[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20180830090319.985-1-feng.tang@intel.com>
Date: Thu, 30 Aug 2018 17:03:19 +0800
From: Feng Tang <feng.tang@...el.com>
To: linux-kernel@...r.kernel.org, x86@...nel.org,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...e.hu>,
"H . Peter Anvin" <hpa@...ux.intel.com>,
Yinghai Lu <yinghai@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
Dave Hansen <dave.hansen@...el.com>,
Andi Kleen <ak@...ux.intel.com>
Cc: Feng Tang <feng.tang@...el.com>
Subject: [PATCH] x86, mm: Reserver some memory for bootmem allocator for NO_BOOTMEM
We hit a kernel panic when enabling earlycon for a platform, the
call trace is:
panic+0xd2/0x220
__alloc_bootmem+0x31/0x34
spp_getpage+0x60/0x8a
fill_pte+0x71/0x130
__set_pte_vaddr+0x1d/0x50
set_pte_vaddr+0x3c/0x60
__native_set_fixmap+0x23/0x30
native_set_fixmap+0x30/0x40
setup_earlycon+0x1e0/0x32f
param_setup_earlycon+0x13/0x22
do_early_param+0x5b/0x90
parse_args+0x1f7/0x300
parse_early_options+0x24/0x28
parse_early_param+0x65/0x73
setup_arch+0x31e/0x9f1
start_kernel+0x58/0x44e
The root cause is that when CONFIG_NO_BOOTMEM=y, before
e820__memblock_setup() is called there is no memory for bootmem
to allocate, and any alloc_bootmem() in this time window will
trigger a panic. Seems this is a known issue as already
mentioned in arch/x86/kernel/setup.c:
"
/* after early param, so could get panic from serial */
memblock_x86_reserve_range_setup_data();
"
By reserving some memory in the linker script, small memory request
could be fulfilled by bootmem allocator. And this memory region
is not wasted, but usable after kernel boot.
Signed-off-by: Feng Tang <feng.tang@...el.com>
---
arch/x86/include/asm/sections.h | 5 +++++
arch/x86/kernel/setup.c | 12 +++++++++++-
arch/x86/kernel/vmlinux.lds.S | 9 +++++++++
3 files changed, 25 insertions(+), 1 deletion(-)
diff --git a/arch/x86/include/asm/sections.h b/arch/x86/include/asm/sections.h
index 4a911a382ade..d568829510e4 100644
--- a/arch/x86/include/asm/sections.h
+++ b/arch/x86/include/asm/sections.h
@@ -9,6 +9,11 @@ extern char __brk_base[], __brk_limit[];
extern struct exception_table_entry __stop___ex_table[];
extern char __end_rodata_aligned[];
+#ifdef CONFIG_NO_BOOTMEM
+extern char __bootmem_start[], __bootmem_end[];
+#endif
+
+
#if defined(CONFIG_X86_64)
extern char __end_rodata_hpage_align[];
extern char __entry_trampoline_start[], __entry_trampoline_end[];
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index b4866badb235..a4e3d9ec1a83 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -829,6 +829,17 @@ void __init setup_arch(char **cmdline_p)
*/
memblock_reserve(0, PAGE_SIZE);
+#ifdef CONFIG_NO_BOOTMEM
+ /*
+ * There is small time window till e820__memblock_setup() is
+ * called, in which __bootmem_alloc() has no available memory
+ * to allocate and will trigger panic. Adding this revered bootmem
+ * can alleviate the situation.
+ */
+ memblock_add(__pa_symbol(__bootmem_start),
+ __bootmem_end - __bootmem_start);
+#endif
+
early_reserve_initrd();
/*
@@ -989,7 +1000,6 @@ void __init setup_arch(char **cmdline_p)
x86_report_nx();
- /* after early param, so could get panic from serial */
memblock_x86_reserve_range_setup_data();
if (acpi_mps_check()) {
diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S
index 8bde0a419f86..fec1965d668f 100644
--- a/arch/x86/kernel/vmlinux.lds.S
+++ b/arch/x86/kernel/vmlinux.lds.S
@@ -367,6 +367,15 @@ SECTIONS
__brk_limit = .;
}
+#ifdef CONFIG_NO_BOOTMEM
+ . = ALIGN(PAGE_SIZE);
+ .bootmem : AT(ADDR(.bootmem) - LOAD_OFFSET) {
+ __bootmem_start = .;
+ . += 128 * 1024; /* 128 KB for early boot mem */
+ __bootmem_end = .;
+ }
+#endif
+
. = ALIGN(PAGE_SIZE); /* keep VO_INIT_SIZE page aligned */
_end = .;
--
2.14.1
Powered by blists - more mailing lists