crashkernel: Maximize the reserved size in osbuild case

At present, the deduction of the proper crashkernel value depends on
detecting platform details. However, more and more platforms are now
deployed using OSTree images, which means kdump-utils cannot retrieve
complete platform information. This often leads to an underestimation of
the memory required by the kdump kernel, increasing the risk of
out-of-memory (OOM) issues.

To mitigate this situation, we choose to maximize the reserved
crashkernel size in the osbuild case, and recommend users to update the
kernel command line and release the excess reserved memory after reboot
(addressed in the next patch).

Signed-off-by: Pingfan Liu <[email protected]>

Author: Pingfan Liu

kdump-lib.sh

@@ -16,6 +16,7 @@ FADUMP_REGISTER_SYS_NODE="/sys/kernel/fadump/registered"
 FADUMP_APPEND_ARGS_SYS_NODE="/sys/kernel/fadump/bootargs_append"
 # shellcheck disable=SC2034
 FENCE_KDUMP_CONFIG_FILE="/etc/sysconfig/fence_kdump"
+maximize_crashkernel=false
 
 is_uki()
 {
@@ -46,7 +47,7 @@ is_aws_aarch64()
 
 is_sme_or_sev_active()
 {
-	journalctl -q --dmesg --grep "^Memory Encryption Features active: AMD (SME|SEV)$" > /dev/null 2>&1
+	$maximize_crashkernel || journalctl -q --dmesg --grep "^Memory Encryption Features active: AMD (SME|SEV)$" > /dev/null 2>&1
 }
 
 has_command()
@@ -273,6 +274,20 @@ is_ostree()
 	test -f /run/ostree-booted
 }
 
+# Decide whether it runs in sandbox or not
+is_osbuild()
+{
+	local _init_comm
+
+	_init_comm=$(ps -p 1 -o comm= 2>/dev/null || echo "unknown")
+	# no real init in isolated env
+	if [[ "$init_comm" != "systemd" ]]; then
+		return 0
+	fi
+
+	return 1
+}
+
 # get ip address or hostname from nfs/ssh config value
 get_remote_host()
 {
@@ -854,12 +869,18 @@ get_recommend_size()
 
 has_mlx5()
 {
-	[[ -d /sys/bus/pci/drivers/mlx5_core ]]
+	$maximize_crashkernel || [[ -d /sys/bus/pci/drivers/mlx5_core ]]
 }
 
 has_aarch64_smmu()
 {
-	ls /sys/devices/platform/arm-smmu-* 1> /dev/null 2>&1
+	$maximize_crashkernel || ls /sys/devices/platform/arm-smmu-* 1> /dev/null 2>&1
+}
+
+is_aarch64_64k_kernel()
+{
+	local _kernel="$1"
+	$maximize_crashkernel || echo "$_kernel" | grep -q 64k
 }
 
 is_memsize()
@@ -1006,6 +1027,9 @@ kdump_get_arch_recommend_crashkernel()
 	local _arch _ck_cmdline _dump_mode
 	local _delta=0
 
+	# osbuild deploys rpm on isolated environment. kdump-utils has no opportunity
+	# to deduce the exact memory cost on the real target.
+	is_osbuild && maximize_crashkernel=true
 	if [[ -z $1 ]]; then
 		if is_fadump_capable; then
 			_dump_mode=fadump
@@ -1033,7 +1057,7 @@ kdump_get_arch_recommend_crashkernel()
 		fi
 
 		# the naming convention of 64k variant suffixes with +64k, e.g. "vmlinuz-5.14.0-312.el9.aarch64+64k"
-		if echo "$_running_kernel" | grep -q 64k; then
+		if is_aarch64_64k_kernel "$_running_kernel"; then
 			# Without smmu, the diff of MemFree between 4K and 64K measured on a high end aarch64 machine is 82M.
 			# Picking up 100M to cover this diff. And finally, we have "2G-4G:356M;4G-64G:420M;64G-:676M"
 			((_delta += 100))