In the 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.
Signed-off-by: Yajun Deng <[email protected]>
---
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(-)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 56d9458276a6..ac0002b2e323 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -692,7 +692,7 @@
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
diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index 66e70ca47680..d560aef6aafa 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -410,8 +410,6 @@ void __init bootmem_init(void)
arm64_hugetlb_cma_reserve();
#endif
- dma_pernuma_cma_reserve();
-
kvm_hyp_reserve();
/*
diff --git a/include/linux/dma-map-ops.h b/include/linux/dma-map-ops.h
index 31f114f486c4..7af9949828ff 100644
--- a/include/linux/dma-map-ops.h
+++ b/include/linux/dma-map-ops.h
@@ -168,12 +168,6 @@ static inline void dma_free_contiguous(struct device *dev, struct page *page,
}
#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);
diff --git a/kernel/dma/Kconfig b/kernel/dma/Kconfig
index 6677d0e64d27..79f83091e3a2 100644
--- a/kernel/dma/Kconfig
+++ b/kernel/dma/Kconfig
@@ -140,10 +140,10 @@ 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.
diff --git a/kernel/dma/contiguous.c b/kernel/dma/contiguous.c
index 6ea80ae42622..26a8e5365fcd 100644
--- a/kernel/dma/contiguous.c
+++ b/kernel/dma/contiguous.c
@@ -128,7 +128,7 @@ static inline __maybe_unused phys_addr_t cma_early_percent_memory(void)
#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_addr_t limit)
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) {
--
2.25.1
On Fri, 12 May 2023 17:42:10 +0800 Yajun Deng <[email protected]> wrote:
> In the 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.
>
> ...
>
> --- a/include/linux/dma-map-ops.h
> +++ b/include/linux/dma-map-ops.h
> @@ -168,12 +168,6 @@ static inline void dma_free_contiguous(struct device *dev, struct page *page,
> }
> #endif /* CONFIG_DMA_CMA*/
>
> -#ifdef CONFIG_DMA_PERNUMA_CMA
> -void dma_pernuma_cma_reserve(void);
> -#else
> -static inline void dma_pernuma_cma_reserve(void) { }
It would be a little nicer to retain this line.
> -#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
> +++ b/kernel/dma/contiguous.c
> @@ -128,7 +128,7 @@ static inline __maybe_unused phys_addr_t cma_early_percent_memory(void)
> #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
And to not add this function?
This looks fine to me. Can you please work with Barry to make sure
the slight different place of the initcall doesn't break anything
for his setup? I doubt it would, but I'd rather have a Tested-by:
tag.
May 15, 2023 5:49 PM, "Christoph Hellwig" <[email protected]> wrote:
> This looks fine to me. Can you please work with Barry to make sure
> the slight different place of the initcall doesn't break anything
> for his setup? I doubt it would, but I'd rather have a Tested-by:
> tag.
Barry's email is no longer in use. I can't reach him.
On Mon, 15 May 2023 11:23:27 +0000
"Yajun Deng" <[email protected]> wrote:
> May 15, 2023 5:49 PM, "Christoph Hellwig" <[email protected]> wrote:
>
> > This looks fine to me. Can you please work with Barry to make sure
> > the slight different place of the initcall doesn't break anything
> > for his setup? I doubt it would, but I'd rather have a Tested-by:
> > tag.
>
> Barry's email is no longer in use. I can't reach him.
Which one? I would hope that his Gmail account is still valid:
Barry Song <[email protected]>
Petr T
Hello, Barry
This patch changed the caller of dma_pernuma_cma_reserve() from
bootmem_init() to dma_contiguous_reserve(), do you think
would there be something wrong?
On 2023/5/12 17:42, Yajun Deng wrote:
> In the 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.
>
> Signed-off-by: Yajun Deng <[email protected]>
> ---
> 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(-)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index 56d9458276a6..ac0002b2e323 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -692,7 +692,7 @@
> 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
> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
> index 66e70ca47680..d560aef6aafa 100644
> --- a/arch/arm64/mm/init.c
> +++ b/arch/arm64/mm/init.c
> @@ -410,8 +410,6 @@ void __init bootmem_init(void)
> arm64_hugetlb_cma_reserve();
> #endif
>
> - dma_pernuma_cma_reserve();
> -
> kvm_hyp_reserve();
>
> /*
> diff --git a/include/linux/dma-map-ops.h b/include/linux/dma-map-ops.h
> index 31f114f486c4..7af9949828ff 100644
> --- a/include/linux/dma-map-ops.h
> +++ b/include/linux/dma-map-ops.h
> @@ -168,12 +168,6 @@ static inline void dma_free_contiguous(struct device *dev, struct page *page,
> }
> #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);
> diff --git a/kernel/dma/Kconfig b/kernel/dma/Kconfig
> index 6677d0e64d27..79f83091e3a2 100644
> --- a/kernel/dma/Kconfig
> +++ b/kernel/dma/Kconfig
> @@ -140,10 +140,10 @@ 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.
> diff --git a/kernel/dma/contiguous.c b/kernel/dma/contiguous.c
> index 6ea80ae42622..26a8e5365fcd 100644
> --- a/kernel/dma/contiguous.c
> +++ b/kernel/dma/contiguous.c
> @@ -128,7 +128,7 @@ static inline __maybe_unused phys_addr_t cma_early_percent_memory(void)
> #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_addr_t limit)
> 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) {
On 2023/5/15 19:38, Petr Tesařík wrote:
> On Mon, 15 May 2023 11:23:27 +0000
> "Yajun Deng" <[email protected]> wrote:
>
>> May 15, 2023 5:49 PM, "Christoph Hellwig" <[email protected]> wrote:
>>
>>> This looks fine to me. Can you please work with Barry to make sure
>>> the slight different place of the initcall doesn't break anything
>>> for his setup? I doubt it would, but I'd rather have a Tested-by:
>>> tag.
>> Barry's email is no longer in use. I can't reach him.
> Which one? I would hope that his Gmail account is still valid:
>
> Barry Song <[email protected]>
Thanks.
>
> Petr T
On Mon, 15 May 2023 13:38:21 +0200 Petr Tesařík <[email protected]> wrote:
> On Mon, 15 May 2023 11:23:27 +0000
> "Yajun Deng" <[email protected]> wrote:
>
> > May 15, 2023 5:49 PM, "Christoph Hellwig" <[email protected]> wrote:
> >
> > > This looks fine to me. Can you please work with Barry to make sure
> > > the slight different place of the initcall doesn't break anything
> > > for his setup? I doubt it would, but I'd rather have a Tested-by:
> > > tag.
> >
> > Barry's email is no longer in use. I can't reach him.
>
> Which one? I would hope that his Gmail account is still valid:
>
> Barry Song <[email protected]>
>
Maybe his kernel.org address works...
I have this patch stuck in limbo for 6.4. I guess I'll carry it over
into the next -rc cycle, see what happens.
fwiw, it has been in -next for six weeks, no known issues.
June 24, 2023 8:40 AM, "Andrew Morton" <[email protected]> wrote:
> On Mon, 15 May 2023 13:38:21 +0200 Petr Tesařík <[email protected]> wrote:
>
>> On Mon, 15 May 2023 11:23:27 +0000
>> "Yajun Deng" <[email protected]> wrote:
>>
>> May 15, 2023 5:49 PM, "Christoph Hellwig" <[email protected]> wrote:
>
> This looks fine to me. Can you please work with Barry to make sure
> the slight different place of the initcall doesn't break anything
> for his setup? I doubt it would, but I'd rather have a Tested-by:
> tag.
>> Barry's email is no longer in use. I can't reach him.
>>
>> Which one? I would hope that his Gmail account is still valid:
>>
>> Barry Song <[email protected]>
>
> Maybe his kernel.org address works...
>
> I have this patch stuck in limbo for 6.4. I guess I'll carry it over
> into the next -rc cycle, see what happens.
>
> fwiw, it has been in -next for six weeks, no known issues.
Hi, Barry, The slight different place of the initcall, does break anything?
On Sun, Jun 25, 2023 at 7:30 PM Yajun Deng <[email protected]> wrote:
>
> June 24, 2023 8:40 AM, "Andrew Morton" <[email protected]> wrote:
>
> > On Mon, 15 May 2023 13:38:21 +0200 Petr Tesařík <[email protected]> wrote:
> >
> >> On Mon, 15 May 2023 11:23:27 +0000
> >> "Yajun Deng" <[email protected]> wrote:
> >>
> >> May 15, 2023 5:49 PM, "Christoph Hellwig" <[email protected]> wrote:
> >
> > This looks fine to me. Can you please work with Barry to make sure
> > the slight different place of the initcall doesn't break anything
> > for his setup? I doubt it would, but I'd rather have a Tested-by:
> > tag.
> >> Barry's email is no longer in use. I can't reach him.
> >>
> >> Which one? I would hope that his Gmail account is still valid:
> >>
> >> Barry Song <[email protected]>
> >
> > Maybe his kernel.org address works...
> >
> > I have this patch stuck in limbo for 6.4. I guess I'll carry it over
> > into the next -rc cycle, see what happens.
> >
> > fwiw, it has been in -next for six weeks, no known issues.
>
> Hi, Barry, The slight different place of the initcall, does break anything?
i don't see a fundamental difference as anyway it is still after
arch_numa_init()
which is really what we depend on.
and i did a test on qemu with the command line:
qemu-system-aarch64 -M virt,gic-version=3 -nographic \
-smp cpus=8 \
-numa node,cpus=0-1,nodeid=0 \
-numa node,cpus=2-3,nodeid=1 \
-numa node,cpus=4-5,nodeid=2 \
-numa node,cpus=6-7,nodeid=3 \
-numa dist,src=0,dst=1,val=12 \
-numa dist,src=0,dst=2,val=20 \
-numa dist,src=0,dst=3,val=22 \
-numa dist,src=1,dst=2,val=22 \
-numa dist,src=2,dst=3,val=12 \
-numa dist,src=1,dst=3,val=24 \
-m 4096M -cpu cortex-a57 -kernel arch/arm64/boot/Image \
-nographic -append "cma_pernuma=32M root=/dev/vda2 rw ip=dhcp
sched_debug irqchip.gicv3_pseudo_nmi=1" \
-drive if=none,file=extra/ubuntu16.04-arm64.img,id=hd0 -device
virtio-blk-device,drive=hd0 \
-net nic -net user,hostfwd=tcp::2222-:22
and in system, i can see all cma areas are correctly reserved:
~# dmesg | grep cma
[ 0.000000] cma: cma_declare_contiguous_nid(size
0x0000000002000000, base 0x0000000000000000, limit 0x0000000000000000
alignment 0x0000000000000000)
[ 0.000000] cma: Reserved 32 MiB at 0x000000007ce00000
[ 0.000000] cma: dma_pernuma_cma_reserve: reserved 32 MiB on node 0
[ 0.000000] cma: cma_declare_contiguous_nid(size
0x0000000002000000, base 0x0000000000000000, limit 0x0000000000000000
alignment 0x0000000000000000)
[ 0.000000] cma: Reserved 32 MiB at 0x00000000bce00000
[ 0.000000] cma: dma_pernuma_cma_reserve: reserved 32 MiB on node 1
[ 0.000000] cma: cma_declare_contiguous_nid(size
0x0000000002000000, base 0x0000000000000000, limit 0x0000000000000000
alignment 0x0000000000000000)
[ 0.000000] cma: Reserved 32 MiB at 0x00000000fce00000
[ 0.000000] cma: dma_pernuma_cma_reserve: reserved 32 MiB on node 2
[ 0.000000] cma: cma_declare_contiguous_nid(size
0x0000000002000000, base 0x0000000000000000, limit 0x0000000000000000
alignment 0x0000000000000000)
[ 0.000000] cma: Reserved 32 MiB at 0x0000000100000000
[ 0.000000] cma: dma_pernuma_cma_reserve: reserved 32 MiB on node 3
[ 0.000000] cma: dma_contiguous_reserve(limit 100000000)
[ 0.000000] cma: dma_contiguous_reserve: reserving 32 MiB for global area
[ 0.000000] cma: cma_declare_contiguous_nid(size
0x0000000002000000, base 0x0000000000000000, limit 0x0000000100000000
alignment 0x0000000000000000)
[ 0.000000] cma: Reserved 32 MiB at 0x00000000fae00000
[ 0.000000] Kernel command line: cma_pernuma=32M root=/dev/vda2 rw
ip=dhcp sched_debug irqchip.gicv3_pseudo_nmi=1
[ 0.000000] Memory: 3848784K/4194304K available (16128K kernel
code, 4152K rwdata, 10244K rodata, 8512K init, 612K bss, 181680K
reserved, 163840K cma-reserved)
[ 0.175309] cma: cma_alloc(cma (____ptrval____), count 128, align 7)
[ 0.179264] cma: cma_alloc(): returned (____ptrval____)
[ 0.179869] cma: cma_alloc(cma (____ptrval____), count 128, align 7)
[ 0.180027] cma: cma_alloc(): returned (____ptrval____)
[ 0.180187] cma: cma_alloc(cma (____ptrval____), count 128, align 7)
[ 0.180374] cma: cma_alloc(): returned (____ptrval____)
so my feeling is that this patch is fine. but I would prefer Yicong
and Tiantao who have a real numa machine
and we can get some real device drivers to call dma APIs to allocate
memory from pernuma cma on arm64
even though it is 99.9% OK.
With their testing done, please feel free to add
Acked-by: Barry Song <[email protected]>
Thanks
Barry
Thanks,
applied to the dma-mapping tree for 6.6.