[PATCH v3 5/5] mm/memory-failure: efi: replay the poisioned page in the next kernel
From: Breno Leitao
Date: Wed Aug 26 2026 - 08:08:19 EST
The bitmap an earlier kernel filled in rides the EFI system table into
this one, but nothing reads it back until now.
Add efi_offline_poisoned_memory(), which vets the table header, copies
the geometry out of it, and hands every frame of every set unit to
hwpoison_boot_pfn(). The bitmap runs to megabytes on a large machine, so
it is mapped and walked a page at a time rather than in one go.
Call it through hwpoison_init_boot() from mm_core_init() right after
memblock_free_all(), the first point at which the recorded frames have
struct pages. They end up in the state a frame poisoned by this kernel
would be in, so the placement check that already understands PG_hwpoison
covers them too.
Suggested-by: Kiryl Shutsemau <kas@xxxxxxxxxx>
Signed-off-by: Breno Leitao <leitao@xxxxxxxxxx>
---
drivers/firmware/efi/poison.c | 113 +++++++++++++++++++++++++++++++++++++++++-
include/linux/efi.h | 2 +
include/linux/mm.h | 5 ++
mm/memory-failure.c | 6 +++
mm/mm_init.c | 1 +
5 files changed, 126 insertions(+), 1 deletion(-)
diff --git a/drivers/firmware/efi/poison.c b/drivers/firmware/efi/poison.c
index d6855e712832c..179843277457f 100644
--- a/drivers/firmware/efi/poison.c
+++ b/drivers/firmware/efi/poison.c
@@ -9,14 +9,20 @@
#define pr_fmt(fmt) "efi: " fmt
-#include <linux/bitmap.h>
#include <linux/efi.h>
#include <linux/io.h>
#include <linux/log2.h>
#include <linux/memblock.h>
+#include <linux/minmax.h>
#include <linux/mm.h>
#include <linux/overflow.h>
+struct efi_poison_geometry {
+ u64 table; /* phys address of the table */
+ u64 unit_size;
+ u64 bitmap_size;
+};
+
static struct linux_efi_poisoned_memory *efi_poison __ro_after_init;
static u64 efi_poison_nbits __ro_after_init;
@@ -124,3 +130,108 @@ void efi_hwpoison_record_pfn(unsigned long pfn)
set_bit(unit, efi_poison->bitmap);
}
+
+static bool __init efi_poison_read_geometry(u64 ppm,
+ struct efi_poison_geometry *g)
+{
+ struct linux_efi_poisoned_memory *pm;
+ bool valid;
+
+ pm = early_memremap(ppm, sizeof(*pm));
+ if (!pm) {
+ pr_warn("Could not map poisoned-memory table\n");
+ return false;
+ }
+
+ valid = efi_poison_table_valid(pm);
+ if (valid) {
+ g->table = ppm;
+ g->unit_size = pm->unit_size;
+ g->bitmap_size = efi_poison_usable_size(pm);
+ } else {
+ /* Keep the runtime side off a table this pass rejected. */
+ efi.poisoned_memory = EFI_INVALID_TABLE_ADDR;
+ }
+
+ early_memunmap(pm, sizeof(*pm));
+
+ return valid;
+}
+
+static unsigned long __init
+efi_poison_offline_unit(const struct efi_poison_geometry *g, u64 unit)
+{
+ unsigned long pfn = PHYS_PFN(unit * g->unit_size);
+ unsigned long i, nr_pages = 0;
+
+ for (i = 0; i < g->unit_size >> PAGE_SHIFT; i++)
+ nr_pages += hwpoison_boot_pfn(pfn + i);
+
+ return nr_pages;
+}
+
+static long __init efi_poison_walk_chunk(const struct efi_poison_geometry *g,
+ u64 off, unsigned long *nr_units)
+{
+ u64 chunk = min_t(u64, PAGE_SIZE, g->bitmap_size - off);
+ unsigned long bit, nbits = chunk * BITS_PER_BYTE;
+ unsigned long *map, nr_pages = 0;
+
+ map = early_memremap(g->table + offsetof(struct linux_efi_poisoned_memory,
+ bitmap) + off, chunk);
+ if (!map)
+ return -1;
+
+ for_each_set_bit(bit, map, nbits) {
+ nr_pages += efi_poison_offline_unit(g, off * BITS_PER_BYTE + bit);
+ (*nr_units)++;
+ }
+
+ early_memunmap(map, chunk);
+
+ return nr_pages;
+}
+
+static long __init efi_poison_walk(const struct efi_poison_geometry *g,
+ unsigned long *nr_units)
+{
+ unsigned long nr_pages = 0;
+ u64 off;
+
+ for (off = 0; off < g->bitmap_size; off += PAGE_SIZE) {
+ long nr = efi_poison_walk_chunk(g, off, nr_units);
+
+ if (nr < 0) {
+ pr_warn("Could not map poisoned-memory bitmap\n");
+ return -1;
+ }
+ nr_pages += nr;
+ }
+
+ return nr_pages;
+}
+
+void __init efi_offline_poisoned_memory(void)
+{
+ struct efi_poison_geometry g;
+ unsigned long nr_units = 0;
+ long nr_pages, expected;
+
+ if (efi.poisoned_memory == EFI_INVALID_TABLE_ADDR)
+ return;
+
+ if (!efi_poison_read_geometry(efi.poisoned_memory, &g))
+ return;
+
+ nr_pages = efi_poison_walk(&g, &nr_units);
+ if (nr_pages < 0)
+ return;
+
+ if (nr_pages)
+ pr_info("poisoned %ld page(s) inherited across kexec\n", nr_pages);
+
+ expected = nr_units * (g.unit_size >> PAGE_SHIFT);
+ if (nr_pages < expected)
+ pr_warn("%ld inherited poisoned page(s) could not be taken out of use\n",
+ expected - nr_pages);
+}
diff --git a/include/linux/efi.h b/include/linux/efi.h
index d579d75372248..03ba40ff70e7e 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -1285,8 +1285,10 @@ struct linux_efi_poisoned_memory {
#define EFI_POISON_UNIT_SIZE SZ_2M
#ifdef CONFIG_EFI_POISONED_MEMORY
+void efi_offline_poisoned_memory(void);
void efi_hwpoison_record_pfn(unsigned long pfn);
#else
+static inline void efi_offline_poisoned_memory(void) { }
static inline void efi_hwpoison_record_pfn(unsigned long pfn) { }
#endif
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 52a00a0e090c7..092220943531e 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -5104,12 +5104,17 @@ extern const struct attribute_group memory_failure_attr_group;
extern void memory_failure_queue(unsigned long pfn, int flags);
void num_poisoned_pages_inc(unsigned long pfn);
void num_poisoned_pages_sub(unsigned long pfn, long i);
+void __init hwpoison_init_boot(void);
bool __init hwpoison_boot_pfn(unsigned long pfn);
#else
static inline void memory_failure_queue(unsigned long pfn, int flags)
{
}
+static inline void hwpoison_init_boot(void)
+{
+}
+
static inline void num_poisoned_pages_inc(unsigned long pfn)
{
}
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 713fb2f8e0332..1d6a472267bdb 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -121,6 +121,12 @@ bool __init hwpoison_boot_pfn(unsigned long pfn)
return true;
}
+/* The EFI table is the only source of inherited poison today. */
+void __init hwpoison_init_boot(void)
+{
+ efi_offline_poisoned_memory();
+}
+
/**
* MF_ATTR_RO - Create sysfs entry for each memory failure statistics.
* @_name: name of the file in the per NUMA sysfs directory.
diff --git a/mm/mm_init.c b/mm/mm_init.c
index ddda9d6837f32..f94fa221da0b8 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -2667,6 +2667,7 @@ void __init mm_core_init(void)
kho_memory_init();
memblock_free_all();
+ hwpoison_init_boot();
mem_init();
kmem_cache_init();
/*
--
2.53.0-Meta