Home » Mailing list archives » linux-kernel » 2008 » September » 17

[PATCH 04/23] AMD IOMMU: add branch hints to completion wait checks

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
[view in full thread]
From: Joerg Roedel
Subject: [PATCH 04/23] AMD IOMMU: add branch hints to completion wait checks
Date: Wednesday, September 17, 2008 - 9:52 am

This patch adds branch hints to the cecks if a completion_wait is
necessary. The completion_waits in the mapping paths are unlikly because
they will only happen on software implementations of AMD IOMMU which
don't exists today or with lazy IO/TLB flushing when the allocator wraps
around the address space. With lazy IO/TLB flushing the completion_wait
in the unmapping path is unlikely too.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c |   12 ++++++------
 1 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 0569098..7e9e4e7 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -876,7 +876,7 @@ static dma_addr_t map_single(struct device *dev, phys_addr_t paddr,
 	if (addr == bad_dma_address)
 		goto out;
 
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 out:
@@ -905,7 +905,7 @@ static void unmap_single(struct device *dev, dma_addr_t dma_addr,
 
 	__unmap_single(iommu, domain->priv, dma_addr, size, dir);
 
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 	spin_unlock_irqrestore(&domain->lock, flags);
@@ -968,7 +968,7 @@ static int map_sg(struct device *dev, struct scatterlist *sglist,
 			goto unmap;
 	}
 
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 out:
@@ -1014,7 +1014,7 @@ static void unmap_sg(struct device *dev, struct scatterlist *sglist,
 		s->dma_address = s->dma_length = 0;
 	}
 
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 	spin_unlock_irqrestore(&domain->lock, flags);
@@ -1061,7 +1061,7 @@ static void *alloc_coherent(struct device *dev, size_t size,
 		goto out;
 	}
 
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 out:
@@ -1093,7 +1093,7 @@ static void free_coherent(struct device *dev, size_t size,
 
 	__unmap_single(iommu, domain->priv, dma_addr, size, DMA_BIDIRECTIONAL);
 
-	if (iommu->need_sync)
+	if (unlikely(iommu->need_sync))
 		iommu_completion_wait(iommu);
 
 	spin_unlock_irqrestore(&domain->lock, flags);
-- 
1.5.6.4


--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
Messages in current thread:
[PATCH 0/23] AMD IOMMU 2.6.28 updates for review, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 01/23] AMD IOMMU: check for invalid device pointers, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 02/23] AMD IOMMU: move TLB flushing to the map/unma ..., Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 04/23] AMD IOMMU: add branch hints to completion wa ..., Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 05/23] AMD IOMMU: align alloc_coherent addresses pr ..., Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 06/23] AMD IOMMU: add event buffer allocation, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 07/23] AMD IOMMU: save pci segment from ACPI tables, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 08/23] AMD IOMMU: save pci_dev instead of devid, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 09/23] AMD IOMMU: add MSI interrupt support, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 10/23] AMD IOMMU: add event handling code, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 11/23] AMD IOMMU: enable event logging, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 12/23] AMD IOMMU: allow IO page faults from devices, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 13/23] AMD IOMMU: add dma_supported callback, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 14/23] AMD IOMMU: don't assing preallocated protect ..., Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 15/23] AMD IOMMU: some set_device_domain cleanups, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 16/23] AMD IOMMU: avoid unnecessary low zone alloca ..., Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 17/23] AMD IOMMU: replace memset with __GFP_ZERO in ..., Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 18/23] AMD IOMMU: simplify dma_mask_to_pages, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 19/23] AMD IOMMU: free domain bitmap with its alloc ..., Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 20/23] AMD IOMMU: remove unnecessary cast to u64 in ..., Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 21/23] AMD IOMMU: calculate IVHD size with a function, Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 22/23] AMD IOMMU: use cmd_buf_size when freeing the ..., Joerg Roedel, (Wed Sep 17, 9:52 am)
[PATCH 23/23] add AMD IOMMU tree to MAINTAINERS file, Joerg Roedel, (Wed Sep 17, 9:52 am)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, FUJITA Tomonori, (Wed Sep 17, 12:20 pm)
Re: [PATCH 18/23] AMD IOMMU: simplify dma_mask_to_pages, FUJITA Tomonori, (Wed Sep 17, 12:20 pm)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, Joerg Roedel, (Wed Sep 17, 12:28 pm)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, FUJITA Tomonori, (Wed Sep 17, 6:29 pm)
Re: [PATCH 18/23] AMD IOMMU: simplify dma_mask_to_pages, Joerg Roedel, (Thu Sep 18, 12:32 am)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, Joerg Roedel, (Thu Sep 18, 3:13 am)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, Joerg Roedel, (Thu Sep 18, 7:03 am)
Re: [PATCH 18/23] AMD IOMMU: simplify dma_mask_to_pages, FUJITA Tomonori, (Thu Sep 18, 8:57 am)
Re: [PATCH 18/23] AMD IOMMU: simplify dma_mask_to_pages, Joerg Roedel, (Thu Sep 18, 9:39 am)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, FUJITA Tomonori, (Thu Sep 18, 4:10 pm)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, Joerg Roedel, (Thu Sep 18, 11:29 pm)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, FUJITA Tomonori, (Fri Sep 19, 3:21 am)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, Joerg Roedel, (Fri Sep 19, 10:43 am)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, FUJITA Tomonori, (Fri Sep 19, 11:40 am)
RE: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, Keshavamurthy, Anil S, (Fri Sep 19, 11:47 am)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, Joerg Roedel, (Fri Sep 19, 12:27 pm)
Re: [PATCH 03/23] AMD IOMMU: implement lazy IO/TLB flushing, Muli Ben-Yehuda, (Sat Sep 20, 10:27 pm)