The quilt patch titled
Subject: dma-contiguous: support per-numa CMA for all architectures
has been removed from the -mm tree. Its filename was
dma-contiguous-support-per-numa-cma-for-all-architectures.patch
This patch was dropped because it is obsolete
------------------------------------------------------
From: Yajun Deng <yajun.deng@xxxxxxxxx>
Subject: dma-contiguous: support per-numa CMA for all architectures
Date: Fri, 12 May 2023 17:42:10 +0800
In commit b7176c261cdb ("dma-contiguous: provide the ability to reserve
per-numa CMA"), Barry adds DMA_PERNUMA_CMA for ARM64.
But this feature is architecture independent, so support per-numa CMA for
all architectures, and enable it by default if NUMA.
Link: https://lkml.kernel.org/r/20230512094210.141540-1-yajun.deng@xxxxxxxxx
Signed-off-by: Yajun Deng <yajun.deng@xxxxxxxxx>
Cc: Anshuman Khandual <anshuman.khandual@xxxxxxx>
Cc: Ard Biesheuvel <ardb@xxxxxxxxxx>
Cc: Baoquan He <bhe@xxxxxxxxxx>
Cc: Barry Song <song.bao.hua@xxxxxxxxxxxxx>
Cc: Borislav Petkov <bp@xxxxxxx>
Cc: Catalin Marinas <catalin.marinas@xxxxxxx>
Cc: Christoph Hellwig <hch@xxxxxx>
Cc: Jonathan Corbet <corbet@xxxxxxx>
Cc: Kim Phillips <kim.phillips@xxxxxxx>
Cc: Marek Szyprowski <m.szyprowski@xxxxxxxxxxx
Cc: Paul E. McKenney <paulmck@xxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Randy Dunlap <rdunlap@xxxxxxxxxxxxx>
Cc: Robin Murphy <robin.murphy@xxxxxxx>
Cc: Steven Rostedt (Google) <rostedt@xxxxxxxxxxx>
Cc: Will Deacon <will@xxxxxxxxxx>
Cc: Yajun Deng <yajun.deng@xxxxxxxxx>
Cc: Zhen Lei <thunder.leizhen@xxxxxxxxxx>
Cc: Minchan Kim <minchan@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
Documentation/admin-guide/kernel-parameters.txt | 2 +-
arch/arm64/mm/init.c | 2 --
include/linux/dma-map-ops.h | 6 ------
kernel/dma/Kconfig | 6 +++---
kernel/dma/contiguous.c | 8 +++++++-
5 files changed, 11 insertions(+), 13 deletions(-)
--- a/arch/arm64/mm/init.c~dma-contiguous-support-per-numa-cma-for-all-architectures
+++ a/arch/arm64/mm/init.c
@@ -434,8 +434,6 @@ void __init bootmem_init(void)
arm64_hugetlb_cma_reserve();
#endif
- dma_pernuma_cma_reserve();
-
kvm_hyp_reserve();
/*
--- a/Documentation/admin-guide/kernel-parameters.txt~dma-contiguous-support-per-numa-cma-for-all-architectures
+++ a/Documentation/admin-guide/kernel-parameters.txt
@@ -696,7 +696,7 @@
kernel/dma/contiguous.c
cma_pernuma=nn[MG]
- [ARM64,KNL,CMA]
+ [KNL,CMA]
Sets the size of kernel per-numa memory area for
contiguous memory allocations. A value of 0 disables
per-numa CMA altogether. And If this option is not
--- a/include/linux/dma-map-ops.h~dma-contiguous-support-per-numa-cma-for-all-architectures
+++ a/include/linux/dma-map-ops.h
@@ -169,12 +169,6 @@ static inline void dma_free_contiguous(s
}
#endif /* CONFIG_DMA_CMA*/
-#ifdef CONFIG_DMA_PERNUMA_CMA
-void dma_pernuma_cma_reserve(void);
-#else
-static inline void dma_pernuma_cma_reserve(void) { }
-#endif /* CONFIG_DMA_PERNUMA_CMA */
-
#ifdef CONFIG_DMA_DECLARE_COHERENT
int dma_declare_coherent_memory(struct device *dev, phys_addr_t phys_addr,
dma_addr_t device_addr, size_t size);
--- a/kernel/dma/contiguous.c~dma-contiguous-support-per-numa-cma-for-all-architectures
+++ a/kernel/dma/contiguous.c
@@ -128,7 +128,7 @@ static inline __maybe_unused phys_addr_t
#endif
#ifdef CONFIG_DMA_PERNUMA_CMA
-void __init dma_pernuma_cma_reserve(void)
+static void __init dma_pernuma_cma_reserve(void)
{
int nid;
@@ -153,6 +153,10 @@ void __init dma_pernuma_cma_reserve(void
(unsigned long long)pernuma_size_bytes / SZ_1M, nid);
}
}
+#else
+static inline void __init dma_pernuma_cma_reserve(void)
+{
+}
#endif
/**
@@ -171,6 +175,8 @@ void __init dma_contiguous_reserve(phys_
phys_addr_t selected_limit = limit;
bool fixed = false;
+ dma_pernuma_cma_reserve();
+
pr_debug("%s(limit %08lx)\n", __func__, (unsigned long)limit);
if (size_cmdline != -1) {
--- a/kernel/dma/Kconfig~dma-contiguous-support-per-numa-cma-for-all-architectures
+++ a/kernel/dma/Kconfig
@@ -147,10 +147,10 @@ if DMA_CMA
config DMA_PERNUMA_CMA
bool "Enable separate DMA Contiguous Memory Area for each NUMA Node"
- default NUMA && ARM64
+ default NUMA
help
- Enable this option to get pernuma CMA areas so that devices like
- ARM64 SMMU can get local memory by DMA coherent APIs.
+ Enable this option to get pernuma CMA areas so that NUMA devices
+ can get local memory by DMA coherent APIs.
You can set the size of pernuma CMA by specifying "cma_pernuma=size"
on the kernel's command line.
_
Patches currently in -mm which might be from yajun.deng@xxxxxxxxx are
