From: Joerg Roedel <joerg.roedel@amd.com>
To: iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org
Cc: Joerg Roedel <joerg.roedel@amd.com>
Subject: [PATCH 01/14] x86/amd-iommu: Make fetch_pte aware of dynamic mapping levels
Date: Fri, 4 Sep 2009 11:40:39 +0200 [thread overview]
Message-ID: <1252057252-32074-2-git-send-email-joerg.roedel@amd.com> (raw)
In-Reply-To: <1252057252-32074-1-git-send-email-joerg.roedel@amd.com>
This patch changes the fetch_pte function in the AMD IOMMU
driver to support dynamic mapping levels.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
arch/x86/include/asm/amd_iommu_types.h | 9 +++++++++
arch/x86/kernel/amd_iommu.c | 24 +++++++++++++-----------
2 files changed, 22 insertions(+), 11 deletions(-)
diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 0c878ca..7fce4ef 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -146,12 +146,21 @@
#define PAGE_MODE_1_LEVEL 0x01
#define PAGE_MODE_2_LEVEL 0x02
#define PAGE_MODE_3_LEVEL 0x03
+#define PAGE_MODE_4_LEVEL 0x04
+#define PAGE_MODE_5_LEVEL 0x05
+#define PAGE_MODE_6_LEVEL 0x06
#define IOMMU_PDE_NL_0 0x000ULL
#define IOMMU_PDE_NL_1 0x200ULL
#define IOMMU_PDE_NL_2 0x400ULL
#define IOMMU_PDE_NL_3 0x600ULL
+#define PM_LEVEL_SHIFT(x) (12 + ((x) * 9))
+#define PM_LEVEL_SIZE(x) (((x) < 6) ? \
+ ((1ULL << PM_LEVEL_SHIFT((x))) - 1): \
+ (0xffffffffffffffffULL))
+#define PM_LEVEL_INDEX(x, a) (((a) >> PM_LEVEL_SHIFT((x))) & 0x1ffULL)
+
#define IOMMU_PTE_L2_INDEX(address) (((address) >> 30) & 0x1ffULL)
#define IOMMU_PTE_L1_INDEX(address) (((address) >> 21) & 0x1ffULL)
#define IOMMU_PTE_L0_INDEX(address) (((address) >> 12) & 0x1ffULL)
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 6c99f50..29bcd35 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -61,6 +61,8 @@ static u64* alloc_pte(struct protection_domain *dom,
static void dma_ops_reserve_addresses(struct dma_ops_domain *dom,
unsigned long start_page,
unsigned int pages);
+static u64 *fetch_pte(struct protection_domain *domain,
+ unsigned long address);
#ifndef BUS_NOTIFY_UNBOUND_DRIVER
#define BUS_NOTIFY_UNBOUND_DRIVER 0x0005
@@ -670,24 +672,24 @@ static int init_unity_mappings_for_device(struct dma_ops_domain *dma_dom,
* This function checks if there is a PTE for a given dma address. If
* there is one, it returns the pointer to it.
*/
-static u64* fetch_pte(struct protection_domain *domain,
+static u64 *fetch_pte(struct protection_domain *domain,
unsigned long address)
{
+ int level;
u64 *pte;
- pte = &domain->pt_root[IOMMU_PTE_L2_INDEX(address)];
+ level = domain->mode - 1;
+ pte = &domain->pt_root[PM_LEVEL_INDEX(level, address)];
- if (!IOMMU_PTE_PRESENT(*pte))
- return NULL;
+ while (level > 0) {
+ if (!IOMMU_PTE_PRESENT(*pte))
+ return NULL;
- pte = IOMMU_PTE_PAGE(*pte);
- pte = &pte[IOMMU_PTE_L1_INDEX(address)];
+ level -= 1;
- if (!IOMMU_PTE_PRESENT(*pte))
- return NULL;
-
- pte = IOMMU_PTE_PAGE(*pte);
- pte = &pte[IOMMU_PTE_L0_INDEX(address)];
+ pte = IOMMU_PTE_PAGE(*pte);
+ pte = &pte[PM_LEVEL_INDEX(level, address)];
+ }
return pte;
}
--
1.6.3.3
next prev parent reply other threads:[~2009-09-04 9:42 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-09-04 9:40 [PATCH 0/14] Page table handling updates for AMD-IOMMU (AMD-Vi) Joerg Roedel
2009-09-04 9:40 ` Joerg Roedel [this message]
2009-09-04 9:40 ` [PATCH 02/14] x86/amd-iommu: Use fetch_pte in iommu_unmap_page Joerg Roedel
2009-09-04 9:40 ` [PATCH 03/14] x86/amd-iommu: Use fetch_pte in amd_iommu_iova_to_phys Joerg Roedel
2009-09-04 9:40 ` [PATCH 04/14] x86/amd-iommu: Add a gneric version of amd_iommu_flush_all_devices Joerg Roedel
2009-09-04 9:40 ` [PATCH 05/14] x86/amd-iommu: Introduce set_dte_entry function Joerg Roedel
2009-09-04 9:40 ` [PATCH 06/14] x86/amd-iommu: Flush domains if address space size was increased Joerg Roedel
2009-09-04 9:40 ` [PATCH 07/14] x86/amd-iommu: Introduce increase_address_space function Joerg Roedel
2009-09-04 9:40 ` [PATCH 08/14] x86/amd-iommu: Change alloc_pte to support 64 bit address space Joerg Roedel
2009-09-04 9:40 ` [PATCH 09/14] x86/amd-iommu: Remove last usages of IOMMU_PTE_L0_INDEX Joerg Roedel
2009-09-04 9:40 ` [PATCH 10/14] x86/amd-iommu: Remove bus_addr check in iommu_map_page Joerg Roedel
2009-09-04 9:40 ` [PATCH 11/14] x86/amd-iommu: Use 2-level page tables for dma_ops domains Joerg Roedel
2009-09-04 9:40 ` [PATCH 12/14] x86/amd-iommu: Remove old page table handling macros Joerg Roedel
2009-09-04 9:40 ` [PATCH 13/14] x86/amd-iommu: Support higher level PTEs in iommu_page_unmap Joerg Roedel
2009-09-04 9:40 ` [PATCH 14/14] x86/amd-iommu: Change iommu_map_page to support multiple page sizes Joerg Roedel
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=1252057252-32074-2-git-send-email-joerg.roedel@amd.com \
--to=joerg.roedel@amd.com \
--cc=iommu@lists.linux-foundation.org \
--cc=linux-kernel@vger.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
all inboxes | Powered by JetHome®