From: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
To: dave.hansen@intel.com
Cc: kirill.shutemov@linux.intel.com, aaron.lu@intel.com,
ardb@google.com, bagasdotme@gmail.com, bp@alien8.de,
keescook@chromium.org, kexec@lists.infradead.org,
linux-kernel@vger.kernel.org, regressions@lists.linux.de,
tglx@linutronix.de, thomas.lendacky@amd.com, x86@kernel.org
Subject: [PATCHv2] x86/boot/compressed: Reserve more memory for page tables
Date: Fri, 15 Sep 2023 10:02:21 +0300 [thread overview]
Message-ID: <20230915070221.10266-1-kirill.shutemov@linux.intel.com> (raw)
In-Reply-To: <20230914170726.4am7xi36m4hdgiyk@box>
The decompressor has a hard limit on the number of page tables it can
allocate. This limit is defined at compile-time and will cause boot
failure if it is reached.
The kernel is very strict and calculates the limit precisely for the
worst-case scenario based on the current configuration. However, it is
easy to forget to adjust the limit when a new use-case arises. The
worst-case scenario is rarely encountered during sanity checks.
In the case of enabling 5-level paging, a use-case was overlooked. The
limit needs to be increased by one to accommodate the additional level.
This oversight went unnoticed until Aaron attempted to run the kernel
via kexec with 5-level paging and unaccepted memory enabled.
Update wost-case calculations to include 5-level paging.
To address this issue, let's allocate some extra space for page tables.
128K should be sufficient for any use-case. The logic can be simplified
by using a single value for all kernel configurations.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reported-by: Aaron Lu <aaron.lu@intel.com>
Fixes: 34bbb0009f3b ("x86/boot/compressed: Enable 5-level paging during decompression stage")
---
arch/x86/boot/compressed/ident_map_64.c | 8 +++++
arch/x86/include/asm/boot.h | 47 +++++++++++++++++--------
2 files changed, 40 insertions(+), 15 deletions(-)
diff --git a/arch/x86/boot/compressed/ident_map_64.c b/arch/x86/boot/compressed/ident_map_64.c
index bcc956c17872..08f93b0401bb 100644
--- a/arch/x86/boot/compressed/ident_map_64.c
+++ b/arch/x86/boot/compressed/ident_map_64.c
@@ -59,6 +59,14 @@ static void *alloc_pgt_page(void *context)
return NULL;
}
+ /* Consumed more tables than expected? */
+ if (pages->pgt_buf_offset == BOOT_PGT_SIZE_WARN) {
+ debug_putstr("pgt_buf running low in " __FILE__ "\n");
+ debug_putstr("Need to raise BOOT_PGT_SIZE?\n");
+ debug_putaddr(pages->pgt_buf_offset);
+ debug_putaddr(pages->pgt_buf_size);
+ }
+
entry = pages->pgt_buf + pages->pgt_buf_offset;
pages->pgt_buf_offset += PAGE_SIZE;
diff --git a/arch/x86/include/asm/boot.h b/arch/x86/include/asm/boot.h
index 9191280d9ea3..215d37f7dde8 100644
--- a/arch/x86/include/asm/boot.h
+++ b/arch/x86/include/asm/boot.h
@@ -40,23 +40,40 @@
#ifdef CONFIG_X86_64
# define BOOT_STACK_SIZE 0x4000
-# define BOOT_INIT_PGT_SIZE (6*4096)
-# ifdef CONFIG_RANDOMIZE_BASE
/*
- * Assuming all cross the 512GB boundary:
- * 1 page for level4
- * (2+2)*4 pages for kernel, param, cmd_line, and randomized kernel
- * 2 pages for first 2M (video RAM: CONFIG_X86_VERBOSE_BOOTUP).
- * Total is 19 pages.
+ * Used by decompressor's startup_32() to allocate page tables for identity
+ * mapping of the 4G of RAM in 4-level paging mode:
+ * - 1 level4 table;
+ * - 1 level3 table;
+ * - 4 level2 table that maps everything with 2M pages;
+ *
+ * The additional level5 table needed for 5-level paging is allocated from
+ * trampoline_32bit memory.
*/
-# ifdef CONFIG_X86_VERBOSE_BOOTUP
-# define BOOT_PGT_SIZE (19*4096)
-# else /* !CONFIG_X86_VERBOSE_BOOTUP */
-# define BOOT_PGT_SIZE (17*4096)
-# endif
-# else /* !CONFIG_RANDOMIZE_BASE */
-# define BOOT_PGT_SIZE BOOT_INIT_PGT_SIZE
-# endif
+# define BOOT_INIT_PGT_SIZE (6*4096)
+
+/*
+ * Total number of page tables kernel_add_identity_map() can allocate,
+ * including page tables consumed by startup_32().
+ *
+ * Worst-case scenario:
+ * - 5-level paging needs 1 level5 table;
+ * - KASLR needs to map kernel, boot_params, cmdline and randomized kernel,
+ * assuming all of them cross 256T boundary:
+ * + 4*2 level4 table;
+ * + 4*2 level3 table;
+ * + 4*2 level2 table;
+ * - X86_VERBOSE_BOOTUP needs to map the first 2M (video RAM):
+ * + 1 level4 table;
+ * + 1 level3 table;
+ * + 1 level2 table;
+ * Total: 28 tables
+ *
+ * Add 4 spare table in case decompressor touches anything beyond what is
+ * accounted above. Warn if it happens.
+ */
+# define BOOT_PGT_SIZE_WARN (28*4096)
+# define BOOT_PGT_SIZE (32*4096)
#else /* !CONFIG_X86_64 */
# define BOOT_STACK_SIZE 0x1000
--
2.41.0
_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec
prev parent reply other threads:[~2023-09-15 7:02 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-14 12:30 [PATCH] x86/boot/compressed: Reserve more memory for page tables Kirill A. Shutemov
2023-09-14 15:51 ` Dave Hansen
2023-09-14 17:07 ` Kirill A. Shutemov
2023-09-15 7:02 ` Kirill A. Shutemov [this message]
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=20230915070221.10266-1-kirill.shutemov@linux.intel.com \
--to=kirill.shutemov@linux.intel.com \
--cc=aaron.lu@intel.com \
--cc=ardb@google.com \
--cc=bagasdotme@gmail.com \
--cc=bp@alien8.de \
--cc=dave.hansen@intel.com \
--cc=keescook@chromium.org \
--cc=kexec@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=regressions@lists.linux.de \
--cc=tglx@linutronix.de \
--cc=thomas.lendacky@amd.com \
--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