support reserving crashkernel above 4G on arm64 kdump
Subsystems (8)
v3 cover support reserving crashkernel above 4G on arm64 kdump
When crashkernel is reserved above 4G in memory, kernel should reserve some amount of low memory for swiotlb and some DMA buffers. So there may be two crash kernel regions, one is below 4G, the other is above 4G.
Crash dump kernel reads more than one crash kernel regions via a dtb property under node /chosen,
linux,usable-memory-range = <BASE1 SIZE1 [BASE2 SIZE2]>.
Besides, we need to modify kexec-tools:
arm64: support more than one crash kernel regions(see [1])
Changes since [v2]
- Split patch "arm64: kdump: support reserving crashkernel above 4G" as two. Put "move reserve_crashkernel_low() into kexec_core.c" in a seperate patch.
Changes since [v1]:
- Move common reserve_crashkernel_low() code into kernel/kexec_core.c.
- Remove memblock_cap_memory_ranges() i added in v1 and implement that in fdt_enforce_memory_region().
There are at most two crash kernel regions, for two crash kernel regions case, we cap the memory range [min(regs[*].start), max(regs[*].end)] and then remove the memory range in the middle.
[1]: http://lists.infradead.org/pipermail/kexec/2019-April/022792.html
[v1]: https://lkml.org/lkml/2019/4/8/628
[v2]: https://lkml.org/lkml/2019/4/9/86
Chen Zhou (4):
x86: kdump: move reserve_crashkernel_low() into kexec_core.c
arm64: kdump: support reserving crashkernel above 4G
arm64: kdump: support more than one crash kernel regions
kdump: update Documentation about crashkernel on arm64
Documentation/admin-guide/kernel-parameters.txt | 4 +- arch/arm64/include/asm/kexec.h | 3 + arch/arm64/kernel/setup.c | 3 + arch/arm64/mm/init.c | 92 +++++++++++++++++++++---- arch/x86/include/asm/kexec.h | 3 + arch/x86/kernel/setup.c | 66 ++---------------- include/linux/kexec.h | 1 + include/linux/memblock.h | 6 ++ kernel/kexec_core.c | 53 ++++++++++++++ mm/memblock.c | 7 +- 10 files changed, 159 insertions(+), 79 deletions(-)
Signature · 7 lines
2.7.4 _______________________________________________ kexec mailing list [email protected] http://lists.infradead.org/mailman/listinfo/kexec
PATCH v3 1/4 x86: kdump: move reserve_crashkernel_low() into kexec_core.c
In preparation for supporting more than one crash kernel regions in arm64 as x86_64 does, move reserve_crashkernel_low() into kexec/kexec_core.c.
--- arch/x86/include/asm/kexec.h | 3 ++ arch/x86/kernel/setup.c | 66 +++++--------------------------------------- include/linux/kexec.h | 1 + kernel/kexec_core.c | 53 +++++++++++++++++++++++++++++++++++ 4 files changed, 64 insertions(+), 59 deletions(-)
arch/x86/include/asm/kexec.h +3 -0
@@ -18,6 +18,9 @@# define KEXEC_CONTROL_CODE_MAX_SIZE 2048 +/* 16M alignment for crash kernel regions */ +#define CRASH_ALIGN (16 << 20) +#ifndef __ASSEMBLY__#include <linux/string.h>
arch/x86/kernel/setup.c +7 -59
84 lines not on this page · raw message
include/linux/kexec.h +1 -0
@@ -281,6 +281,7 @@ extern void __crash_kexec(struct pt_regs *);extern void crash_kexec (struct pt_regs *);int kexec_should_crash (struct task_struct *);int kexec_crash_loaded (void ); +int __init reserve_crashkernel_low (void );void crash_save_cpu (struct pt_regs *regs,int cpu);extern int kimage_crash_copy_vmcoreinfo (struct kimage *image);
kernel/kexec_core.c +53 -0
65 lines not on this page · raw message
Signature · 1 line
2.7.4
Re: [PATCH v3 1/4] x86: kdump: move reserve_crashkernel_low() into kexec_core.c
Quoted 11 lines · Chen Zhou
In preparation for supporting more than one crash kernel regions in arm64 as
x86_64does, movereserve_crashkernel_low()intokexec/kexec_core.c.Signed-off-by Chen Zhou <[email protected]>--- arch/x86/include/asm/kexec.h | 3 ++ arch/x86/kernel/setup.c | 66 +++++--------------------------------------- include/linux/kexec.h | 1 + kernel/kexec_core.c | 53 +++++++++++++++++++++++++++++++++++ 4 files changed, 64 insertions(+), 59 deletions(-)
No objections for this to be merged via the ARM tree, as long as x86 functionality is kept intact.
Thanks,
Ingo
_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec
Re: [PATCH v3 1/4] x86: kdump: move reserve_crashkernel_low() into kexec_core.c
Hi Ingo,
Quoted 15 lines · Ingo Molnar
Quoted 11 lines · Chen Zhou
In preparation for supporting more than one crash kernel regions in arm64 as
x86_64does, movereserve_crashkernel_low()intokexec/kexec_core.c.Signed-off-by Chen Zhou <[email protected]>--- arch/x86/include/asm/kexec.h | 3 ++ arch/x86/kernel/setup.c | 66 +++++--------------------------------------- include/linux/kexec.h | 1 + kernel/kexec_core.c | 53 +++++++++++++++++++++++++++++++++++ 4 files changed, 64 insertions(+), 59 deletions(-)
Quoted 2 lines
No objections for this to be merged via the ARM tree, as long as x86 functionality is kept intact.
This patch has no affect on x86.
Thanks,
Chen Zhou
Quoted 11 lines
Thanks,
Ingo
_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec
Re: [PATCH v3 1/4] x86: kdump: move reserve_crashkernel_low() into kexec_core.c
Quoted 18 lines · Chen Zhou
Hi Ingo,
Quoted 15 lines · Ingo Molnar
Quoted 11 lines · Chen Zhou
In preparation for supporting more than one crash kernel regions in arm64 as
x86_64does, movereserve_crashkernel_low()intokexec/kexec_core.c.Signed-off-by Chen Zhou <[email protected]>--- arch/x86/include/asm/kexec.h | 3 ++ arch/x86/kernel/setup.c | 66 +++++--------------------------------------- include/linux/kexec.h | 1 + kernel/kexec_core.c | 53 +++++++++++++++++++++++++++++++++++ 4 files changed, 64 insertions(+), 59 deletions(-)
Quoted 4 lines
Quoted 2 lines
No objections for this to be merged via the ARM tree, as long as x86 functionality is kept intact.
This patch has no affect on x86.
In *principle*.
In practice the series does change x86 code:
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
arch/x86/kernel/setup.c | 66 +++++--------------------------------------- include/linux/kexec.h | 1 + kernel/kexec_core.c | 53 +++++++++++++++++++++++++++++++++++
which is, *hopefully*, an identity transformation. :-)
I.e. Ack, but only if it doesn't break anything. :-)
Thanks,
Ingo
_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec
PATCH v3 2/4 arm64: kdump: support reserving crashkernel above 4G
In reply to Chen ZhouWhen crashkernel is reserved above 4G in memory, kernel should reserve some amount of low memory for swiotlb and some DMA buffers.
Kernel would try to allocate at least 256M below 4G automatically as x86_64 if crashkernel is above 4G. Meanwhile, support crashkernel=X,[high,low] in arm64.
--- arch/arm64/include/asm/kexec.h | 3 +++ arch/arm64/kernel/setup.c | 3 +++ arch/arm64/mm/init.c | 26 +++++++++++++++++++++----- 3 files changed, 27 insertions(+), 5 deletions(-)
arch/arm64/include/asm/kexec.h +3 -0
@@ -28,6 +28,9 @@#define KEXEC_ARCH KEXEC_ARCH_AARCH64 +/* 2M alignment for crash kernel regions */ +#define CRASH_ALIGN SZ_2M +#ifndef __ASSEMBLY__/**
arch/arm64/kernel/setup.c +3 -0
@@ -243,6 +243,9 @@ static void __init request_standard_resources(void)request_resource (res, &kernel_data);#ifdef CONFIG_KEXEC_CORE /* Userspace will find "Crash kernel" region in /proc/iomem. */ +if (crashk_low_res.end && crashk_low_res.start >= res->start && + crashk_low_res.end <= res->end) +request_resource (res, &crashk_low_res);if (crashk_res.end && crashk_res.start >= res->start && crashk_res.end <= res->end)request_resource (res, &crashk_res);
arch/arm64/mm/init.c +21 -5
@@ -64,6 +64,7 @@ EXPORT_SYMBOL(memstart_addr);phys_addr_t arm64_dma_phys_limit__ro_after_init ;#ifdef CONFIG_KEXEC_CORE +/* * reserve_crashkernel() - reserves memory for crash kernel *
@@ -74,20 +75,30 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init;static void __init reserve_crashkernel (void ) {unsigned long long crash_base, crash_size; +bool high = false;int ret; ret =parse_crashkernel (boot_command_line,memblock_phys_mem_size (), &crash_size, &crash_base);/* no crashkernel= or invalid value specified */ -if (ret || !crash_size) -return ; +if (ret || !crash_size) { +/* crashkernel=X,high */ + ret =parse_crashkernel_high (boot_command_line, +memblock_phys_mem_size (), + &crash_size, &crash_base); +if (ret || !crash_size) +return ; + high = true; + } crash_size =PAGE_ALIGN (crash_size);if (crash_base ==0 ) {/* Current arm64 boot protocol requires 2MB alignment */ - crash_base =memblock_find_in_range (0 ,ARCH_LOW_ADDRESS_LIMIT , - crash_size,SZ_2M ); + crash_base =memblock_find_in_range (0 , + high ?memblock_end_of_DRAM () + :ARCH_LOW_ADDRESS_LIMIT , + crash_size,CRASH_ALIGN );if (crash_base ==0 ) {pr_warn ("cannot allocate crashkernel (size:0x%llx)\n" , crash_size);
@@ -105,13 +116,18 @@ static void __init reserve_crashkernel(void)return ; } -if (!IS_ALIGNED (crash_base,SZ_2M )) { +if (!IS_ALIGNED (crash_base,CRASH_ALIGN )) {pr_warn ("cannot reserve crashkernel: base address is not 2MB aligned\n" );return ; } }memblock_reserve (crash_base, crash_size); +if (crash_base >=SZ_4G &&reserve_crashkernel_low ()) { +memblock_free (crash_base, crash_size); +return ; + } +pr_info ("crashkernel reserved: 0x%016llx - 0x%016llx (%lld MB)\n" , crash_base, crash_base + crash_size, crash_size >>20 );
Signature · 1 line
2.7.4
PATCH v3 3/4 arm64: kdump: support more than one crash kernel regions
In reply to Chen ZhouAfter commit (arm64: kdump: support reserving crashkernel above 4G), there may be two crash kernel regions, one is below 4G, the other is above 4G.
Crash dump kernel reads more than one crash kernel regions via a dtb property under node /chosen,
linux,usable-memory-range = <BASE1 SIZE1 [BASE2 SIZE2]>
--- arch/arm64/mm/init.c | 66 ++++++++++++++++++++++++++++++++++++++++-------- include/linux/memblock.h | 6 +++++ mm/memblock.c | 7 ++--- 3 files changed, 66 insertions(+), 13 deletions(-)
arch/arm64/mm/init.c +56 -10
@@ -65,6 +65,11 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init;#ifdef CONFIG_KEXEC_CORE +/* at most two crash kernel regions, low_region and high_region */ +#define CRASH_MAX_USABLE_RANGES 2 +#define LOW_REGION_IDX 0 +#define HIGH_REGION_IDX 1 +/* * reserve_crashkernel() - reserves memory for crash kernel *
@@ -297,8 +302,8 @@ static int __init early_init_dt_scan_usablemem(unsigned long node, const char *uname, int depth, void *data) { struct memblock_region *usablemem = data; -const __be32 *reg; -int len; +const __be32 *reg, *endp; +int len, nr =0 ;if (depth !=1 ||strcmp (uname,"chosen" ) !=0 )return 0 ;
@@ -307,22 +312,63 @@ static int __init early_init_dt_scan_usablemem(unsigned long node,if (!reg || (len < (dt_root_addr_cells + dt_root_size_cells)))return 1 ; - usablemem->base =dt_mem_next_cell (dt_root_addr_cells, ®); - usablemem->size =dt_mem_next_cell (dt_root_size_cells, ®); + endp = reg + (len /sizeof (__be32)); +while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) { + usablemem[nr].base =dt_mem_next_cell (dt_root_addr_cells, ®); + usablemem[nr].size =dt_mem_next_cell (dt_root_size_cells, ®); + +if (++nr >=CRASH_MAX_USABLE_RANGES ) +break ; + }return 1 ; }static void __init fdt_enforce_memory_region (void ) { -struct memblock_region reg = { - .size =0 , - }; +int i, cnt =0 ; +struct memblock_region regs[CRASH_MAX_USABLE_RANGES ]; + +memset (regs,0 ,sizeof (regs)); +of_scan_flat_dt (early_init_dt_scan_usablemem, regs); + +for (i =0 ; i <CRASH_MAX_USABLE_RANGES ; i++) +if (regs[i].size) + cnt++; +else +break ; + +if (cnt -1 ==LOW_REGION_IDX ) +memblock_cap_memory_range (regs[LOW_REGION_IDX ].base, + regs[LOW_REGION_IDX ].size); +else if (cnt -1 ==HIGH_REGION_IDX ) { +/* +* Two crash kernel regions, cap the memory range +* [regs[LOW_REGION_IDX].base, regs[HIGH_REGION_IDX].end] +* and then remove the memory range in the middle. +*/ +int start_rgn, end_rgn, i, ret; +phys_addr_t mid_base, mid_size; + + mid_base = regs[LOW_REGION_IDX ].base + regs[LOW_REGION_IDX ].size; + mid_size = regs[HIGH_REGION_IDX ].base - mid_base; + ret =memblock_isolate_range (&memblock.memory, mid_base, + mid_size, &start_rgn, &end_rgn); -of_scan_flat_dt (early_init_dt_scan_usablemem, ®); +if (ret) +return ; -if (reg.size) -memblock_cap_memory_range (reg.base, reg.size); +memblock_cap_memory_range (regs[LOW_REGION_IDX ].base, + regs[HIGH_REGION_IDX ].base - + regs[LOW_REGION_IDX ].base + + regs[HIGH_REGION_IDX ].size); +for (i = end_rgn -1 ; i >= start_rgn; i--) { +if (!memblock_is_nomap (&memblock.memory.regions[i])) +memblock_remove_region (&memblock.memory, i); + } +memblock_remove_range (&memblock.reserved, mid_base, + mid_base + mid_size); + } }void __init arm64_memblock_init (void )
include/linux/memblock.h +6 -0
@@ -110,9 +110,15 @@ void memblock_discard(void);phys_addr_t memblock_find_in_range (phys_addr_t start,phys_addr_t end,phys_addr_t size,phys_addr_t align); +void memblock_remove_region (struct memblock_type *type,unsigned long r);void memblock_allow_resize (void );int memblock_add_node (phys_addr_t base,phys_addr_t size,int nid);int memblock_add (phys_addr_t base,phys_addr_t size); +int memblock_isolate_range (struct memblock_type *type, +phys_addr_t base,phys_addr_t size, +int *start_rgn,int *end_rgn); +int memblock_remove_range (struct memblock_type *type, +phys_addr_t base,phys_addr_t size);int memblock_remove (phys_addr_t base,phys_addr_t size);int memblock_free (phys_addr_t base,phys_addr_t size);int memblock_reserve (phys_addr_t base,phys_addr_t size);
mm/memblock.c +4 -3
@@ -357,7 +357,8 @@ phys_addr_t __init_memblock memblock_find_in_range(phys_addr_t start,return ret; } -static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r) +void __init_memblock memblock_remove_region(struct memblock_type *type, +unsigned long r) { type->total_size -= type->regions[r].size; memmove(&type->regions[r], &type->regions[r + 1],
@@ -724,7 +725,7 @@ int __init_memblock memblock_add(phys_addr_t base, phys_addr_t size)* Return: * 0 on success, -errno on failure. */ -static int __init_memblock memblock_isolate_range(struct memblock_type *type, +int __init_memblock memblock_isolate_range(struct memblock_type *type,phys_addr_t base,phys_addr_t size,int *start_rgn,int *end_rgn) {
@@ -784,7 +785,7 @@ static int __init_memblock memblock_isolate_range(struct memblock_type *type,return 0 ; } -static int __init_memblock memblock_remove_range(struct memblock_type *type, +int __init_memblock memblock_remove_range(struct memblock_type *type,phys_addr_t base,phys_addr_t size) {int start_rgn, end_rgn;
Signature · 7 lines
2.7.4 _______________________________________________ kexec mailing list [email protected] http://lists.infradead.org/mailman/listinfo/kexec
Re: [PATCH v3 3/4] arm64: kdump: support more than one crash kernel regions
Hi,
Quoted 57 lines · Chen Zhou
After commit (arm64: kdump: support reserving crashkernel above 4G), there may be two crash kernel regions, one is below 4G, the other is above 4G.
Crash dump kernel reads more than one crash kernel regions via a dtb property under node /chosen,
linux,usable-memory-range = <BASE1 SIZE1 [BASE2 SIZE2]>Signed-off-by Chen Zhou <[email protected]>--- arch/arm64/mm/init.c | 66 ++++++++++++++++++++++++++++++++++++++++-------- include/linux/memblock.h | 6 +++++ mm/memblock.c | 7 ++--- 3 files changed, 66 insertions(+), 13 deletions(-)arch/arm64/mm/init.c +15
-4@@ -65,6 +65,11 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init;#ifdef CONFIG_KEXEC_CORE +/* at most two crash kernel regions, low_region and high_region */ +#define CRASH_MAX_USABLE_RANGES 2 +#define LOW_REGION_IDX 0 +#define HIGH_REGION_IDX 1 +/* * reserve_crashkernel() - reserves memory for crash kernel * @@ -297,8 +302,8 @@ static int __init early_init_dt_scan_usablemem(unsigned long node, const char *uname, int depth, void *data) { struct memblock_region *usablemem = data; -const __be32 *reg; -int len; +const __be32 *reg, *endp; +int len, nr =0 ;if (depth !=1 ||strcmp (uname,"chosen" ) !=0 )return 0 ;@@ -307,22 +312,63 @@ static int __init early_init_dt_scan_usablemem(unsigned long node,if (!reg || (len < (dt_root_addr_cells + dt_root_size_cells)))return 1 ; - usablemem->base =dt_mem_next_cell (dt_root_addr_cells, ®); - usablemem->size =dt_mem_next_cell (dt_root_size_cells, ®); + endp = reg + (len /sizeof (__be32)); +while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) { + usablemem[nr].base =dt_mem_next_cell (dt_root_addr_cells, ®); + usablemem[nr].size =dt_mem_next_cell (dt_root_size_cells, ®); + +if (++nr >=CRASH_MAX_USABLE_RANGES ) +break ; + }
arch/arm64/mm/init.c static int __init early_init_dt_scan_usablemem(unsigned long node, · line 330
Quoted 10 lines
arch/arm64/mm/init.c +2
-3return 1 ; }static void __init fdt_enforce_memory_region (void ) { -struct memblock_region reg = { - .size =0 , - }; +int i, cnt =0 ; +struct memblock_region regs[CRASH_MAX_USABLE_RANGES ];
I only now noticed that fdt_enforce_memory_region() uses memblock_region to pass the ranges around. If we'd switch to memblock_type instead, the implementation of memblock_cap_memory_ranges() would be really straightforward. Can you check if the below patch works for you?
From e476d584098e31273af573e1a78e308880c5cf28 Mon Sep 17 00:00:00 2001
From: Mike Rapoport <[email protected]>
Date: Wed, 10 Apr 2019 16:02:32 +0300
Subject: [PATCH] memblock: extend memblock_cap_memory_range to multiple ranges
The memblock_cap_memory_range() removes all the memory except the range passed to it. Extend this function to recieve memblock_type with the regions that should be kept. This allows switching to simple iteration over memblock arrays with 'for_each_mem_range' to remove the unneeded memory.
Enable use of this function in arm64 for reservation of multile regions for the crash kernel.
--- arch/arm64/mm/init.c | 34 ++++++++++++++++++++++++---------- include/linux/memblock.h | 2 +- mm/memblock.c | 45 ++++++++++++++++++++++----------------------- 3 files changed, 47 insertions(+), 34 deletions(-)
arch/arm64/mm/init.c +24 -10
@@ -64,6 +64,10 @@ EXPORT_SYMBOL(memstart_addr);phys_addr_t arm64_dma_phys_limit__ro_after_init ;#ifdef CONFIG_KEXEC_CORE + +/* at most two crash kernel regions, low_region and high_region */ +#define CRASH_MAX_USABLE_RANGES 2 +/* * reserve_crashkernel() - reserves memory for crash kernel *
@@ -280,9 +284,9 @@ early_param("mem", early_mem);static int __init early_init_dt_scan_usablemem (unsigned long node,const char *uname,int depth,void *data) { -struct memblock_region *usablemem = data; -const __be32 *reg; -int len; +struct memblock_type *usablemem = data; +const __be32 *reg, *endp; +int len, nr =0 ;if (depth !=1 ||strcmp (uname,"chosen" ) !=0 )return 0 ;
@@ -291,22 +295,32 @@ static int __init early_init_dt_scan_usablemem(unsigned long node,if (!reg || (len < (dt_root_addr_cells + dt_root_size_cells)))return 1 ; - usablemem->base =dt_mem_next_cell (dt_root_addr_cells, ®); - usablemem->size =dt_mem_next_cell (dt_root_size_cells, ®); + endp = reg + (len /sizeof (__be32)); +while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) { +unsigned long base =dt_mem_next_cell (dt_root_addr_cells, ®); +unsigned long size =dt_mem_next_cell (dt_root_size_cells, ®); +if (memblock_add_range (usablemem, base, size,NUMA_NO_NODE , +MEMBLOCK_NONE )) +return 0 ; +if (++nr >=CRASH_MAX_USABLE_RANGES ) +break ; + }return 1 ; }static void __init fdt_enforce_memory_region (void ) { -struct memblock_region reg = { - .size =0 , +struct memblock_region usable_regions[CRASH_MAX_USABLE_RANGES ]; +struct memblock_type usablemem = { + .max =CRASH_MAX_USABLE_RANGES , + .regions = usable_regions, }; -of_scan_flat_dt (early_init_dt_scan_usablemem, ®); +of_scan_flat_dt (early_init_dt_scan_usablemem, &usablemem); -if (reg.size) -memblock_cap_memory_range (reg.base, reg.size); +if (usablemem.cnt) +memblock_cap_memory_ranges (&usablemem); }void __init arm64_memblock_init (void )
include/linux/memblock.h +1 -1
@@ -404,7 +404,7 @@ phys_addr_t memblock_mem_size(unsigned long limit_pfn);phys_addr_t memblock_start_of_DRAM (void );phys_addr_t memblock_end_of_DRAM (void );void memblock_enforce_memory_limit (phys_addr_t memory_limit); -void memblock_cap_memory_range (phys_addr_t base,phys_addr_t size); +void memblock_cap_memory_ranges (struct memblock_type *regions_to_keep);void memblock_mem_limit_remove_map (phys_addr_t limit);bool memblock_is_memory (phys_addr_t addr);bool memblock_is_map_memory (phys_addr_t addr);
mm/memblock.c +22 -23
65 lines not on this page · raw message
Signature · 7 lines
2.7.4 -- Sincerely yours, Mike.
Re: [PATCH v3 3/4] arm64: kdump: support more than one crash kernel regions
Hi Mike,
This overall looks well.
Replacing memblock_cap_memory_range() with memblock_cap_memory_ranges() was what i wanted to do in v1, sorry for don't express that clearly.
But there are some issues as below. After fixing this, it can work correctly.
Quoted 115 lines · Mike Rapoport
Hi,
Quoted 57 lines · Chen Zhou
After commit (arm64: kdump: support reserving crashkernel above 4G), there may be two crash kernel regions, one is below 4G, the other is above 4G.
Crash dump kernel reads more than one crash kernel regions via a dtb property under node /chosen,
linux,usable-memory-range = <BASE1 SIZE1 [BASE2 SIZE2]>Signed-off-by Chen Zhou <[email protected]>--- arch/arm64/mm/init.c | 66 ++++++++++++++++++++++++++++++++++++++++-------- include/linux/memblock.h | 6 +++++ mm/memblock.c | 7 ++--- 3 files changed, 66 insertions(+), 13 deletions(-)arch/arm64/mm/init.c +15
-4@@ -65,6 +65,11 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init;#ifdef CONFIG_KEXEC_CORE +/* at most two crash kernel regions, low_region and high_region */ +#define CRASH_MAX_USABLE_RANGES 2 +#define LOW_REGION_IDX 0 +#define HIGH_REGION_IDX 1 +/* * reserve_crashkernel() - reserves memory for crash kernel * @@ -297,8 +302,8 @@ static int __init early_init_dt_scan_usablemem(unsigned long node, const char *uname, int depth, void *data) { struct memblock_region *usablemem = data; -const __be32 *reg; -int len; +const __be32 *reg, *endp; +int len, nr =0 ;if (depth !=1 ||strcmp (uname,"chosen" ) !=0 )return 0 ;@@ -307,22 +312,63 @@ static int __init early_init_dt_scan_usablemem(unsigned long node,if (!reg || (len < (dt_root_addr_cells + dt_root_size_cells)))return 1 ; - usablemem->base =dt_mem_next_cell (dt_root_addr_cells, ®); - usablemem->size =dt_mem_next_cell (dt_root_size_cells, ®); + endp = reg + (len /sizeof (__be32)); +while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) { + usablemem[nr].base =dt_mem_next_cell (dt_root_addr_cells, ®); + usablemem[nr].size =dt_mem_next_cell (dt_root_size_cells, ®); + +if (++nr >=CRASH_MAX_USABLE_RANGES ) +break ; + }arch/arm64/mm/init.c static int __init early_init_dt_scan_usablemem(unsigned long node, · line 330
Quoted 10 lines
arch/arm64/mm/init.c +2
-3return 1 ; }static void __init fdt_enforce_memory_region (void ) { -struct memblock_region reg = { - .size =0 , - }; +int i, cnt =0 ; +struct memblock_region regs[CRASH_MAX_USABLE_RANGES ];I only now noticed that
fdt_enforce_memory_region()usesmemblock_regionto pass the ranges around. If we'd switch tomemblock_typeinstead, the implementation ofmemblock_cap_memory_ranges()would be really straightforward. Can you check if the below patch works for you?Quoted 1 line
From
e476d584098e31273af573e1a78e308880c5cf28Mon Sep 17 00:00:00 2001From: Mike Rapoport <[email protected]>
Date: Wed, 10 Apr 2019 16:02:32 +0300
Subject: [PATCH] memblock: extendmemblock_cap_memory_rangeto multiple rangesThe
memblock_cap_memory_range()removes all the memory except the range passed to it. Extend this function to recievememblock_typewith the regions that should be kept. This allows switching to simple iteration over memblock arrays with 'for_each_mem_range' to remove the unneeded memory.Enable use of this function in arm64 for reservation of multile regions for the crash kernel.
Signed-off-by Mike Rapoport <[email protected]>--- arch/arm64/mm/init.c | 34 ++++++++++++++++++++++++---------- include/linux/memblock.h | 2 +- mm/memblock.c | 45 ++++++++++++++++++++++----------------------- 3 files changed, 47 insertions(+), 34 deletions(-)mm/memblock.c +1
-16-void __init memblock_cap_memory_range (phys_addr_t base,phys_addr_t size) +void __init memblock_cap_memory_ranges (struct memblock_type *regions_to_keep) { -int start_rgn, end_rgn; -int i, ret; - -if (!size) -return ; - - ret =memblock_isolate_range (&memblock.memory, base, size, - &start_rgn, &end_rgn); -if (ret) -return ; - -/* remove all the MAP regions */ -for (i = memblock.memory.cnt -1 ; i >= end_rgn; i--) -if (!memblock_is_nomap (&memblock.memory.regions[i])) -memblock_remove_region (&memblock.memory, i);
mm/memblock.c void __init memblock_enforce_memory_limit(phys_addr_t limit) · line 1616
Quoted 10 lines
mm/memblock.c +6
-3+phys_addr_t start, end; +u64 i; -for (i = start_rgn -1 ; i >=0 ; i--) -if (!memblock_is_nomap (&memblock.memory.regions[i])) -memblock_remove_region (&memblock.memory, i); +/* truncate memory while skipping NOMAP regions */ +for_each_mem_range (i, &memblock.memory, regions_to_keep,NUMA_NO_NODE , +MEMBLOCK_NONE , &start, &end,NULL ) +memblock_remove (start, end);
1. use memblock_remove(start, size) instead of memblock_remove(start, end).
2. There is a another hidden issue. We couldn't mix __next_mem_range()(called by for_each_mem_range) operation with remove operation because __next_mem_range() records the index of last time. If we do remove between __next_mem_range(), the index may be mess.
Therefore, we could do remove operation after for_each_mem_range like this, solution A:
void __init memblock_cap_memory_ranges(struct memblock_type *regions_to_keep) {
- phys_addr_t start, end;
- u64 i;
+ phys_addr_t start[INIT_MEMBLOCK_RESERVED_REGIONS * 2];
+ phys_addr_t end[INIT_MEMBLOCK_RESERVED_REGIONS * 2];
+ u64 i, nr = 0;
/* truncate memory while skipping NOMAP regions */ for_each_mem_range(i, &memblock.memory, regions_to_keep, NUMA_NO_NODE,
- MEMBLOCK_NONE, &start, &end, NULL)
- memblock_remove(start, end);
+ MEMBLOCK_NONE, &start[nr], &end[nr], NULL)
+ nr++;
+ for (i = 0; i < nr; i++)
+ memblock_remove(start[i], end[i] - start[i]);
/* truncate the reserved regions */
+ nr = 0;
for_each_mem_range(i, &memblock.reserved, regions_to_keep, NUMA_NO_NODE,
- MEMBLOCK_NONE, &start, &end, NULL)
- memblock_remove_range(&memblock.reserved, start, end);
+ MEMBLOCK_NONE, &start[nr], &end[nr], NULL)
+ nr++;
+ for (i = 0; i < nr; i++)
+ memblock_remove_range(&memblock.reserved, start[i],
+ end[i] - start[i]);
}
But a warning occurs when compiling:
CALL scripts/atomic/check-atomics.sh CALL scripts/checksyscalls.sh CHK include/generated/compile.h CC mm/memblock.o
mm/memblock.c: In function ‘memblock_cap_memory_ranges’:mm/memblock.c:1635:1: warning: the frame size of 36912 bytes is larger than 2048 bytes [-Wframe-larger-than=]
}
another solution is my implementation in v1, solution B:
+void __init memblock_cap_memory_ranges(struct memblock_type *regions_to_keep) +{
+ int start_rgn[INIT_MEMBLOCK_REGIONS], end_rgn[INIT_MEMBLOCK_REGIONS]; + int i, j, ret, nr = 0; + memblock_region *regs = regions_to_keep->regions;
+
+ nr = regions_to_keep -> cnt;
+ if (!nr)
+ return;
+
+ /* remove all the MAP regions */
+ for (i = memblock.memory.cnt - 1; i >= end_rgn[nr - 1]; i--)
+ if (!memblock_is_nomap(&memblock.memory.regions[i]))
+ memblock_remove_region(&memblock.memory, i);
+
+ for (i = nr - 1; i > 0; i--)
+ for (j = start_rgn[i] - 1; j >= end_rgn[i - 1]; j--)
+ if (!memblock_is_nomap(&memblock.memory.regions[j]))
+ memblock_remove_region(&memblock.memory, j);
+
+ for (i = start_rgn[0] - 1; i >= 0; i--)
+ if (!memblock_is_nomap(&memblock.memory.regions[i]))
+ memblock_remove_region(&memblock.memory, i);
+
+ /* truncate the reserved regions */
+ memblock_remove_range(&memblock.reserved, 0, regs[0].base);
+
+ for (i = nr - 1; i > 0; i--)
+ memblock_remove_range(&memblock.reserved,
+ regs[i - 1].base + regs[i - 1].size,
+ regs[i].base - regs[i - 1].base - regs[i - 1].size);
+
+ memblock_remove_range(&memblock.reserved,
+ regs[nr - 1].base + regs[nr - 1].size, PHYS_ADDR_MAX);
+}
solution A: phys_addr_t start[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; phys_addr_t end[INIT_MEMBLOCK_RESERVED_REGIONS * 2];
start, end is physical addr
solution B: int start_rgn[INIT_MEMBLOCK_REGIONS], end_rgn[INIT_MEMBLOCK_REGIONS];start_rgn, end_rgn is rgn index
Solution B do less remove operations and with no warning comparing to solution A. I think solution B is better, could you give some suggestions?
mm/memblock.c void __init memblock_enforce_memory_limit(phys_addr_t limit) · line 1621
Quoted 8 lines
mm/memblock.c +3
-3/* truncate the reserved regions */ -memblock_remove_range (&memblock.reserved,0 , base); -memblock_remove_range (&memblock.reserved, - base + size,PHYS_ADDR_MAX ); +for_each_mem_range (i, &memblock.reserved, regions_to_keep,NUMA_NO_NODE , +MEMBLOCK_NONE , &start, &end,NULL ) +memblock_remove_range (&memblock.reserved, start, end);
There are the same issues as above.
Quoted 19 lines
}
void
__initmemblock_mem_limit_remove_map(phys_addr_t limit){mm/memblock.c +10
-0+struct memblock_region rgn = { + .base =0 , + }; + +struct memblock_type region_to_keep = { + .cnt =1 , + .max =1 , + .regions = &rgn, + }; +phys_addr_t max_addr;if (!limit)
mm/memblock.c void __init memblock_mem_limit_remove_map(phys_addr_t limit) · line 1651
Quoted 9 lines
mm/memblock.c +2
-1@@ -1646,7 +1644,8 @@ void __init memblock_mem_limit_remove_map(phys_addr_t limit)if (max_addr ==PHYS_ADDR_MAX )return ; -memblock_cap_memory_range (0 , max_addr); + region_to_keep.regions[0 ].size = max_addr; +memblock_cap_memory_ranges (®ion_to_keep); } static int __init_memblock memblock_search(struct memblock_type *type, phys_addr_t addr)
Thanks,
Chen Zhou
_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec
Re: [PATCH v3 3/4] arm64: kdump: support more than one crash kernel regions
Hi Mike,
Quoted 176 lines · Chen Zhou
Hi Mike,
This overall looks well.
Replacingmemblock_cap_memory_range()withmemblock_cap_memory_ranges()was what i wanted to do in v1, sorry for don't express that clearly.But there are some issues as below. After fixing this, it can work correctly.
Quoted 115 lines · Mike Rapoport
Hi,
Quoted 57 lines · Chen Zhou
After commit (arm64: kdump: support reserving crashkernel above 4G), there may be two crash kernel regions, one is below 4G, the other is above 4G.
Crash dump kernel reads more than one crash kernel regions via a dtb property under node /chosen,
linux,usable-memory-range = <BASE1 SIZE1 [BASE2 SIZE2]>Signed-off-by Chen Zhou <[email protected]>--- arch/arm64/mm/init.c | 66 ++++++++++++++++++++++++++++++++++++++++-------- include/linux/memblock.h | 6 +++++ mm/memblock.c | 7 ++--- 3 files changed, 66 insertions(+), 13 deletions(-)arch/arm64/mm/init.c +15
-4@@ -65,6 +65,11 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init;#ifdef CONFIG_KEXEC_CORE +/* at most two crash kernel regions, low_region and high_region */ +#define CRASH_MAX_USABLE_RANGES 2 +#define LOW_REGION_IDX 0 +#define HIGH_REGION_IDX 1 +/* * reserve_crashkernel() - reserves memory for crash kernel * @@ -297,8 +302,8 @@ static int __init early_init_dt_scan_usablemem(unsigned long node, const char *uname, int depth, void *data) { struct memblock_region *usablemem = data; -const __be32 *reg; -int len; +const __be32 *reg, *endp; +int len, nr =0 ;if (depth !=1 ||strcmp (uname,"chosen" ) !=0 )return 0 ;@@ -307,22 +312,63 @@ static int __init early_init_dt_scan_usablemem(unsigned long node,if (!reg || (len < (dt_root_addr_cells + dt_root_size_cells)))return 1 ; - usablemem->base =dt_mem_next_cell (dt_root_addr_cells, ®); - usablemem->size =dt_mem_next_cell (dt_root_size_cells, ®); + endp = reg + (len /sizeof (__be32)); +while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) { + usablemem[nr].base =dt_mem_next_cell (dt_root_addr_cells, ®); + usablemem[nr].size =dt_mem_next_cell (dt_root_size_cells, ®); + +if (++nr >=CRASH_MAX_USABLE_RANGES ) +break ; + }arch/arm64/mm/init.c static int __init early_init_dt_scan_usablemem(unsigned long node, · line 330
Quoted 10 lines
arch/arm64/mm/init.c +2
-3return 1 ; }static void __init fdt_enforce_memory_region (void ) { -struct memblock_region reg = { - .size =0 , - }; +int i, cnt =0 ; +struct memblock_region regs[CRASH_MAX_USABLE_RANGES ];I only now noticed that
fdt_enforce_memory_region()usesmemblock_regionto pass the ranges around. If we'd switch tomemblock_typeinstead, the implementation ofmemblock_cap_memory_ranges()would be really straightforward. Can you check if the below patch works for you?Quoted 1 line
From
e476d584098e31273af573e1a78e308880c5cf28Mon Sep 17 00:00:00 2001From: Mike Rapoport <[email protected]>
Date: Wed, 10 Apr 2019 16:02:32 +0300
Subject: [PATCH] memblock: extendmemblock_cap_memory_rangeto multiple rangesThe
memblock_cap_memory_range()removes all the memory except the range passed to it. Extend this function to recievememblock_typewith the regions that should be kept. This allows switching to simple iteration over memblock arrays with 'for_each_mem_range' to remove the unneeded memory.Enable use of this function in arm64 for reservation of multile regions for the crash kernel.
Signed-off-by Mike Rapoport <[email protected]>--- arch/arm64/mm/init.c | 34 ++++++++++++++++++++++++---------- include/linux/memblock.h | 2 +- mm/memblock.c | 45 ++++++++++++++++++++++----------------------- 3 files changed, 47 insertions(+), 34 deletions(-)mm/memblock.c +1
-16-void __init memblock_cap_memory_range (phys_addr_t base,phys_addr_t size) +void __init memblock_cap_memory_ranges (struct memblock_type *regions_to_keep) { -int start_rgn, end_rgn; -int i, ret; - -if (!size) -return ; - - ret =memblock_isolate_range (&memblock.memory, base, size, - &start_rgn, &end_rgn); -if (ret) -return ; - -/* remove all the MAP regions */ -for (i = memblock.memory.cnt -1 ; i >= end_rgn; i--) -if (!memblock_is_nomap (&memblock.memory.regions[i])) -memblock_remove_region (&memblock.memory, i);mm/memblock.c void __init memblock_enforce_memory_limit(phys_addr_t limit) · line 1616
Quoted 10 lines
mm/memblock.c +6
-3+phys_addr_t start, end; +u64 i; -for (i = start_rgn -1 ; i >=0 ; i--) -if (!memblock_is_nomap (&memblock.memory.regions[i])) -memblock_remove_region (&memblock.memory, i); +/* truncate memory while skipping NOMAP regions */ +for_each_mem_range (i, &memblock.memory, regions_to_keep,NUMA_NO_NODE , +MEMBLOCK_NONE , &start, &end,NULL ) +memblock_remove (start, end);1. use
memblock_remove(start, size)instead ofmemblock_remove(start, end).2. There is a another hidden issue. We couldn't mix __next_mem_range()(called by
for_each_mem_range) operation with remove operation because__next_mem_range()records the index of last time. If we do remove between__next_mem_range(), the index may be mess.Therefore, we could do remove operation after
for_each_mem_rangelike this, solution A:
void__initmemblock_cap_memory_ranges(struct memblock_type *regions_to_keep){(quoted diff) +13
-6- phys_addr_t start, end; - u64 i; + phys_addr_t start[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; + phys_addr_t end[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; + u64 i, nr = 0; /* truncate memory while skipping NOMAP regions */ for_each_mem_range(i, &memblock.memory, regions_to_keep, NUMA_NO_NODE, - MEMBLOCK_NONE, &start, &end, NULL) - memblock_remove(start, end); + MEMBLOCK_NONE, &start[nr], &end[nr], NULL) + nr++; + for (i = 0; i < nr; i++) + memblock_remove(start[i], end[i] - start[i]); /* truncate the reserved regions */ + nr = 0; for_each_mem_range(i, &memblock.reserved, regions_to_keep, NUMA_NO_NODE, - MEMBLOCK_NONE, &start, &end, NULL) - memblock_remove_range(&memblock.reserved, start, end); + MEMBLOCK_NONE, &start[nr], &end[nr], NULL) + nr++; + for (i = 0; i < nr; i++) + memblock_remove_range(&memblock.reserved, start[i], + end[i] - start[i]); }But a warning occurs when compiling:
CALL scripts/atomic/check-atomics.sh CALL scripts/checksyscalls.sh CHK include/generated/compile.h CC mm/memblock.o
Quoted 6 lines
mm/memblock.c: In function ‘memblock_cap_memory_ranges’:mm/memblock.c:1635:1: warning: the frame size of 36912 bytes is larger than 2048 bytes [-Wframe-larger-than=]
}another solution is my implementation in v1, solution B:
+void__initmemblock_cap_memory_ranges(struct memblock_type *regions_to_keep)
----------
Review
Quoted 8 lines
(quoted diff) +8
-0+{ + int start_rgn[INIT_MEMBLOCK_REGIONS], end_rgn[INIT_MEMBLOCK_REGIONS]; + int i, j, ret, nr = 0; + memblock_region *regs = regions_to_keep->regions; + + nr = regions_to_keep -> cnt; + if (!nr) + return;
----------
Sorry, i sent the drafts by mistake. I mixed the drafts with my tested version. These lines replace with below.
+ int start_rgn[INIT_MEMBLOCK_REGIONS], end_rgn[INIT_MEMBLOCK_REGIONS]; + int i, j, ret, nr = 0; + struct memblock_region *regs = regions_to_keep->regions;
+
+ for (i = 0; i < regions_to_keep->cnt; i++) {
+ ret = memblock_isolate_range(&memblock.memory, regs[i].base,
+ regs[i].size, &start_rgn[i], &end_rgn[i]);
+ if (ret)
+ break;
+ nr++;
+ }
+ if (!nr)
+ return;
Thanks,
Chen Zhou
Quoted 74 lines
(quoted diff) +26
-0+ + /* remove all the MAP regions */ + for (i = memblock.memory.cnt - 1; i >= end_rgn[nr - 1]; i--) + if (!memblock_is_nomap(&memblock.memory.regions[i])) + memblock_remove_region(&memblock.memory, i); + + for (i = nr - 1; i > 0; i--) + for (j = start_rgn[i] - 1; j >= end_rgn[i - 1]; j--) + if (!memblock_is_nomap(&memblock.memory.regions[j])) + memblock_remove_region(&memblock.memory, j); + + for (i = start_rgn[0] - 1; i >= 0; i--) + if (!memblock_is_nomap(&memblock.memory.regions[i])) + memblock_remove_region(&memblock.memory, i); + + /* truncate the reserved regions */ + memblock_remove_range(&memblock.reserved, 0, regs[0].base); + + for (i = nr - 1; i > 0; i--) + memblock_remove_range(&memblock.reserved, + regs[i - 1].base + regs[i - 1].size, + regs[i].base - regs[i - 1].base - regs[i - 1].size); + + memblock_remove_range(&memblock.reserved, + regs[nr - 1].base + regs[nr - 1].size, PHYS_ADDR_MAX); +}solution A: phys_addr_t start[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; phys_addr_t end[INIT_MEMBLOCK_RESERVED_REGIONS * 2];start, end is physical addr
solution B: int start_rgn[INIT_MEMBLOCK_REGIONS], end_rgn[INIT_MEMBLOCK_REGIONS];
start_rgn,end_rgnis rgn indexSolution B do less remove operations and with no warning comparing to solution A. I think solution B is better, could you give some suggestions?
mm/memblock.c void __init memblock_enforce_memory_limit(phys_addr_t limit) · line 1621
Quoted 8 lines
mm/memblock.c +3
-3/* truncate the reserved regions */ -memblock_remove_range (&memblock.reserved,0 , base); -memblock_remove_range (&memblock.reserved, - base + size,PHYS_ADDR_MAX ); +for_each_mem_range (i, &memblock.reserved, regions_to_keep,NUMA_NO_NODE , +MEMBLOCK_NONE , &start, &end,NULL ) +memblock_remove_range (&memblock.reserved, start, end);There are the same issues as above.
Quoted 26 lines
}
void
__initmemblock_mem_limit_remove_map(phys_addr_t limit){mm/memblock.c +12
-1+struct memblock_region rgn = { + .base =0 , + }; + +struct memblock_type region_to_keep = { + .cnt =1 , + .max =1 , + .regions = &rgn, + }; +phys_addr_t max_addr;if (!limit)@@ -1646,7 +1644,8 @@ void __init memblock_mem_limit_remove_map(phys_addr_t limit)if (max_addr ==PHYS_ADDR_MAX )return ; -memblock_cap_memory_range (0 , max_addr); + region_to_keep.regions[0 ].size = max_addr; +memblock_cap_memory_ranges (®ion_to_keep); }
Quoted 6 lines
mm/memblock.c void __init memblock_mem_limit_remove_map(phys_addr_t limit) · line 1651
Quoted 2 lines
mm/memblock.c +0
-0static int __init_memblock memblock_search(struct memblock_type *type, phys_addr_t addr)Thanks,
Chen Zhou
_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec
Re: [PATCH v3 3/4] arm64: kdump: support more than one crash kernel regions
In reply to Chen ZhouHi,
Quoted 5 lines · Chen Zhou
Hi Mike,
This overall looks well.
Replacingmemblock_cap_memory_range()withmemblock_cap_memory_ranges()was what i wanted to do in v1, sorry for don't express that clearly.
I didn't object to memblock_cap_memory_ranges() in general, I was worried about it's complexity and I hoped that we could find a simpler solution.
Quoted 131 lines
But there are some issues as below. After fixing this, it can work correctly.
Quoted 115 lines · Mike Rapoport
Hi,
Quoted 57 lines · Chen Zhou
After commit (arm64: kdump: support reserving crashkernel above 4G), there may be two crash kernel regions, one is below 4G, the other is above 4G.
Crash dump kernel reads more than one crash kernel regions via a dtb property under node /chosen,
linux,usable-memory-range = <BASE1 SIZE1 [BASE2 SIZE2]>Signed-off-by Chen Zhou <[email protected]>--- arch/arm64/mm/init.c | 66 ++++++++++++++++++++++++++++++++++++++++-------- include/linux/memblock.h | 6 +++++ mm/memblock.c | 7 ++--- 3 files changed, 66 insertions(+), 13 deletions(-)arch/arm64/mm/init.c +15
-4@@ -65,6 +65,11 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init;#ifdef CONFIG_KEXEC_CORE +/* at most two crash kernel regions, low_region and high_region */ +#define CRASH_MAX_USABLE_RANGES 2 +#define LOW_REGION_IDX 0 +#define HIGH_REGION_IDX 1 +/* * reserve_crashkernel() - reserves memory for crash kernel * @@ -297,8 +302,8 @@ static int __init early_init_dt_scan_usablemem(unsigned long node, const char *uname, int depth, void *data) { struct memblock_region *usablemem = data; -const __be32 *reg; -int len; +const __be32 *reg, *endp; +int len, nr =0 ;if (depth !=1 ||strcmp (uname,"chosen" ) !=0 )return 0 ;@@ -307,22 +312,63 @@ static int __init early_init_dt_scan_usablemem(unsigned long node,if (!reg || (len < (dt_root_addr_cells + dt_root_size_cells)))return 1 ; - usablemem->base =dt_mem_next_cell (dt_root_addr_cells, ®); - usablemem->size =dt_mem_next_cell (dt_root_size_cells, ®); + endp = reg + (len /sizeof (__be32)); +while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) { + usablemem[nr].base =dt_mem_next_cell (dt_root_addr_cells, ®); + usablemem[nr].size =dt_mem_next_cell (dt_root_size_cells, ®); + +if (++nr >=CRASH_MAX_USABLE_RANGES ) +break ; + }arch/arm64/mm/init.c static int __init early_init_dt_scan_usablemem(unsigned long node, · line 330
Quoted 10 lines
arch/arm64/mm/init.c +2
-3return 1 ; }static void __init fdt_enforce_memory_region (void ) { -struct memblock_region reg = { - .size =0 , - }; +int i, cnt =0 ; +struct memblock_region regs[CRASH_MAX_USABLE_RANGES ];I only now noticed that
fdt_enforce_memory_region()usesmemblock_regionto pass the ranges around. If we'd switch tomemblock_typeinstead, the implementation ofmemblock_cap_memory_ranges()would be really straightforward. Can you check if the below patch works for you?Quoted 1 line
From
e476d584098e31273af573e1a78e308880c5cf28Mon Sep 17 00:00:00 2001From: Mike Rapoport <[email protected]>
Date: Wed, 10 Apr 2019 16:02:32 +0300
Subject: [PATCH] memblock: extendmemblock_cap_memory_rangeto multiple rangesThe
memblock_cap_memory_range()removes all the memory except the range passed to it. Extend this function to recievememblock_typewith the regions that should be kept. This allows switching to simple iteration over memblock arrays with 'for_each_mem_range' to remove the unneeded memory.Enable use of this function in arm64 for reservation of multile regions for the crash kernel.
Signed-off-by Mike Rapoport <[email protected]>--- arch/arm64/mm/init.c | 34 ++++++++++++++++++++++++---------- include/linux/memblock.h | 2 +- mm/memblock.c | 45 ++++++++++++++++++++++----------------------- 3 files changed, 47 insertions(+), 34 deletions(-)mm/memblock.c +1
-16-void __init memblock_cap_memory_range (phys_addr_t base,phys_addr_t size) +void __init memblock_cap_memory_ranges (struct memblock_type *regions_to_keep) { -int start_rgn, end_rgn; -int i, ret; - -if (!size) -return ; - - ret =memblock_isolate_range (&memblock.memory, base, size, - &start_rgn, &end_rgn); -if (ret) -return ; - -/* remove all the MAP regions */ -for (i = memblock.memory.cnt -1 ; i >= end_rgn; i--) -if (!memblock_is_nomap (&memblock.memory.regions[i])) -memblock_remove_region (&memblock.memory, i);mm/memblock.c void __init memblock_enforce_memory_limit(phys_addr_t limit) · line 1616
Quoted 10 lines
mm/memblock.c +6
-3+phys_addr_t start, end; +u64 i; -for (i = start_rgn -1 ; i >=0 ; i--) -if (!memblock_is_nomap (&memblock.memory.regions[i])) -memblock_remove_region (&memblock.memory, i); +/* truncate memory while skipping NOMAP regions */ +for_each_mem_range (i, &memblock.memory, regions_to_keep,NUMA_NO_NODE , +MEMBLOCK_NONE , &start, &end,NULL ) +memblock_remove (start, end);1. use
memblock_remove(start, size)instead ofmemblock_remove(start, end).
Quoted 3 lines
2. There is a another hidden issue. We couldn't mix __next_mem_range()(called by
for_each_mem_range) operation with remove operation because__next_mem_range()records the index of last time. If we do remove between__next_mem_range(), the index may be mess.
Oops, I've really missed that :)
Quoted 78 lines
Therefore, we could do remove operation after
for_each_mem_rangelike this, solution A:
void__initmemblock_cap_memory_ranges(struct memblock_type *regions_to_keep){(quoted diff) +13
-6- phys_addr_t start, end; - u64 i; + phys_addr_t start[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; + phys_addr_t end[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; + u64 i, nr = 0; /* truncate memory while skipping NOMAP regions */ for_each_mem_range(i, &memblock.memory, regions_to_keep, NUMA_NO_NODE, - MEMBLOCK_NONE, &start, &end, NULL) - memblock_remove(start, end); + MEMBLOCK_NONE, &start[nr], &end[nr], NULL) + nr++; + for (i = 0; i < nr; i++) + memblock_remove(start[i], end[i] - start[i]); /* truncate the reserved regions */ + nr = 0; for_each_mem_range(i, &memblock.reserved, regions_to_keep, NUMA_NO_NODE, - MEMBLOCK_NONE, &start, &end, NULL) - memblock_remove_range(&memblock.reserved, start, end); + MEMBLOCK_NONE, &start[nr], &end[nr], NULL) + nr++; + for (i = 0; i < nr; i++) + memblock_remove_range(&memblock.reserved, start[i], + end[i] - start[i]); }But a warning occurs when compiling:
CALL scripts/atomic/check-atomics.sh CALL scripts/checksyscalls.sh CHK include/generated/compile.h CC mm/memblock.o
mm/memblock.c: In function ‘memblock_cap_memory_ranges’:mm/memblock.c:1635:1: warning: the frame size of 36912 bytes is larger than 2048 bytes [-Wframe-larger-than=]
}another solution is my implementation in v1, solution B:
(quoted diff) +35
-0+void __init memblock_cap_memory_ranges(struct memblock_type *regions_to_keep) +{ + int start_rgn[INIT_MEMBLOCK_REGIONS], end_rgn[INIT_MEMBLOCK_REGIONS]; + int i, j, ret, nr = 0; + memblock_region *regs = regions_to_keep->regions; + + nr = regions_to_keep -> cnt; + if (!nr) + return; + + /* remove all the MAP regions */ + for (i = memblock.memory.cnt - 1; i >= end_rgn[nr - 1]; i--) + if (!memblock_is_nomap(&memblock.memory.regions[i])) + memblock_remove_region(&memblock.memory, i); + + for (i = nr - 1; i > 0; i--) + for (j = start_rgn[i] - 1; j >= end_rgn[i - 1]; j--) + if (!memblock_is_nomap(&memblock.memory.regions[j])) + memblock_remove_region(&memblock.memory, j); + + for (i = start_rgn[0] - 1; i >= 0; i--) + if (!memblock_is_nomap(&memblock.memory.regions[i])) + memblock_remove_region(&memblock.memory, i); + + /* truncate the reserved regions */ + memblock_remove_range(&memblock.reserved, 0, regs[0].base); + + for (i = nr - 1; i > 0; i--) + memblock_remove_range(&memblock.reserved, + regs[i - 1].base + regs[i - 1].size, + regs[i].base - regs[i - 1].base - regs[i - 1].size); + + memblock_remove_range(&memblock.reserved, + regs[nr - 1].base + regs[nr - 1].size, PHYS_ADDR_MAX); +}solution A: phys_addr_t start[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; phys_addr_t end[INIT_MEMBLOCK_RESERVED_REGIONS * 2];
Quoted 7 lines
start, end is physical addr
solution B: int start_rgn[INIT_MEMBLOCK_REGIONS], end_rgn[INIT_MEMBLOCK_REGIONS];
start_rgn,end_rgnis rgn indexSolution B do less remove operations and with no warning comparing to solution A. I think solution B is better, could you give some suggestions?
Solution B is indeed better that solution A, but I'm still worried by relatively large arrays on stack and the amount of loops :(
The very least we could do is to call memblock_cap_memory_range() to drop the memory before and after the ranges we'd like to keep.
Quoted 43 lines
mm/memblock.c void __init memblock_enforce_memory_limit(phys_addr_t limit) · line 1621
Quoted 8 lines
mm/memblock.c +3
-3/* truncate the reserved regions */ -memblock_remove_range (&memblock.reserved,0 , base); -memblock_remove_range (&memblock.reserved, - base + size,PHYS_ADDR_MAX ); +for_each_mem_range (i, &memblock.reserved, regions_to_keep,NUMA_NO_NODE , +MEMBLOCK_NONE , &start, &end,NULL ) +memblock_remove_range (&memblock.reserved, start, end);There are the same issues as above.
Quoted 19 lines
}
void
__initmemblock_mem_limit_remove_map(phys_addr_t limit){mm/memblock.c +10
-0+struct memblock_region rgn = { + .base =0 , + }; + +struct memblock_type region_to_keep = { + .cnt =1 , + .max =1 , + .regions = &rgn, + }; +phys_addr_t max_addr;if (!limit)mm/memblock.c void __init memblock_mem_limit_remove_map(phys_addr_t limit) · line 1651
Quoted 9 lines
mm/memblock.c +2
-1@@ -1646,7 +1644,8 @@ void __init memblock_mem_limit_remove_map(phys_addr_t limit)if (max_addr ==PHYS_ADDR_MAX )return ; -memblock_cap_memory_range (0 , max_addr); + region_to_keep.regions[0 ].size = max_addr; +memblock_cap_memory_ranges (®ion_to_keep); } static int __init_memblock memblock_search(struct memblock_type *type, phys_addr_t addr)Thanks,
Chen Zhou
Signature · 2 lines
Sincerely yours, Mike.
Re: [PATCH v3 3/4] arm64: kdump: support more than one crash kernel regions
Hi Mike,
Quoted 232 lines · Mike Rapoport
Hi,
Quoted 5 lines · Chen Zhou
Hi Mike,
This overall looks well.
Replacingmemblock_cap_memory_range()withmemblock_cap_memory_ranges()was what i wanted to do in v1, sorry for don't express that clearly.I didn't object to
memblock_cap_memory_ranges()in general, I was worried about it's complexity and I hoped that we could find a simpler solution.Quoted 131 lines
But there are some issues as below. After fixing this, it can work correctly.
Quoted 115 lines · Mike Rapoport
Hi,
Quoted 57 lines · Chen Zhou
After commit (arm64: kdump: support reserving crashkernel above 4G), there may be two crash kernel regions, one is below 4G, the other is above 4G.
Crash dump kernel reads more than one crash kernel regions via a dtb property under node /chosen,
linux,usable-memory-range = <BASE1 SIZE1 [BASE2 SIZE2]>Signed-off-by Chen Zhou <[email protected]>--- arch/arm64/mm/init.c | 66 ++++++++++++++++++++++++++++++++++++++++-------- include/linux/memblock.h | 6 +++++ mm/memblock.c | 7 ++--- 3 files changed, 66 insertions(+), 13 deletions(-)arch/arm64/mm/init.c +15
-4@@ -65,6 +65,11 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init;#ifdef CONFIG_KEXEC_CORE +/* at most two crash kernel regions, low_region and high_region */ +#define CRASH_MAX_USABLE_RANGES 2 +#define LOW_REGION_IDX 0 +#define HIGH_REGION_IDX 1 +/* * reserve_crashkernel() - reserves memory for crash kernel * @@ -297,8 +302,8 @@ static int __init early_init_dt_scan_usablemem(unsigned long node, const char *uname, int depth, void *data) { struct memblock_region *usablemem = data; -const __be32 *reg; -int len; +const __be32 *reg, *endp; +int len, nr =0 ;if (depth !=1 ||strcmp (uname,"chosen" ) !=0 )return 0 ;@@ -307,22 +312,63 @@ static int __init early_init_dt_scan_usablemem(unsigned long node,if (!reg || (len < (dt_root_addr_cells + dt_root_size_cells)))return 1 ; - usablemem->base =dt_mem_next_cell (dt_root_addr_cells, ®); - usablemem->size =dt_mem_next_cell (dt_root_size_cells, ®); + endp = reg + (len /sizeof (__be32)); +while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) { + usablemem[nr].base =dt_mem_next_cell (dt_root_addr_cells, ®); + usablemem[nr].size =dt_mem_next_cell (dt_root_size_cells, ®); + +if (++nr >=CRASH_MAX_USABLE_RANGES ) +break ; + }arch/arm64/mm/init.c static int __init early_init_dt_scan_usablemem(unsigned long node, · line 330
Quoted 10 lines
arch/arm64/mm/init.c +2
-3return 1 ; }static void __init fdt_enforce_memory_region (void ) { -struct memblock_region reg = { - .size =0 , - }; +int i, cnt =0 ; +struct memblock_region regs[CRASH_MAX_USABLE_RANGES ];I only now noticed that
fdt_enforce_memory_region()usesmemblock_regionto pass the ranges around. If we'd switch tomemblock_typeinstead, the implementation ofmemblock_cap_memory_ranges()would be really straightforward. Can you check if the below patch works for you?Quoted 1 line
From
e476d584098e31273af573e1a78e308880c5cf28Mon Sep 17 00:00:00 2001From: Mike Rapoport <[email protected]>
Date: Wed, 10 Apr 2019 16:02:32 +0300
Subject: [PATCH] memblock: extendmemblock_cap_memory_rangeto multiple rangesThe
memblock_cap_memory_range()removes all the memory except the range passed to it. Extend this function to recievememblock_typewith the regions that should be kept. This allows switching to simple iteration over memblock arrays with 'for_each_mem_range' to remove the unneeded memory.Enable use of this function in arm64 for reservation of multile regions for the crash kernel.
Signed-off-by Mike Rapoport <[email protected]>--- arch/arm64/mm/init.c | 34 ++++++++++++++++++++++++---------- include/linux/memblock.h | 2 +- mm/memblock.c | 45 ++++++++++++++++++++++----------------------- 3 files changed, 47 insertions(+), 34 deletions(-)mm/memblock.c +1
-16-void __init memblock_cap_memory_range (phys_addr_t base,phys_addr_t size) +void __init memblock_cap_memory_ranges (struct memblock_type *regions_to_keep) { -int start_rgn, end_rgn; -int i, ret; - -if (!size) -return ; - - ret =memblock_isolate_range (&memblock.memory, base, size, - &start_rgn, &end_rgn); -if (ret) -return ; - -/* remove all the MAP regions */ -for (i = memblock.memory.cnt -1 ; i >= end_rgn; i--) -if (!memblock_is_nomap (&memblock.memory.regions[i])) -memblock_remove_region (&memblock.memory, i);mm/memblock.c void __init memblock_enforce_memory_limit(phys_addr_t limit) · line 1616
Quoted 10 lines
mm/memblock.c +6
-3+phys_addr_t start, end; +u64 i; -for (i = start_rgn -1 ; i >=0 ; i--) -if (!memblock_is_nomap (&memblock.memory.regions[i])) -memblock_remove_region (&memblock.memory, i); +/* truncate memory while skipping NOMAP regions */ +for_each_mem_range (i, &memblock.memory, regions_to_keep,NUMA_NO_NODE , +MEMBLOCK_NONE , &start, &end,NULL ) +memblock_remove (start, end);1. use
memblock_remove(start, size)instead ofmemblock_remove(start, end).Quoted 3 lines
2. There is a another hidden issue. We couldn't mix __next_mem_range()(called by
for_each_mem_range) operation with remove operation because__next_mem_range()records the index of last time. If we do remove between__next_mem_range(), the index may be mess.Oops, I've really missed that :)
Quoted 83 lines
Therefore, we could do remove operation after
for_each_mem_rangelike this, solution A:
void__initmemblock_cap_memory_ranges(struct memblock_type *regions_to_keep){(quoted diff) +13
-6- phys_addr_t start, end; - u64 i; + phys_addr_t start[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; + phys_addr_t end[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; + u64 i, nr = 0; /* truncate memory while skipping NOMAP regions */ for_each_mem_range(i, &memblock.memory, regions_to_keep, NUMA_NO_NODE, - MEMBLOCK_NONE, &start, &end, NULL) - memblock_remove(start, end); + MEMBLOCK_NONE, &start[nr], &end[nr], NULL) + nr++; + for (i = 0; i < nr; i++) + memblock_remove(start[i], end[i] - start[i]); /* truncate the reserved regions */ + nr = 0; for_each_mem_range(i, &memblock.reserved, regions_to_keep, NUMA_NO_NODE, - MEMBLOCK_NONE, &start, &end, NULL) - memblock_remove_range(&memblock.reserved, start, end); + MEMBLOCK_NONE, &start[nr], &end[nr], NULL) + nr++; + for (i = 0; i < nr; i++) + memblock_remove_range(&memblock.reserved, start[i], + end[i] - start[i]); }But a warning occurs when compiling:
CALL scripts/atomic/check-atomics.sh CALL scripts/checksyscalls.sh CHK include/generated/compile.h CC mm/memblock.o
mm/memblock.c: In function ‘memblock_cap_memory_ranges’:mm/memblock.c:1635:1: warning: the frame size of 36912 bytes is larger than 2048 bytes [-Wframe-larger-than=]
}another solution is my implementation in v1, solution B:
(quoted diff) +35
-0+void __init memblock_cap_memory_ranges(struct memblock_type *regions_to_keep) +{ + int start_rgn[INIT_MEMBLOCK_REGIONS], end_rgn[INIT_MEMBLOCK_REGIONS]; + int i, j, ret, nr = 0; + memblock_region *regs = regions_to_keep->regions; + + nr = regions_to_keep -> cnt; + if (!nr) + return; + + /* remove all the MAP regions */ + for (i = memblock.memory.cnt - 1; i >= end_rgn[nr - 1]; i--) + if (!memblock_is_nomap(&memblock.memory.regions[i])) + memblock_remove_region(&memblock.memory, i); + + for (i = nr - 1; i > 0; i--) + for (j = start_rgn[i] - 1; j >= end_rgn[i - 1]; j--) + if (!memblock_is_nomap(&memblock.memory.regions[j])) + memblock_remove_region(&memblock.memory, j); + + for (i = start_rgn[0] - 1; i >= 0; i--) + if (!memblock_is_nomap(&memblock.memory.regions[i])) + memblock_remove_region(&memblock.memory, i); + + /* truncate the reserved regions */ + memblock_remove_range(&memblock.reserved, 0, regs[0].base); + + for (i = nr - 1; i > 0; i--) + memblock_remove_range(&memblock.reserved, + regs[i - 1].base + regs[i - 1].size, + regs[i].base - regs[i - 1].base - regs[i - 1].size); + + memblock_remove_range(&memblock.reserved, + regs[nr - 1].base + regs[nr - 1].size, PHYS_ADDR_MAX); +}solution A: phys_addr_t start[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; phys_addr_t end[INIT_MEMBLOCK_RESERVED_REGIONS * 2];start, end is physical addr
solution B: int start_rgn[INIT_MEMBLOCK_REGIONS], end_rgn[INIT_MEMBLOCK_REGIONS];
start_rgn,end_rgnis rgn index
Quoted 8 lines
Quoted 2 lines
Solution B do less remove operations and with no warning comparing to solution A. I think solution B is better, could you give some suggestions?
Solution B is indeed better that solution A, but I'm still worried by relatively large arrays on stack and the amount of loops :(
The very least we could do is to call
memblock_cap_memory_range()to drop the memory before and after the ranges we'd like to keep.
1. relatively large arrays
As my said above, the start_rgn, end_rgn is rgn index, we could use unsigned char type.
2. loops
Loops always exist, and the solution with fewer loops may be just encapsulated well.
Thanks,
Chen Zhou
Quoted 38 lines
Quoted 38 lines
mm/memblock.c void __init memblock_enforce_memory_limit(phys_addr_t limit) · line 1621
Quoted 9 lines
mm/memblock.c +3
-3/* truncate the reserved regions */ -memblock_remove_range (&memblock.reserved,0 , base); -memblock_remove_range (&memblock.reserved, - base + size,PHYS_ADDR_MAX ); +for_each_mem_range (i, &memblock.reserved, regions_to_keep,NUMA_NO_NODE , +MEMBLOCK_NONE , &start, &end,NULL ) +memblock_remove_range (&memblock.reserved, start, end);There are the same issues as above.
Quoted 26 lines
}
void
__initmemblock_mem_limit_remove_map(phys_addr_t limit){mm/memblock.c +12
-1+struct memblock_region rgn = { + .base =0 , + }; + +struct memblock_type region_to_keep = { + .cnt =1 , + .max =1 , + .regions = &rgn, + }; +phys_addr_t max_addr;if (!limit)@@ -1646,7 +1644,8 @@ void __init memblock_mem_limit_remove_map(phys_addr_t limit)if (max_addr ==PHYS_ADDR_MAX )return ; -memblock_cap_memory_range (0 , max_addr); + region_to_keep.regions[0 ].size = max_addr; +memblock_cap_memory_ranges (®ion_to_keep); }
Quoted 7 lines
Quoted 6 lines
mm/memblock.c void __init memblock_mem_limit_remove_map(phys_addr_t limit) · line 1651
Quoted 2 lines
mm/memblock.c +0
-0static int __init_memblock memblock_search(struct memblock_type *type, phys_addr_t addr)Thanks,
Chen Zhou
_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec
Re: [PATCH v3 3/4] arm64: kdump: support more than one crash kernel regions
Hi,
Quoted 14 lines · Chen Zhou
Hi Mike,
Quoted 11 lines · Mike Rapoport
Quoted 10 lines
solution A: phys_addr_t start[INIT_MEMBLOCK_RESERVED_REGIONS * 2]; phys_addr_t end[INIT_MEMBLOCK_RESERVED_REGIONS * 2];start, end is physical addr
solution B: int start_rgn[INIT_MEMBLOCK_REGIONS], end_rgn[INIT_MEMBLOCK_REGIONS];
start_rgn,end_rgnis rgn indexSolution B do less remove operations and with no warning comparing to solution A. I think solution B is better, could you give some suggestions?
Quoted 8 lines
Quoted 5 lines
Solution B is indeed better that solution A, but I'm still worried by relatively large arrays on stack and the amount of loops :(
The very least we could do is to call
memblock_cap_memory_range()to drop the memory before and after the ranges we'd like to keep.1. relatively large arrays
As my said above, thestart_rgn,end_rgnis rgn index, we could use unsigned char type.
Let's stick to int for now
Quoted 2 lines
2. loops
Loops always exist, and the solution with fewer loops may be just encapsulated well.
Of course the loops are there, I just hoped we could get rid of the nested loop and get away with single passes in all the cases. Apparently it's not the case :(
Quoted 3 lines
Thanks,
Chen Zhou
Signature · 8 lines
Sincerely yours, Mike. _______________________________________________ kexec mailing list [email protected] http://lists.infradead.org/mailman/listinfo/kexec
Re: [PATCH v3 3/4] arm64: kdump: support more than one crash kernel regions
In reply to Chen ZhouHi,
Quoted 7 lines · Chen Zhou
After commit (arm64: kdump: support reserving crashkernel above 4G), there may be two crash kernel regions, one is below 4G, the other is above 4G.
Crash dump kernel reads more than one crash kernel regions via a dtb property under node /chosen,
linux,usable-memory-range = <BASE1 SIZE1 [BASE2 SIZE2]>
Somehow I've missed that previously, but how is this supposed to work on EFI systems?
Quoted 161 lines
Signed-off-by Chen Zhou <[email protected]>--- arch/arm64/mm/init.c | 66 ++++++++++++++++++++++++++++++++++++++++-------- include/linux/memblock.h | 6 +++++ mm/memblock.c | 7 ++--- 3 files changed, 66 insertions(+), 13 deletions(-)arch/arm64/mm/init.c +56
-10@@ -65,6 +65,11 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init;#ifdef CONFIG_KEXEC_CORE +/* at most two crash kernel regions, low_region and high_region */ +#define CRASH_MAX_USABLE_RANGES 2 +#define LOW_REGION_IDX 0 +#define HIGH_REGION_IDX 1 +/* * reserve_crashkernel() - reserves memory for crash kernel * @@ -297,8 +302,8 @@ static int __init early_init_dt_scan_usablemem(unsigned long node, const char *uname, int depth, void *data) { struct memblock_region *usablemem = data; -const __be32 *reg; -int len; +const __be32 *reg, *endp; +int len, nr =0 ;if (depth !=1 ||strcmp (uname,"chosen" ) !=0 )return 0 ;@@ -307,22 +312,63 @@ static int __init early_init_dt_scan_usablemem(unsigned long node,if (!reg || (len < (dt_root_addr_cells + dt_root_size_cells)))return 1 ; - usablemem->base =dt_mem_next_cell (dt_root_addr_cells, ®); - usablemem->size =dt_mem_next_cell (dt_root_size_cells, ®); + endp = reg + (len /sizeof (__be32)); +while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) { + usablemem[nr].base =dt_mem_next_cell (dt_root_addr_cells, ®); + usablemem[nr].size =dt_mem_next_cell (dt_root_size_cells, ®); + +if (++nr >=CRASH_MAX_USABLE_RANGES ) +break ; + }return 1 ; }static void __init fdt_enforce_memory_region (void ) { -struct memblock_region reg = { - .size =0 , - }; +int i, cnt =0 ; +struct memblock_region regs[CRASH_MAX_USABLE_RANGES ]; + +memset (regs,0 ,sizeof (regs)); +of_scan_flat_dt (early_init_dt_scan_usablemem, regs); + +for (i =0 ; i <CRASH_MAX_USABLE_RANGES ; i++) +if (regs[i].size) + cnt++; +else +break ; + +if (cnt -1 ==LOW_REGION_IDX ) +memblock_cap_memory_range (regs[LOW_REGION_IDX ].base, + regs[LOW_REGION_IDX ].size); +else if (cnt -1 ==HIGH_REGION_IDX ) { +/* +* Two crash kernel regions, cap the memory range +* [regs[LOW_REGION_IDX].base, regs[HIGH_REGION_IDX].end] +* and then remove the memory range in the middle. +*/ +int start_rgn, end_rgn, i, ret; +phys_addr_t mid_base, mid_size; + + mid_base = regs[LOW_REGION_IDX ].base + regs[LOW_REGION_IDX ].size; + mid_size = regs[HIGH_REGION_IDX ].base - mid_base; + ret =memblock_isolate_range (&memblock.memory, mid_base, + mid_size, &start_rgn, &end_rgn); -of_scan_flat_dt (early_init_dt_scan_usablemem, ®); +if (ret) +return ; -if (reg.size) -memblock_cap_memory_range (reg.base, reg.size); +memblock_cap_memory_range (regs[LOW_REGION_IDX ].base, + regs[HIGH_REGION_IDX ].base - + regs[LOW_REGION_IDX ].base + + regs[HIGH_REGION_IDX ].size); +for (i = end_rgn -1 ; i >= start_rgn; i--) { +if (!memblock_is_nomap (&memblock.memory.regions[i])) +memblock_remove_region (&memblock.memory, i); + } +memblock_remove_range (&memblock.reserved, mid_base, + mid_base + mid_size); + } }void __init arm64_memblock_init (void )include/linux/memblock.h +6
-0@@ -110,9 +110,15 @@ void memblock_discard(void);phys_addr_t memblock_find_in_range (phys_addr_t start,phys_addr_t end,phys_addr_t size,phys_addr_t align); +void memblock_remove_region (struct memblock_type *type,unsigned long r);void memblock_allow_resize (void );int memblock_add_node (phys_addr_t base,phys_addr_t size,int nid);int memblock_add (phys_addr_t base,phys_addr_t size); +int memblock_isolate_range (struct memblock_type *type, +phys_addr_t base,phys_addr_t size, +int *start_rgn,int *end_rgn); +int memblock_remove_range (struct memblock_type *type, +phys_addr_t base,phys_addr_t size);int memblock_remove (phys_addr_t base,phys_addr_t size);int memblock_free (phys_addr_t base,phys_addr_t size);int memblock_reserve (phys_addr_t base,phys_addr_t size);mm/memblock.c +4
-3@@ -357,7 +357,8 @@ phys_addr_t __init_memblock memblock_find_in_range(phys_addr_t start,return ret; } -static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r) +void __init_memblock memblock_remove_region(struct memblock_type *type, +unsigned long r) { type->total_size -= type->regions[r].size; memmove(&type->regions[r], &type->regions[r + 1],@@ -724,7 +725,7 @@ int __init_memblock memblock_add(phys_addr_t base, phys_addr_t size)* Return: * 0 on success, -errno on failure. */ -static int __init_memblock memblock_isolate_range(struct memblock_type *type, +int __init_memblock memblock_isolate_range(struct memblock_type *type,phys_addr_t base,phys_addr_t size,int *start_rgn,int *end_rgn) {@@ -784,7 +785,7 @@ static int __init_memblock memblock_isolate_range(struct memblock_type *type,return 0 ; } -static int __init_memblock memblock_remove_range(struct memblock_type *type, +int __init_memblock memblock_remove_range(struct memblock_type *type,phys_addr_t base,phys_addr_t size) {int start_rgn, end_rgn;Signature · 1 line
2.7.4
Signature · 8 lines
Sincerely yours, Mike. _______________________________________________ kexec mailing list [email protected] http://lists.infradead.org/mailman/listinfo/kexec
Re: [PATCH v3 3/4] arm64: kdump: support more than one crash kernel regions
Hi Mike,
Quoted 7 lines · Mike Rapoport
Hi,
Quoted 4 lines · Chen Zhou
After commit (arm64: kdump: support reserving crashkernel above 4G), there may be two crash kernel regions, one is below 4G, the other is above 4G.
Quoted 6 lines
Quoted 3 lines
Crash dump kernel reads more than one crash kernel regions via a dtb property under node /chosen,
linux,usable-memory-range = <BASE1 SIZE1 [BASE2 SIZE2]>Somehow I've missed that previously, but how is this supposed to work on EFI systems?
Whatever the way in which the systems work, there is FDT pointer(__fdt_pointer) in arm64 kernel and file /sys/firmware/fdt will be created in late_initcall.
Kexec-tools read and update file /sys/firmware/fdt in EFI systems to support kdump to boot capture kernel.
For supporting more than one crash kernel regions, kexec-tools make changes accordingly. Details are in below:
http://lists.infradead.org/pipermail/kexec/2019-April/022792.html
Thanks,
Chen Zhou
Quoted 154 lines
Quoted 154 lines
Signed-off-by Chen Zhou <[email protected]>--- arch/arm64/mm/init.c | 66 ++++++++++++++++++++++++++++++++++++++++-------- include/linux/memblock.h | 6 +++++ mm/memblock.c | 7 ++--- 3 files changed, 66 insertions(+), 13 deletions(-)arch/arm64/mm/init.c +56
-10@@ -65,6 +65,11 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init;#ifdef CONFIG_KEXEC_CORE +/* at most two crash kernel regions, low_region and high_region */ +#define CRASH_MAX_USABLE_RANGES 2 +#define LOW_REGION_IDX 0 +#define HIGH_REGION_IDX 1 +/* * reserve_crashkernel() - reserves memory for crash kernel * @@ -297,8 +302,8 @@ static int __init early_init_dt_scan_usablemem(unsigned long node, const char *uname, int depth, void *data) { struct memblock_region *usablemem = data; -const __be32 *reg; -int len; +const __be32 *reg, *endp; +int len, nr =0 ;if (depth !=1 ||strcmp (uname,"chosen" ) !=0 )return 0 ;@@ -307,22 +312,63 @@ static int __init early_init_dt_scan_usablemem(unsigned long node,if (!reg || (len < (dt_root_addr_cells + dt_root_size_cells)))return 1 ; - usablemem->base =dt_mem_next_cell (dt_root_addr_cells, ®); - usablemem->size =dt_mem_next_cell (dt_root_size_cells, ®); + endp = reg + (len /sizeof (__be32)); +while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) { + usablemem[nr].base =dt_mem_next_cell (dt_root_addr_cells, ®); + usablemem[nr].size =dt_mem_next_cell (dt_root_size_cells, ®); + +if (++nr >=CRASH_MAX_USABLE_RANGES ) +break ; + }return 1 ; }static void __init fdt_enforce_memory_region (void ) { -struct memblock_region reg = { - .size =0 , - }; +int i, cnt =0 ; +struct memblock_region regs[CRASH_MAX_USABLE_RANGES ]; + +memset (regs,0 ,sizeof (regs)); +of_scan_flat_dt (early_init_dt_scan_usablemem, regs); + +for (i =0 ; i <CRASH_MAX_USABLE_RANGES ; i++) +if (regs[i].size) + cnt++; +else +break ; + +if (cnt -1 ==LOW_REGION_IDX ) +memblock_cap_memory_range (regs[LOW_REGION_IDX ].base, + regs[LOW_REGION_IDX ].size); +else if (cnt -1 ==HIGH_REGION_IDX ) { +/* +* Two crash kernel regions, cap the memory range +* [regs[LOW_REGION_IDX].base, regs[HIGH_REGION_IDX].end] +* and then remove the memory range in the middle. +*/ +int start_rgn, end_rgn, i, ret; +phys_addr_t mid_base, mid_size; + + mid_base = regs[LOW_REGION_IDX ].base + regs[LOW_REGION_IDX ].size; + mid_size = regs[HIGH_REGION_IDX ].base - mid_base; + ret =memblock_isolate_range (&memblock.memory, mid_base, + mid_size, &start_rgn, &end_rgn); -of_scan_flat_dt (early_init_dt_scan_usablemem, ®); +if (ret) +return ; -if (reg.size) -memblock_cap_memory_range (reg.base, reg.size); +memblock_cap_memory_range (regs[LOW_REGION_IDX ].base, + regs[HIGH_REGION_IDX ].base - + regs[LOW_REGION_IDX ].base + + regs[HIGH_REGION_IDX ].size); +for (i = end_rgn -1 ; i >= start_rgn; i--) { +if (!memblock_is_nomap (&memblock.memory.regions[i])) +memblock_remove_region (&memblock.memory, i); + } +memblock_remove_range (&memblock.reserved, mid_base, + mid_base + mid_size); + } }void __init arm64_memblock_init (void )include/linux/memblock.h +6
-0@@ -110,9 +110,15 @@ void memblock_discard(void);phys_addr_t memblock_find_in_range (phys_addr_t start,phys_addr_t end,phys_addr_t size,phys_addr_t align); +void memblock_remove_region (struct memblock_type *type,unsigned long r);void memblock_allow_resize (void );int memblock_add_node (phys_addr_t base,phys_addr_t size,int nid);int memblock_add (phys_addr_t base,phys_addr_t size); +int memblock_isolate_range (struct memblock_type *type, +phys_addr_t base,phys_addr_t size, +int *start_rgn,int *end_rgn); +int memblock_remove_range (struct memblock_type *type, +phys_addr_t base,phys_addr_t size);int memblock_remove (phys_addr_t base,phys_addr_t size);int memblock_free (phys_addr_t base,phys_addr_t size);int memblock_reserve (phys_addr_t base,phys_addr_t size);mm/memblock.c +3
-2@@ -357,7 +357,8 @@ phys_addr_t __init_memblock memblock_find_in_range(phys_addr_t start,return ret; } -static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r) +void __init_memblock memblock_remove_region(struct memblock_type *type, +unsigned long r) { type->total_size -= type->regions[r].size; memmove(&type->regions[r], &type->regions[r + 1],@@ -724,7 +725,7 @@ int __init_memblock memblock_add(phys_addr_t base, phys_addr_t size)* Return: * 0 on success, -errno on failure. */ -static int __init_memblock memblock_isolate_range(struct memblock_type *type, +int __init_memblock memblock_isolate_range(struct memblock_type *type,phys_addr_t base,phys_addr_t size,int *start_rgn,int *end_rgn) {@@ -784,7 +785,7 @@ static int __init_memblock memblock_isolate_range(struct memblock_type *type,return 0 ; }
Quoted 9 lines
Quoted 8 lines
mm/memblock.c +1
-1-static int __init_memblock memblock_remove_range(struct memblock_type *type, +int __init_memblock memblock_remove_range(struct memblock_type *type,phys_addr_t base,phys_addr_t size) {int start_rgn, end_rgn;Signature · 1 line
2.7.4
_______________________________________________
kexec mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/kexec
Re: [PATCH v3 3/4] arm64: kdump: support more than one crash kernel regions
Quoted 20 lines · Chen Zhou
Hi Mike,
Quoted 7 lines · Mike Rapoport
Hi,
Quoted 4 lines · Chen Zhou
After commit (arm64: kdump: support reserving crashkernel above 4G), there may be two crash kernel regions, one is below 4G, the other is above 4G.
Quoted 6 lines
Quoted 3 lines
Crash dump kernel reads more than one crash kernel regions via a dtb property under node /chosen,
linux,usable-memory-range = <BASE1 SIZE1 [BASE2 SIZE2]>Somehow I've missed that previously, but how is this supposed to work on EFI systems?
Whatever the way in which the systems work, there is FDT
pointer(__fdt_pointer)in arm64 kernel and file/sys/firmware/fdtwill be created inlate_initcall.
Quoted 6 lines
Kexec-tools read and update file
/sys/firmware/fdtin EFI systems to support kdump to boot capture kernel.For supporting more than one crash kernel regions, kexec-tools make changes accordingly. Details are in below:
http://lists.infradead.org/pipermail/kexec/2019-April/022792.html
Thanks for the clarification!
Quoted 167 lines
Thanks,
Chen ZhouQuoted 163 lines
Quoted 162 lines
Signed-off-by Chen Zhou <[email protected]>--- arch/arm64/mm/init.c | 66 ++++++++++++++++++++++++++++++++++++++++-------- include/linux/memblock.h | 6 +++++ mm/memblock.c | 7 ++--- 3 files changed, 66 insertions(+), 13 deletions(-)arch/arm64/mm/init.c +56
-10@@ -65,6 +65,11 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init;#ifdef CONFIG_KEXEC_CORE +/* at most two crash kernel regions, low_region and high_region */ +#define CRASH_MAX_USABLE_RANGES 2 +#define LOW_REGION_IDX 0 +#define HIGH_REGION_IDX 1 +/* * reserve_crashkernel() - reserves memory for crash kernel * @@ -297,8 +302,8 @@ static int __init early_init_dt_scan_usablemem(unsigned long node, const char *uname, int depth, void *data) { struct memblock_region *usablemem = data; -const __be32 *reg; -int len; +const __be32 *reg, *endp; +int len, nr =0 ;if (depth !=1 ||strcmp (uname,"chosen" ) !=0 )return 0 ;@@ -307,22 +312,63 @@ static int __init early_init_dt_scan_usablemem(unsigned long node,if (!reg || (len < (dt_root_addr_cells + dt_root_size_cells)))return 1 ; - usablemem->base =dt_mem_next_cell (dt_root_addr_cells, ®); - usablemem->size =dt_mem_next_cell (dt_root_size_cells, ®); + endp = reg + (len /sizeof (__be32)); +while ((endp - reg) >= (dt_root_addr_cells + dt_root_size_cells)) { + usablemem[nr].base =dt_mem_next_cell (dt_root_addr_cells, ®); + usablemem[nr].size =dt_mem_next_cell (dt_root_size_cells, ®); + +if (++nr >=CRASH_MAX_USABLE_RANGES ) +break ; + }return 1 ; }static void __init fdt_enforce_memory_region (void ) { -struct memblock_region reg = { - .size =0 , - }; +int i, cnt =0 ; +struct memblock_region regs[CRASH_MAX_USABLE_RANGES ]; + +memset (regs,0 ,sizeof (regs)); +of_scan_flat_dt (early_init_dt_scan_usablemem, regs); + +for (i =0 ; i <CRASH_MAX_USABLE_RANGES ; i++) +if (regs[i].size) + cnt++; +else +break ; + +if (cnt -1 ==LOW_REGION_IDX ) +memblock_cap_memory_range (regs[LOW_REGION_IDX ].base, + regs[LOW_REGION_IDX ].size); +else if (cnt -1 ==HIGH_REGION_IDX ) { +/* +* Two crash kernel regions, cap the memory range +* [regs[LOW_REGION_IDX].base, regs[HIGH_REGION_IDX].end] +* and then remove the memory range in the middle. +*/ +int start_rgn, end_rgn, i, ret; +phys_addr_t mid_base, mid_size; + + mid_base = regs[LOW_REGION_IDX ].base + regs[LOW_REGION_IDX ].size; + mid_size = regs[HIGH_REGION_IDX ].base - mid_base; + ret =memblock_isolate_range (&memblock.memory, mid_base, + mid_size, &start_rgn, &end_rgn); -of_scan_flat_dt (early_init_dt_scan_usablemem, ®); +if (ret) +return ; -if (reg.size) -memblock_cap_memory_range (reg.base, reg.size); +memblock_cap_memory_range (regs[LOW_REGION_IDX ].base, + regs[HIGH_REGION_IDX ].base - + regs[LOW_REGION_IDX ].base + + regs[HIGH_REGION_IDX ].size); +for (i = end_rgn -1 ; i >= start_rgn; i--) { +if (!memblock_is_nomap (&memblock.memory.regions[i])) +memblock_remove_region (&memblock.memory, i); + } +memblock_remove_range (&memblock.reserved, mid_base, + mid_base + mid_size); + } }void __init arm64_memblock_init (void )include/linux/memblock.h +6
-0@@ -110,9 +110,15 @@ void memblock_discard(void);phys_addr_t memblock_find_in_range (phys_addr_t start,phys_addr_t end,phys_addr_t size,phys_addr_t align); +void memblock_remove_region (struct memblock_type *type,unsigned long r);void memblock_allow_resize (void );int memblock_add_node (phys_addr_t base,phys_addr_t size,int nid);int memblock_add (phys_addr_t base,phys_addr_t size); +int memblock_isolate_range (struct memblock_type *type, +phys_addr_t base,phys_addr_t size, +int *start_rgn,int *end_rgn); +int memblock_remove_range (struct memblock_type *type, +phys_addr_t base,phys_addr_t size);int memblock_remove (phys_addr_t base,phys_addr_t size);int memblock_free (phys_addr_t base,phys_addr_t size);int memblock_reserve (phys_addr_t base,phys_addr_t size);mm/memblock.c +4
-3@@ -357,7 +357,8 @@ phys_addr_t __init_memblock memblock_find_in_range(phys_addr_t start,return ret; } -static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r) +void __init_memblock memblock_remove_region(struct memblock_type *type, +unsigned long r) { type->total_size -= type->regions[r].size; memmove(&type->regions[r], &type->regions[r + 1],@@ -724,7 +725,7 @@ int __init_memblock memblock_add(phys_addr_t base, phys_addr_t size)* Return: * 0 on success, -errno on failure. */ -static int __init_memblock memblock_isolate_range(struct memblock_type *type, +int __init_memblock memblock_isolate_range(struct memblock_type *type,phys_addr_t base,phys_addr_t size,int *start_rgn,int *end_rgn) {@@ -784,7 +785,7 @@ static int __init_memblock memblock_isolate_range(struct memblock_type *type,return 0 ; } -static int __init_memblock memblock_remove_range(struct memblock_type *type, +int __init_memblock memblock_remove_range(struct memblock_type *type,phys_addr_t base,phys_addr_t size) {int start_rgn, end_rgn;Signature · 1 line
2.7.4
Signature · 8 lines
Sincerely yours, Mike. _______________________________________________ kexec mailing list [email protected] http://lists.infradead.org/mailman/listinfo/kexec
PATCH v3 4/4 kdump: update Documentation about crashkernel on arm64
In reply to Chen ZhouNow we support crashkernel=X,[high,low] on arm64, update the Documentation.
--- Documentation/admin-guide/kernel-parameters.txt | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
Documentation/admin-guide/kernel-parameters.txt +2 -2
@@ -715,14 +715,14 @@ Documentation/kdump/kdump.txt for an example. crashkernel=size[KMG],high - [KNL, x86_64] range could be above 4G. Allow kernel + [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. It will be ignored if crashkernel=X is specified. crashkernel=size[KMG],low - [KNL, x86_64] range under 4G. When crashkernel=X,high + [KNL, x86_64, arm64] 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
Signature · 7 lines
2.7.4 _______________________________________________ kexec mailing list [email protected] http://lists.infradead.org/mailman/listinfo/kexec