Re: [PATCH] Document/kexec: Generalize crash hotplug description
From: Sourabh Jain
Date: Fri Aug 09 2024 - 07:03:58 EST
Hello Baoquan,
On 09/08/24 07:18, Baoquan He wrote:
On 08/05/24 at 10:38am, Sourabh Jain wrote:
Commit 79365026f869 ("crash: add a new kexec flag for hotplug support")
generalizes the crash hotplug support to allow architectures to update
multiple kexec segments on CPU/Memory hotplug and not just elfcorehdr.
Therefore, update the relevant kernel documentation to reflect the same.
No functional change.
Cc: Petr Tesarik <petr@xxxxxxxxxxx>
Cc: Hari Bathini <hbathini@xxxxxxxxxxxxx>
Cc: kexec@xxxxxxxxxxxxxxxxxxx
Cc: linux-kernel@xxxxxxxxxxxxxxx
Cc: linuxppc-dev@xxxxxxxxxxxxxxxx
Cc: x86@xxxxxxxxxx
Signed-off-by: Sourabh Jain <sourabhjain@xxxxxxxxxxxxx>
---
Discussion about the documentation update:
https://lore.kernel.org/all/68d0328d-531a-4a2b-ab26-c97fd8a12e8b@xxxxxxxxxxxxx/
---
.../ABI/testing/sysfs-devices-memory | 6 ++--
.../ABI/testing/sysfs-devices-system-cpu | 6 ++--
.../admin-guide/mm/memory-hotplug.rst | 5 ++--
Documentation/core-api/cpu_hotplug.rst | 10 ++++---
kernel/crash_core.c | 29 ++++++++++++-------
5 files changed, 33 insertions(+), 23 deletions(-)
The overall looks good to me, except of concern from Petr. Thanks.
Thanks for the review. I will make the suggested changes in v2.
Additionally I will also generalize the error message
"kexec_trylock() failed, elfcorehdr may be inaccurate " from
functions crash_handle_hotplug_event() and crash_check_hotplug_support()
to "kexec_trylock() failed, kdump image may be inaccurate"
- Sourabh Jain
diff --git a/Documentation/ABI/testing/sysfs-devices-memory b/Documentation/ABI/testing/sysfs-devices-memory
index a95e0f17c35a..421acc8e2c6b 100644
--- a/Documentation/ABI/testing/sysfs-devices-memory
+++ b/Documentation/ABI/testing/sysfs-devices-memory
@@ -115,6 +115,6 @@ What: /sys/devices/system/memory/crash_hotplug
Date: Aug 2023
Contact: Linux kernel mailing list <linux-kernel@xxxxxxxxxxxxxxx>
Description:
- (RO) indicates whether or not the kernel directly supports
- modifying the crash elfcorehdr for memory hot un/plug and/or
- on/offline changes.
+ (RO) indicates whether or not the kernel update of kexec
+ segments on memory hot un/plug and/or on/offline events,
+ avoiding the need to reload kdump kernel.
diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu
index 325873385b71..f4ada1cd2f96 100644
--- a/Documentation/ABI/testing/sysfs-devices-system-cpu
+++ b/Documentation/ABI/testing/sysfs-devices-system-cpu
@@ -703,9 +703,9 @@ What: /sys/devices/system/cpu/crash_hotplug
Date: Aug 2023
Contact: Linux kernel mailing list <linux-kernel@xxxxxxxxxxxxxxx>
Description:
- (RO) indicates whether or not the kernel directly supports
- modifying the crash elfcorehdr for CPU hot un/plug and/or
- on/offline changes.
+ (RO) indicates whether or not the kernel update of kexec
+ segments on CPU hot un/plug and/or on/offline events,
+ avoiding the need to reload kdump kernel.
What: /sys/devices/system/cpu/enabled
Date: Nov 2022
diff --git a/Documentation/admin-guide/mm/memory-hotplug.rst b/Documentation/admin-guide/mm/memory-hotplug.rst
index 098f14d83e99..cb2c080f400c 100644
--- a/Documentation/admin-guide/mm/memory-hotplug.rst
+++ b/Documentation/admin-guide/mm/memory-hotplug.rst
@@ -294,8 +294,9 @@ The following files are currently defined:
``crash_hotplug`` read-only: when changes to the system memory map
occur due to hot un/plug of memory, this file contains
'1' if the kernel updates the kdump capture kernel memory
- map itself (via elfcorehdr), or '0' if userspace must update
- the kdump capture kernel memory map.
+ map itself (via elfcorehdr and other relevant kexec
+ segments), or '0' if userspace must update the kdump
+ capture kernel memory map.
Availability depends on the CONFIG_MEMORY_HOTPLUG kernel
configuration option.
diff --git a/Documentation/core-api/cpu_hotplug.rst b/Documentation/core-api/cpu_hotplug.rst
index dcb0e379e5e8..a21dbf261be7 100644
--- a/Documentation/core-api/cpu_hotplug.rst
+++ b/Documentation/core-api/cpu_hotplug.rst
@@ -737,8 +737,9 @@ can process the event further.
When changes to the CPUs in the system occur, the sysfs file
/sys/devices/system/cpu/crash_hotplug contains '1' if the kernel
-updates the kdump capture kernel list of CPUs itself (via elfcorehdr),
-or '0' if userspace must update the kdump capture kernel list of CPUs.
+updates the kdump capture kernel list of CPUs itself (via elfcorehdr and
+other relevant kexec segment), or '0' if userspace must update the kdump
+capture kernel list of CPUs.
The availability depends on the CONFIG_HOTPLUG_CPU kernel configuration
option.
@@ -750,8 +751,9 @@ file can be used in a udev rule as follows:
SUBSYSTEM=="cpu", ATTRS{crash_hotplug}=="1", GOTO="kdump_reload_end"
For a CPU hot un/plug event, if the architecture supports kernel updates
-of the elfcorehdr (which contains the list of CPUs), then the rule skips
-the unload-then-reload of the kdump capture kernel.
+of the elfcorehdr (which contains the list of CPUs) and other relevant
+kexec segments, then the rule skips the unload-then-reload of the kdump
+capture kernel.
Kernel Inline Documentations Reference
======================================
diff --git a/kernel/crash_core.c b/kernel/crash_core.c
index 63cf89393c6e..64dad01e260b 100644
--- a/kernel/crash_core.c
+++ b/kernel/crash_core.c
@@ -520,18 +520,25 @@ int crash_check_hotplug_support(void)
}
/*
- * To accurately reflect hot un/plug changes of cpu and memory resources
- * (including onling and offlining of those resources), the elfcorehdr
- * (which is passed to the crash kernel via the elfcorehdr= parameter)
- * must be updated with the new list of CPUs and memories.
+ * To accurately reflect hot un/plug changes of CPU and Memory resources
+ * (including onling and offlining of those resources), the relevant
+ * kexec segments must be updated with latest CPU and Memory resources.
*
- * In order to make changes to elfcorehdr, two conditions are needed:
- * First, the segment containing the elfcorehdr must be large enough
- * to permit a growing number of resources; the elfcorehdr memory size
- * is based on NR_CPUS_DEFAULT and CRASH_MAX_MEMORY_RANGES.
- * Second, purgatory must explicitly exclude the elfcorehdr from the
- * list of segments it checks (since the elfcorehdr changes and thus
- * would require an update to purgatory itself to update the digest).
+ * Architectures must ensure two things for all segments that need
+ * updating during hotplug events:
+ *
+ * 1. Segments must be large enough to accommodate a growing number of
+ * resources.
+ * 2. Exclude the segments from SHA verification.
+ *
+ * For example, on most architectures, the elfcorehdr (which is passed
+ * to the crash kernel via the elfcorehdr= parameter) must include the
+ * new list of CPUs and memory. To make changes to the elfcorehdr, it
+ * should be large enough to permit a growing number of CPU and Memory
+ * resources. One can estimate the elfcorehdr memory size based on
+ * NR_CPUS_DEFAULT and CRASH_MAX_MEMORY_RANGES. The elfcorehdr is
+ * excluded from SHA verification by default if the architecture
+ * supports crash hotplug.
*/
static void crash_handle_hotplug_event(unsigned int hp_action, unsigned int cpu, void *arg)
{
--
2.45.2