All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrii Anisov <andrii.anisov@gmail.com>
Cc: xen-devel@lists.xenproject.org,
	Julien Grall <julien.grall@arm.com>,
	Stefano Stabellini <sstabellini@kernel.org>,
	Andrii Anisov <andrii_anisov@epam.com>
Subject: [RFC 16/16] gic: vgic: align frequently accessed data by cache line size
Date: Wed, 28 Nov 2018 23:32:11 +0200	[thread overview]
Message-ID: <1543440731-21947-17-git-send-email-andrii.anisov@gmail.com> (raw)
In-Reply-To: <1543440731-21947-1-git-send-email-andrii.anisov@gmail.com>

From: Andrii Anisov <andrii_anisov@epam.com>

Cache line size assumed 64 byte as for H3. Align the `struct
pending_irq` and allocate lrs shadow aligned to cache line size.

Signed-off-by: Andrii Anisov <andrii_anisov@epam.com>
---
 xen/arch/arm/domain.c        | 4 ++++
 xen/arch/arm/vgic.c          | 9 +++++++++
 xen/include/asm-arm/config.h | 2 +-
 xen/include/asm-arm/gic.h    | 2 +-
 xen/include/asm-arm/vgic.h   | 2 +-
 5 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/xen/arch/arm/domain.c b/xen/arch/arm/domain.c
index 8e886b7..8bcb667 100644
--- a/xen/arch/arm/domain.c
+++ b/xen/arch/arm/domain.c
@@ -558,6 +558,10 @@ int arch_vcpu_create(struct vcpu *v)
     v->arch.saved_context.sp = (register_t)v->arch.cpu_info;
     v->arch.saved_context.pc = (register_t)continue_new_vcpu;
 
+    v->arch.gic.v2.lr = xzalloc_bytes(sizeof(uint32_t) * gic_number_lines());
+    if ( v->arch.gic.v2.lr == NULL )
+        return -ENOMEM;
+
     /* Idle VCPUs don't need the rest of this setup */
     if ( is_idle_vcpu(v) )
         return rc;
diff --git a/xen/arch/arm/vgic.c b/xen/arch/arm/vgic.c
index 7691310..bedfb99 100644
--- a/xen/arch/arm/vgic.c
+++ b/xen/arch/arm/vgic.c
@@ -166,6 +166,15 @@ int domain_vgic_init(struct domain *d, unsigned int nr_spis)
 
     d->arch.vgic.pending_irqs =
         xzalloc_array(struct pending_irq, d->arch.vgic.nr_spis);
+
+    if ( sizeof(struct pending_irq) != dcache_line_bytes )
+    {
+        printk ("sizeof(struct pending_irq) = %lu  is not equal to cacheline"
+                "size %lu. Is it expected?\n", sizeof(struct pending_irq),
+                dcache_line_bytes);
+        WARN();
+    }
+
     if ( d->arch.vgic.pending_irqs == NULL )
         return -ENOMEM;
 
diff --git a/xen/include/asm-arm/config.h b/xen/include/asm-arm/config.h
index bc89e84..4f3669f 100644
--- a/xen/include/asm-arm/config.h
+++ b/xen/include/asm-arm/config.h
@@ -28,7 +28,7 @@
 
 #define CONFIG_ARM 1
 
-#define CONFIG_ARM_L1_CACHE_SHIFT 7 /* XXX */
+#define CONFIG_ARM_L1_CACHE_SHIFT 6 /* XXX */
 
 #define CONFIG_SMP 1
 
diff --git a/xen/include/asm-arm/gic.h b/xen/include/asm-arm/gic.h
index add2566..fe44d3a 100644
--- a/xen/include/asm-arm/gic.h
+++ b/xen/include/asm-arm/gic.h
@@ -186,7 +186,7 @@ struct gic_v2 {
     uint32_t hcr;
     uint32_t vmcr;
     uint32_t apr;
-    uint32_t lr[64];
+    uint32_t *lr;
     uint64_t lr_update_mask;
 };
 
diff --git a/xen/include/asm-arm/vgic.h b/xen/include/asm-arm/vgic.h
index a27a1a9..d4ec96f 100644
--- a/xen/include/asm-arm/vgic.h
+++ b/xen/include/asm-arm/vgic.h
@@ -99,7 +99,7 @@ struct pending_irq
      * TODO: when implementing irq migration, taking only the current
      * vgic lock is not going to be enough. */
     struct list_head lr_queue;
-};
+}__cacheline_aligned;
 
 #define NR_INTERRUPT_PER_RANK   32
 #define INTERRUPT_RANK_MASK (NR_INTERRUPT_PER_RANK - 1)
-- 
2.7.4


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

  parent reply	other threads:[~2018-11-28 21:32 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <1543440731-21947-1-git-send-email-andrii.anisov@gmail.com>
2018-11-28 21:31 ` [RFC 01/16] xen/arm: Re-enable interrupt later in the trap path Andrii Anisov
2018-11-28 22:00   ` Julien Grall
2019-07-30 17:34     ` [Xen-devel] " Andrii Anisov
2019-07-30 20:46       ` Julien Grall
2018-11-28 21:31 ` [RFC 02/16] hack: drop GIC v3 support Andrii Anisov
2018-11-29 12:08   ` Andre Przywara
2018-11-28 21:31 ` [RFC 03/16] vgic: move pause_flags check out of vgic spinlock Andrii Anisov
2018-11-28 22:02   ` Julien Grall
2018-11-28 21:31 ` [RFC 04/16] vgic: move irq_to_pending out of lock Andrii Anisov
2018-11-29 11:07   ` Julien Grall
2018-11-28 21:32 ` [RFC 05/16] gic-vgic: Drop an excessive clear_lrs Andrii Anisov
2018-12-11 14:33   ` Julien Grall
2018-12-12 11:01     ` Andrii Anisov
2018-12-12 11:55       ` Julien Grall
2018-11-28 21:32 ` [RFC 06/16] gic: drop interrupts enabling on interrupts processing Andrii Anisov
2018-11-28 22:06   ` Julien Grall
2018-12-12 12:10     ` Julien Grall
2018-11-28 21:32 ` [RFC 07/16] gic-vgic:vgic: do not keep disabled IRQs in any of queues Andrii Anisov
2018-11-28 21:32 ` [RFC 08/16] gic: separate ppi processing Andrii Anisov
2018-12-12 12:37   ` Andrii Anisov
2018-12-12 12:46     ` Julien Grall
2018-12-12 12:52       ` Andrii Anisov
2018-12-12 13:46         ` Julien Grall
2018-11-28 21:32 ` [RFC 09/16] gic-vgic:vgic: avoid excessive conversions Andrii Anisov
2018-11-29 10:23   ` Julien Grall
2018-11-28 21:32 ` [RFC 10/16] gic:vgic:gic-vgic: introduce non-atomic bitops Andrii Anisov
2018-11-29 12:14   ` Andre Przywara
2018-11-29 16:07     ` Julien Grall
2018-12-03 12:33     ` Andrii Anisov
2018-12-03 12:58       ` Julien Grall
2018-12-03 13:05         ` Andrii Anisov
2018-12-03 13:08           ` Julien Grall
2018-12-03 16:37       ` Andre Przywara
2018-11-28 21:32 ` [RFC 11/16] irq: skip action avalability check for guest's IRQ Andrii Anisov
2018-12-11 14:48   ` Julien Grall
2018-12-12 11:30     ` Andrii Anisov
2018-12-12 11:59       ` Julien Grall
2018-12-12 13:51         ` Andrii Anisov
2018-12-12 14:49           ` Julien Grall
2018-12-12 14:58             ` Andrii Anisov
2018-12-12 15:08               ` Julien Grall
2018-12-12 15:57                 ` Andrii Anisov
2018-11-28 21:32 ` [RFC 12/16] gic-v2: Write HCR only on change Andrii Anisov
2018-11-29 16:36   ` Julien Grall
2018-11-28 21:32 ` [RFC 13/16] gic-vgic: skip irqs locking Andrii Anisov
2018-12-12 12:07   ` Julien Grall
2018-12-12 12:35     ` Andrii Anisov
2018-12-12 12:44       ` Julien Grall
2018-12-12 12:47         ` Andrii Anisov
2018-11-28 21:32 ` [RFC 14/16] hack: arm/domain: simplify context restore from idle vcpu Andrii Anisov
2018-11-29 15:53   ` Julien Grall
2018-11-29 16:00   ` Wei Liu
2018-11-28 21:32 ` [RFC 15/16] hack: move gicv2 LRs reads and writes out of spinlocks Andrii Anisov
2018-11-28 21:58   ` Julien Grall
2018-11-28 21:32 ` Andrii Anisov [this message]
2018-11-29 16:24   ` [RFC 16/16] gic: vgic: align frequently accessed data by cache line size Julien Grall
2018-11-29  7:40 ` [RFC 00/16] Old GIC (gic-vgic) optimizations for GICV2 Andrii Anisov
2018-11-29 11:00   ` Julien Grall
2018-11-29 12:07   ` Andre Przywara

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=1543440731-21947-17-git-send-email-andrii.anisov@gmail.com \
    --to=andrii.anisov@gmail.com \
    --cc=andrii_anisov@epam.com \
    --cc=julien.grall@arm.com \
    --cc=sstabellini@kernel.org \
    --cc=xen-devel@lists.xenproject.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.