From: Palmer Dabbelt <palmerdabbelt@google.com>
To: sashal@kernel.org
Cc: linux-kernel@vger.kernel.org, stable@vger.kernel.org,
sashal@kernel.org, linux-riscv@lists.infradead.org
Subject: Re: [PATCH AUTOSEL 4.19 18/19] RISC-V: Upgrade smp_mb__after_spinlock() to iorw,iorw
Date: Sun, 26 Jul 2020 11:36:58 -0700 (PDT) [thread overview]
Message-ID: <mhng-5bf9e67f-f6f9-4c9b-9d56-afd0a6e21cd6@palmerdabbelt-glaptop1> (raw)
In-Reply-To: <20200720213851.407715-18-sashal@kernel.org>
On Mon, 20 Jul 2020 14:38:49 PDT (-0700), sashal@kernel.org wrote:
> From: Palmer Dabbelt <palmerdabbelt@google.com>
>
> [ Upstream commit 38b7c2a3ffb1fce8358ddc6006cfe5c038ff9963 ]
>
> While digging through the recent mmiowb preemption issue it came up that
> we aren't actually preventing IO from crossing a scheduling boundary.
> While it's a bit ugly to overload smp_mb__after_spinlock() with this
> behavior, it's what PowerPC is doing so there's some precedent.
>
> Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
> Signed-off-by: Sasha Levin <sashal@kernel.org>
> ---
> arch/riscv/include/asm/barrier.h | 10 +++++++++-
> 1 file changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/arch/riscv/include/asm/barrier.h b/arch/riscv/include/asm/barrier.h
> index d4628e4b3a5ea..f4c92c91aa047 100644
> --- a/arch/riscv/include/asm/barrier.h
> +++ b/arch/riscv/include/asm/barrier.h
> @@ -69,8 +69,16 @@ do { \
> * The AQ/RL pair provides a RCpc critical section, but there's not really any
> * way we can take advantage of that here because the ordering is only enforced
> * on that one lock. Thus, we're just doing a full fence.
> + *
> + * Since we allow writeX to be called from preemptive regions we need at least
> + * an "o" in the predecessor set to ensure device writes are visible before the
> + * task is marked as available for scheduling on a new hart. While I don't see
> + * any concrete reason we need a full IO fence, it seems safer to just upgrade
> + * this in order to avoid any IO crossing a scheduling boundary. In both
> + * instances the scheduler pairs this with an mb(), so nothing is necessary on
> + * the new hart.
> */
> -#define smp_mb__after_spinlock() RISCV_FENCE(rw,rw)
> +#define smp_mb__after_spinlock() RISCV_FENCE(iorw,iorw)
>
> #include <asm-generic/barrier.h>
While I don't think it hurts to have this, IIRC we didn't have the generic
mmiowb spinlock stuff back then so it doesn't really fix the problem. That
said, I'm pretty sure 4.19 doesn't make it to userspace so backports are really
an academic discussion at this point. Whatever's less work for everyone is
fine on my end for 4.19.
next prev parent reply other threads:[~2020-07-26 18:37 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-20 21:38 [PATCH AUTOSEL 4.19 01/19] HID: i2c-hid: add Mediacom FlexBook edge13 to descriptor override Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 02/19] HID: alps: support devices with report id 2 Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 03/19] HID: steam: fixes race in handling device list Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 04/19] HID: apple: Disable Fn-key key-re-mapping on clone keyboards Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 05/19] dmaengine: tegra210-adma: Fix runtime PM imbalance on error Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 06/19] Input: add `SW_MACHINE_COVER` Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 07/19] spi: mediatek: use correct SPI_CFG2_REG MACRO Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 08/19] regmap: dev_get_regmap_match(): fix string comparison Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 09/19] hwmon: (aspeed-pwm-tacho) Avoid possible buffer overflow Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 10/19] dmaengine: ioat setting ioat timeout as module parameter Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 11/19] Input: synaptics - enable InterTouch for ThinkPad X1E 1st gen Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 12/19] Input: elan_i2c - only increment wakeup count on touch Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 13/19] usb: gadget: udc: gr_udc: fix memleak on error handling path in gr_ep_init() Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 14/19] hwmon: (adm1275) Make sure we are reading enough data for different chips Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 15/19] hwmon: (scmi) Fix potential buffer overflow in scmi_hwmon_probe() Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 16/19] arm64: Use test_tsk_thread_flag() for checking TIF_SINGLESTEP Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 17/19] x86: math-emu: Fix up 'cmp' insn for clang ias Sasha Levin
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 18/19] RISC-V: Upgrade smp_mb__after_spinlock() to iorw,iorw Sasha Levin
2020-07-26 18:36 ` Palmer Dabbelt [this message]
2020-07-20 21:38 ` [PATCH AUTOSEL 4.19 19/19] SUNRPC reverting d03727b248d0 ("NFSv4 fix CLOSE not waiting for direct IO compeletion") Sasha Levin
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=mhng-5bf9e67f-f6f9-4c9b-9d56-afd0a6e21cd6@palmerdabbelt-glaptop1 \
--to=palmerdabbelt@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-riscv@lists.infradead.org \
--cc=sashal@kernel.org \
--cc=stable@vger.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;
as well as URLs for NNTP newsgroup(s).