<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="/rss.xsl.xml"?>
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/">
<channel>
    <title>Changes in Makefile</title>
    <description></description>
    <language>en</language>
    <copyright>Copyright 2015</copyright>
    <generator>Java</generator><item>
        <title>de6c85bf - dma-mapping: clearly mark DMA ops as an architecture feature</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#de6c85bf</link>
        <description>dma-mapping: clearly mark DMA ops as an architecture featureDMA ops are a helper for architectures and not for drivers to overridethe DMA implementation.Unfortunately driver authors keep ignoring this.  Make the fact moreclear by renaming the symbol to ARCH_HAS_DMA_OPS and having the two driversoverriding their dma_ops depend on that.  These drivers should probably bemarked broken, but we can give them a bit of a grace period for that.Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;Reviewed-by: Thomas Gleixner &lt;tglx@linutronix.de&gt;Acked-by: Sakari Ailus &lt;sakari.ailus@linux.intel.com&gt; # for IPU6Acked-by: Robin Murphy &lt;robin.murphy@arm.com&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Wed, 28 Aug 2024 06:02:47 +0000</pubDate>
        <dc:creator>Christoph Hellwig &lt;hch@lst.de&gt;</dc:creator>
    </item>
<item>
        <title>b5c58b2f - dma-mapping: direct calls for dma-iommu</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#b5c58b2f</link>
        <description>dma-mapping: direct calls for dma-iommuDirectly call into dma-iommu just like we have been doing for dma-directfor a while.  This avoids the indirect call overhead for IOMMU ops andremoves the need to have DMA ops entirely for many common configurations.Signed-off-by: Leon Romanovsky &lt;leonro@nvidia.com&gt;Signed-off-by: Leon Romanovsky &lt;leon@kernel.org&gt;Acked-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;Acked-by: Robin Murphy &lt;robin.murphy@arm.com&gt;Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Wed, 24 Jul 2024 18:04:49 +0000</pubDate>
        <dc:creator>Leon Romanovsky &lt;leonro@nvidia.com&gt;</dc:creator>
    </item>
<item>
        <title>f5ff79fd - dma-mapping: remove CONFIG_DMA_REMAP</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#f5ff79fd</link>
        <description>dma-mapping: remove CONFIG_DMA_REMAPCONFIG_DMA_REMAP is used to build a few helpers around the corevmalloc code, and to use them in case there is a highmem page indma-direct, and to make dma coherent allocations be able to usenon-contiguous pages allocations for DMA allocations in the dma-iommulayer.Right now it needs to be explicitly selected by architectures, andis only done so by architectures that require remapping to dealwith devices that are not DMA coherent.  Make it unconditional forbuilds with CONFIG_MMU as it is very little extra code, but makesit much more likely that large DMA allocations succeed on x86.This fixes hot plugging a NVMe thunderbolt SSD for me, which triesto allocate a 1MB buffer that is otherwise hard to obtain due tomemory fragmentation on a heavily used laptop.Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;Reviewed-by: Robin Murphy &lt;robin.murphy@arm.com&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Sat, 26 Feb 2022 15:40:21 +0000</pubDate>
        <dc:creator>Christoph Hellwig &lt;hch@lst.de&gt;</dc:creator>
    </item>
<item>
        <title>65789daa - dma-mapping: add benchmark support for streaming DMA APIs</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#65789daa</link>
        <description>dma-mapping: add benchmark support for streaming DMA APIsNowadays, there are increasing requirements to benchmark the performanceof dma_map and dma_unmap particually while the device is attached to anIOMMU.This patch enables the support. Users can run specified number of threadsto do dma_map_page and dma_unmap_page on a specific NUMA node with thespecified duration. Then dma_map_benchmark will calculate the averagelatency for map and unmap.A difficulity for this benchmark is that dma_map/unmap APIs must run ona particular device. Each device might have different backend of IOMMU ornon-IOMMU.So we use the driver_override to bind dma_map_benchmark to a particualdevice by:For platform devices:echo dma_map_benchmark &gt; /sys/bus/platform/devices/xxx/driver_overrideecho xxx &gt; /sys/bus/platform/drivers/xxx/unbindecho xxx &gt; /sys/bus/platform/drivers/dma_map_benchmark/bindFor PCI devices:echo dma_map_benchmark &gt; /sys/bus/pci/devices/0000:00:01.0/driver_overrideecho 0000:00:01.0 &gt; /sys/bus/pci/drivers/xxx/unbindecho 0000:00:01.0 &gt; /sys/bus/pci/drivers/dma_map_benchmark/bindCc: Will Deacon &lt;will@kernel.org&gt;Cc: Shuah Khan &lt;shuah@kernel.org&gt;Cc: Christoph Hellwig &lt;hch@lst.de&gt;Cc: Marek Szyprowski &lt;m.szyprowski@samsung.com&gt;Cc: Robin Murphy &lt;robin.murphy@arm.com&gt;Signed-off-by: Barry Song &lt;song.bao.hua@hisilicon.com&gt;[hch: folded in two fixes from Colin Ian King &lt;colin.king@canonical.com&gt;]Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Mon, 16 Nov 2020 06:08:47 +0000</pubDate>
        <dc:creator>Barry Song &lt;song.bao.hua@hisilicon.com&gt;</dc:creator>
    </item>
<item>
        <title>172292be - dma-mapping: remove dma_virt_ops</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#172292be</link>
        <description>dma-mapping: remove dma_virt_opsNow that the RDMA core deals with devices that only do DMA mapping inlower layers properly, there is no user for dma_virt_ops and it can beremoved.Link: https://lore.kernel.org/r/20201106181941.1878556-11-hch@lst.deSigned-off-by: Christoph Hellwig &lt;hch@lst.de&gt;Signed-off-by: Jason Gunthorpe &lt;jgg@nvidia.com&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Fri, 06 Nov 2020 18:19:41 +0000</pubDate>
        <dc:creator>Christoph Hellwig &lt;hch@lst.de&gt;</dc:creator>
    </item>
<item>
        <title>545d2927 - dma-mapping: move dma_common_{mmap,get_sgtable} out of mapping.c</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#545d2927</link>
        <description>dma-mapping: move dma_common_{mmap,get_sgtable} out of mapping.cAdd a new file that contains helpers for misc DMA ops, which is onlybuilt when CONFIG_DMA_OPS is set.Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;Reviewed-by: Robin Murphy &lt;robin.murphy@arm.com&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Tue, 18 Aug 2020 07:30:44 +0000</pubDate>
        <dc:creator>Christoph Hellwig &lt;hch@lst.de&gt;</dc:creator>
    </item>
<item>
        <title>2f9237d4 - dma-mapping: make support for dma ops optional</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#2f9237d4</link>
        <description>dma-mapping: make support for dma ops optionalAvoid the overhead of the dma ops support for tiny builds that onlyuse the direct mapping.Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;Tested-by: Alexey Kardashevskiy &lt;aik@ozlabs.ru&gt;Reviewed-by: Alexey Kardashevskiy &lt;aik@ozlabs.ru&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Wed, 08 Jul 2020 07:30:00 +0000</pubDate>
        <dc:creator>Christoph Hellwig &lt;hch@lst.de&gt;</dc:creator>
    </item>
<item>
        <title>e860c299 - dma-remap: separate DMA atomic pools from direct remap code</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#e860c299</link>
        <description>dma-remap: separate DMA atomic pools from direct remap codeDMA atomic pools will be needed beyond only CONFIG_DMA_DIRECT_REMAP soseparate them out into their own file.This also adds a new Kconfig option that can be subsequently used foroptions, such as CONFIG_AMD_MEM_ENCRYPT, that will utilize the coherentpools but do not have a dependency on direct remapping.For this patch alone, there is no functional change introduced.Reviewed-by: Christoph Hellwig &lt;hch@lst.de&gt;Signed-off-by: David Rientjes &lt;rientjes@google.com&gt;[hch: fixup copyrights and remove unused includes]Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Wed, 15 Apr 2020 00:04:52 +0000</pubDate>
        <dc:creator>David Rientjes &lt;rientjes@google.com&gt;</dc:creator>
    </item>
<item>
        <title>ff4c25f2 - dma-mapping: improve selection of dma_declare_coherent availability</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#ff4c25f2</link>
        <description>dma-mapping: improve selection of dma_declare_coherent availabilityThis API is primarily used through DT entries, but two architecturesand two drivers call it directly.  So instead of selecting the configsymbol for random architectures pull it in implicitly for the actualusers.  Also rename the Kconfig option to describe the feature better.Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;Acked-by: Paul Burton &lt;paul.burton@mips.com&gt; # MIPSAcked-by: Lee Jones &lt;lee.jones@linaro.org&gt;Reviewed-by: Greg Kroah-Hartman &lt;gregkh@linuxfoundation.org&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Sun, 03 Feb 2019 19:12:02 +0000</pubDate>
        <dc:creator>Christoph Hellwig &lt;hch@lst.de&gt;</dc:creator>
    </item>
<item>
        <title>90ac706e - dma-mapping: factor out dummy DMA ops</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#90ac706e</link>
        <description>dma-mapping: factor out dummy DMA opsThe dummy DMA ops are currently used by arm64 for any device which hasan invalid ACPI description and is thus barred from using DMA due to notknowing whether is is cache-coherent or not. Factor these out intogeneral dma-mapping code so that they can be referenced from othercommon code paths. In the process, we can prune all the optionalcallbacks which just do the same thing as the default behaviour, andfill in .map_resource for completeness.Signed-off-by: Robin Murphy &lt;robin.murphy@arm.com&gt;[hch: moved to a separate source file]Reviewed-by: Rafael J. Wysocki &lt;rafael.j.wysocki@intel.com&gt;Acked-by: Jesper Dangaard Brouer &lt;brouer@redhat.com&gt;Tested-by: Jesper Dangaard Brouer &lt;brouer@redhat.com&gt;Tested-by: Tony Luck &lt;tony.luck@intel.com&gt;Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Thu, 06 Dec 2018 21:14:44 +0000</pubDate>
        <dc:creator>Robin Murphy &lt;robin.murphy@arm.com&gt;</dc:creator>
    </item>
<item>
        <title>3731c3d4 - dma-mapping: always build the direct mapping code</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#3731c3d4</link>
        <description>dma-mapping: always build the direct mapping codeAll architectures except for sparc64 use the dma-direct code in someform, and even for sparc64 we had the discussion of a direct mappingmode a while ago.  In preparation for directly calling the directmapping code don&apos;t bother having it optionally but always build thecode in.  This is a minor hardship for some powerpc and arm configsthat don&apos;t pull it in yet (although they should in a relase ot two),and sparc64 which currently doesn&apos;t need it at all, but it willreduce the ifdef mess we&apos;d otherwise need significantly.Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;Acked-by: Jesper Dangaard Brouer &lt;brouer@redhat.com&gt;Tested-by: Jesper Dangaard Brouer &lt;brouer@redhat.com&gt;Tested-by: Tony Luck &lt;tony.luck@intel.com&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Thu, 06 Dec 2018 20:50:26 +0000</pubDate>
        <dc:creator>Christoph Hellwig &lt;hch@lst.de&gt;</dc:creator>
    </item>
<item>
        <title>f0edfea8 - dma-mapping: move the remap helpers to a separate file</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#f0edfea8</link>
        <description>dma-mapping: move the remap helpers to a separate fileThe dma remap code only makes sense for not cache coherent architectures(or possibly the corner case of highmem CMA allocations) and currentlyis only used by arm, arm64, csky and xtensa.  Split it out into aseparate file with a separate Kconfig symbol, which gets the rightcopyright notice given that this code was written by Laura Abbottworking for Code Aurora at that point.Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;Acked-by: Laura Abbott &lt;labbott@redhat.com&gt;Reviewed-by: Robin Murphy &lt;robin.murphy@arm.com&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Fri, 24 Aug 2018 08:31:08 +0000</pubDate>
        <dc:creator>Christoph Hellwig &lt;hch@lst.de&gt;</dc:creator>
    </item>
<item>
        <title>bc3ec75d - dma-mapping: merge direct and noncoherent ops</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#bc3ec75d</link>
        <description>dma-mapping: merge direct and noncoherent opsAll the cache maintainance is already stubbed out when not enabled,but merging the two allows us to nicely handle the case wherecache maintainance is required for some devices, but not others.Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;Acked-by: Paul Burton &lt;paul.burton@mips.com&gt; # MIPS parts

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Sat, 08 Sep 2018 09:22:43 +0000</pubDate>
        <dc:creator>Christoph Hellwig &lt;hch@lst.de&gt;</dc:creator>
    </item>
<item>
        <title>cf65a0f6 - dma-mapping: move all DMA mapping code to kernel/dma</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/kernel/dma/Makefile#cf65a0f6</link>
        <description>dma-mapping: move all DMA mapping code to kernel/dmaCurrently the code is split over various files with dma- prefixes in thelib/ and drives/base directories, and the number of files keeps growing.Move them into a single directory to keep the code together and removethe file name prefixes.  To match the irq infrastructure this directoryis placed under the kernel/ directory.Signed-off-by: Christoph Hellwig &lt;hch@lst.de&gt;

            List of files:
            /linux-6.15/kernel/dma/Makefile</description>
        <pubDate>Tue, 12 Jun 2018 17:01:45 +0000</pubDate>
        <dc:creator>Christoph Hellwig &lt;hch@lst.de&gt;</dc:creator>
    </item>
</channel>
</rss>
