From: Gregory CLEMENT <gregory.clement@bootlin.com>
To: Bart Van Assche <bvanassche@acm.org>,
"Martin K . Petersen" <martin.petersen@oracle.com>
Cc: linux-scsi@vger.kernel.org,
"Bart Van Assche" <bvanassche@acm.org>,
"Neil Armstrong" <neil.armstrong@linaro.org>,
"孙魁 (Kui Sun)" <kui.sun@unisoc.com>,
"André Draszik" <andre.draszik@linaro.org>,
"Sebastian Andrzej Siewior" <bigeasy@linutronix.de>
Subject: Re: [PATCH] ufs: core: Revert "Delegate the interrupt service routine to a threaded IRQ handler"
Date: Thu, 16 Jul 2026 13:37:17 +0200 [thread overview]
Message-ID: <87se5j5fo2.fsf@BLaptop.bootlin.com> (raw)
In-Reply-To: <b70eb60a01f971bed68c42c5b555929db5f835df.1784135511.git.bvanassche@acm.org>
Hello Bart,
> There have been multiple reports of performance regressions caused by
> commit 3c7ac40d7322. Hence this revert.
>
> This patch reverts most of the following commits:
> * 3c7ac40d7322 ("scsi: ufs: core: Delegate the interrupt service routine to
> a threaded IRQ handler")
> * 6475cfb81fc4 ("scsi: ufs: core: Avoid IRQ thread wakeup during active UIC
> command")
>
> This patch preserves the following commits:
> * 034d319c8899 ("scsi: ufs: core: Fix interrupt handling for MCQ Mode")
> * eabcac808ca3 ("scsi: ufs: core: Fix IRQ lock inversion for the SCSI host
> lock")
>
> Neil, please check whether this patch series fixes the stalls that you
> reported: [f2fs-dev] [PATCH 0/4] Reduce the time spent in interrupt context
> (https://sourceforge.net/p/linux-f2fs/mailman/linux-f2fs-devel/thread/4b337c7f-2fbd-47b6-9c8c-587e7b5513f1%40kernel.org/#msg59345049).
>
This patch appears to be a good solution to fix the issue I reported
with PREEMPT_TR enabled. I don't currently have the hardware available
to test it, but as soon as I do, I will test this patch.
Thanks,
Gregory
> Cc: Neil Armstrong <neil.armstrong@linaro.org>
> Cc: 孙魁 (Kui Sun) <kui.sun@unisoc.com>
> Cc: André Draszik <andre.draszik@linaro.org>
> Cc: Gregory CLEMENT <gregory.clement@bootlin.com>
> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
> Fixes: 3c7ac40d7322 ("scsi: ufs: core: Delegate the interrupt service routine to a threaded IRQ handler")
> Signed-off-by: Bart Van Assche <bvanassche@acm.org>
> ---
> drivers/ufs/core/ufshcd.c | 39 +++------------------------------------
> 1 file changed, 3 insertions(+), 36 deletions(-)
>
> diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
> index 79331c2bd38d..56c96a3b10e2 100644
> --- a/drivers/ufs/core/ufshcd.c
> +++ b/drivers/ufs/core/ufshcd.c
> @@ -7356,7 +7356,7 @@ static irqreturn_t ufshcd_sl_intr(struct ufs_hba *hba, u32 intr_status)
> }
>
> /**
> - * ufshcd_threaded_intr - Threaded interrupt service routine
> + * ufshcd_intr - Main interrupt service routine
> * @irq: irq number
> * @__hba: pointer to adapter instance
> *
> @@ -7364,7 +7364,7 @@ static irqreturn_t ufshcd_sl_intr(struct ufs_hba *hba, u32 intr_status)
> * IRQ_HANDLED - If interrupt is valid
> * IRQ_NONE - If invalid interrupt
> */
> -static irqreturn_t ufshcd_threaded_intr(int irq, void *__hba)
> +static irqreturn_t ufshcd_intr(int irq, void *__hba)
> {
> u32 last_intr_status, intr_status, enabled_intr_status = 0;
> irqreturn_t retval = IRQ_NONE;
> @@ -7403,38 +7403,6 @@ static irqreturn_t ufshcd_threaded_intr(int irq, void *__hba)
> return retval;
> }
>
> -/**
> - * ufshcd_intr - Main interrupt service routine
> - * @irq: irq number
> - * @__hba: pointer to adapter instance
> - *
> - * Return:
> - * IRQ_HANDLED - If interrupt is valid
> - * IRQ_WAKE_THREAD - If handling is moved to threaded handled
> - * IRQ_NONE - If invalid interrupt
> - */
> -static irqreturn_t ufshcd_intr(int irq, void *__hba)
> -{
> - struct ufs_hba *hba = __hba;
> - u32 intr_status, enabled_intr_status;
> -
> - /*
> - * Handle interrupt in thread if MCQ or ESI is disabled,
> - * and no active UIC command.
> - */
> - if ((!hba->mcq_enabled || !hba->mcq_esi_enabled) &&
> - !hba->active_uic_cmd)
> - return IRQ_WAKE_THREAD;
> -
> - intr_status = ufshcd_readl(hba, REG_INTERRUPT_STATUS);
> - enabled_intr_status = intr_status & ufshcd_readl(hba, REG_INTERRUPT_ENABLE);
> -
> - ufshcd_writel(hba, intr_status, REG_INTERRUPT_STATUS);
> -
> - /* Directly handle interrupts since MCQ ESI handlers does the hard job */
> - return ufshcd_sl_intr(hba, enabled_intr_status);
> -}
> -
> static int ufshcd_clear_tm_cmd(struct ufs_hba *hba, int tag)
> {
> int err = 0;
> @@ -11233,8 +11201,7 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
> ufshcd_readl(hba, REG_INTERRUPT_ENABLE);
>
> /* IRQ registration */
> - err = devm_request_threaded_irq(dev, irq, ufshcd_intr, ufshcd_threaded_intr,
> - IRQF_ONESHOT | IRQF_SHARED, UFSHCD, hba);
> + err = devm_request_irq(dev, irq, ufshcd_intr, IRQF_SHARED, UFSHCD, hba);
> if (err) {
> dev_err(hba->dev, "request irq failed\n");
> goto out_disable;
--
Grégory CLEMENT, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
next prev parent reply other threads:[~2026-07-16 11:37 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-15 17:12 [PATCH] ufs: core: Revert "Delegate the interrupt service routine to a threaded IRQ handler" Bart Van Assche
2026-07-16 11:37 ` Gregory CLEMENT [this message]
2026-07-16 12:41 ` Sebastian Andrzej Siewior
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=87se5j5fo2.fsf@BLaptop.bootlin.com \
--to=gregory.clement@bootlin.com \
--cc=andre.draszik@linaro.org \
--cc=bigeasy@linutronix.de \
--cc=bvanassche@acm.org \
--cc=kui.sun@unisoc.com \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=neil.armstrong@linaro.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