[PATCH -next v4 0/2] support allocating crashkernel above 4G explicitly on riscv

[PATCH -next v4 0/2] support allocating crashkernel above 4G explicitly on riscv

[Chen Jiahao]

On riscv, the current crash kernel allocation logic is trying to
allocate within 32bit addressible memory region by default, if
failed, try to allocate without 4G restriction.

In need of saving DMA zone memory while allocating a relatively large
crash kernel region, allocating the reserved memory top down in
high memory, without overlapping the DMA zone, is a mature solution.
Hence this patchset introduces the parameter option crashkernel=X,[high,low].

One can reserve the crash kernel from high memory above DMA zone range
by explicitly passing "crashkernel=X,high"; or reserve a memory range
below 4G with "crashkernel=X,low". Besides, there are few rules need
to take notice:
1. "crashkernel=X,[high,low]" will be ignored if "crashkernel=size"
   is specified.
2. "crashkernel=X,low" is valid only when "crashkernel=X,high" is passed
   and there is enough memory to be allocated under 4G.
3. When allocating crashkernel above 4G and no "crashkernel=X,low" is
   specified, a 128M low memory will be allocated automatically for
   swiotlb bounce buffer.
See Documentation/admin-guide/kernel-parameters.txt for more information.

To verify loading the crashkernel, adapted kexec-tools is attached below:
https://github.com/chenjh005/kexec-tools/tree/build-test-riscv-v2

Following test cases have been performed as expected:
1) crashkernel=256M                          //low=256M
2) crashkernel=1G                            //low=1G
3) crashkernel=4G                            //high=4G, low=128M(default)
4) crashkernel=4G crashkernel=256M,high      //high=4G, low=128M(default), high is ignored
5) crashkernel=4G crashkernel=256M,low       //high=4G, low=128M(default), low is ignored
6) crashkernel=4G,high                       //high=4G, low=128M(default)
7) crashkernel=256M,low                      //low=0M, invalid
8) crashkernel=4G,high crashkernel=256M,low  //high=4G, low=256M
9) crashkernel=4G,high crashkernel=4G,low    //high=0M, low=0M, invalid
10) crashkernel=512M@0xd0000000              //low=512M
11) crashkernel=1G@0xe0000000                //high=0M, low=0M, no enough low memory, failed

Changes since [v4]:
1. Update some imprecise code comments for cmdline parsing.

Changes since [v3]:
1. Update to print warning and return explicitly on failure when
   crashkernel=size@offset is specified. Not changing the result
   in this case but making the logic more straightforward.
2. Some minor cleanup.

Changes since [v2]:
1. Update the allocation logic to ensure the high crashkernel
   region is reserved strictly above dma32_phys_limit.
2. Clean up some minor format problems.

Chen Jiahao (2):
  riscv: kdump: Implement crashkernel=X,[high,low]
  docs: kdump: Update the crashkernel description for riscv

 .../admin-guide/kernel-parameters.txt         | 15 ++--
 arch/riscv/kernel/setup.c                     |  5 ++
 arch/riscv/mm/init.c                          | 71 +++++++++++++++++--
 3 files changed, 79 insertions(+), 12 deletions(-)

-- 
2.31.1

[PATCH -next v4 1/2] riscv: kdump: Implement crashkernel=X,[high,low]

[Chen Jiahao]

On riscv, the current crash kernel allocation logic is trying to
allocate within 32bit addressible memory region by default, if
failed, try to allocate without 4G restriction.

In need of saving DMA zone memory while allocating a relatively large
crash kernel region, allocating the reserved memory top down in
high memory, without overlapping the DMA zone, is a mature solution.
Here introduce the parameter option crashkernel=X,[high,low].

One can reserve the crash kernel from high memory above DMA zone range
by explicitly passing "crashkernel=X,high"; or reserve a memory range
below 4G with "crashkernel=X,low".

Signed-off-by: Chen Jiahao <chenjiahao16@huawei.com>
Acked-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Zhen Lei <thunder.leizhen@huawei.com>
---
 arch/riscv/kernel/setup.c |  5 +++
 arch/riscv/mm/init.c      | 71 ++++++++++++++++++++++++++++++++++++---
 2 files changed, 71 insertions(+), 5 deletions(-)

diff --git a/arch/riscv/kernel/setup.c b/arch/riscv/kernel/setup.c
index 5d3184cbf518..ea84e5047c23 100644
--- a/arch/riscv/kernel/setup.c
+++ b/arch/riscv/kernel/setup.c
@@ -176,6 +176,11 @@ static void __init init_resources(void)
 		if (ret < 0)
 			goto error;
 	}
+	if (crashk_low_res.start != crashk_low_res.end) {
+		ret = add_resource(&iomem_resource, &crashk_low_res);
+		if (ret < 0)
+			goto error;
+	}
 #endif
 
 #ifdef CONFIG_CRASH_DUMP
diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c
index 478d6763a01a..70f6cb281ed1 100644
--- a/arch/riscv/mm/init.c
+++ b/arch/riscv/mm/init.c
@@ -1152,6 +1152,28 @@ static inline void setup_vm_final(void)
 }
 #endif /* CONFIG_MMU */
 
+/* Reserve 128M low memory by default for swiotlb buffer */
+#define DEFAULT_CRASH_KERNEL_LOW_SIZE	(128UL << 20)
+
+static int __init reserve_crashkernel_low(unsigned long long low_size)
+{
+	unsigned long long low_base;
+
+	low_base = memblock_phys_alloc_range(low_size, PMD_SIZE, 0, dma32_phys_limit);
+	if (!low_base) {
+		pr_err("cannot allocate crashkernel low memory (size:0x%llx).\n", low_size);
+		return -ENOMEM;
+	}
+
+	pr_info("crashkernel low memory reserved: 0x%016llx - 0x%016llx (%lld MB)\n",
+		low_base, low_base + low_size, low_size >> 20);
+
+	crashk_low_res.start = low_base;
+	crashk_low_res.end = low_base + low_size - 1;
+
+	return 0;
+}
+
 /*
  * reserve_crashkernel() - reserves memory for crash kernel
  *
@@ -1163,8 +1185,12 @@ static void __init reserve_crashkernel(void)
 {
 	unsigned long long crash_base = 0;
 	unsigned long long crash_size = 0;
+	unsigned long long crash_low_size = 0;
 	unsigned long search_start = memblock_start_of_DRAM();
 	unsigned long search_end = memblock_end_of_DRAM();
+	unsigned long search_low_max = (unsigned long)dma32_phys_limit;
+	char *cmdline = boot_command_line;
+	bool fixed_base = false;
 
 	int ret = 0;
 
@@ -1180,14 +1206,34 @@ static void __init reserve_crashkernel(void)
 		return;
 	}
 
-	ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(),
+	ret = parse_crashkernel(cmdline, memblock_phys_mem_size(),
 				&crash_size, &crash_base);
-	if (ret || !crash_size)
+	if (ret == -ENOENT) {
+		/* Fallback to crashkernel=X,[high,low] */
+		ret = parse_crashkernel_high(cmdline, 0, &crash_size, &crash_base);
+		if (ret || !crash_size)
+			return;
+
+		/*
+		 * crashkernel=Y,low is valid only when crashkernel=X,high
+		 * is passed.
+		 */
+		ret = parse_crashkernel_low(cmdline, 0, &crash_low_size, &crash_base);
+		if (ret == -ENOENT)
+			crash_low_size = DEFAULT_CRASH_KERNEL_LOW_SIZE;
+		else if (ret)
+			return;
+
+		search_start = search_low_max;
+	} else if (ret || !crash_size) {
+		/* Invalid argument value specified */
 		return;
+	}
 
 	crash_size = PAGE_ALIGN(crash_size);
 
 	if (crash_base) {
+		fixed_base = true;
 		search_start = crash_base;
 		search_end = crash_base + crash_size;
 	}
@@ -1201,16 +1247,31 @@ static void __init reserve_crashkernel(void)
 	 */
 	crash_base = memblock_phys_alloc_range(crash_size, PMD_SIZE,
 					       search_start,
-					       min(search_end, (unsigned long) SZ_4G));
+					       min(search_end, search_low_max));
 	if (crash_base == 0) {
-		/* Try again without restricting region to 32bit addressible memory */
+		if (fixed_base) {
+			pr_warn("crashkernel: allocating failed with given size@offset\n");
+			return;
+		}
+
+		/* Try again above the region of 32bit addressible memory */
 		crash_base = memblock_phys_alloc_range(crash_size, PMD_SIZE,
-						search_start, search_end);
+						       max(search_start, search_low_max),
+						       search_end);
 		if (crash_base == 0) {
 			pr_warn("crashkernel: couldn't allocate %lldKB\n",
 				crash_size >> 10);
 			return;
 		}
+
+		if (!crash_low_size)
+			crash_low_size = DEFAULT_CRASH_KERNEL_LOW_SIZE;
+	}
+
+	if ((crash_base > dma32_phys_limit - crash_low_size) &&
+	    crash_low_size && reserve_crashkernel_low(crash_low_size)) {
+		memblock_phys_free(crash_base, crash_size);
+		return;
 	}
 
 	pr_info("crashkernel: reserved 0x%016llx - 0x%016llx (%lld MB)\n",
-- 
2.31.1

[PATCH -next v4 2/2] docs: kdump: Update the crashkernel description for riscv

[Chen Jiahao]

Now "crashkernel=" parameter on riscv has been updated to support
crashkernel=X,[high,low]. Through which we can reserve memory region
above/within 32bit addressible DMA zone.

Here update the parameter description accordingly.

Signed-off-by: Chen Jiahao <chenjiahao16@huawei.com>
Reviewed-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Zhen Lei <thunder.leizhen@huawei.com>
---
 Documentation/admin-guide/kernel-parameters.txt | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 54702bd488eb..41865aae9eaa 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -866,7 +866,7 @@
 			memory region [offset, offset + size] for that kernel
 			image. If '@offset' is omitted, then a suitable offset
 			is selected automatically.
-			[KNL, X86-64, ARM64] Select a region under 4G first, and
+			[KNL, X86-64, ARM64, RISCV] Select a region under 4G first, and
 			fall back to reserve region above 4G when '@offset'
 			hasn't been specified.
 			See Documentation/admin-guide/kdump/kdump.rst for further details.
@@ -879,14 +879,14 @@
 			Documentation/admin-guide/kdump/kdump.rst for an example.
 
 	crashkernel=size[KMG],high
-			[KNL, X86-64, ARM64] range could be above 4G. Allow kernel
-			to allocate physical memory region from top, so could
-			be above 4G if system have more than 4G ram installed.
-			Otherwise memory region will be allocated below 4G, if
-			available.
+			[KNL, X86-64, ARM64, RISCV] range could be above 4G.
+			Allow kernel to allocate physical memory region from top,
+			so could be above 4G if system have more than 4G ram
+			installed. Otherwise memory region will be allocated
+			below 4G, if available.
 			It will be ignored if crashkernel=X is specified.
 	crashkernel=size[KMG],low
-			[KNL, X86-64, ARM64] range under 4G. When crashkernel=X,high
+			[KNL, X86-64, ARM64, RISCV] range under 4G. When crashkernel=X,high
 			is passed, kernel could allocate physical memory region
 			above 4G, that cause second kernel crash on system
 			that require some amount of low memory, e.g. swiotlb
@@ -897,6 +897,7 @@
 			size is	platform dependent.
 			  --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB)
 			  --> arm64: 128MiB
+			  --> riscv: 128MiB
 			This one lets the user specify own low range under 4G
 			for second kernel instead.
 			0: to disable low allocation.
-- 
2.31.1

Re: [PATCH -next v4 1/2] riscv: kdump: Implement crashkernel=X,[high,low]

[Simon Horman]

On Mon, Apr 10, 2023 at 09:05:52PM +0800, Chen Jiahao wrote:
> On riscv, the current crash kernel allocation logic is trying to
> allocate within 32bit addressible memory region by default, if
> failed, try to allocate without 4G restriction.
> 
> In need of saving DMA zone memory while allocating a relatively large
> crash kernel region, allocating the reserved memory top down in
> high memory, without overlapping the DMA zone, is a mature solution.
> Here introduce the parameter option crashkernel=X,[high,low].
> 
> One can reserve the crash kernel from high memory above DMA zone range
> by explicitly passing "crashkernel=X,high"; or reserve a memory range
> below 4G with "crashkernel=X,low".
> 
> Signed-off-by: Chen Jiahao <chenjiahao16@huawei.com>
> Acked-by: Guo Ren <guoren@kernel.org>
> Reviewed-by: Zhen Lei <thunder.leizhen@huawei.com>

Reviewed-by: Simon Horman <horms@kernel.org>

Re: [PATCH -next v4 0/2] support allocating crashkernel above 4G explicitly on riscv

[chenjiahao (C)]

On 2023/4/10 21:05, Chen Jiahao wrote:
> On riscv, the current crash kernel allocation logic is trying to
> allocate within 32bit addressible memory region by default, if
> failed, try to allocate without 4G restriction.
>
> In need of saving DMA zone memory while allocating a relatively large
> crash kernel region, allocating the reserved memory top down in
> high memory, without overlapping the DMA zone, is a mature solution.
> Hence this patchset introduces the parameter option crashkernel=X,[high,low].
>
> One can reserve the crash kernel from high memory above DMA zone range
> by explicitly passing "crashkernel=X,high"; or reserve a memory range
> below 4G with "crashkernel=X,low". Besides, there are few rules need
> to take notice:
> 1. "crashkernel=X,[high,low]" will be ignored if "crashkernel=size"
>     is specified.
> 2. "crashkernel=X,low" is valid only when "crashkernel=X,high" is passed
>     and there is enough memory to be allocated under 4G.
> 3. When allocating crashkernel above 4G and no "crashkernel=X,low" is
>     specified, a 128M low memory will be allocated automatically for
>     swiotlb bounce buffer.
> See Documentation/admin-guide/kernel-parameters.txt for more information.
>
> To verify loading the crashkernel, adapted kexec-tools is attached below:
> https://github.com/chenjh005/kexec-tools/tree/build-test-riscv-v2
>
> Following test cases have been performed as expected:
> 1) crashkernel=256M                          //low=256M
> 2) crashkernel=1G                            //low=1G
> 3) crashkernel=4G                            //high=4G, low=128M(default)
> 4) crashkernel=4G crashkernel=256M,high      //high=4G, low=128M(default), high is ignored
> 5) crashkernel=4G crashkernel=256M,low       //high=4G, low=128M(default), low is ignored
> 6) crashkernel=4G,high                       //high=4G, low=128M(default)
> 7) crashkernel=256M,low                      //low=0M, invalid
> 8) crashkernel=4G,high crashkernel=256M,low  //high=4G, low=256M
> 9) crashkernel=4G,high crashkernel=4G,low    //high=0M, low=0M, invalid
> 10) crashkernel=512M@0xd0000000              //low=512M
> 11) crashkernel=1G@0xe0000000                //high=0M, low=0M, no enough low memory, failed
>
> Changes since [v4]:
> 1. Update some imprecise code comments for cmdline parsing.
>
> Changes since [v3]:
> 1. Update to print warning and return explicitly on failure when
>     crashkernel=size@offset is specified. Not changing the result
>     in this case but making the logic more straightforward.
> 2. Some minor cleanup.
>
> Changes since [v2]:
> 1. Update the allocation logic to ensure the high crashkernel
>     region is reserved strictly above dma32_phys_limit.
> 2. Clean up some minor format problems.
>
> Chen Jiahao (2):
>    riscv: kdump: Implement crashkernel=X,[high,low]
>    docs: kdump: Update the crashkernel description for riscv
>
>   .../admin-guide/kernel-parameters.txt         | 15 ++--
>   arch/riscv/kernel/setup.c                     |  5 ++
>   arch/riscv/mm/init.c                          | 71 +++++++++++++++++--
>   3 files changed, 79 insertions(+), 12 deletions(-)

Hi folks,

The current version has fix all problems up-to-date, please let me know
if there is any other bug, or somewhere not quiet correct. Thanks for
looking into this.

Best Regards,
Jiahao

Re: [PATCH -next v4 1/2] riscv: kdump: Implement crashkernel=X,[high,low]

[Baoquan He]

On 04/10/23 at 09:05pm, Chen Jiahao wrote:
> On riscv, the current crash kernel allocation logic is trying to
> allocate within 32bit addressible memory region by default, if
> failed, try to allocate without 4G restriction.
> 
> In need of saving DMA zone memory while allocating a relatively large
> crash kernel region, allocating the reserved memory top down in
> high memory, without overlapping the DMA zone, is a mature solution.
> Here introduce the parameter option crashkernel=X,[high,low].
> 
> One can reserve the crash kernel from high memory above DMA zone range
> by explicitly passing "crashkernel=X,high"; or reserve a memory range
> below 4G with "crashkernel=X,low".
> 
> Signed-off-by: Chen Jiahao <chenjiahao16@huawei.com>
> Acked-by: Guo Ren <guoren@kernel.org>
> Reviewed-by: Zhen Lei <thunder.leizhen@huawei.com>
> ---
>  arch/riscv/kernel/setup.c |  5 +++
>  arch/riscv/mm/init.c      | 71 ++++++++++++++++++++++++++++++++++++---
>  2 files changed, 71 insertions(+), 5 deletions(-)
> 
> diff --git a/arch/riscv/kernel/setup.c b/arch/riscv/kernel/setup.c
> index 5d3184cbf518..ea84e5047c23 100644
> --- a/arch/riscv/kernel/setup.c
> +++ b/arch/riscv/kernel/setup.c
> @@ -176,6 +176,11 @@ static void __init init_resources(void)
>  		if (ret < 0)
>  			goto error;
>  	}
> +	if (crashk_low_res.start != crashk_low_res.end) {
> +		ret = add_resource(&iomem_resource, &crashk_low_res);
> +		if (ret < 0)
> +			goto error;
> +	}
>  #endif
>  
>  #ifdef CONFIG_CRASH_DUMP
> diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c
> index 478d6763a01a..70f6cb281ed1 100644
> --- a/arch/riscv/mm/init.c
> +++ b/arch/riscv/mm/init.c
> @@ -1152,6 +1152,28 @@ static inline void setup_vm_final(void)
>  }
>  #endif /* CONFIG_MMU */
>  
> +/* Reserve 128M low memory by default for swiotlb buffer */
> +#define DEFAULT_CRASH_KERNEL_LOW_SIZE	(128UL << 20)
> +
> +static int __init reserve_crashkernel_low(unsigned long long low_size)
> +{
> +	unsigned long long low_base;
> +
> +	low_base = memblock_phys_alloc_range(low_size, PMD_SIZE, 0, dma32_phys_limit);
> +	if (!low_base) {
> +		pr_err("cannot allocate crashkernel low memory (size:0x%llx).\n", low_size);
> +		return -ENOMEM;
> +	}
> +
> +	pr_info("crashkernel low memory reserved: 0x%016llx - 0x%016llx (%lld MB)\n",
> +		low_base, low_base + low_size, low_size >> 20);
> +
> +	crashk_low_res.start = low_base;
> +	crashk_low_res.end = low_base + low_size - 1;
> +
> +	return 0;
> +}
> +
>  /*
>   * reserve_crashkernel() - reserves memory for crash kernel
>   *
> @@ -1163,8 +1185,12 @@ static void __init reserve_crashkernel(void)
>  {
>  	unsigned long long crash_base = 0;
>  	unsigned long long crash_size = 0;
> +	unsigned long long crash_low_size = 0;
>  	unsigned long search_start = memblock_start_of_DRAM();
>  	unsigned long search_end = memblock_end_of_DRAM();
> +	unsigned long search_low_max = (unsigned long)dma32_phys_limit;
> +	char *cmdline = boot_command_line;
> +	bool fixed_base = false;
>  
>  	int ret = 0;
>  
> @@ -1180,14 +1206,34 @@ static void __init reserve_crashkernel(void)
>  		return;
>  	}
>  
> -	ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(),
> +	ret = parse_crashkernel(cmdline, memblock_phys_mem_size(),
>  				&crash_size, &crash_base);
> -	if (ret || !crash_size)
> +	if (ret == -ENOENT) {
> +		/* Fallback to crashkernel=X,[high,low] */
> +		ret = parse_crashkernel_high(cmdline, 0, &crash_size, &crash_base);
> +		if (ret || !crash_size)
> +			return;
> +
> +		/*
> +		 * crashkernel=Y,low is valid only when crashkernel=X,high
> +		 * is passed.
> +		 */
> +		ret = parse_crashkernel_low(cmdline, 0, &crash_low_size, &crash_base);
> +		if (ret == -ENOENT)
> +			crash_low_size = DEFAULT_CRASH_KERNEL_LOW_SIZE;
> +		else if (ret)
> +			return;
> +
> +		search_start = search_low_max;
> +	} else if (ret || !crash_size) {
> +		/* Invalid argument value specified */
>  		return;
> +	}
>  
>  	crash_size = PAGE_ALIGN(crash_size);
>  
>  	if (crash_base) {
> +		fixed_base = true;
>  		search_start = crash_base;
>  		search_end = crash_base + crash_size;
>  	}
> @@ -1201,16 +1247,31 @@ static void __init reserve_crashkernel(void)
>  	 */
>  	crash_base = memblock_phys_alloc_range(crash_size, PMD_SIZE,
>  					       search_start,
> -					       min(search_end, (unsigned long) SZ_4G));
> +					       min(search_end, search_low_max));

Here, it seems not right in case crashkernel=,high is specified. In that
case, search_start == search_low_max, then the min(search_end,
search_low_max) will get search_low_max too. Then you make the fallback
in below code block to try to get crashkernel reservation above 4G. This
doesn't comply with the crashkernel=,high grammer which has been
implemented in other architectures.

For crashkernel=,high, user explicitly require memory reservation above
4G. Why does crashkernel=,high is needed? E.g on big end server with
huge memory, while the low memory under 4G is limited and precious.
Hence, user want to put the main crashkernel reservation above 4G to
contain kdump kernel/initrd and run user space program, while with few
low memory for pci device driver. E.g crashkernel=2G,high, it won't
impact much if there's huge memory above 4G and get crashkernel
reservation there. However, it impacts a lot if it reserves memory
below 4G.

I would strongly suggest that risc-v also reserve memory from above 4G
for crashkernel=,high, then fallback to below 4G. That's consistent with
crashkernel=,high grammer.

>  	if (crash_base == 0) {
> -		/* Try again without restricting region to 32bit addressible memory */
> +		if (fixed_base) {
> +			pr_warn("crashkernel: allocating failed with given size@offset\n");
> +			return;
> +		}
> +
> +		/* Try again above the region of 32bit addressible memory */
>  		crash_base = memblock_phys_alloc_range(crash_size, PMD_SIZE,
> -						search_start, search_end);
> +						       max(search_start, search_low_max),
> +						       search_end);
>  		if (crash_base == 0) {
>  			pr_warn("crashkernel: couldn't allocate %lldKB\n",
>  				crash_size >> 10);
>  			return;
>  		}
> +
> +		if (!crash_low_size)
> +			crash_low_size = DEFAULT_CRASH_KERNEL_LOW_SIZE;
> +	}
> +
> +	if ((crash_base > dma32_phys_limit - crash_low_size) &&
> +	    crash_low_size && reserve_crashkernel_low(crash_low_size)) {
> +		memblock_phys_free(crash_base, crash_size);
> +		return;
>  	}
>  
>  	pr_info("crashkernel: reserved 0x%016llx - 0x%016llx (%lld MB)\n",
> -- 
> 2.31.1
> 

Re: [PATCH -next v4 1/2] riscv: kdump: Implement crashkernel=X,[high,low]

[chenjiahao (C)]

On 2023/4/27 10:13, Baoquan He wrote:
> On 04/10/23 at 09:05pm, Chen Jiahao wrote:
>> On riscv, the current crash kernel allocation logic is trying to
>> allocate within 32bit addressible memory region by default, if
>> failed, try to allocate without 4G restriction.
>>
>> In need of saving DMA zone memory while allocating a relatively large
>> crash kernel region, allocating the reserved memory top down in
>> high memory, without overlapping the DMA zone, is a mature solution.
>> Here introduce the parameter option crashkernel=X,[high,low].
>>
>> One can reserve the crash kernel from high memory above DMA zone range
>> by explicitly passing "crashkernel=X,high"; or reserve a memory range
>> below 4G with "crashkernel=X,low".
>>
>> Signed-off-by: Chen Jiahao <chenjiahao16@huawei.com>
>> Acked-by: Guo Ren <guoren@kernel.org>
>> Reviewed-by: Zhen Lei <thunder.leizhen@huawei.com>
>> ---
>>   arch/riscv/kernel/setup.c |  5 +++
>>   arch/riscv/mm/init.c      | 71 ++++++++++++++++++++++++++++++++++++---
>>   2 files changed, 71 insertions(+), 5 deletions(-)
>>
>> diff --git a/arch/riscv/kernel/setup.c b/arch/riscv/kernel/setup.c
>> index 5d3184cbf518..ea84e5047c23 100644
>> --- a/arch/riscv/kernel/setup.c
>> +++ b/arch/riscv/kernel/setup.c
>> @@ -176,6 +176,11 @@ static void __init init_resources(void)
>>   		if (ret < 0)
>>   			goto error;
>>   	}
>> +	if (crashk_low_res.start != crashk_low_res.end) {
>> +		ret = add_resource(&iomem_resource, &crashk_low_res);
>> +		if (ret < 0)
>> +			goto error;
>> +	}
>>   #endif
>>   
>>   #ifdef CONFIG_CRASH_DUMP
>> diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c
>> index 478d6763a01a..70f6cb281ed1 100644
>> --- a/arch/riscv/mm/init.c
>> +++ b/arch/riscv/mm/init.c
>> @@ -1152,6 +1152,28 @@ static inline void setup_vm_final(void)
>>   }
>>   #endif /* CONFIG_MMU */
>>   
>> +/* Reserve 128M low memory by default for swiotlb buffer */
>> +#define DEFAULT_CRASH_KERNEL_LOW_SIZE	(128UL << 20)
>> +
>> +static int __init reserve_crashkernel_low(unsigned long long low_size)
>> +{
>> +	unsigned long long low_base;
>> +
>> +	low_base = memblock_phys_alloc_range(low_size, PMD_SIZE, 0, dma32_phys_limit);
>> +	if (!low_base) {
>> +		pr_err("cannot allocate crashkernel low memory (size:0x%llx).\n", low_size);
>> +		return -ENOMEM;
>> +	}
>> +
>> +	pr_info("crashkernel low memory reserved: 0x%016llx - 0x%016llx (%lld MB)\n",
>> +		low_base, low_base + low_size, low_size >> 20);
>> +
>> +	crashk_low_res.start = low_base;
>> +	crashk_low_res.end = low_base + low_size - 1;
>> +
>> +	return 0;
>> +}
>> +
>>   /*
>>    * reserve_crashkernel() - reserves memory for crash kernel
>>    *
>> @@ -1163,8 +1185,12 @@ static void __init reserve_crashkernel(void)
>>   {
>>   	unsigned long long crash_base = 0;
>>   	unsigned long long crash_size = 0;
>> +	unsigned long long crash_low_size = 0;
>>   	unsigned long search_start = memblock_start_of_DRAM();
>>   	unsigned long search_end = memblock_end_of_DRAM();
>> +	unsigned long search_low_max = (unsigned long)dma32_phys_limit;
>> +	char *cmdline = boot_command_line;
>> +	bool fixed_base = false;
>>   
>>   	int ret = 0;
>>   
>> @@ -1180,14 +1206,34 @@ static void __init reserve_crashkernel(void)
>>   		return;
>>   	}
>>   
>> -	ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(),
>> +	ret = parse_crashkernel(cmdline, memblock_phys_mem_size(),
>>   				&crash_size, &crash_base);
>> -	if (ret || !crash_size)
>> +	if (ret == -ENOENT) {
>> +		/* Fallback to crashkernel=X,[high,low] */
>> +		ret = parse_crashkernel_high(cmdline, 0, &crash_size, &crash_base);
>> +		if (ret || !crash_size)
>> +			return;
>> +
>> +		/*
>> +		 * crashkernel=Y,low is valid only when crashkernel=X,high
>> +		 * is passed.
>> +		 */
>> +		ret = parse_crashkernel_low(cmdline, 0, &crash_low_size, &crash_base);
>> +		if (ret == -ENOENT)
>> +			crash_low_size = DEFAULT_CRASH_KERNEL_LOW_SIZE;
>> +		else if (ret)
>> +			return;
>> +
>> +		search_start = search_low_max;
>> +	} else if (ret || !crash_size) {
>> +		/* Invalid argument value specified */
>>   		return;
>> +	}
>>   
>>   	crash_size = PAGE_ALIGN(crash_size);
>>   
>>   	if (crash_base) {
>> +		fixed_base = true;
>>   		search_start = crash_base;
>>   		search_end = crash_base + crash_size;
>>   	}
>> @@ -1201,16 +1247,31 @@ static void __init reserve_crashkernel(void)
>>   	 */
>>   	crash_base = memblock_phys_alloc_range(crash_size, PMD_SIZE,
>>   					       search_start,
>> -					       min(search_end, (unsigned long) SZ_4G));
>> +					       min(search_end, search_low_max));
> Here, it seems not right in case crashkernel=,high is specified. In that
> case, search_start == search_low_max, then the min(search_end,
> search_low_max) will get search_low_max too. Then you make the fallback
> in below code block to try to get crashkernel reservation above 4G. This
> doesn't comply with the crashkernel=,high grammer which has been
> implemented in other architectures.
>
> For crashkernel=,high, user explicitly require memory reservation above
> 4G. Why does crashkernel=,high is needed? E.g on big end server with
> huge memory, while the low memory under 4G is limited and precious.
> Hence, user want to put the main crashkernel reservation above 4G to
> contain kdump kernel/initrd and run user space program, while with few
> low memory for pci device driver. E.g crashkernel=2G,high, it won't
> impact much if there's huge memory above 4G and get crashkernel
> reservation there. However, it impacts a lot if it reserves memory
> below 4G.
>
> I would strongly suggest that risc-v also reserve memory from above 4G
> for crashkernel=,high, then fallback to below 4G. That's consistent with
> crashkernel=,high grammer.

Sorry for late response.

I have got the point here. So with the original implication of "crashkernel=,high",
there is even no need to try reserving low memory under 4G. I have arranged another
version of patchset, in which I updated the allocation logic in that case.

For example, when "crashkernel=1G,high" is specified, the previous logic is like:
alloc range: crash_size: 0x40000000 (1G), crash_base: 4G_limit,
              crash_max: 4G_limit
alloc range high: crash_size: 0x40000000 (1G), crash_base: 4G_limit,
                   crash_max: memblock_range_end
alloc range low: low_size: 0x8000000 (128MB,default), crash_base: 0x0,
                  crash_max: 4G_limit

After revision, the logic is like:
alloc range: crash_size: 0x40000000 (1G), crash_base: memblock_range_start,
              crash_max: memblock_range_end
alloc range low: low_size: 0x8000000 (128MB,default), crash_base: 0x0,
                  crash_max: 4G_limit

Please let me know if there is any problem exist.

Thanks,
Jiahao


>
>>   	if (crash_base == 0) {
>> -		/* Try again without restricting region to 32bit addressible memory */
>> +		if (fixed_base) {
>> +			pr_warn("crashkernel: allocating failed with given size@offset\n");
>> +			return;
>> +		}
>> +
>> +		/* Try again above the region of 32bit addressible memory */
>>   		crash_base = memblock_phys_alloc_range(crash_size, PMD_SIZE,
>> -						search_start, search_end);
>> +						       max(search_start, search_low_max),
>> +						       search_end);
>>   		if (crash_base == 0) {
>>   			pr_warn("crashkernel: couldn't allocate %lldKB\n",
>>   				crash_size >> 10);
>>   			return;
>>   		}
>> +
>> +		if (!crash_low_size)
>> +			crash_low_size = DEFAULT_CRASH_KERNEL_LOW_SIZE;
>> +	}
>> +
>> +	if ((crash_base > dma32_phys_limit - crash_low_size) &&
>> +	    crash_low_size && reserve_crashkernel_low(crash_low_size)) {
>> +		memblock_phys_free(crash_base, crash_size);
>> +		return;
>>   	}
>>   
>>   	pr_info("crashkernel: reserved 0x%016llx - 0x%016llx (%lld MB)\n",
>> -- 
>> 2.31.1
>>

Re: [PATCH -next v4 1/2] riscv: kdump: Implement crashkernel=X,[high,low]

[Baoquan He]

On 05/11/23 at 04:47pm, chenjiahao (C) wrote:
......
> > > @@ -1163,8 +1185,12 @@ static void __init reserve_crashkernel(void)
> > >   {
> > >   	unsigned long long crash_base = 0;
> > >   	unsigned long long crash_size = 0;
> > > +	unsigned long long crash_low_size = 0;
> > >   	unsigned long search_start = memblock_start_of_DRAM();
> > >   	unsigned long search_end = memblock_end_of_DRAM();
> > > +	unsigned long search_low_max = (unsigned long)dma32_phys_limit;
> > > +	char *cmdline = boot_command_line;
> > > +	bool fixed_base = false;
> > >   	int ret = 0;
> > > @@ -1180,14 +1206,34 @@ static void __init reserve_crashkernel(void)
> > >   		return;
> > >   	}
> > > -	ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(),
> > > +	ret = parse_crashkernel(cmdline, memblock_phys_mem_size(),
> > >   				&crash_size, &crash_base);
> > > -	if (ret || !crash_size)
> > > +	if (ret == -ENOENT) {
> > > +		/* Fallback to crashkernel=X,[high,low] */
> > > +		ret = parse_crashkernel_high(cmdline, 0, &crash_size, &crash_base);
> > > +		if (ret || !crash_size)
> > > +			return;
> > > +
> > > +		/*
> > > +		 * crashkernel=Y,low is valid only when crashkernel=X,high
> > > +		 * is passed.
> > > +		 */
> > > +		ret = parse_crashkernel_low(cmdline, 0, &crash_low_size, &crash_base);
> > > +		if (ret == -ENOENT)
> > > +			crash_low_size = DEFAULT_CRASH_KERNEL_LOW_SIZE;
> > > +		else if (ret)
> > > +			return;
> > > +
> > > +		search_start = search_low_max;
> > > +	} else if (ret || !crash_size) {
> > > +		/* Invalid argument value specified */
> > >   		return;
> > > +	}
> > >   	crash_size = PAGE_ALIGN(crash_size);
> > >   	if (crash_base) {
> > > +		fixed_base = true;
> > >   		search_start = crash_base;
> > >   		search_end = crash_base + crash_size;
> > >   	}
> > > @@ -1201,16 +1247,31 @@ static void __init reserve_crashkernel(void)
> > >   	 */
> > >   	crash_base = memblock_phys_alloc_range(crash_size, PMD_SIZE,
> > >   					       search_start,
> > > -					       min(search_end, (unsigned long) SZ_4G));
> > > +					       min(search_end, search_low_max));
> > Here, it seems not right in case crashkernel=,high is specified. In that
> > case, search_start == search_low_max, then the min(search_end,
> > search_low_max) will get search_low_max too. Then you make the fallback
> > in below code block to try to get crashkernel reservation above 4G. This
> > doesn't comply with the crashkernel=,high grammer which has been
> > implemented in other architectures.
> > 
> > For crashkernel=,high, user explicitly require memory reservation above
> > 4G. Why does crashkernel=,high is needed? E.g on big end server with
> > huge memory, while the low memory under 4G is limited and precious.
> > Hence, user want to put the main crashkernel reservation above 4G to
> > contain kdump kernel/initrd and run user space program, while with few
> > low memory for pci device driver. E.g crashkernel=2G,high, it won't
> > impact much if there's huge memory above 4G and get crashkernel
> > reservation there. However, it impacts a lot if it reserves memory
> > below 4G.
> > 
> > I would strongly suggest that risc-v also reserve memory from above 4G
> > for crashkernel=,high, then fallback to below 4G. That's consistent with
> > crashkernel=,high grammer.
> 
> Sorry for late response.
> 
> I have got the point here. So with the original implication of "crashkernel=,high",
> there is even no need to try reserving low memory under 4G. I have arranged another
> version of patchset, in which I updated the allocation logic in that case.
> 
> For example, when "crashkernel=1G,high" is specified, the previous logic is like:
> alloc range: crash_size: 0x40000000 (1G), crash_base: 4G_limit,
>              crash_max: 4G_limit
> alloc range high: crash_size: 0x40000000 (1G), crash_base: 4G_limit,
>                   crash_max: memblock_range_end
> alloc range low: low_size: 0x8000000 (128MB,default), crash_base: 0x0,
>                  crash_max: 4G_limit
> 
> After revision, the logic is like:
> alloc range: crash_size: 0x40000000 (1G), crash_base: memblock_range_start,
>              crash_max: memblock_range_end
> alloc range low: low_size: 0x8000000 (128MB,default), crash_base: 0x0,
>                  crash_max: 4G_limit
> 
> Please let me know if there is any problem exist.

Sorry for late reply.

Hmm, it doesn't seem completely correct. I will comment in your v5
patch. Please see over there.