LoongArch: vDSO: Explicitly include asm/vdso/vdso.h Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 09/34] MIPS: vdso: Add include guard to asm/vdso/vdso.h Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 10/34] MIPS: vdso: Explicitly include asm/vdso/vdso.h Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 11/34] random: vDSO: Add explicit includes Thomas Weißschuh
2025-11-07 23:59   ` Jason A. Donenfeld
2025-11-06 10:02 ` [PATCH v5 12/34] vdso/gettimeofday: " Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 13/34] vdso/helpers: Explicitly include vdso/processor.h Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 14/34] vdso/datapage: Remove inclusion of gettimeofday.h Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 15/34] vdso/datapage: Trim down unnecessary includes Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 16/34] random: vDSO: trim vDSO includes Thomas Weißschuh
2025-11-07 23:49   ` Jason A. Donenfeld
2025-11-08  0:00     ` Jason A. Donenfeld
2025-11-06 10:02 ` [PATCH v5 17/34] random: vDSO: remove ifdeffery Thomas Weißschuh
2025-11-07 23:37   ` Jason A. Donenfeld
2025-11-10  8:45     ` Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 18/34] random: vDSO: split out datapage update into helper functions Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 19/34] random: vDSO: only access vDSO datapage after random_init() Thomas Weißschuh
2025-11-07 23:46   ` Jason A. Donenfeld
2025-11-10  9:04     ` Thomas Weißschuh
2025-11-10 10:37       ` Jason A. Donenfeld
2025-11-10 11:24         ` Thomas Weißschuh
2025-11-10 11:40           ` Jason A. Donenfeld
2025-11-11  8:55             ` Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 20/34] s390/time: Set up vDSO datapage later Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 21/34] vdso/datastore: Reduce scope of some variables in vvar_fault() Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 22/34] vdso/datastore: Drop inclusion of linux/mmap_lock.h Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 23/34] vdso/datastore: Allocate data pages dynamically Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 24/34] sparc64: vdso: Link with -z noexecstack Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 25/34] sparc64: vdso: Remove obsolete "fake section table" reservation Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 26/34] sparc64: vdso: Replace code patching with runtime conditional Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 27/34] sparc64: vdso: Move hardware counter read into header Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 28/34] sparc64: vdso: Move syscall fallbacks " Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 29/34] sparc64: vdso: Introduce vdso/processor.h Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 30/34] sparc64: vdso: Switch to the generic vDSO library Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 31/34] sparc64: vdso2c: Drop sym_vvar_start handling Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 32/34] sparc64: vdso2c: Remove symbol handling Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 33/34] sparc64: vdso: Implement clock_gettime64() Thomas Weißschuh
2025-11-06 10:02 ` [PATCH v5 34/34] clocksource: remove ARCH_CLOCKSOURCE_DATA Thomas Weißschuh
2025-11-06 12:55 ` [PATCH v5 00/34] sparc64: vdso: Switch to the generic vDSO library Aithal, Srikanth
2025-11-06 14:02 ` Mark Brown
2025-11-06 17:20 ` Arnd Bergmann
2025-11-08  0:17 ` Andy Lutomirski
2025-11-08 10:17   ` Arnd Bergmann
2025-11-09  3:23     ` Maciej W. Rozycki
2025-11-09 16:08       ` Arnd Bergmann [this message]
2026-01-16 12:10   ` Thomas Weißschuh

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=7eb1a661-e5bf-45cb-a3cf-7337fb0c4173@app.fastmail.com \
    --to=arnd@arndb.de \
    --cc=Jason@zx2c4.com \
    --cc=agordeev@linux.ibm.com \
    --cc=andreas@gaisler.com \
    --cc=arnd@kernel.org \
    --cc=borntraeger@linux.ibm.com \
    --cc=catalin.marinas@arm.com \
    --cc=chenhuacai@kernel.org \
    --cc=christophe.leroy@csgroup.eu \
    --cc=davem@davemloft.net \
    --cc=glaubitz@physik.fu-berlin.de \
    --cc=gor@linux.ibm.com \
    --cc=hca@linux.ibm.com \
    --cc=jstultz@google.com \
    --cc=kernel@xen0n.name \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=linux-mips@vger.kernel.org \
    --cc=linux-s390@vger.kernel.org \
    --cc=linux@armlinux.org.uk \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=loongarch@lists.linux.dev \
    --cc=luto@kernel.org \
    --cc=macro@orcam.me.uk \
    --cc=maddy@linux.ibm.com \
    --cc=mpe@ellerman.id.au \
    --cc=nick.alcock@oracle.com \
    --cc=npiggin@gmail.com \
    --cc=sboyd@kernel.org \
    --cc=shuah@kernel.org \
    --cc=sln@onemain.com \
    --cc=sparclinux@vger.kernel.org \
    --cc=svens@linux.ibm.com \
    --cc=tglx@linutronix.de \
    --cc=thomas.weissschuh@linutronix.de \
    --cc=tsbogend@alpha.franken.de \
    --cc=tytso@mit.edu \
    --cc=vincenzo.frascino@arm.com \
    --cc=will@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox