* [PATCH linux-next v3 08/14] ppc, crash: enforce KEXEC and KEXEC_FILE to select CRASH_DUMP
2024-01-24 5:12 [PATCH linux-next v3 00/14] Split crash out from kexec and clean up related config items Baoquan He
` (2 preceding siblings ...)
(?)
@ 2024-01-24 5:12 ` Baoquan He
0 siblings, 0 replies; 6+ messages in thread
From: Baoquan He @ 2024-01-24 5:12 UTC (permalink / raw)
To: linux-kernel
Cc: kexec, x86, linux-arm-kernel, linuxppc-dev, linux-s390, linux-sh,
linux-mips, linux-riscv, loongarch, akpm, ebiederm, hbathini,
piliu, viro, Baoquan He
In PowerPC, the crash dumping and kexec reboot share code in
arch_kexec_locate_mem_hole(), in which struct crash_mem is used.
Here enfoce enforce KEXEC and KEXEC_FILE to select CRASH_DUMP for now.
Signed-off-by: Baoquan He <bhe@redhat.com>
---
arch/powerpc/Kconfig | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index e66fd9923250..31f013e636e3 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -608,6 +608,10 @@ config PPC64_SUPPORTS_MEMORY_FAILURE
config ARCH_SUPPORTS_KEXEC
def_bool PPC_BOOK3S || PPC_E500 || (44x && !SMP)
+config ARCH_SELECTS_KEXEC
+ def_bool y
+ select CRASH_DUMP
+
config ARCH_SUPPORTS_KEXEC_FILE
def_bool PPC64
@@ -618,6 +622,7 @@ config ARCH_SELECTS_KEXEC_FILE
def_bool y
depends on KEXEC_FILE
select KEXEC_ELF
+ select CRASH_DUMP
select HAVE_IMA_KEXEC if IMA
config PPC64_BIG_ENDIAN_ELF_ABI_V2
--
2.41.0
_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH linux-next v3 08/14] ppc, crash: enforce KEXEC and KEXEC_FILE to select CRASH_DUMP
@ 2024-01-24 5:12 ` Baoquan He
0 siblings, 0 replies; 6+ messages in thread
From: Baoquan He @ 2024-01-24 5:12 UTC (permalink / raw)
To: linux-kernel
Cc: kexec, x86, linux-arm-kernel, linuxppc-dev, linux-s390, linux-sh,
linux-mips, linux-riscv, loongarch, akpm, ebiederm, hbathini,
piliu, viro, Baoquan He
In PowerPC, the crash dumping and kexec reboot share code in
arch_kexec_locate_mem_hole(), in which struct crash_mem is used.
Here enfoce enforce KEXEC and KEXEC_FILE to select CRASH_DUMP for now.
Signed-off-by: Baoquan He <bhe@redhat.com>
---
arch/powerpc/Kconfig | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index e66fd9923250..31f013e636e3 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -608,6 +608,10 @@ config PPC64_SUPPORTS_MEMORY_FAILURE
config ARCH_SUPPORTS_KEXEC
def_bool PPC_BOOK3S || PPC_E500 || (44x && !SMP)
+config ARCH_SELECTS_KEXEC
+ def_bool y
+ select CRASH_DUMP
+
config ARCH_SUPPORTS_KEXEC_FILE
def_bool PPC64
@@ -618,6 +622,7 @@ config ARCH_SELECTS_KEXEC_FILE
def_bool y
depends on KEXEC_FILE
select KEXEC_ELF
+ select CRASH_DUMP
select HAVE_IMA_KEXEC if IMA
config PPC64_BIG_ENDIAN_ELF_ABI_V2
--
2.41.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH linux-next v3 08/14] ppc, crash: enforce KEXEC and KEXEC_FILE to select CRASH_DUMP
@ 2024-01-24 5:12 ` Baoquan He
0 siblings, 0 replies; 6+ messages in thread
From: Baoquan He @ 2024-01-24 5:12 UTC (permalink / raw)
To: linux-kernel
Cc: kexec, x86, linux-arm-kernel, linuxppc-dev, linux-s390, linux-sh,
linux-mips, linux-riscv, loongarch, akpm, ebiederm, hbathini,
piliu, viro, Baoquan He
In PowerPC, the crash dumping and kexec reboot share code in
arch_kexec_locate_mem_hole(), in which struct crash_mem is used.
Here enfoce enforce KEXEC and KEXEC_FILE to select CRASH_DUMP for now.
Signed-off-by: Baoquan He <bhe@redhat.com>
---
arch/powerpc/Kconfig | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index e66fd9923250..31f013e636e3 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -608,6 +608,10 @@ config PPC64_SUPPORTS_MEMORY_FAILURE
config ARCH_SUPPORTS_KEXEC
def_bool PPC_BOOK3S || PPC_E500 || (44x && !SMP)
+config ARCH_SELECTS_KEXEC
+ def_bool y
+ select CRASH_DUMP
+
config ARCH_SUPPORTS_KEXEC_FILE
def_bool PPC64
@@ -618,6 +622,7 @@ config ARCH_SELECTS_KEXEC_FILE
def_bool y
depends on KEXEC_FILE
select KEXEC_ELF
+ select CRASH_DUMP
select HAVE_IMA_KEXEC if IMA
config PPC64_BIG_ENDIAN_ELF_ABI_V2
--
2.41.0
_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH linux-next v3 08/14] ppc, crash: enforce KEXEC and KEXEC_FILE to select CRASH_DUMP
@ 2024-01-24 5:12 ` Baoquan He
0 siblings, 0 replies; 6+ messages in thread
From: Baoquan He @ 2024-01-24 5:12 UTC (permalink / raw)
To: linux-kernel
Cc: linux-s390, Baoquan He, piliu, linux-sh, x86, kexec, linux-mips,
ebiederm, loongarch, hbathini, linux-riscv, linuxppc-dev, akpm,
linux-arm-kernel, viro
In PowerPC, the crash dumping and kexec reboot share code in
arch_kexec_locate_mem_hole(), in which struct crash_mem is used.
Here enfoce enforce KEXEC and KEXEC_FILE to select CRASH_DUMP for now.
Signed-off-by: Baoquan He <bhe@redhat.com>
---
arch/powerpc/Kconfig | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index e66fd9923250..31f013e636e3 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -608,6 +608,10 @@ config PPC64_SUPPORTS_MEMORY_FAILURE
config ARCH_SUPPORTS_KEXEC
def_bool PPC_BOOK3S || PPC_E500 || (44x && !SMP)
+config ARCH_SELECTS_KEXEC
+ def_bool y
+ select CRASH_DUMP
+
config ARCH_SUPPORTS_KEXEC_FILE
def_bool PPC64
@@ -618,6 +622,7 @@ config ARCH_SELECTS_KEXEC_FILE
def_bool y
depends on KEXEC_FILE
select KEXEC_ELF
+ select CRASH_DUMP
select HAVE_IMA_KEXEC if IMA
config PPC64_BIG_ENDIAN_ELF_ABI_V2
--
2.41.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH linux-next v3 08/14] ppc, crash: enforce KEXEC and KEXEC_FILE to select CRASH_DUMP
@ 2024-01-24 5:12 ` Baoquan He
0 siblings, 0 replies; 6+ messages in thread
From: Baoquan He @ 2024-01-24 5:12 UTC (permalink / raw)
To: linux-kernel
Cc: kexec, x86, linux-arm-kernel, linuxppc-dev, linux-s390, linux-sh,
linux-mips, linux-riscv, loongarch, akpm, ebiederm, hbathini,
piliu, viro, Baoquan He
In PowerPC, the crash dumping and kexec reboot share code in
arch_kexec_locate_mem_hole(), in which struct crash_mem is used.
Here enfoce enforce KEXEC and KEXEC_FILE to select CRASH_DUMP for now.
Signed-off-by: Baoquan He <bhe@redhat.com>
---
arch/powerpc/Kconfig | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index e66fd9923250..31f013e636e3 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -608,6 +608,10 @@ config PPC64_SUPPORTS_MEMORY_FAILURE
config ARCH_SUPPORTS_KEXEC
def_bool PPC_BOOK3S || PPC_E500 || (44x && !SMP)
+config ARCH_SELECTS_KEXEC
+ def_bool y
+ select CRASH_DUMP
+
config ARCH_SUPPORTS_KEXEC_FILE
def_bool PPC64
@@ -618,6 +622,7 @@ config ARCH_SELECTS_KEXEC_FILE
def_bool y
depends on KEXEC_FILE
select KEXEC_ELF
+ select CRASH_DUMP
select HAVE_IMA_KEXEC if IMA
config PPC64_BIG_ENDIAN_ELF_ABI_V2
--
2.41.0
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH linux-next v3 08/14] ppc, crash: enforce KEXEC and KEXEC_FILE to select CRASH_DUMP
@ 2024-01-27 11:51 kernel test robot
0 siblings, 0 replies; 6+ messages in thread
From: kernel test robot @ 2024-01-27 11:51 UTC (permalink / raw)
To: oe-kbuild; +Cc: lkp
::::::
:::::: Manual check reason: "only kconfig file changed"
::::::
BCC: lkp@intel.com
CC: oe-kbuild-all@lists.linux.dev
In-Reply-To: <20240124051254.67105-9-bhe@redhat.com>
References: <20240124051254.67105-9-bhe@redhat.com>
TO: Baoquan He <bhe@redhat.com>
Hi Baoquan,
kernel test robot noticed the following build errors:
[auto build test ERROR on next-20240123]
url: https://github.com/intel-lab-lkp/linux/commits/Baoquan-He/kexec-split-crashkernel-reservation-code-out-from-crash_core-c/20240124-131952
base: next-20240123
patch link: https://lore.kernel.org/r/20240124051254.67105-9-bhe%40redhat.com
patch subject: [PATCH linux-next v3 08/14] ppc, crash: enforce KEXEC and KEXEC_FILE to select CRASH_DUMP
:::::: branch date: 3 days ago
:::::: commit date: 3 days ago
config: powerpc-allnoconfig (https://download.01.org/0day-ci/archive/20240127/202401271906.ORpeFe0T-lkp@intel.com/config)
compiler: powerpc-linux-gcc (GCC) 13.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240127/202401271906.ORpeFe0T-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/r/202401271906.ORpeFe0T-lkp@intel.com/
All error/warnings (new ones prefixed by >>):
kernel/crash_core.c: In function 'kimage_crash_copy_vmcoreinfo':
>> kernel/crash_core.c:43:18: error: invalid use of undefined type 'struct kimage'
43 | if (image->type != KEXEC_TYPE_CRASH)
| ^~
>> kernel/crash_core.c:43:28: error: 'KEXEC_TYPE_CRASH' undeclared (first use in this function); did you mean 'KEXEC_ON_CRASH'?
43 | if (image->type != KEXEC_TYPE_CRASH)
| ^~~~~~~~~~~~~~~~
| KEXEC_ON_CRASH
kernel/crash_core.c:43:28: note: each undeclared identifier is reported only once for each function it appears in
>> kernel/crash_core.c:55:27: error: implicit declaration of function 'kimage_alloc_control_pages' [-Werror=implicit-function-declaration]
55 | vmcoreinfo_page = kimage_alloc_control_pages(image, 0);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~
>> kernel/crash_core.c:55:25: warning: assignment to 'struct page *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
55 | vmcoreinfo_page = kimage_alloc_control_pages(image, 0);
| ^
kernel/crash_core.c:66:14: error: invalid use of undefined type 'struct kimage'
66 | image->vmcoreinfo_data_copy = safecopy;
| ^~
kernel/crash_core.c: At top level:
>> kernel/crash_core.c:74:5: error: redefinition of 'kexec_should_crash'
74 | int kexec_should_crash(struct task_struct *p)
| ^~~~~~~~~~~~~~~~~~
In file included from kernel/crash_core.c:12:
include/linux/kexec.h:474:19: note: previous definition of 'kexec_should_crash' with type 'int(struct task_struct *)'
474 | static inline int kexec_should_crash(struct task_struct *p) { return 0; }
| ^~~~~~~~~~~~~~~~~~
>> kernel/crash_core.c:92:5: error: redefinition of 'kexec_crash_loaded'
92 | int kexec_crash_loaded(void)
| ^~~~~~~~~~~~~~~~~~
include/linux/kexec.h:475:19: note: previous definition of 'kexec_crash_loaded' with type 'int(void)'
475 | static inline int kexec_crash_loaded(void) { return 0; }
| ^~~~~~~~~~~~~~~~~~
kernel/crash_core.c: In function 'kexec_crash_loaded':
>> kernel/crash_core.c:94:18: error: 'kexec_crash_image' undeclared (first use in this function); did you mean 'kexec_crash_loaded'?
94 | return !!kexec_crash_image;
| ^~~~~~~~~~~~~~~~~
| kexec_crash_loaded
kernel/crash_core.c: At top level:
>> kernel/crash_core.c:103:16: error: redefinition of '__crash_kexec'
103 | void __noclone __crash_kexec(struct pt_regs *regs)
| ^~~~~~~~~~~~~
include/linux/kexec.h:472:20: note: previous definition of '__crash_kexec' with type 'void(struct pt_regs *)'
472 | static inline void __crash_kexec(struct pt_regs *regs) { }
| ^~~~~~~~~~~~~
kernel/crash_core.c: In function '__crash_kexec':
kernel/crash_core.c:114:21: error: 'kexec_crash_image' undeclared (first use in this function); did you mean 'kexec_crash_loaded'?
114 | if (kexec_crash_image) {
| ^~~~~~~~~~~~~~~~~
| kexec_crash_loaded
>> kernel/crash_core.c:117:25: error: implicit declaration of function 'crash_setup_regs' [-Werror=implicit-function-declaration]
117 | crash_setup_regs(&fixed_regs, regs);
| ^~~~~~~~~~~~~~~~
>> kernel/crash_core.c:120:25: error: implicit declaration of function 'machine_kexec' [-Werror=implicit-function-declaration]
120 | machine_kexec(kexec_crash_image);
| ^~~~~~~~~~~~~
kernel/crash_core.c: At top level:
>> kernel/crash_core.c:128:1: warning: ignoring attribute 'noinline' because it conflicts with attribute 'gnu_inline' [-Wattributes]
128 | {
| ^
include/linux/kexec.h:473:20: note: previous declaration here
473 | static inline void crash_kexec(struct pt_regs *regs) { }
| ^~~~~~~~~~~
>> kernel/crash_core.c:127:18: error: redefinition of 'crash_kexec'
127 | __bpf_kfunc void crash_kexec(struct pt_regs *regs)
| ^~~~~~~~~~~
include/linux/kexec.h:473:20: note: previous definition of 'crash_kexec' with type 'void(struct pt_regs *)'
473 | static inline void crash_kexec(struct pt_regs *regs) { }
| ^~~~~~~~~~~
kernel/crash_core.c: In function '__crash_shrink_memory':
>> kernel/crash_core.c:356:9: error: implicit declaration of function 'crash_free_reserved_phys_range' [-Werror=implicit-function-declaration]
356 | crash_free_reserved_phys_range(ram_res->start, ram_res->end);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
kernel/crash_core.c: In function 'crash_shrink_memory':
kernel/crash_core.c:370:13: error: 'kexec_crash_image' undeclared (first use in this function); did you mean 'kexec_crash_loaded'?
370 | if (kexec_crash_image) {
| ^~~~~~~~~~~~~~~~~
| kexec_crash_loaded
In file included from include/linux/kernel.h:27,
from arch/powerpc/include/asm/page.h:11,
from arch/powerpc/include/asm/thread_info.h:13,
from include/linux/thread_info.h:60,
from arch/powerpc/include/asm/processor.h:41,
from include/linux/sched.h:13,
from include/linux/utsname.h:6,
from kernel/crash_core.c:9:
>> kernel/crash_core.c:377:38: error: 'KEXEC_CRASH_MEM_ALIGN' undeclared (first use in this function)
377 | new_size = roundup(new_size, KEXEC_CRASH_MEM_ALIGN);
| ^~~~~~~~~~~~~~~~~~~~~
include/linux/math.h:61:16: note: in definition of macro 'roundup'
61 | typeof(y) __y = y; \
| ^
kernel/crash_core.c: In function 'crash_save_cpu':
>> kernel/crash_core.c:437:36: error: 'KEXEC_CORE_NOTE_NAME' undeclared (first use in this function); did you mean 'CRASH_CORE_NOTE_NAME'?
437 | buf = append_elf_note(buf, KEXEC_CORE_NOTE_NAME, NT_PRSTATUS,
| ^~~~~~~~~~~~~~~~~~~~
| CRASH_CORE_NOTE_NAME
kernel/crash_core.c: In function 'kexec_crash_loaded':
>> kernel/crash_core.c:95:1: warning: control reaches end of non-void function [-Wreturn-type]
95 | }
| ^
cc1: some warnings being treated as errors
--
>> arch/powerpc/kernel/crash_dump.c:101:6: error: redefinition of 'is_kdump_kernel'
101 | bool is_kdump_kernel(void)
| ^~~~~~~~~~~~~~~
In file included from arch/powerpc/kernel/crash_dump.c:12:
include/linux/crash_dump.h:64:20: note: previous definition of 'is_kdump_kernel' with type 'bool(void)' {aka '_Bool(void)'}
64 | static inline bool is_kdump_kernel(void)
| ^~~~~~~~~~~~~~~
Kconfig warnings: (for reference only)
WARNING: unmet direct dependencies detected for CRASH_DUMP
Depends on [n]: ARCH_SUPPORTS_CRASH_DUMP [=y] && KEXEC_CORE [=n]
Selected by [y]:
- ARCH_SELECTS_KEXEC [=y]
vim +43 kernel/crash_core.c
636be0013152f6 Baoquan He 2024-01-24 35
636be0013152f6 Baoquan He 2024-01-24 36 int kimage_crash_copy_vmcoreinfo(struct kimage *image)
636be0013152f6 Baoquan He 2024-01-24 37 {
636be0013152f6 Baoquan He 2024-01-24 38 struct page *vmcoreinfo_page;
636be0013152f6 Baoquan He 2024-01-24 39 void *safecopy;
636be0013152f6 Baoquan He 2024-01-24 40
636be0013152f6 Baoquan He 2024-01-24 41 if (!IS_ENABLED(CONFIG_CRASH_DUMP))
636be0013152f6 Baoquan He 2024-01-24 42 return 0;
636be0013152f6 Baoquan He 2024-01-24 @43 if (image->type != KEXEC_TYPE_CRASH)
636be0013152f6 Baoquan He 2024-01-24 44 return 0;
636be0013152f6 Baoquan He 2024-01-24 45
636be0013152f6 Baoquan He 2024-01-24 46 /*
636be0013152f6 Baoquan He 2024-01-24 47 * For kdump, allocate one vmcoreinfo safe copy from the
636be0013152f6 Baoquan He 2024-01-24 48 * crash memory. as we have arch_kexec_protect_crashkres()
636be0013152f6 Baoquan He 2024-01-24 49 * after kexec syscall, we naturally protect it from write
636be0013152f6 Baoquan He 2024-01-24 50 * (even read) access under kernel direct mapping. But on
636be0013152f6 Baoquan He 2024-01-24 51 * the other hand, we still need to operate it when crash
636be0013152f6 Baoquan He 2024-01-24 52 * happens to generate vmcoreinfo note, hereby we rely on
636be0013152f6 Baoquan He 2024-01-24 53 * vmap for this purpose.
636be0013152f6 Baoquan He 2024-01-24 54 */
636be0013152f6 Baoquan He 2024-01-24 @55 vmcoreinfo_page = kimage_alloc_control_pages(image, 0);
636be0013152f6 Baoquan He 2024-01-24 56 if (!vmcoreinfo_page) {
636be0013152f6 Baoquan He 2024-01-24 57 pr_warn("Could not allocate vmcoreinfo buffer\n");
636be0013152f6 Baoquan He 2024-01-24 58 return -ENOMEM;
636be0013152f6 Baoquan He 2024-01-24 59 }
636be0013152f6 Baoquan He 2024-01-24 60 safecopy = vmap(&vmcoreinfo_page, 1, VM_MAP, PAGE_KERNEL);
636be0013152f6 Baoquan He 2024-01-24 61 if (!safecopy) {
636be0013152f6 Baoquan He 2024-01-24 62 pr_warn("Could not vmap vmcoreinfo buffer\n");
636be0013152f6 Baoquan He 2024-01-24 63 return -ENOMEM;
636be0013152f6 Baoquan He 2024-01-24 64 }
636be0013152f6 Baoquan He 2024-01-24 65
636be0013152f6 Baoquan He 2024-01-24 66 image->vmcoreinfo_data_copy = safecopy;
636be0013152f6 Baoquan He 2024-01-24 67 crash_update_vmcoreinfo_safecopy(safecopy);
636be0013152f6 Baoquan He 2024-01-24 68
636be0013152f6 Baoquan He 2024-01-24 69 return 0;
636be0013152f6 Baoquan He 2024-01-24 70 }
636be0013152f6 Baoquan He 2024-01-24 71
636be0013152f6 Baoquan He 2024-01-24 72
636be0013152f6 Baoquan He 2024-01-24 73
636be0013152f6 Baoquan He 2024-01-24 @74 int kexec_should_crash(struct task_struct *p)
636be0013152f6 Baoquan He 2024-01-24 75 {
636be0013152f6 Baoquan He 2024-01-24 76 /*
636be0013152f6 Baoquan He 2024-01-24 77 * If crash_kexec_post_notifiers is enabled, don't run
636be0013152f6 Baoquan He 2024-01-24 78 * crash_kexec() here yet, which must be run after panic
636be0013152f6 Baoquan He 2024-01-24 79 * notifiers in panic().
636be0013152f6 Baoquan He 2024-01-24 80 */
636be0013152f6 Baoquan He 2024-01-24 81 if (crash_kexec_post_notifiers)
636be0013152f6 Baoquan He 2024-01-24 82 return 0;
636be0013152f6 Baoquan He 2024-01-24 83 /*
636be0013152f6 Baoquan He 2024-01-24 84 * There are 4 panic() calls in make_task_dead() path, each of which
636be0013152f6 Baoquan He 2024-01-24 85 * corresponds to each of these 4 conditions.
636be0013152f6 Baoquan He 2024-01-24 86 */
636be0013152f6 Baoquan He 2024-01-24 87 if (in_interrupt() || !p->pid || is_global_init(p) || panic_on_oops)
636be0013152f6 Baoquan He 2024-01-24 88 return 1;
636be0013152f6 Baoquan He 2024-01-24 89 return 0;
636be0013152f6 Baoquan He 2024-01-24 90 }
636be0013152f6 Baoquan He 2024-01-24 91
636be0013152f6 Baoquan He 2024-01-24 @92 int kexec_crash_loaded(void)
636be0013152f6 Baoquan He 2024-01-24 93 {
636be0013152f6 Baoquan He 2024-01-24 @94 return !!kexec_crash_image;
636be0013152f6 Baoquan He 2024-01-24 @95 }
636be0013152f6 Baoquan He 2024-01-24 96 EXPORT_SYMBOL_GPL(kexec_crash_loaded);
636be0013152f6 Baoquan He 2024-01-24 97
636be0013152f6 Baoquan He 2024-01-24 98 /*
636be0013152f6 Baoquan He 2024-01-24 99 * No panic_cpu check version of crash_kexec(). This function is called
636be0013152f6 Baoquan He 2024-01-24 100 * only when panic_cpu holds the current CPU number; this is the only CPU
636be0013152f6 Baoquan He 2024-01-24 101 * which processes crash_kexec routines.
636be0013152f6 Baoquan He 2024-01-24 102 */
636be0013152f6 Baoquan He 2024-01-24 @103 void __noclone __crash_kexec(struct pt_regs *regs)
636be0013152f6 Baoquan He 2024-01-24 104 {
636be0013152f6 Baoquan He 2024-01-24 105 /* Take the kexec_lock here to prevent sys_kexec_load
636be0013152f6 Baoquan He 2024-01-24 106 * running on one cpu from replacing the crash kernel
636be0013152f6 Baoquan He 2024-01-24 107 * we are using after a panic on a different cpu.
636be0013152f6 Baoquan He 2024-01-24 108 *
636be0013152f6 Baoquan He 2024-01-24 109 * If the crash kernel was not located in a fixed area
636be0013152f6 Baoquan He 2024-01-24 110 * of memory the xchg(&kexec_crash_image) would be
636be0013152f6 Baoquan He 2024-01-24 111 * sufficient. But since I reuse the memory...
636be0013152f6 Baoquan He 2024-01-24 112 */
636be0013152f6 Baoquan He 2024-01-24 113 if (kexec_trylock()) {
636be0013152f6 Baoquan He 2024-01-24 114 if (kexec_crash_image) {
636be0013152f6 Baoquan He 2024-01-24 115 struct pt_regs fixed_regs;
636be0013152f6 Baoquan He 2024-01-24 116
636be0013152f6 Baoquan He 2024-01-24 @117 crash_setup_regs(&fixed_regs, regs);
636be0013152f6 Baoquan He 2024-01-24 118 crash_save_vmcoreinfo();
636be0013152f6 Baoquan He 2024-01-24 119 machine_crash_shutdown(&fixed_regs);
636be0013152f6 Baoquan He 2024-01-24 @120 machine_kexec(kexec_crash_image);
636be0013152f6 Baoquan He 2024-01-24 121 }
636be0013152f6 Baoquan He 2024-01-24 122 kexec_unlock();
636be0013152f6 Baoquan He 2024-01-24 123 }
636be0013152f6 Baoquan He 2024-01-24 124 }
636be0013152f6 Baoquan He 2024-01-24 125 STACK_FRAME_NON_STANDARD(__crash_kexec);
636be0013152f6 Baoquan He 2024-01-24 126
636be0013152f6 Baoquan He 2024-01-24 @127 __bpf_kfunc void crash_kexec(struct pt_regs *regs)
636be0013152f6 Baoquan He 2024-01-24 @128 {
636be0013152f6 Baoquan He 2024-01-24 129 int old_cpu, this_cpu;
636be0013152f6 Baoquan He 2024-01-24 130
636be0013152f6 Baoquan He 2024-01-24 131 /*
636be0013152f6 Baoquan He 2024-01-24 132 * Only one CPU is allowed to execute the crash_kexec() code as with
636be0013152f6 Baoquan He 2024-01-24 133 * panic(). Otherwise parallel calls of panic() and crash_kexec()
636be0013152f6 Baoquan He 2024-01-24 134 * may stop each other. To exclude them, we use panic_cpu here too.
636be0013152f6 Baoquan He 2024-01-24 135 */
636be0013152f6 Baoquan He 2024-01-24 136 old_cpu = PANIC_CPU_INVALID;
636be0013152f6 Baoquan He 2024-01-24 137 this_cpu = raw_smp_processor_id();
636be0013152f6 Baoquan He 2024-01-24 138
636be0013152f6 Baoquan He 2024-01-24 139 if (atomic_try_cmpxchg(&panic_cpu, &old_cpu, this_cpu)) {
636be0013152f6 Baoquan He 2024-01-24 140 /* This is the 1st CPU which comes here, so go ahead. */
636be0013152f6 Baoquan He 2024-01-24 141 __crash_kexec(regs);
636be0013152f6 Baoquan He 2024-01-24 142
636be0013152f6 Baoquan He 2024-01-24 143 /*
636be0013152f6 Baoquan He 2024-01-24 144 * Reset panic_cpu to allow another panic()/crash_kexec()
636be0013152f6 Baoquan He 2024-01-24 145 * call.
636be0013152f6 Baoquan He 2024-01-24 146 */
636be0013152f6 Baoquan He 2024-01-24 147 atomic_set(&panic_cpu, PANIC_CPU_INVALID);
636be0013152f6 Baoquan He 2024-01-24 148 }
636be0013152f6 Baoquan He 2024-01-24 149 }
636be0013152f6 Baoquan He 2024-01-24 150
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2024-01-27 11:51 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-01-27 11:51 [PATCH linux-next v3 08/14] ppc, crash: enforce KEXEC and KEXEC_FILE to select CRASH_DUMP kernel test robot
-- strict thread matches above, loose matches on Subject: below --
2024-01-24 5:12 [PATCH linux-next v3 00/14] Split crash out from kexec and clean up related config items Baoquan He
2024-01-24 5:12 ` [PATCH linux-next v3 08/14] ppc, crash: enforce KEXEC and KEXEC_FILE to select CRASH_DUMP Baoquan He
2024-01-24 5:12 ` Baoquan He
2024-01-24 5:12 ` Baoquan He
2024-01-24 5:12 ` Baoquan He
2024-01-24 5:12 ` Baoquan He
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.