efi: Fixes for EFI_MEMORY_SP memory on RISC-V and ARM64
v2 cover efi: Fixes for EFI_MEMORY_SP memory on RISC-V and ARM64
Two small fixes to enable the use soft-reserved/special-purpose memory (EFI_MEMORY_SP) with dax_kmem on RISC-V (and ARM64, I think, though I don't have a platform to test it on).
Patch 1 fixes a trivial integer narrowing bug. Patch 2 prevents adding memblocks for soft-reserved memory so that it can later be hotplugged by dax_kmem.
Tested on a RISC-V platform that presents a range of EFI_MEMORY_SP with Bjorn's MEMORY_HOTPLUG series[0] applied.
[0]: https://lore.kernel.org/lkml/[email protected]/
v1->v2: address comments from Ard
Andrew Bresticker (2):
efi: runtime: Fix potential overflow of soft-reserved region size
efi: Don't add memblocks for soft-reserved memory
drivers/firmware/efi/arm-runtime.c | 2 +- drivers/firmware/efi/efi-init.c | 19 ++++++++++--------- drivers/firmware/efi/riscv-runtime.c | 2 +- 3 files changed, 12 insertions(+), 11 deletions(-)
Signature · 1 line
2.34.1
PATCH v2 1/2 efi: runtime: Fix potential overflow of soft-reserved region size
md_size will have been narrowed if we have >= 4GB worth of pages in a soft-reserved region.
---
v2: use u64 instead of unsigned long
--- drivers/firmware/efi/arm-runtime.c | 2 +- drivers/firmware/efi/riscv-runtime.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/firmware/efi/arm-runtime.c +1 -1
@@ -107,7 +107,7 @@ static int __init arm_enable_runtime_services(void)efi_memory_desc_t *md;for_each_efi_memory_desc (md) { -int md_size = md->num_pages <<EFI_PAGE_SHIFT ; +u64 md_size = md->num_pages <<EFI_PAGE_SHIFT ;struct resource *res;if (!(md->attribute &EFI_MEMORY_SP ))
drivers/firmware/efi/riscv-runtime.c +1 -1
@@ -85,7 +85,7 @@ static int __init riscv_enable_runtime_services(void)efi_memory_desc_t *md;for_each_efi_memory_desc (md) { -int md_size = md->num_pages <<EFI_PAGE_SHIFT ; +u64 md_size = md->num_pages <<EFI_PAGE_SHIFT ;struct resource *res;if (!(md->attribute &EFI_MEMORY_SP ))
Signature · 1 line
2.34.1
PATCH v2 2/2 efi: Don't add memblocks for soft-reserved memory
In reply to Andrew BrestickerAdding memblocks for soft-reserved regions prevents them from later being hotplugged in by dax_kmem.
---
v2: only skip adding memblocks for soft-reserved mem
--- drivers/firmware/efi/efi-init.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-)
drivers/firmware/efi/efi-init.c +10 -9
@@ -143,15 +143,6 @@ static __init int is_usable_memory(efi_memory_desc_t *md)case EFI_BOOT_SERVICES_DATA :case EFI_CONVENTIONAL_MEMORY :case EFI_PERSISTENT_MEMORY : -/* -* Special purpose memory is 'soft reserved', which means it -* is set aside initially, but can be hotplugged back in or -* be assigned to the dax driver after boot. -*/ -if (efi_soft_reserve_enabled () && - (md->attribute &EFI_MEMORY_SP )) -return false; -/* * According to the spec, these regions are no longer reserved * after calling ExitBootServices(). However, we can only use
@@ -196,6 +187,16 @@ static __init void reserve_regions(void) size = npages <<PAGE_SHIFT ;if (is_memory (md)) { +/* +* Special purpose memory is 'soft reserved', which +* means it is set aside initially. Don't add a memblock +* for it now so that it can be hotplugged back in or +* be assigned to the dax driver after boot. +*/ +if (efi_soft_reserve_enabled () && + (md->attribute &EFI_MEMORY_SP )) +continue ; +early_init_dt_add_memory_arch (paddr, size);if (!is_usable_memory (md))
Signature · 1 line
2.34.1
Ard Biesheuvel
maintainerQuoted 12 lines · Andrew Bresticker
Two small fixes to enable the use soft-reserved/special-purpose memory (
EFI_MEMORY_SP) withdax_kmemon RISC-V (and ARM64, I think, though I don't have a platform to test it on).Patch 1 fixes a trivial integer narrowing bug. Patch 2 prevents adding memblocks for soft-reserved memory so that it can later be hotplugged by
dax_kmem.Tested on a RISC-V platform that presents a range of
EFI_MEMORY_SPwith Bjorn'sMEMORY_HOTPLUGseries[0] applied.
Quoted 8 lines
[0]: https://lore.kernel.org/lkml/[email protected]/
v1->v2: address comments from ArdAndrew Bresticker (2):
efi: runtime: Fix potential overflow of soft-reserved region size
efi: Don't add memblocks for soft-reserved memory
Thanks, I'll take these both as fixes.
Quoted 15 lines · Ard Biesheuvel
Quoted 14 lines · Andrew Bresticker
Two small fixes to enable the use soft-reserved/special-purpose memory (
EFI_MEMORY_SP) withdax_kmemon RISC-V (and ARM64, I think, though I don't have a platform to test it on).Patch 1 fixes a trivial integer narrowing bug. Patch 2 prevents adding memblocks for soft-reserved memory so that it can later be hotplugged by
dax_kmem.Tested on a RISC-V platform that presents a range of
EFI_MEMORY_SPwith Bjorn'sMEMORY_HOTPLUGseries[0] applied.
Quoted 8 lines
Quoted 6 lines
v1->v2: address comments from ArdAndrew Bresticker (2):
efi: runtime: Fix potential overflow of soft-reserved region size
efi: Don't add memblocks for soft-reserved memoryThanks, I'll take these both as fixes.
Thanks!