From: Anthony Yznaga <anthony.yznaga@oracle.com>
To: linux-mm@kvack.org, linux-kernel@vger.kernel.org
Cc: tglx@linutronix.de, mingo@redhat.com, bp@alien8.de,
x86@kernel.org, hpa@zytor.com, dave.hansen@linux.intel.com,
luto@kernel.org, peterz@infradead.org, rppt@kernel.org,
akpm@linux-foundation.org, ebiederm@xmission.com,
keescook@chromium.org, graf@amazon.com, jason.zeng@intel.com,
lei.l.li@intel.com, steven.sistare@oracle.com,
fam.zheng@bytedance.com, mgalaxy@akamai.com,
kexec@lists.infradead.org
Subject: [RFC v3 20/21] x86/KASLR: PKRAM: support physical kaslr
Date: Wed, 26 Apr 2023 17:08:56 -0700 [thread overview]
Message-ID: <1682554137-13938-21-git-send-email-anthony.yznaga@oracle.com> (raw)
In-Reply-To: <1682554137-13938-1-git-send-email-anthony.yznaga@oracle.com>
Avoid regions of memory that contain preserved pages when computing
slots used to select where to put the decompressed kernel.
Signed-off-by: Anthony Yznaga <anthony.yznaga@oracle.com>
---
arch/x86/boot/compressed/Makefile | 3 ++
arch/x86/boot/compressed/kaslr.c | 10 +++-
arch/x86/boot/compressed/misc.h | 10 ++++
arch/x86/boot/compressed/pkram.c | 110 ++++++++++++++++++++++++++++++++++++++
mm/pkram.c | 2 +-
5 files changed, 132 insertions(+), 3 deletions(-)
create mode 100644 arch/x86/boot/compressed/pkram.c
diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile
index 6b6cfe607bdb..d9a5af94a797 100644
--- a/arch/x86/boot/compressed/Makefile
+++ b/arch/x86/boot/compressed/Makefile
@@ -103,6 +103,9 @@ ifdef CONFIG_X86_64
vmlinux-objs-$(CONFIG_AMD_MEM_ENCRYPT) += $(obj)/mem_encrypt.o
vmlinux-objs-y += $(obj)/pgtable_64.o
vmlinux-objs-$(CONFIG_AMD_MEM_ENCRYPT) += $(obj)/sev.o
+ifdef CONFIG_RANDOMIZE_BASE
+ vmlinux-objs-$(CONFIG_PKRAM) += $(obj)/pkram.o
+endif
endif
vmlinux-objs-$(CONFIG_ACPI) += $(obj)/acpi.o
diff --git a/arch/x86/boot/compressed/kaslr.c b/arch/x86/boot/compressed/kaslr.c
index 454757fbdfe5..047b8b9a0799 100644
--- a/arch/x86/boot/compressed/kaslr.c
+++ b/arch/x86/boot/compressed/kaslr.c
@@ -436,6 +436,7 @@ static bool mem_avoid_overlap(struct mem_vector *img,
struct setup_data *ptr;
u64 earliest = img->start + img->size;
bool is_overlapping = false;
+ struct mem_vector avoid;
for (i = 0; i < MEM_AVOID_MAX; i++) {
if (mem_overlaps(img, &mem_avoid[i]) &&
@@ -449,8 +450,6 @@ static bool mem_avoid_overlap(struct mem_vector *img,
/* Avoid all entries in the setup_data linked list. */
ptr = (struct setup_data *)(unsigned long)boot_params->hdr.setup_data;
while (ptr) {
- struct mem_vector avoid;
-
avoid.start = (unsigned long)ptr;
avoid.size = sizeof(*ptr) + ptr->len;
@@ -475,6 +474,12 @@ static bool mem_avoid_overlap(struct mem_vector *img,
ptr = (struct setup_data *)(unsigned long)ptr->next;
}
+ if (pkram_has_overlap(img, &avoid) && (avoid.start < earliest)) {
+ *overlap = avoid;
+ earliest = overlap->start;
+ is_overlapping = true;
+ }
+
return is_overlapping;
}
@@ -836,6 +841,7 @@ void choose_random_location(unsigned long input,
return;
}
+ pkram_init();
boot_params->hdr.loadflags |= KASLR_FLAG;
if (IS_ENABLED(CONFIG_X86_32))
diff --git a/arch/x86/boot/compressed/misc.h b/arch/x86/boot/compressed/misc.h
index 20118fb7c53b..01ff5e507064 100644
--- a/arch/x86/boot/compressed/misc.h
+++ b/arch/x86/boot/compressed/misc.h
@@ -124,6 +124,16 @@ static inline void console_init(void)
{ }
#endif
+#ifdef CONFIG_PKRAM
+void pkram_init(void);
+int pkram_has_overlap(struct mem_vector *entry, struct mem_vector *overlap);
+#else
+static inline void pkram_init(void) { }
+static inline int pkram_has_overlap(struct mem_vector *entry,
+ struct mem_vector *overlap)
+{ return 0; }
+#endif
+
#ifdef CONFIG_AMD_MEM_ENCRYPT
void sev_enable(struct boot_params *bp);
void snp_check_features(void);
diff --git a/arch/x86/boot/compressed/pkram.c b/arch/x86/boot/compressed/pkram.c
new file mode 100644
index 000000000000..19267ca2ce8e
--- /dev/null
+++ b/arch/x86/boot/compressed/pkram.c
@@ -0,0 +1,110 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include "misc.h"
+
+#define PKRAM_MAGIC 0x706B726D
+
+struct pkram_super_block {
+ __u32 magic;
+
+ __u64 node_pfn;
+ __u64 region_list_pfn;
+ __u64 nr_regions;
+};
+
+struct pkram_region {
+ phys_addr_t base;
+ phys_addr_t size;
+};
+
+struct pkram_region_list {
+ __u64 prev_pfn;
+ __u64 next_pfn;
+
+ struct pkram_region regions[0];
+};
+
+#define PKRAM_REGIONS_LIST_MAX \
+ ((PAGE_SIZE-sizeof(struct pkram_region_list))/sizeof(struct pkram_region))
+
+static u64 pkram_sb_pfn;
+static struct pkram_super_block *pkram_sb;
+
+void pkram_init(void)
+{
+ struct pkram_super_block *sb;
+ char arg[32];
+
+ if (cmdline_find_option("pkram", arg, sizeof(arg)) > 0) {
+ if (kstrtoull(arg, 16, &pkram_sb_pfn) != 0)
+ return;
+ } else
+ return;
+
+ sb = (struct pkram_super_block *)(pkram_sb_pfn << PAGE_SHIFT);
+ if (sb->magic != PKRAM_MAGIC) {
+ debug_putstr("PKRAM: invalid super block\n");
+ return;
+ }
+
+ pkram_sb = sb;
+}
+
+static struct pkram_region *pkram_first_region(struct pkram_super_block *sb,
+ struct pkram_region_list **rlp, int *idx)
+{
+ if (!sb || !sb->region_list_pfn)
+ return NULL;
+
+ *rlp = (struct pkram_region_list *)(sb->region_list_pfn << PAGE_SHIFT);
+ *idx = 0;
+
+ return &(*rlp)->regions[0];
+}
+
+static struct pkram_region *pkram_next_region(struct pkram_region_list **rlp, int *idx)
+{
+ struct pkram_region_list *rl = *rlp;
+ int i = *idx;
+
+ i++;
+ if (i >= PKRAM_REGIONS_LIST_MAX) {
+ if (!rl->next_pfn) {
+ debug_putstr("PKRAM: no more pkram_region_list pages\n");
+ return NULL;
+ }
+ rl = (struct pkram_region_list *)(rl->next_pfn << PAGE_SHIFT);
+ *rlp = rl;
+ i = 0;
+ }
+ *idx = i;
+
+ if (rl->regions[i].size == 0)
+ return NULL;
+
+ return &rl->regions[i];
+}
+
+int pkram_has_overlap(struct mem_vector *entry, struct mem_vector *overlap)
+{
+ struct pkram_region_list *rl;
+ struct pkram_region *r;
+ int idx;
+
+ r = pkram_first_region(pkram_sb, &rl, &idx);
+
+ while (r) {
+ if (r->base + r->size <= entry->start) {
+ r = pkram_next_region(&rl, &idx);
+ continue;
+ }
+ if (r->base >= entry->start + entry->size)
+ return 0;
+
+ overlap->start = r->base;
+ overlap->size = r->size;
+ return 1;
+ }
+
+ return 0;
+}
diff --git a/mm/pkram.c b/mm/pkram.c
index f38236e5d836..a3e045b8dfe4 100644
--- a/mm/pkram.c
+++ b/mm/pkram.c
@@ -96,7 +96,7 @@ struct pkram_region_list {
__u64 prev_pfn;
__u64 next_pfn;
- struct pkram_region regions[0];
+ struct pkram_region regions[];
};
#define PKRAM_REGIONS_LIST_MAX \
--
1.9.4
next prev parent reply other threads:[~2023-04-27 0:10 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-27 0:08 [RFC v3 00/21] Preserved-over-Kexec RAM Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 01/21] mm: add PKRAM API stubs and Kconfig Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 02/21] mm: PKRAM: implement node load and save functions Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 03/21] mm: PKRAM: implement object " Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 04/21] mm: PKRAM: implement folio stream operations Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 05/21] mm: PKRAM: implement byte " Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 06/21] mm: PKRAM: link nodes by pfn before reboot Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 07/21] mm: PKRAM: introduce super block Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 08/21] PKRAM: track preserved pages in a physical mapping pagetable Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 09/21] PKRAM: pass a list of preserved ranges to the next kernel Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 10/21] PKRAM: prepare for adding preserved ranges to memblock reserved Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 11/21] mm: PKRAM: reserve preserved memory at boot Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 12/21] PKRAM: free the preserved ranges list Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 13/21] PKRAM: prevent inadvertent use of a stale superblock Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 14/21] PKRAM: provide a way to ban pages from use by PKRAM Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 15/21] kexec: PKRAM: prevent kexec clobbering preserved pages in some cases Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 16/21] PKRAM: provide a way to check if a memory range has preserved pages Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 17/21] kexec: PKRAM: avoid clobbering already " Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 18/21] mm: PKRAM: allow preserved memory to be freed from userspace Anthony Yznaga
2023-04-27 0:08 ` [RFC v3 19/21] PKRAM: disable feature when running the kdump kernel Anthony Yznaga
2023-04-27 0:08 ` Anthony Yznaga [this message]
2023-04-27 0:08 ` [RFC v3 21/21] x86/boot/compressed/64: use 1GB pages for mappings Anthony Yznaga
2023-04-27 18:40 ` H. Peter Anvin
2023-04-27 22:38 ` Anthony Yznaga
2023-05-26 13:57 ` [RFC v3 00/21] Preserved-over-Kexec RAM Gowans, James
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1682554137-13938-21-git-send-email-anthony.yznaga@oracle.com \
--to=anthony.yznaga@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=bp@alien8.de \
--cc=dave.hansen@linux.intel.com \
--cc=ebiederm@xmission.com \
--cc=fam.zheng@bytedance.com \
--cc=graf@amazon.com \
--cc=hpa@zytor.com \
--cc=jason.zeng@intel.com \
--cc=keescook@chromium.org \
--cc=kexec@lists.infradead.org \
--cc=lei.l.li@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@kernel.org \
--cc=mgalaxy@akamai.com \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=rppt@kernel.org \
--cc=steven.sistare@oracle.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox