efi: Fix EFI boot with kexec handover (KHO)
Subsystems (5)
v2 cover efi: Fix EFI boot with kexec handover (KHO)
This patch series fixes a kernel panic that occurs when booting with both EFI and KHO (Kexec HandOver) enabled.
The issue arises because EFI's reserve_regions() clears all memory regions with memblock_remove(0, PHYS_ADDR_MAX) before rebuilding them from EFI data. This destroys KHO scratch regions that were set up early during device tree scanning, causing a panic as the kernel has no valid memory regions for early allocations.
The first patch introduces is_kho_boot() to allow early boot components to reliably detect if the kernel was booted via KHO-enabled kexec. The existing kho_is_enabled() only checks the command line and doesn't verify if an actual KHO FDT was passed.
The second patch modifies EFI's reserve_regions() to selectively remove only non-KHO memory regions when KHO is active, preserving the critical scratch regions while still allowing EFI to rebuild its memory map.
The patchset was developed/tested on arm64.
On a side note, I have noticed that kho_populate() calls memblock_set_kho_scratch_only(), but the kho cmdline option is not checked until much later. Therefore, memblock will use only the scratch regions that were passed from the outgoing kernel, even if the incoming kernel doesn't explicitly want that. I am not sure if this is done on purpose, but in any case we can discuss this in another patch, as it is orthogonal to this one.
Main Changes in v2 (smaller changes can be found in individual patches):
- Introduce is_kho_boot()
- Replaced manual loop with for_each_mem_region macro
Evangelos Petrongonas (2):
kexec: introduce is_kho_boot()
efi: Support booting with kexec handover (KHO)
drivers/firmware/efi/efi-init.c | 28 ++++++++++++++++++++++++---- include/linux/kexec_handover.h | 6 ++++++ kernel/kexec_handover.c | 20 ++++++++++++++++++++ 3 files changed, 50 insertions(+), 4 deletions(-)
Signature · 11 lines
2.47.3 Amazon Web Services Development Center Germany GmbH Tamara-Danz-Str. 13 10243 Berlin Geschaeftsfuehrung: Christian Schlaeger, Jonathan Weiss Eingetragen am Amtsgericht Charlottenburg unter HRB 257764 B Sitz: Berlin Ust-ID: DE 365 538 597
PATCH v2 1/2 kexec: introduce is_kho_boot()
During early initialisation, after a kexec, other components, like EFI need to know if a KHO enabled kexec is performed. The kho_is_enabled function is not enough as in the early stages, it only reflects whether the cmdline has KHO enabled, not if an actual KHO FDT exists.
Extend the KHO API with is_kho_boot() to provide a way for components to check if a KHO enabled kexec is performed.
--- include/linux/kexec_handover.h | 6 ++++++ kernel/kexec_handover.c | 20 ++++++++++++++++++++ 2 files changed, 26 insertions(+)
include/linux/kexec_handover.h +6 -0
@@ -40,6 +40,7 @@ struct kho_serialization;#ifdef CONFIG_KEXEC_HANDOVER bool kho_is_enabled (void ); +bool is_kho_boot (void );int kho_preserve_folio (struct folio *folio);int kho_preserve_phys (phys_addr_t phys,size_t size);
@@ -60,6 +61,11 @@ static inline bool kho_is_enabled(void)return false; } +static inline bool is_kho_boot (void ) +{ +return false; +} +static inline int kho_preserve_folio (struct folio *folio) {return -EOPNOTSUPP ;
kernel/kexec_handover.c +20 -0
@@ -925,6 +925,26 @@ static const void *kho_get_fdt(void)return kho_in.fdt_phys ?phys_to_virt (kho_in.fdt_phys) :NULL ; } +/** +* is_kho_boot - check if current kernel was booted via KHO-enabled +* kexec +* +* This function checks if the current kernel was loaded through a kexec +* operation with KHO enabled, by verifying both that KHO is enabled via +* cmdline AND that a valid KHO FDT was passed. +* +* Note: This function returns reliable results only after +* kho_populate() has been called during early boot. Before that, +* it may return false even if KHO data is present. +* +* Return: true if booted via KHO-enabled kexec, false otherwise +*/ +bool is_kho_boot (void ) +{ +return !!kho_get_fdt () && kho_enable; +} +EXPORT_SYMBOL_GPL (is_kho_boot); +/** * kho_retrieve_subtree - retrieve a preserved sub FDT by its name. * @name: the name of the sub FDT passed to kho_add_subtree().
Signature · 12 lines
2.47.3 Amazon Web Services Development Center Germany GmbH Tamara-Danz-Str. 13 10243 Berlin Geschaeftsfuehrung: Christian Schlaeger, Jonathan Weiss Eingetragen am Amtsgericht Charlottenburg unter HRB 257764 B Sitz: Berlin Ust-ID: DE 365 538 597
PATCH v2 2/2 efi: Support booting with kexec handover (KHO)
In reply to Evangelos PetrongonasWhen KHO (Kexec HandOver) is enabled, it sets up scratch memory regions early during device tree scanning. After kexec, the new kernel exclusively uses this region for memory allocations during boot up to the initialization of the page allocator
However, when booting with EFI, EFI's reserve_regions() uses memblock_remove(0, PHYS_ADDR_MAX) to clear all memory regions before rebuilding them from EFI data. This destroys KHO scratch regions and their flags, thus causing a kernel panic, as there are no scratch memory regions.
Instead of wholesale removal, iterate through memory regions and only remove non-KHO ones. This preserves KHO scratch regions, which are good known memory, while still allowing EFI to rebuild its memory map.
---
Changes in v2:
- Replaced the for loop with for_each_mem_region
- Fixed comment indentation
- Amended commit message to specify that scratch regions are known good regions
drivers/firmware/efi/efi-init.c | 28 ++++++++++++++++++++++++---- 1 file changed, 24 insertions(+), 4 deletions(-)
drivers/firmware/efi/efi-init.c +24 -4
@@ -12,6 +12,7 @@#include <linux/efi.h> #include <linux/fwnode.h> #include <linux/init.h> +#include <linux/kexec_handover.h> #include <linux/memblock.h> #include <linux/mm_types.h> #include <linux/of.h>
@@ -164,12 +165,31 @@ static __init void reserve_regions(void)pr_info ("Processing EFI memory map:\n" );/* -* Discard memblocks discovered so far: if there are any at this -* point, they originate from memory nodes in the DT, and UEFI -* uses its own memory map instead. +* Discard memblocks discovered so far except for KHO scratch +* regions. Most memblocks at this point originate from memory nodes +* in the DT and UEFI uses its own memory map instead. However, if +* KHO is enabled, scratch regions must be preserved. */ memblock_dump_all (); -memblock_remove (0 ,PHYS_ADDR_MAX ); + +if (is_kho_boot ()) { +struct memblock_region *r; + +/* Remove all non-KHO regions */ +for_each_mem_region (r) { +if (!memblock_is_kho_scratch (r)) { +memblock_remove (r->base, r->size); + r--; + } + } + }else { +/* +* KHO is disabled. Discard memblocks discovered so far: +* if there are any at this point, they originate from memory +* nodes in the DT, and UEFI uses its own memory map instead. +*/ +memblock_remove (0 ,PHYS_ADDR_MAX ); + }for_each_efi_memory_desc (md) { paddr = md->phys_addr;
Signature · 12 lines
2.47.3 Amazon Web Services Development Center Germany GmbH Tamara-Danz-Str. 13 10243 Berlin Geschaeftsfuehrung: Christian Schlaeger, Jonathan Weiss Eingetragen am Amtsgericht Charlottenburg unter HRB 257764 B Sitz: Berlin Ust-ID: DE 365 538 597
Re: [PATCH v2 2/2] efi: Support booting with kexec handover (KHO)
Quoted 39 lines · Evangelos Petrongonas
When KHO (Kexec HandOver) is enabled, it sets up scratch memory regions early during device tree scanning. After kexec, the new kernel exclusively uses this region for memory allocations during boot up to the initialization of the page allocator
However, when booting with EFI, EFI's
reserve_regions()usesmemblock_remove(0, PHYS_ADDR_MAX)to clear all memory regions before rebuilding them from EFI data. This destroys KHO scratch regions and their flags, thus causing a kernel panic, as there are no scratch memory regions.Instead of wholesale removal, iterate through memory regions and only remove non-KHO ones. This preserves KHO scratch regions, which are good known memory, while still allowing EFI to rebuild its memory map.
Signed-off-by Evangelos Petrongonas <[email protected]>---
Changes in v2:
- Replaced the for loop withfor_each_mem_region
- Fixed comment indentation
- Amended commit message to specify that scratch regions are known good regionsdrivers/firmware/efi/efi-init.c | 28 ++++++++++++++++++++++++---- 1 file changed, 24 insertions(+), 4 deletions(-)drivers/firmware/efi/efi-init.c +1
-0@@ -12,6 +12,7 @@#include <linux/efi.h> #include <linux/fwnode.h> #include <linux/init.h> +#include <linux/kexec_handover.h> #include <linux/memblock.h> #include <linux/mm_types.h> #include <linux/of.h>
drivers/firmware/efi/efi-init.c static __init void reserve_regions(void) · line 171
Quoted 10 lines
drivers/firmware/efi/efi-init.c +4
-3@@ -164,12 +165,31 @@ static __init void reserve_regions(void)pr_info ("Processing EFI memory map:\n" );/* -* Discard memblocks discovered so far: if there are any at this -* point, they originate from memory nodes in the DT, and UEFI -* uses its own memory map instead. +* Discard memblocks discovered so far except for KHO scratch +* regions. Most memblocks at this point originate from memory nodes +* in the DT and UEFI uses its own memory map instead. However, if +* KHO is enabled, scratch regions must be preserved.
I'd add that KHO scratch regions are good know memory here as well. With that
Quoted 27 lines
drivers/firmware/efi/efi-init.c +19
-1*/ memblock_dump_all (); -memblock_remove (0 ,PHYS_ADDR_MAX ); + +if (is_kho_boot ()) { +struct memblock_region *r; + +/* Remove all non-KHO regions */ +for_each_mem_region (r) { +if (!memblock_is_kho_scratch (r)) { +memblock_remove (r->base, r->size); + r--; + } + } + }else { +/* +* KHO is disabled. Discard memblocks discovered so far: +* if there are any at this point, they originate from memory +* nodes in the DT, and UEFI uses its own memory map instead. +*/ +memblock_remove (0 ,PHYS_ADDR_MAX ); + }for_each_efi_memory_desc (md) { paddr = md->phys_addr;Signature · 1 line
2.47.3
Signature · 2 lines
Sincerely yours, Mike.
Mike Rapoport
maintainerQuoted 21 lines · Evangelos Petrongonas
This patch series fixes a kernel panic that occurs when booting with both EFI and KHO (Kexec HandOver) enabled.
The issue arises because EFI's
reserve_regions()clears all memory regions withmemblock_remove(0, PHYS_ADDR_MAX)before rebuilding them from EFI data. This destroys KHO scratch regions that were set up early during device tree scanning, causing a panic as the kernel has no valid memory regions for early allocations.The first patch introduces
is_kho_boot()to allow early boot components to reliably detect if the kernel was booted via KHO-enabled kexec. The existingkho_is_enabled()only checks the command line and doesn't verify if an actual KHO FDT was passed.The second patch modifies EFI's
reserve_regions()to selectively remove only non-KHO memory regions when KHO is active, preserving the critical scratch regions while still allowing EFI to rebuild its memory map.The patchset was developed/tested on arm64.
Quoted 7 lines
On a side note, I have noticed that
kho_populate()callsmemblock_set_kho_scratch_only(), but thekhocmdline option is not checked until much later. Therefore, memblock will use only the scratch regions that were passed from the outgoing kernel, even if the incoming kernel doesn't explicitly want that. I am not sure if this is done on purpose, but in any case we can discuss this in another patch, as it is orthogonal to this one.
kho_populate runs earlier than we parse the command line, so there is an implicit assumption that we are going through KHO-enabled kexec if FDT was passed to the new kernel.
I believe the best way is to document that and make it more explicit that kho command line parameter only affects the "out" part.
Quoted 15 lines
Main Changes in v2 (smaller changes can be found in individual patches):
- Introduceis_kho_boot()
- Replaced manual loop withfor_each_mem_regionmacroEvangelos Petrongonas (2):
kexec: introduceis_kho_boot()
efi: Support booting with kexec handover (KHO)drivers/firmware/efi/efi-init.c | 28 ++++++++++++++++++++++++---- include/linux/kexec_handover.h | 6 ++++++ kernel/kexec_handover.c | 20 ++++++++++++++++++++ 3 files changed, 50 insertions(+), 4 deletions(-)Signature · 1 line
2.47.3
Signature · 2 lines
Sincerely yours, Mike.
Re: [PATCH v2 1/2] kexec: introduce is_kho_boot()
In reply to Evangelos PetrongonasQuoted 53 lines · Evangelos Petrongonas
During early initialisation, after a kexec, other components, like EFI need to know if a KHO enabled kexec is performed. The
kho_is_enabledfunction is not enough as in the early stages, it only reflects whether the cmdline has KHO enabled, not if an actual KHO FDT exists.Extend the KHO API with
is_kho_boot()to provide a way for components to check if a KHO enabled kexec is performed.Signed-off-by Evangelos Petrongonas <[email protected]>--- include/linux/kexec_handover.h | 6 ++++++ kernel/kexec_handover.c | 20 ++++++++++++++++++++ 2 files changed, 26 insertions(+)include/linux/kexec_handover.h +6
-0@@ -40,6 +40,7 @@ struct kho_serialization;#ifdef CONFIG_KEXEC_HANDOVER bool kho_is_enabled (void ); +bool is_kho_boot (void );int kho_preserve_folio (struct folio *folio);int kho_preserve_phys (phys_addr_t phys,size_t size);@@ -60,6 +61,11 @@ static inline bool kho_is_enabled(void)return false; } +static inline bool is_kho_boot (void ) +{ +return false; +} +static inline int kho_preserve_folio (struct folio *folio) {return -EOPNOTSUPP ;kernel/kexec_handover.c +7
-0@@ -925,6 +925,26 @@ static const void *kho_get_fdt(void)return kho_in.fdt_phys ?phys_to_virt (kho_in.fdt_phys) :NULL ; } +/** +* is_kho_boot - check if current kernel was booted via KHO-enabled +* kexec +* +* This function checks if the current kernel was loaded through a kexec +* operation with KHO enabled, by verifying both that KHO is enabled via +* cmdline AND that a valid KHO FDT was passed.
kernel/kexec_handover.c static const void *kho_get_fdt(void) · line 944
Quoted 10 lines
kernel/kexec_handover.c +10
-0+* +* Note: This function returns reliable results only after +* kho_populate() has been called during early boot. Before that, +* it may return false even if KHO data is present. +* +* Return: true if booted via KHO-enabled kexec, false otherwise +*/ +bool is_kho_boot (void ) +{ +return !!kho_get_fdt () && kho_enable;
Let's drop '&& kho_enable' and presume that a kernel that booted with KHO FDT goes through KHO revival.
With that change feel free to add
Quoted 8 lines
kernel/kexec_handover.c +3
-0+} +EXPORT_SYMBOL_GPL (is_kho_boot); +/** * kho_retrieve_subtree - retrieve a preserved sub FDT by its name. * @name: the name of the sub FDT passed to kho_add_subtree(). Signature · 1 line
2.47.3
Signature · 2 lines
Sincerely yours, Mike.