[PATCH 0/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[PATCH 0/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[Chen Zhou]

If the memory reserved for crash dump kernel falled in ZONE_DMA32,
the devices in crash dump kernel need to use ZONE_DMA will alloc fail.
           
Fix this by reserving low memory in ZONE_DMA if CONFIG_ZONE_DMA is
enabled, otherwise, reserving in ZONE_DMA32.

Patch 1 updates the comments about the ZONE_DMA.
Patch 2 fix kdump broken.

Chen Zhou (2):
  arm64: mm: update the comments about ZONE_DMA
  arm64: mm: fix kdump broken with ZONE_DMA reintroduced

 arch/arm64/mm/init.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 1/2] arm64: mm: update the comments about ZONE_DMA

[Chen Zhou]

Since patchset "arm64: Default to 32-bit wide ZONE_DMA", ZONE_DMA's
size is fine-tuned. In the absence of addressing limited masters,
ZONE_DMA will span the whole 32-bit address space, otherwise, in
the case of the Raspberry Pi 4, it'll only span the 30-bit address
space. Update the comments.

Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
---
 arch/arm64/mm/init.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index 75addb36354a..7b9809e39927 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -53,10 +53,11 @@ s64 memstart_addr __ro_after_init = -1;
 EXPORT_SYMBOL(memstart_addr);
 
 /*
- * We create both ZONE_DMA and ZONE_DMA32. ZONE_DMA covers the first 1G of
- * memory as some devices, namely the Raspberry Pi 4, have peripherals with
- * this limited view of the memory. ZONE_DMA32 will cover the rest of the 32
- * bit addressable memory area.
+ * We create both ZONE_DMA and ZONE_DMA32. ZONE_DMA's size is fine-tuned.
+ * In the absence of addressing limited masters, ZONE_DMA will span the
+ * whole 32-bit address space, otherwise, in the case of the Raspberry Pi 4,
+ * it'll only span the 30-bit address space. ZONE_DMA32 will cover the rest
+ * of the 32 bit addressable memory area.
  */
 phys_addr_t arm64_dma_phys_limit __ro_after_init;
 static phys_addr_t arm64_dma32_phys_limit __ro_after_init;
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 1/2] arm64: mm: update the comments about ZONE_DMA

[Nicolas Saenz Julienne]

[-- Attachment #1.1: Type: text/plain, Size: 1637 bytes --]

On Sat, 2020-12-26 at 11:35 +0800, Chen Zhou wrote:
> Since patchset "arm64: Default to 32-bit wide ZONE_DMA", ZONE_DMA's
> size is fine-tuned. In the absence of addressing limited masters,
> ZONE_DMA will span the whole 32-bit address space, otherwise, in
> the case of the Raspberry Pi 4, it'll only span the 30-bit address
> space. Update the comments.
> 
> Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
> ---

Reviewed-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>

>  arch/arm64/mm/init.c | 9 +++++----
>  1 file changed, 5 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
> index 75addb36354a..7b9809e39927 100644
> --- a/arch/arm64/mm/init.c
> +++ b/arch/arm64/mm/init.c
> @@ -53,10 +53,11 @@ s64 memstart_addr __ro_after_init = -1;
>  EXPORT_SYMBOL(memstart_addr);
>  
> 
>  /*
> - * We create both ZONE_DMA and ZONE_DMA32. ZONE_DMA covers the first 1G of
> - * memory as some devices, namely the Raspberry Pi 4, have peripherals with
> - * this limited view of the memory. ZONE_DMA32 will cover the rest of the 32
> - * bit addressable memory area.
> + * We create both ZONE_DMA and ZONE_DMA32. ZONE_DMA's size is fine-tuned.
> + * In the absence of addressing limited masters, ZONE_DMA will span the
> + * whole 32-bit address space, otherwise, in the case of the Raspberry Pi 4,
> + * it'll only span the 30-bit address space. ZONE_DMA32 will cover the rest
> + * of the 32 bit addressable memory area.
>   */
>  phys_addr_t arm64_dma_phys_limit __ro_after_init;
>  static phys_addr_t arm64_dma32_phys_limit __ro_after_init;



[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

[-- Attachment #2: Type: text/plain, Size: 176 bytes --]

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH 2/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[Chen Zhou]

If the memory reserved for crash dump kernel falled in ZONE_DMA32,
the devices in crash dump kernel need to use ZONE_DMA will alloc fail.

Fix this by reserving low memory in ZONE_DMA if CONFIG_ZONE_DMA is
enabled, otherwise, reserving in ZONE_DMA32.

Fixes: bff3b04460a8 ("arm64: mm: reserve CMA and crashkernel in ZONE_DMA32")
Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
---
 arch/arm64/mm/init.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index 7b9809e39927..5074e945f1a6 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -85,7 +85,8 @@ static void __init reserve_crashkernel(void)
 
 	if (crash_base == 0) {
 		/* Current arm64 boot protocol requires 2MB alignment */
-		crash_base = memblock_find_in_range(0, arm64_dma32_phys_limit,
+		crash_base = memblock_find_in_range(0,
+				arm64_dma_phys_limit ? : arm64_dma32_phys_limit,
 				crash_size, SZ_2M);
 		if (crash_base == 0) {
 			pr_warn("cannot allocate crashkernel (size:0x%llx)\n",
-- 
2.20.1


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 2/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[Nicolas Saenz Julienne]

[-- Attachment #1.1: Type: text/plain, Size: 1582 bytes --]

Hi Chen, thanks for looking at this.

On Sat, 2020-12-26 at 11:35 +0800, Chen Zhou wrote:
> If the memory reserved for crash dump kernel falled in ZONE_DMA32,
> the devices in crash dump kernel need to use ZONE_DMA will alloc fail.
> 
> Fix this by reserving low memory in ZONE_DMA if CONFIG_ZONE_DMA is
> enabled, otherwise, reserving in ZONE_DMA32.
> 
> Fixes: bff3b04460a8 ("arm64: mm: reserve CMA and crashkernel in ZONE_DMA32")

I'm not so sure this counts as a fix, if someone backports it it'll probably
break things as it depends on the series that dynamically sizes DMA zones.

> Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
> ---

Why not doing the same with CMA? You'll probably have to move the
dma_contiguous_reserve() call into bootmem_init() so as to make sure that
arm64_dma_phys_limit is populated.

Regards,
Nicolas

>  arch/arm64/mm/init.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
> index 7b9809e39927..5074e945f1a6 100644
> --- a/arch/arm64/mm/init.c
> +++ b/arch/arm64/mm/init.c
> @@ -85,7 +85,8 @@ static void __init reserve_crashkernel(void)
>  
> 
>  	if (crash_base == 0) {
>  		/* Current arm64 boot protocol requires 2MB alignment */
> -		crash_base = memblock_find_in_range(0, arm64_dma32_phys_limit,
> +		crash_base = memblock_find_in_range(0,
> +				arm64_dma_phys_limit ? : arm64_dma32_phys_limit,
>  				crash_size, SZ_2M);
>  		if (crash_base == 0) {
>  			pr_warn("cannot allocate crashkernel (size:0x%llx)\n",



[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

[-- Attachment #2: Type: text/plain, Size: 176 bytes --]

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 2/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[chenzhou]

Hi Nicolas,

Thanks for your review.


On 2020/12/26 18:34, Nicolas Saenz Julienne wrote:
> Hi Chen, thanks for looking at this.
>
> On Sat, 2020-12-26 at 11:35 +0800, Chen Zhou wrote:
>> If the memory reserved for crash dump kernel falled in ZONE_DMA32,
>> the devices in crash dump kernel need to use ZONE_DMA will alloc fail.
>>
>> Fix this by reserving low memory in ZONE_DMA if CONFIG_ZONE_DMA is
>> enabled, otherwise, reserving in ZONE_DMA32.
>>
>> Fixes: bff3b04460a8 ("arm64: mm: reserve CMA and crashkernel in ZONE_DMA32")
> I'm not so sure this counts as a fix, if someone backports it it'll probably
> break things as it depends on the series that dynamically sizes DMA zones.
I write this just because kdump is broken from this commit.
>
>> Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
>> ---
> Why not doing the same with CMA? You'll probably have to move the
> dma_contiguous_reserve() call into bootmem_init() so as to make sure that
> arm64_dma_phys_limit is populated.
You are right, CMA also need this. I will do this in next version.

Thanks,
Chen Zhou
>
> Regards,
> Nicolas
>
>>  arch/arm64/mm/init.c | 3 ++-
>>  1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
>> index 7b9809e39927..5074e945f1a6 100644
>> --- a/arch/arm64/mm/init.c
>> +++ b/arch/arm64/mm/init.c
>> @@ -85,7 +85,8 @@ static void __init reserve_crashkernel(void)
>>  
>>
>>  	if (crash_base == 0) {
>>  		/* Current arm64 boot protocol requires 2MB alignment */
>> -		crash_base = memblock_find_in_range(0, arm64_dma32_phys_limit,
>> +		crash_base = memblock_find_in_range(0,
>> +				arm64_dma_phys_limit ? : arm64_dma32_phys_limit,
>>  				crash_size, SZ_2M);
>>  		if (crash_base == 0) {
>>  			pr_warn("cannot allocate crashkernel (size:0x%llx)\n",
>


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 2/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[Catalin Marinas]

On Mon, Dec 28, 2020 at 03:32:35PM +0800, chenzhou wrote:
> On 2020/12/26 18:34, Nicolas Saenz Julienne wrote:
> > On Sat, 2020-12-26 at 11:35 +0800, Chen Zhou wrote:
> >> If the memory reserved for crash dump kernel falled in ZONE_DMA32,
> >> the devices in crash dump kernel need to use ZONE_DMA will alloc fail.
> >>
> >> Fix this by reserving low memory in ZONE_DMA if CONFIG_ZONE_DMA is
> >> enabled, otherwise, reserving in ZONE_DMA32.
> >>
> >> Fixes: bff3b04460a8 ("arm64: mm: reserve CMA and crashkernel in ZONE_DMA32")
> > I'm not so sure this counts as a fix, if someone backports it it'll probably
> > break things as it depends on the series that dynamically sizes DMA zones.
> 
> I write this just because kdump is broken from this commit.

Commit bff3b04460a8 actually fixed a regression introduced by the
limited 30-bit ZONE_DMA, under the assumption that RPi4 is the only
platform that cares about this. I agree a kdump kernel will fail to get
memory in the low zone but it's not clear to me whether the regression
you are seeing happened in 5.11-rc1 or you it's been around since 5.5.

We may have got away before because max_zone_phys() assumed an implicit
DMA offset equal to the start of the DRAM (well, power of two). This
assumption was removed in 5.11 by commit 791ab8b2e3db ("arm64: Ignore
any DMA offsets in the max_zone_phys() calculation").

So your patch or the clean-up I posted only makes sense with the dynamic
ZONE_DMA patches in 5.11-rc1. For any earlier version, I don't have any
suggestion other than not enabling CONFIG_ZONE_DMA in the kdump kernel
(or backporting all of Nicolas' patches).

I'll post my clean-up separately to go into 5.11 but won't have a fixes
tag for bff3b04460a8.

-- 
Catalin

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 2/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[Catalin Marinas]

On Sat, Dec 26, 2020 at 11:34:58AM +0100, Nicolas Saenz Julienne wrote:
> On Sat, 2020-12-26 at 11:35 +0800, Chen Zhou wrote:
> > If the memory reserved for crash dump kernel falled in ZONE_DMA32,
> > the devices in crash dump kernel need to use ZONE_DMA will alloc fail.
> > 
> > Fix this by reserving low memory in ZONE_DMA if CONFIG_ZONE_DMA is
> > enabled, otherwise, reserving in ZONE_DMA32.
> > 
> > Fixes: bff3b04460a8 ("arm64: mm: reserve CMA and crashkernel in ZONE_DMA32")
> 
> I'm not so sure this counts as a fix, if someone backports it it'll probably
> break things as it depends on the series that dynamically sizes DMA zones.
> 
> > Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
> > ---
> 
> Why not doing the same with CMA? You'll probably have to move the
> dma_contiguous_reserve() call into bootmem_init() so as to make sure that
> arm64_dma_phys_limit is populated.

Do we need the arm64_dma32_phys_limit at all? I can see the
(arm64_dma_phys_limit ? : arm64_dma32_phys_limit) pattern in several
places but I think we can just live with the arm64_dma_phys_limit.

Also, I don't think we need any early ARCH_LOW_ADDRESS_LIMIT. It's only
used by memblock_alloc_low() and that's called from swiotlb_init()
after arm64_dma_phys_limit was initialised.

What about something like below (on top of you ARCH_LOW_ADDRESS_LIMIT
fix but I can revert that)? I haven't tested it in all configurations
yet.

diff --git a/arch/arm64/include/asm/processor.h b/arch/arm64/include/asm/processor.h
index 69ad25fbeae4..ca2cd75d3286 100644
--- a/arch/arm64/include/asm/processor.h
+++ b/arch/arm64/include/asm/processor.h
@@ -94,8 +94,7 @@
 #endif /* CONFIG_ARM64_FORCE_52BIT */
 
 extern phys_addr_t arm64_dma_phys_limit;
-extern phys_addr_t arm64_dma32_phys_limit;
-#define ARCH_LOW_ADDRESS_LIMIT	((arm64_dma_phys_limit ? : arm64_dma32_phys_limit) - 1)
+#define ARCH_LOW_ADDRESS_LIMIT	(arm64_dma_phys_limit - 1)
 
 struct debug_info {
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index 7deddf56f7c3..596a94bf5ed6 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -59,7 +59,6 @@ EXPORT_SYMBOL(memstart_addr);
  * bit addressable memory area.
  */
 phys_addr_t arm64_dma_phys_limit __ro_after_init;
-phys_addr_t arm64_dma32_phys_limit __ro_after_init;
 
 #ifdef CONFIG_KEXEC_CORE
 /*
@@ -84,7 +83,7 @@ static void __init reserve_crashkernel(void)
 
 	if (crash_base == 0) {
 		/* Current arm64 boot protocol requires 2MB alignment */
-		crash_base = memblock_find_in_range(0, arm64_dma32_phys_limit,
+		crash_base = memblock_find_in_range(0, arm64_dma_phys_limit,
 				crash_size, SZ_2M);
 		if (crash_base == 0) {
 			pr_warn("cannot allocate crashkernel (size:0x%llx)\n",
@@ -196,6 +195,7 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max)
 	unsigned long max_zone_pfns[MAX_NR_ZONES]  = {0};
 	unsigned int __maybe_unused acpi_zone_dma_bits;
 	unsigned int __maybe_unused dt_zone_dma_bits;
+	phys_addr_t dma32_phys_limit = max_zone_phys(32);
 
 #ifdef CONFIG_ZONE_DMA
 	acpi_zone_dma_bits = fls64(acpi_iort_dma_get_max_cpu_address());
@@ -205,8 +205,12 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max)
 	max_zone_pfns[ZONE_DMA] = PFN_DOWN(arm64_dma_phys_limit);
 #endif
 #ifdef CONFIG_ZONE_DMA32
-	max_zone_pfns[ZONE_DMA32] = PFN_DOWN(arm64_dma32_phys_limit);
+	max_zone_pfns[ZONE_DMA32] = PFN_DOWN(dma32_phys_limit);
+	if (!arm64_dma_phys_limit)
+		arm64_dma_phys_limit = dma32_phys_limit;
 #endif
+	if (!arm64_dma_phys_limit)
+		arm64_dma_phys_limit = PHYS_MASK + 1;
 	max_zone_pfns[ZONE_NORMAL] = max;
 
 	free_area_init(max_zone_pfns);
@@ -394,16 +398,9 @@ void __init arm64_memblock_init(void)
 
 	early_init_fdt_scan_reserved_mem();
 
-	if (IS_ENABLED(CONFIG_ZONE_DMA32))
-		arm64_dma32_phys_limit = max_zone_phys(32);
-	else
-		arm64_dma32_phys_limit = PHYS_MASK + 1;
-
 	reserve_elfcorehdr();
 
 	high_memory = __va(memblock_end_of_DRAM() - 1) + 1;
-
-	dma_contiguous_reserve(arm64_dma32_phys_limit);
 }
 
 void __init bootmem_init(void)
@@ -438,6 +435,11 @@ void __init bootmem_init(void)
 	sparse_init();
 	zone_sizes_init(min, max);
 
+	/*
+	 * Reserve the CMA area after arm64_dma_phys_limit was initialised.
+	 */
+	dma_contiguous_reserve(arm64_dma_phys_limit);
+
 	/*
 	 * request_standard_resources() depends on crashkernel's memory being
 	 * reserved, so do it here.
@@ -455,7 +457,7 @@ void __init bootmem_init(void)
 void __init mem_init(void)
 {
 	if (swiotlb_force == SWIOTLB_FORCE ||
-	    max_pfn > PFN_DOWN(arm64_dma_phys_limit ? : arm64_dma32_phys_limit))
+	    max_pfn > PFN_DOWN(arm64_dma_phys_limit))
 		swiotlb_init(1);
 	else
 		swiotlb_force = SWIOTLB_NO_FORCE;

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 2/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[chenzhou]

On 2021/1/7 22:25, Catalin Marinas wrote:
> On Sat, Dec 26, 2020 at 11:34:58AM +0100, Nicolas Saenz Julienne wrote:
>> On Sat, 2020-12-26 at 11:35 +0800, Chen Zhou wrote:
>>> If the memory reserved for crash dump kernel falled in ZONE_DMA32,
>>> the devices in crash dump kernel need to use ZONE_DMA will alloc fail.
>>>
>>> Fix this by reserving low memory in ZONE_DMA if CONFIG_ZONE_DMA is
>>> enabled, otherwise, reserving in ZONE_DMA32.
>>>
>>> Fixes: bff3b04460a8 ("arm64: mm: reserve CMA and crashkernel in ZONE_DMA32")
>> I'm not so sure this counts as a fix, if someone backports it it'll probably
>> break things as it depends on the series that dynamically sizes DMA zones.
>>
>>> Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
>>> ---
>> Why not doing the same with CMA? You'll probably have to move the
>> dma_contiguous_reserve() call into bootmem_init() so as to make sure that
>> arm64_dma_phys_limit is populated.
> Do we need the arm64_dma32_phys_limit at all? I can see the
> (arm64_dma_phys_limit ? : arm64_dma32_phys_limit) pattern in several
> places but I think we can just live with the arm64_dma_phys_limit.
Yes, arm64_dma_phys_limit is enough.
>
> Also, I don't think we need any early ARCH_LOW_ADDRESS_LIMIT. It's only
> used by memblock_alloc_low() and that's called from swiotlb_init()
> after arm64_dma_phys_limit was initialised.
>
> What about something like below (on top of you ARCH_LOW_ADDRESS_LIMIT
> fix but I can revert that)? I haven't tested it in all configurations
> yet.
>
> diff --git a/arch/arm64/include/asm/processor.h b/arch/arm64/include/asm/processor.h
> index 69ad25fbeae4..ca2cd75d3286 100644
> --- a/arch/arm64/include/asm/processor.h
> +++ b/arch/arm64/include/asm/processor.h
> @@ -94,8 +94,7 @@
>  #endif /* CONFIG_ARM64_FORCE_52BIT */
>  
>  extern phys_addr_t arm64_dma_phys_limit;
> -extern phys_addr_t arm64_dma32_phys_limit;
> -#define ARCH_LOW_ADDRESS_LIMIT	((arm64_dma_phys_limit ? : arm64_dma32_phys_limit) - 1)
> +#define ARCH_LOW_ADDRESS_LIMIT	(arm64_dma_phys_limit - 1)
>  
>  struct debug_info {
>  #ifdef CONFIG_HAVE_HW_BREAKPOINT
> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
> index 7deddf56f7c3..596a94bf5ed6 100644
> --- a/arch/arm64/mm/init.c
> +++ b/arch/arm64/mm/init.c
> @@ -59,7 +59,6 @@ EXPORT_SYMBOL(memstart_addr);
>   * bit addressable memory area.
>   */
>  phys_addr_t arm64_dma_phys_limit __ro_after_init;
> -phys_addr_t arm64_dma32_phys_limit __ro_after_init;
>  
>  #ifdef CONFIG_KEXEC_CORE
>  /*
> @@ -84,7 +83,7 @@ static void __init reserve_crashkernel(void)
>  
>  	if (crash_base == 0) {
>  		/* Current arm64 boot protocol requires 2MB alignment */
> -		crash_base = memblock_find_in_range(0, arm64_dma32_phys_limit,
> +		crash_base = memblock_find_in_range(0, arm64_dma_phys_limit,
>  				crash_size, SZ_2M);
>  		if (crash_base == 0) {
>  			pr_warn("cannot allocate crashkernel (size:0x%llx)\n",
> @@ -196,6 +195,7 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max)
>  	unsigned long max_zone_pfns[MAX_NR_ZONES]  = {0};
>  	unsigned int __maybe_unused acpi_zone_dma_bits;
>  	unsigned int __maybe_unused dt_zone_dma_bits;
> +	phys_addr_t dma32_phys_limit = max_zone_phys(32);
>  
>  #ifdef CONFIG_ZONE_DMA
>  	acpi_zone_dma_bits = fls64(acpi_iort_dma_get_max_cpu_address());
> @@ -205,8 +205,12 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max)
>  	max_zone_pfns[ZONE_DMA] = PFN_DOWN(arm64_dma_phys_limit);
>  #endif
>  #ifdef CONFIG_ZONE_DMA32
> -	max_zone_pfns[ZONE_DMA32] = PFN_DOWN(arm64_dma32_phys_limit);
> +	max_zone_pfns[ZONE_DMA32] = PFN_DOWN(dma32_phys_limit);
> +	if (!arm64_dma_phys_limit)
> +		arm64_dma_phys_limit = dma32_phys_limit;
>  #endif
> +	if (!arm64_dma_phys_limit)
> +		arm64_dma_phys_limit = PHYS_MASK + 1;
>  	max_zone_pfns[ZONE_NORMAL] = max;
>  
>  	free_area_init(max_zone_pfns);
> @@ -394,16 +398,9 @@ void __init arm64_memblock_init(void)
>  
>  	early_init_fdt_scan_reserved_mem();
>  
> -	if (IS_ENABLED(CONFIG_ZONE_DMA32))
> -		arm64_dma32_phys_limit = max_zone_phys(32);
> -	else
> -		arm64_dma32_phys_limit = PHYS_MASK + 1;
> -
>  	reserve_elfcorehdr();
>  
>  	high_memory = __va(memblock_end_of_DRAM() - 1) + 1;
> -
> -	dma_contiguous_reserve(arm64_dma32_phys_limit);
>  }
>  
>  void __init bootmem_init(void)
> @@ -438,6 +435,11 @@ void __init bootmem_init(void)
>  	sparse_init();
>  	zone_sizes_init(min, max);
>  
> +	/*
> +	 * Reserve the CMA area after arm64_dma_phys_limit was initialised.
> +	 */
> +	dma_contiguous_reserve(arm64_dma_phys_limit);
> +
>  	/*
>  	 * request_standard_resources() depends on crashkernel's memory being
>  	 * reserved, so do it here.
> @@ -455,7 +457,7 @@ void __init bootmem_init(void)
>  void __init mem_init(void)
>  {
>  	if (swiotlb_force == SWIOTLB_FORCE ||
> -	    max_pfn > PFN_DOWN(arm64_dma_phys_limit ? : arm64_dma32_phys_limit))
> +	    max_pfn > PFN_DOWN(arm64_dma_phys_limit))
>  		swiotlb_init(1);
>  	else
>  		swiotlb_force = SWIOTLB_NO_FORCE;
> .
>


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 0/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[Will Deacon]

On Sat, Dec 26, 2020 at 11:35:55AM +0800, Chen Zhou wrote:
> If the memory reserved for crash dump kernel falled in ZONE_DMA32,
> the devices in crash dump kernel need to use ZONE_DMA will alloc fail.
>            
> Fix this by reserving low memory in ZONE_DMA if CONFIG_ZONE_DMA is
> enabled, otherwise, reserving in ZONE_DMA32.
> 
> Patch 1 updates the comments about the ZONE_DMA.
> Patch 2 fix kdump broken.
> 
> Chen Zhou (2):
>   arm64: mm: update the comments about ZONE_DMA
>   arm64: mm: fix kdump broken with ZONE_DMA reintroduced

Please can you repost this if it is still needed after the changes that
landed in -rc4?

Will

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 0/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[Catalin Marinas]

On Wed, Jan 20, 2021 at 12:40:55PM +0000, Will Deacon wrote:
> On Sat, Dec 26, 2020 at 11:35:55AM +0800, Chen Zhou wrote:
> > If the memory reserved for crash dump kernel falled in ZONE_DMA32,
> > the devices in crash dump kernel need to use ZONE_DMA will alloc fail.
> >            
> > Fix this by reserving low memory in ZONE_DMA if CONFIG_ZONE_DMA is
> > enabled, otherwise, reserving in ZONE_DMA32.
> > 
> > Patch 1 updates the comments about the ZONE_DMA.
> > Patch 2 fix kdump broken.
> > 
> > Chen Zhou (2):
> >   arm64: mm: update the comments about ZONE_DMA
> >   arm64: mm: fix kdump broken with ZONE_DMA reintroduced
> 
> Please can you repost this if it is still needed after the changes that
> landed in -rc4?

I don't think this series is needed anymore but I'll let Chen confirm.

-- 
Catalin

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 0/2] arm64: mm: fix kdump broken with ZONE_DMA reintroduced

[chenzhou]

Hi Will, Catalin,


On 2021/1/20 21:07, Catalin Marinas wrote:
> On Wed, Jan 20, 2021 at 12:40:55PM +0000, Will Deacon wrote:
>> On Sat, Dec 26, 2020 at 11:35:55AM +0800, Chen Zhou wrote:
>>> If the memory reserved for crash dump kernel falled in ZONE_DMA32,
>>> the devices in crash dump kernel need to use ZONE_DMA will alloc fail.
>>>            
>>> Fix this by reserving low memory in ZONE_DMA if CONFIG_ZONE_DMA is
>>> enabled, otherwise, reserving in ZONE_DMA32.
>>>
>>> Patch 1 updates the comments about the ZONE_DMA.
>>> Patch 2 fix kdump broken.
>>>
>>> Chen Zhou (2):
>>>   arm64: mm: update the comments about ZONE_DMA
>>>   arm64: mm: fix kdump broken with ZONE_DMA reintroduced
>> Please can you repost this if it is still needed after the changes that
>> landed in -rc4?
> I don't think this series is needed anymore but I'll let Chen confirm.
Yes, we don't need this series, Catalin has solved the issue i mentioned.

I will repost the series "support reserving crashkernel above 4G on arm64 kdump"
based on the latest code.

Thanks,
Chen Zhou


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel