From: Marc Zyngier <maz@kernel.org>
To: linux-kernel@vger.kernel.org,
linux-arm-kernel@lists.infradead.org, linux-acpi@vger.kernel.org
Cc: Thomas Gleixner <tglx@linutronix.de>,
Mark Rutland <mark.rutland@arm.com>,
Will Deacon <will@kernel.org>,
"Rafael J. Wysocki" <rafael@kernel.org>,
Rob Herring <robh@kernel.org>,
Saravana Kannan <saravanak@google.com>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Sven Peter <sven@kernel.org>, Janne Grunau <j@jannau.net>,
Suzuki K Poulose <suzuki.poulose@arm.com>,
James Clark <james.clark@linaro.org>
Subject: [PATCH v2 10/25] irqchip/gic-v3: Switch high priority PPIs over to handle_percpu_devid_irq()
Date: Mon, 15 Sep 2025 09:56:47 +0100 [thread overview]
Message-ID: <20250915085702.519996-11-maz@kernel.org> (raw)
In-Reply-To: <20250915085702.519996-1-maz@kernel.org>
It so appears that handle_percpu_devid_irq() is extremely similar to
handle_percpu_devid_fasteoi_nmi(), and that the differences do no
justify the horrid machinery we have in the GICv3 driver to handle
the flow handler switch.
Let's stick with the standard flow handler, even for NMIs.
Suggested-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
drivers/irqchip/irq-gic-v3.c | 54 ++----------------------------------
1 file changed, 2 insertions(+), 52 deletions(-)
diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c
index 71c278ddd1e39..a14fd4c269692 100644
--- a/drivers/irqchip/irq-gic-v3.c
+++ b/drivers/irqchip/irq-gic-v3.c
@@ -230,9 +230,6 @@ static void __init gic_prio_init(void)
!cpus_have_group0);
}
-/* rdist_nmi_refs[n] == number of cpus having the rdist interrupt n set as NMI */
-static refcount_t *rdist_nmi_refs;
-
static struct gic_kvm_info gic_v3_kvm_info __initdata;
static DEFINE_PER_CPU(bool, has_rss);
@@ -608,24 +605,6 @@ static u32 __gic_get_ppi_index(irq_hw_number_t hwirq)
}
}
-static u32 __gic_get_rdist_index(irq_hw_number_t hwirq)
-{
- switch (__get_intid_range(hwirq)) {
- case SGI_RANGE:
- case PPI_RANGE:
- return hwirq;
- case EPPI_RANGE:
- return hwirq - EPPI_BASE_INTID + 32;
- default:
- unreachable();
- }
-}
-
-static u32 gic_get_rdist_index(struct irq_data *d)
-{
- return __gic_get_rdist_index(d->hwirq);
-}
-
static int gic_irq_nmi_setup(struct irq_data *d)
{
struct irq_desc *desc = irq_to_desc(d->irq);
@@ -646,20 +625,8 @@ static int gic_irq_nmi_setup(struct irq_data *d)
return -EINVAL;
/* desc lock should already be held */
- if (gic_irq_in_rdist(d)) {
- u32 idx = gic_get_rdist_index(d);
-
- /*
- * Setting up a percpu interrupt as NMI, only switch handler
- * for first NMI
- */
- if (!refcount_inc_not_zero(&rdist_nmi_refs[idx])) {
- refcount_set(&rdist_nmi_refs[idx], 1);
- desc->handle_irq = handle_percpu_devid_fasteoi_nmi;
- }
- } else {
+ if (!gic_irq_in_rdist(d))
desc->handle_irq = handle_fasteoi_nmi;
- }
gic_irq_set_prio(d, dist_prio_nmi);
@@ -686,15 +653,8 @@ static void gic_irq_nmi_teardown(struct irq_data *d)
return;
/* desc lock should already be held */
- if (gic_irq_in_rdist(d)) {
- u32 idx = gic_get_rdist_index(d);
-
- /* Tearing down NMI, only switch handler for last NMI */
- if (refcount_dec_and_test(&rdist_nmi_refs[idx]))
- desc->handle_irq = handle_percpu_devid_irq;
- } else {
+ if (!gic_irq_in_rdist(d))
desc->handle_irq = handle_fasteoi_irq;
- }
gic_irq_set_prio(d, dist_prio_irq);
}
@@ -2079,19 +2039,9 @@ static const struct gic_quirk gic_quirks[] = {
static void gic_enable_nmi_support(void)
{
- int i;
-
if (!gic_prio_masking_enabled() || nmi_support_forbidden)
return;
- rdist_nmi_refs = kcalloc(gic_data.ppi_nr + SGI_NR,
- sizeof(*rdist_nmi_refs), GFP_KERNEL);
- if (!rdist_nmi_refs)
- return;
-
- for (i = 0; i < gic_data.ppi_nr + SGI_NR; i++)
- refcount_set(&rdist_nmi_refs[i], 0);
-
pr_info("Pseudo-NMIs enabled using %s ICC_PMR_EL1 synchronisation\n",
gic_has_relaxed_pmr_sync() ? "relaxed" : "forced");
--
2.39.2
next prev parent reply other threads:[~2025-09-15 8:57 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-15 8:56 [PATCH v2 00/25] genirq: Add support for percpu_devid IRQ affinity Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 01/25] irqdomain: Add firmware info reporting interface Marc Zyngier
2025-09-16 15:14 ` Jonathan Cameron
2025-09-16 16:00 ` Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 02/25] ACPI: irq: Add IRQ affinity " Marc Zyngier
2025-09-15 10:16 ` Rafael J. Wysocki
2025-09-15 8:56 ` [PATCH v2 03/25] of/irq: " Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 04/25] platform: Add firmware-agnostic irq and affinity retrieval interface Marc Zyngier
2025-09-16 15:23 ` Jonathan Cameron
2025-09-15 8:56 ` [PATCH v2 05/25] irqchip/gic-v3: Add FW info retrieval support Marc Zyngier
2025-09-16 15:34 ` Jonathan Cameron
2025-09-16 15:59 ` Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 06/25] irqchip/apple-aic: " Marc Zyngier
2025-09-15 15:07 ` Sven Peter
2025-09-15 8:56 ` [PATCH v2 07/25] coresight: trbe: Convert to new IRQ affinity retrieval API Marc Zyngier
2025-09-15 9:30 ` Suzuki K Poulose
2025-09-15 8:56 ` [PATCH v2 08/25] perf: arm_pmu: " Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 09/25] perf: arm_spe_pmu: " Marc Zyngier
2025-09-15 8:56 ` Marc Zyngier [this message]
2025-09-15 8:56 ` [PATCH v2 11/25] genirq: Kill handle_percpu_devid_fasteoi_nmi() Marc Zyngier
2025-09-16 15:38 ` Jonathan Cameron
2025-09-21 15:32 ` Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 12/25] genirq: Merge irqaction::{dev_id,percpu_dev_id} Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 13/25] genirq: Factor-in percpu irqaction creation Marc Zyngier
2025-09-16 16:12 ` Jonathan Cameron
2025-09-15 8:56 ` [PATCH v2 14/25] genirq: Add affinity to percpu_devid interrupt requests Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 15/25] genirq: Update request_percpu_nmi() to take an affinity Marc Zyngier
2025-09-16 16:19 ` Jonathan Cameron
2025-09-15 8:56 ` [PATCH v2 16/25] genirq: Allow per-cpu interrupt sharing for non-overlapping affinities Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 17/25] genirq: Add request_percpu_irq_affinity() helper Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 18/25] perf: arm_pmu: Request specific affinities for percpu NMI/IRQ Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 19/25] perf: arm_spe_pmu: Request specific affinities for percpu IRQ Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 20/25] coresight: trbe: " Marc Zyngier
2025-09-15 9:37 ` Suzuki K Poulose
2025-09-15 8:56 ` [PATCH v2 21/25] irqchip/gic-v3: Drop support for custom PPI partitions Marc Zyngier
2025-09-15 8:56 ` [PATCH v2 22/25] irqchip/apple-aic: Drop support for custom PMU irq partitions Marc Zyngier
2025-09-15 8:57 ` [PATCH v2 23/25] irqchip: Kill irq-partition-percpu Marc Zyngier
2025-09-15 8:57 ` [PATCH v2 24/25] genirq: Kill irq_{g,s}et_percpu_devid_partition() Marc Zyngier
2025-09-15 8:57 ` [PATCH v2 25/25] perf: arm_pmu: Kill last use of per-CPU cpu_armpmu pointer Marc Zyngier
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=20250915085702.519996-11-maz@kernel.org \
--to=maz@kernel.org \
--cc=gregkh@linuxfoundation.org \
--cc=j@jannau.net \
--cc=james.clark@linaro.org \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=rafael@kernel.org \
--cc=robh@kernel.org \
--cc=saravanak@google.com \
--cc=suzuki.poulose@arm.com \
--cc=sven@kernel.org \
--cc=tglx@linutronix.de \
--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;
as well as URLs for NNTP newsgroup(s).