All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Jan Beulich" <JBeulich@suse.com>
To: xen-devel <xen-devel@lists.xenproject.org>
Cc: Yang Z Zhang <yang.z.zhang@intel.com>, Kevin Tian <kevin.tian@intel.com>
Subject: [PATCH v3 09/10] VT-d: use qword MMIO access for MSI address writes
Date: Fri, 05 Jun 2015 12:26:07 +0100	[thread overview]
Message-ID: <5571A36F02000078000814C6@mail.emea.novell.com> (raw)
In-Reply-To: <55719F9D0200007800081425@mail.emea.novell.com>

[-- Attachment #1: Type: text/plain, Size: 1717 bytes --]

Also make dmar_{read,write}q() actually do what their names suggest (we
don't need to be concerned of 32-bit restrictions anymore).

Signed-off-by: Jan Beulich <jbeulich@suse.com>

--- a/xen/drivers/passthrough/vtd/iommu.c
+++ b/xen/drivers/passthrough/vtd/iommu.c
@@ -1054,8 +1054,7 @@ static void dma_msi_set_affinity(struct 
 
     spin_lock_irqsave(&iommu->register_lock, flags);
     dmar_writel(iommu->reg, DMAR_FEDATA_REG, msg.data);
-    dmar_writel(iommu->reg, DMAR_FEADDR_REG, msg.address_lo);
-    dmar_writel(iommu->reg, DMAR_FEUADDR_REG, msg.address_hi);
+    dmar_writeq(iommu->reg, DMAR_FEADDR_REG, msg.address);
     spin_unlock_irqrestore(&iommu->register_lock, flags);
 }
 
--- a/xen/drivers/passthrough/vtd/iommu.h
+++ b/xen/drivers/passthrough/vtd/iommu.h
@@ -51,17 +51,10 @@
 #define    DMAR_IRTA_REG   0xB8    /* intr remap */
 
 #define OFFSET_STRIDE        (9)
-#define dmar_readl(dmar, reg) readl(dmar + reg)
-#define dmar_writel(dmar, reg, val) writel(val, dmar + reg)
-#define dmar_readq(dmar, reg) ({ \
-        u32 lo, hi; \
-        lo = dmar_readl(dmar, reg); \
-        hi = dmar_readl(dmar, reg + 4); \
-        (((u64) hi) << 32) + lo; })
-#define dmar_writeq(dmar, reg, val) do {\
-        dmar_writel(dmar, reg, (u32)val); \
-        dmar_writel(dmar, reg + 4, (u32)((u64) val >> 32)); \
-    } while (0)
+#define dmar_readl(dmar, reg) readl((dmar) + (reg))
+#define dmar_readq(dmar, reg) readq((dmar) + (reg))
+#define dmar_writel(dmar, reg, val) writel(val, (dmar) + (reg))
+#define dmar_writeq(dmar, reg, val) writeq(val, (dmar) + (reg))
 
 #define VER_MAJOR(v)        (((v) & 0xf0) >> 4)
 #define VER_MINOR(v)        ((v) & 0x0f)




[-- Attachment #2: VT-d-MMIO-qword.patch --]
[-- Type: text/plain, Size: 1765 bytes --]

VT-d: use qword MMIO access for MSI address writes

Also make dmar_{read,write}q() actually do what their names suggest (we
don#t need to be concerned of 32-bit restrictions anymore).

Signed-off-by: Jan Beulich <jbeulich@suse.com>

--- a/xen/drivers/passthrough/vtd/iommu.c
+++ b/xen/drivers/passthrough/vtd/iommu.c
@@ -1054,8 +1054,7 @@ static void dma_msi_set_affinity(struct 
 
     spin_lock_irqsave(&iommu->register_lock, flags);
     dmar_writel(iommu->reg, DMAR_FEDATA_REG, msg.data);
-    dmar_writel(iommu->reg, DMAR_FEADDR_REG, msg.address_lo);
-    dmar_writel(iommu->reg, DMAR_FEUADDR_REG, msg.address_hi);
+    dmar_writeq(iommu->reg, DMAR_FEADDR_REG, msg.address);
     spin_unlock_irqrestore(&iommu->register_lock, flags);
 }
 
--- a/xen/drivers/passthrough/vtd/iommu.h
+++ b/xen/drivers/passthrough/vtd/iommu.h
@@ -51,17 +51,10 @@
 #define    DMAR_IRTA_REG   0xB8    /* intr remap */
 
 #define OFFSET_STRIDE        (9)
-#define dmar_readl(dmar, reg) readl(dmar + reg)
-#define dmar_writel(dmar, reg, val) writel(val, dmar + reg)
-#define dmar_readq(dmar, reg) ({ \
-        u32 lo, hi; \
-        lo = dmar_readl(dmar, reg); \
-        hi = dmar_readl(dmar, reg + 4); \
-        (((u64) hi) << 32) + lo; })
-#define dmar_writeq(dmar, reg, val) do {\
-        dmar_writel(dmar, reg, (u32)val); \
-        dmar_writel(dmar, reg + 4, (u32)((u64) val >> 32)); \
-    } while (0)
+#define dmar_readl(dmar, reg) readl((dmar) + (reg))
+#define dmar_readq(dmar, reg) readq((dmar) + (reg))
+#define dmar_writel(dmar, reg, val) writel(val, (dmar) + (reg))
+#define dmar_writeq(dmar, reg, val) writeq(val, (dmar) + (reg))
 
 #define VER_MAJOR(v)        (((v) & 0xf0) >> 4)
 #define VER_MINOR(v)        ((v) & 0x0f)

[-- Attachment #3: Type: text/plain, Size: 126 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

  parent reply	other threads:[~2015-06-05 11:26 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-05 11:09 [PATCH v3 00/10] x86/MSI-X: XSA-120, 128..131 follow-up Jan Beulich
2015-06-05 11:20 ` [PATCH v3 01/10] x86/MSI-X: be more careful during teardown Jan Beulich
2015-06-05 11:20 ` [PATCH v3 02/10] x86/MSI-X: access MSI-X table only after having enabled MSI-X Jan Beulich
2015-06-05 13:01   ` Andrew Cooper
2015-06-05 11:21 ` [PATCH v3 03/10] x86/MSI-X: reduce fiddling with control register during restore Jan Beulich
2015-06-05 11:21 ` [PATCH v3 04/10] x86/MSI-X: cleanup Jan Beulich
2015-06-05 11:22 ` [PATCH v3 05/10] x86/MSI: track host and guest masking separately Jan Beulich
2015-06-05 13:05   ` Andrew Cooper
2016-04-01  7:40   ` Li, Liang Z
2016-04-01  8:47     ` Jan Beulich
2016-04-01  9:21       ` Li, Liang Z
2016-04-01  9:33         ` Jan Beulich
2015-06-05 11:23 ` [PATCH v3 06/10] x86/vMSI-X: cleanup Jan Beulich
2015-06-05 13:07   ` Andrew Cooper
2015-06-05 11:24 ` [PATCH v3 07/10] x86/vMSI-X: support qword MMIO access Jan Beulich
2015-06-05 15:34   ` Andrew Cooper
2015-06-05 11:25 ` [PATCH v3 08/10] x86/MSI-X: use qword MMIO access for address writes Jan Beulich
2015-06-05 15:37   ` Andrew Cooper
2015-06-05 11:26 ` Jan Beulich [this message]
2015-06-05 15:39   ` [PATCH v3 09/10] VT-d: use qword MMIO access for MSI " Andrew Cooper
2015-06-05 15:46     ` Jan Beulich
2015-06-11  7:45   ` Tian, Kevin
2015-06-05 11:28 ` [PATCH v3 10/10] x86/MSI-X: provide hypercall interface for mask-all control Jan Beulich
2015-06-05 15:57   ` Andrew Cooper
2015-06-05 16:17     ` Jan Beulich
2015-06-11  8:35   ` Jan Beulich
2015-06-11  9:51     ` Andrew Cooper
2015-06-19 13:00       ` Jan Beulich
2015-06-19 13:05         ` Konrad Rzeszutek Wilk
2015-06-19 14:52           ` Jan Beulich
2015-06-19 14:07         ` Roger Pau Monné
2015-06-19 14:58           ` Jan Beulich
2015-06-22 17:02             ` Roger Pau Monné
2015-06-23  7:20               ` Jan Beulich
2015-06-23  7:29                 ` Roger Pau Monné
2015-06-23  8:13                   ` Jan Beulich
2015-06-22 11:25           ` Jan Beulich
2015-06-12 13:21     ` Konrad Rzeszutek Wilk
2015-06-12 13:51       ` Jan Beulich
2015-06-12 14:17         ` Konrad Rzeszutek Wilk

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=5571A36F02000078000814C6@mail.emea.novell.com \
    --to=jbeulich@suse.com \
    --cc=kevin.tian@intel.com \
    --cc=xen-devel@lists.xenproject.org \
    --cc=yang.z.zhang@intel.com \
    /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.