All of lore.kernel.org
 help / color / mirror / Atom feed
From: Fenghua Yu <fenghua.yu@intel.com>
To: David Woodhouse <dwmw2@infradead.org>,
	"'Linus Torvalds'" <torvalds@linux-foundation.org>,
	"'Stephen Rothwell'" <sfr@canb.auug.org.au>,
	"'Andrew Morton'" <akpm@linux-foundation.org>,
	"'Ingo Molnar'" <mingo@elte.hu>,
	"'Christopher Wright'" <chrisw@redhat.com>,
	"'Allen Kay'" <allen.m.kay@intel.com>
Cc: "'lkml'" <linux-kernel@vger.kernel.org>,
	"'iommu'" <iommu@lists.linux-foundation.org>
Subject: [PATCH 1/2] IOMMU Identity Mapping Support: iommu_identity_mapping definition
Date: Thu, 18 Jun 2009 11:05:14 -0700	[thread overview]
Message-ID: <20090618180514.GA24082@linux-os.sc.intel.com> (raw)
In-Reply-To: <20090514175944.GA5168@linux-os.sc.intel.com>

IOMMU Identity Mapping Support: iommu_identity_mapping definition

Identity mapping for IOMMU defines a single domain to 1:1 map all pci devices
to all usable memory.

This will reduces map/unmap overhead in DMA API's and improve IOMMU performance.
On 10Gb network cards, Netperf shows no performance degradation compared to
non-IOMMU performance.

This method may lose some of DMA remapping benefits like isolation.

The first patch defines iommu_identity_mapping varialbe which controls the
identity mapping code and is 0 by default.

Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>

---

 Documentation/kernel-parameters.txt |    1 +
 arch/ia64/include/asm/iommu.h       |    2 ++
 arch/ia64/kernel/pci-dma.c          |    1 +
 arch/x86/include/asm/iommu.h        |    1 +
 arch/x86/kernel/pci-dma.c           |    5 +++++
 5 files changed, 10 insertions(+)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 4b78797..6b1d7b4 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -988,6 +988,7 @@ and is between 256 and 4096 characters. It is defined in the file
 		forcesac
 		soft
 		pt	[x86, IA64]
+		identity [x86, IA64]
 
 	io7=		[HW] IO7 for Marvel based alpha systems
 			See comment before marvel_specify_io7 in
diff --git a/arch/ia64/include/asm/iommu.h b/arch/ia64/include/asm/iommu.h
index 745e095..cc6d59f 100644
--- a/arch/ia64/include/asm/iommu.h
+++ b/arch/ia64/include/asm/iommu.h
@@ -11,8 +11,10 @@ extern int force_iommu, no_iommu;
 extern int iommu_detected;
 #ifdef	CONFIG_DMAR
 extern int iommu_pass_through;
+extern int iommu_identity_mapping;
 #else
 #define iommu_pass_through	(0)
+#define iommu_identity_mapping  (0)
 #endif
 extern void iommu_dma_init(void);
 extern void machvec_init(const char *name);
diff --git a/arch/ia64/kernel/pci-dma.c b/arch/ia64/kernel/pci-dma.c
index 0569596..15b8555 100644
--- a/arch/ia64/kernel/pci-dma.c
+++ b/arch/ia64/kernel/pci-dma.c
@@ -33,6 +33,7 @@ int force_iommu __read_mostly;
 #endif
 
 int iommu_pass_through;
+int iommu_identity_mapping;
 
 /* Dummy device used for NULL arguments (normally ISA). Better would
    be probably a smaller DMA mask, but this is bug-to-bug compatible
diff --git a/arch/x86/include/asm/iommu.h b/arch/x86/include/asm/iommu.h
index fd6d21b..d2aee4a 100644
--- a/arch/x86/include/asm/iommu.h
+++ b/arch/x86/include/asm/iommu.h
@@ -7,6 +7,7 @@ extern struct dma_map_ops nommu_dma_ops;
 extern int force_iommu, no_iommu;
 extern int iommu_detected;
 extern int iommu_pass_through;
+extern int iommu_identity_mapping;
 
 /* 10 seconds */
 #define DMAR_OPERATION_TIMEOUT ((cycles_t) tsc_khz*10*1000)
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 049005e..489179a 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -33,6 +33,7 @@ int no_iommu __read_mostly;
 int iommu_detected __read_mostly = 0;
 
 int iommu_pass_through;
+int iommu_identity_mapping;
 
 dma_addr_t bad_dma_address __read_mostly = 0;
 EXPORT_SYMBOL(bad_dma_address);
@@ -208,6 +209,10 @@ static __init int iommu_setup(char *p)
 			forbid_dac = -1;
 			return 1;
 		}
+		if (!strncmp(p, "identity", 8)) {
+			iommu_identity_mapping = 1;
+			return 1;
+		}
 #ifdef CONFIG_SWIOTLB
 		if (!strncmp(p, "soft", 4))
 			swiotlb = 1;

  reply	other threads:[~2009-06-18 18:05 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20090327212241.234500000@intel.com>
2009-03-28 14:24 ` [patch 0/4] Intel IOMMU Supspend/Resume Support Andrew Lutomirski
2009-03-30 23:01 ` David Woodhouse
     [not found] ` <20090327212321.520992000@intel.com>
2009-04-03 12:37   ` [patch 4/4] Intel IOMMU Suspend/Resume Support - Code Clean Up David Woodhouse
     [not found] ` <20090327212321.070229000@intel.com>
2009-04-16  0:19   ` [PATCH] Intel IOMMU Pass Through Support Fenghua Yu
2009-04-16  2:13     ` Han, Weidong
2009-04-19 10:05     ` David Woodhouse
2009-04-20 17:27       ` Yu, Fenghua
2009-05-13 23:13         ` [PATCH] Fix Intel IOMMU Compilation Warnings on IA64 Fenghua Yu
2009-05-13 23:13           ` Fenghua Yu
2009-05-14 15:17           ` David Woodhouse
2009-05-14 15:17             ` David Woodhouse
2009-05-14 15:31             ` Matthew Wilcox
2009-05-14 15:31               ` Matthew Wilcox
2009-05-14 17:59             ` Fenghua Yu
2009-05-14 17:59               ` Fenghua Yu
2009-06-18 18:05               ` Fenghua Yu [this message]
2009-06-18 18:08                 ` [PATCH 1/2] IOMMU Identity Mapping Support: iommu_identity_mapping definition Muli Ben-Yehuda
2009-06-18 18:13                   ` Chris Wright
2009-06-18 18:14                   ` Yu, Fenghua
2009-06-18 18:25                     ` Muli Ben-Yehuda
2009-06-18 18:31                       ` Chris Wright
2009-06-18 18:41                         ` Muli Ben-Yehuda
2009-06-18 18:50                           ` Yu, Fenghua
2009-06-18 18:51                           ` Chris Wright
2009-06-18 19:09                             ` Yu, Fenghua
2009-06-25  0:38                     ` [PATCH] IA64 Compilation Error Fix for Intel IOMMU Identity Mapping Support Fenghua Yu
2009-06-25  1:00                       ` FUJITA Tomonori
2009-06-25  4:16                         ` [PATCH v2] " Fenghua Yu
2009-06-25  4:48                           ` FUJITA Tomonori
2009-06-25  7:11                             ` David Woodhouse
2009-06-25 21:52                               ` David Woodhouse
2009-06-25 21:56                                 ` Yu, Fenghua
2009-06-26 18:21                                   ` David Woodhouse
2009-06-25 22:00                                 ` Linus Torvalds
2009-06-25 22:46                                   ` Tony Luck
2009-06-25 23:43                                 ` Chris Wright
2009-06-26  1:35                                   ` Linus Torvalds
2009-06-26  1:52                                     ` Chris Wright
2009-06-26  2:00                                       ` Linus Torvalds
2009-06-26  2:08                                         ` Chris Wright
2009-06-26 11:15                                           ` David Woodhouse
2009-06-27  0:03                                             ` Chris Wright
2009-06-27 11:44                                         ` David Woodhouse
2009-06-18 18:13                 ` [PATCH 1/2] IOMMU Identity Mapping Support: iommu_identity_mapping definition Chris Wright
2009-06-18 18:28                   ` Yu, Fenghua
2009-06-18 18:34                     ` Chris Wright
2009-07-04 18:40                   ` David Woodhouse
2009-05-20 17:42         ` [PATCH] Time out for possible dead loops during queued invalidation wait Fenghua Yu
2009-05-27  5:51           ` Andrew Morton
2009-05-27 22:40             ` Yu, Fenghua
2009-05-27 22:48               ` Andrew Morton
2009-05-27 23:25                 ` Yu, Fenghua
2009-05-27 23:51                   ` Andrew Morton
2009-05-28  0:47                     ` Yu, Fenghua
2009-06-18 18:05               ` [PATCH 2/2] IOMMU Identity Mapping Support: Intel IOMMU implementation Fenghua Yu
2009-06-18 19:15                 ` Chris Wright
2009-06-18 19:40                   ` Yu, Fenghua
2009-06-18 20:02                     ` Chris Wright
2009-06-19 20:47                     ` [PATCH v2] IOMMU Identity Mapping Support (drivers/pci/intel_iommu.c) Fenghua Yu
2009-04-30 23:29     ` [PATCH] Intel IOMMU Pass Through Support Andrew Morton
2009-04-30 23:37       ` Randy Dunlap
2009-05-01  0:00         ` Andrew Morton
2009-05-01  0:57           ` Fenghua Yu
2009-05-01  0:05         ` Fenghua Yu
2009-05-01  0:14           ` Andrew Morton

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=20090618180514.GA24082@linux-os.sc.intel.com \
    --to=fenghua.yu@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=allen.m.kay@intel.com \
    --cc=chrisw@redhat.com \
    --cc=dwmw2@infradead.org \
    --cc=iommu@lists.linux-foundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=sfr@canb.auug.org.au \
    --cc=torvalds@linux-foundation.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.