rpms/kexec-tools
6
0
Fork 0
Code Issues Pull Requests Releases Activity

Remove kdump-utils and makedumpfile

Browse Source 
Resolves: RHEL-49401

Signed-off-by: Lichen Liu <lichliu@redhat.com>
This commit is contained in:
Lichen Liu 2024-07-18 15:18:42 +08:00
parent 7419c13d17
commit d5198e30fe
No known key found for this signature in database
GPG Key ID: 2ED8215EF57B3D6C
19 changed files with 0 additions and 1469 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

69
0001-Use-grep-q-cmd-instead-of-cmd-grep-q.patch
View File

@ -1,69 +0,0 @@
From 98087d78eda2ca58a3b55ee5dd9e0e7bad8467ef Mon Sep 17 00:00:00 2001
From: Tao Liu <ltao@redhat.com>
Date: Tue, 25 Jun 2024 16:35:16 +1200
Subject: [PATCH] Use "grep -q <<< $(cmd)" instead of "cmd | grep -q"
The following line of code's exit value should be the exit value of
grep by default:
cmd | grep -q
However it will not always work as expected, a 141 exit code may be
returned, see the following debug info:
//--- a/usr/lib/dracut/modules.d/99kdumpbase/module-setup.sh
//+++ b/usr/lib/dracut/modules.d/99kdumpbase/module-setup.sh
//@@ -55,6 +55,11 @@ depends() {
// _dep="$_dep ssh-client"
// fi
//
//+ dracut --list-modules 1>&2
//+ echo $? 1>&2
//+ dracut --list-modules | grep -q lvmthinpool-monitor
//+ echo $? ${PIPESTATUS[0]} ${PIPESTATUS[1]} 1>&2
//+
// if is_lvm2_thinp_dump_target; then
// if dracut --list-modules | grep -q lvmthinpool-monitor; then
// add_opt_module lvmthinpool-monitor
$ kdumpctl rebuild
kdump: Rebuilding /boot/initramfs-6.10.0-0.rc4.11.el10.x86_64kdump.img
...snip...
lvmmerge
lvmthinpool-monitor
..snip...
uefi-lib
0
141 141 0
The reason is, grep exits immediately when there is a match but the
first cmd still keep writing to the pipe. Since there is no reader,
a SIGPIPE signal is generated. As a result, a 141 exit code will be
returned.
Let's use the following line of code instead, which have the same effect
but works as expected:
grep -q <<< $(cmd)
Signed-off-by: Tao Liu <ltao@redhat.com>
---
dracut-module-setup.sh | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/dracut-module-setup.sh b/dracut-module-setup.sh
index 89d9959..76fb5b2 100755
--- a/dracut-module-setup.sh
+++ b/dracut-module-setup.sh
@@ -56,7 +56,7 @@ depends() {
fi
if is_lvm2_thinp_dump_target; then
- if dracut --list-modules | grep -q lvmthinpool-monitor; then
+ if grep -q lvmthinpool-monitor <<< $(dracut --list-modules); then
add_opt_module lvmthinpool-monitor
else
dwarning "Required lvmthinpool-monitor modules is missing! Please upgrade dracut >= 057."
--
2.40.1

81
0001-dracut-module-setup-Fix-missing-systemd-system.conf-.patch
View File

@ -1,81 +0,0 @@
From a078cd8dfe9fe7956ef04c04c97b7c7354808833 Mon Sep 17 00:00:00 2001
From: Philipp Rudo <prudo@redhat.com>
Date: Thu, 11 Apr 2024 16:58:34 +0200
Subject: [PATCH 1/7] dracut-module-setup: Fix missing systemd/system.conf
error
There is a bug report for RHEL10 about a grep error reading
grep: /var/tmp/dracut.DiZuKp/initramfs/etc/systemd/system.conf*: No such file or directory
that shows up when rebuilding the initrd. This is caused by systemd
v255 that allows installing the default systemd config files to
/usr/lib/systemd instead of /etc/systemd [1][2] which is done for RHEL.
So unless a user manually adds /etc/systemd/system.conf the file no
longer exists.
However the test that requires the call to grep is somewhat wonky. IIUC
the test is there so we don't overwrite a setting the user might have
made. In my opinion this only makes sense as long as the timeout set is
larger than what we would set. But this part of the logic is missing.
So fix the error message by removing the test and add our config
unconditionally.
While at it rename the created drop-ins to 99-kdump.conf to follow
the recommended naming convention and to make sure that our value takes
precedence.
Note: In case the test is still needed we can fall back to use
'systemd-analyse cat-config' that automatically considers all potential
locations for the config and its drop-ins.
[1] 6495361c7d ("meson: add build option for install path of main config files")
[2] 6378f257e7 ("various: use new config loader instead of config_parse_config_file()")
Signed-off-by: Philipp Rudo <prudo@redhat.com>
---
dracut-module-setup.sh | 23 +++++++++++++----------
1 file changed, 13 insertions(+), 10 deletions(-)
diff --git a/dracut-module-setup.sh b/dracut-module-setup.sh
index 7e1cb9f..acc0b28 100755
--- a/dracut-module-setup.sh
+++ b/dracut-module-setup.sh
@@ -973,21 +973,24 @@ kdump_install_random_seed() {
kdump_install_systemd_conf() {
# Kdump turns out to require longer default systemd mount timeout
- # than 1st kernel(90s by default), we use default 300s for kdump.
- if ! grep -q -r "^[[:space:]]*DefaultTimeoutStartSec=" "${initdir}/etc/systemd/system.conf"*; then
- mkdir -p "${initdir}/etc/systemd/system.conf.d"
- echo "[Manager]" > "${initdir}/etc/systemd/system.conf.d/kdump.conf"
- echo "DefaultTimeoutStartSec=300s" >> "${initdir}/etc/systemd/system.conf.d/kdump.conf"
- fi
+ # than 1st kernel(45s by default), we use default 300s for kdump.
+ mkdir -p "${initdir}/etc/systemd/system.conf.d"
+ cat > "${initdir}/etc/systemd/system.conf.d/99-kdump.conf" << EOF
+[Manager]
+DefaultTimeoutStartSec=300s
+EOF
+
# Forward logs to console directly, and don't read Kmsg, this avoids
# unneccessary memory consumption and make console output more useful.
# Only do so for non fadump image.
mkdir -p "${initdir}/etc/systemd/journald.conf.d"
- echo "[Journal]" > "${initdir}/etc/systemd/journald.conf.d/kdump.conf"
- echo "Storage=volatile" >> "${initdir}/etc/systemd/journald.conf.d/kdump.conf"
- echo "ReadKMsg=no" >> "${initdir}/etc/systemd/journald.conf.d/kdump.conf"
- echo "ForwardToConsole=yes" >> "${initdir}/etc/systemd/journald.conf.d/kdump.conf"
+ cat > "${initdir}/etc/systemd/journald.conf.d/99-kdump.conf" << EOF
+[Journal]
+Storage=volatile
+ReadKMsg=no
+ForwardToConsole=yes
+EOF
}
remove_cpu_online_rule() {
--
2.44.0

120
0001-kdumpctl-Drop-default-kexec-d-option.patch
View File

@ -1,120 +0,0 @@
From 32f9ada4bfa9f9690bdcdb21de40568c81a5ea80 Mon Sep 17 00:00:00 2001
From: Philipp Rudo <prudo@redhat.com>
Date: Fri, 14 Jun 2024 11:48:24 +0200
Subject: [PATCH 1/7] kdumpctl: Drop default kexec '-d' option
Kernel commits cbc2fe9d9cb2 ("kexec_file: add kexec_file flag to control
debug printing") and a85ee18c7900 ("kexec_file: print out debugging
message if required") added debug messages to the kexec_file_load system
call when option -d is provided to the kexec user space tool. As
kexec_file_load is the default and option -d is set by default these
messages are always printed when a crash kernel is loaded. This not only
clutters the kernel log but also potentially leaks confidential kernel
information to users. As the messages are printed to the kernel log, not
stderr, the redirection to /var/log/kdump.log won't catch them. This
will become even more problematic as for RHEL10 the kernel will be built
without support for the kexec_load system call. So kexec_file_load will
be the only choice in the future.
The redirection also caused confusion in a recent bug report. There a
user moved a working /etc/sysconfig/kdump from ppc to s390 with
KEXEC_ARGS containing the --dt-no-old-root option. This option is arch
specific and does not exist on s390. Thus the kexec-tools failed with an
'unrecognized option' error followed by the usage(). The problem was
that the 'unrecognized option' error is printed to stderr, which got
redirected to /var/log/kdump.log, while the usage() is printed to
stdout, which ended up in the systemd journal. This caused confusion as
the user only checked the journal and found the usage() without any
error message.
Thus remove the default -d option and the redirection of stderr to
/var/log/kdump.log for the kexec-tools user space tool.
This commit ultimately reverts 88a8b94 ("kdumpctl: add the '-d' option to
enable the kexec loading debugging messages").
Signed-off-by: Philipp Rudo <prudo@redhat.com>
---
kdumpctl | 21 +++------------------
kexec-kdump-howto.txt | 7 +------
2 files changed, 4 insertions(+), 24 deletions(-)
diff --git a/kdumpctl b/kdumpctl
index 30eb27d..8dc56e5 100755
--- a/kdumpctl
+++ b/kdumpctl
@@ -5,7 +5,6 @@ KDUMP_KERNELVER=""
KDUMP_KERNEL=""
KDUMP_COMMANDLINE=""
KEXEC_ARGS=""
-KDUMP_LOG_PATH="/var/log"
MKDUMPRD="/sbin/mkdumprd -f"
MKFADUMPRD="/sbin/mkfadumprd"
DRACUT_MODULES_FILE="/usr/lib/dracut/modules.txt"
@@ -17,7 +16,7 @@ TARGET_INITRD=""
#kdump shall be the default dump mode
DEFAULT_DUMP_MODE="kdump"
-standard_kexec_args="-d -p"
+standard_kexec_args="-p"
# Some default values in case /etc/sysconfig/kdump doesn't include
KDUMP_COMMANDLINE_REMOVE="hugepages hugepagesz slub_debug"
@@ -696,7 +695,7 @@ function load_kdump_kernel_key()
# as the currently running kernel.
load_kdump()
{
- local ret uki
+ local uki
KEXEC_ARGS=$(prepare_kexec_args "${KEXEC_ARGS}")
KDUMP_COMMANDLINE=$(prepare_cmdline "${KDUMP_COMMANDLINE}" "${KDUMP_COMMANDLINE_REMOVE}" "${KDUMP_COMMANDLINE_APPEND}")
@@ -713,26 +712,12 @@ load_kdump()
ddebug "$KEXEC $KEXEC_ARGS $standard_kexec_args --command-line=$KDUMP_COMMANDLINE --initrd=$TARGET_INITRD $KDUMP_KERNEL"
- # The '12' represents an intermediate temporary file descriptor
- # to store the standard error file descriptor '2', and later
- # restore the error file descriptor with the file descriptor '12'
- # and release it.
- exec 12>&2
- exec 2>> $KDUMP_LOG_PATH/kdump.log
- chmod 600 $KDUMP_LOG_PATH/kdump.log
- PS4='+ $(date "+%Y-%m-%d %H:%M:%S") ${BASH_SOURCE}@${LINENO}: '
- set -x
-
# shellcheck disable=SC2086
$KEXEC $KEXEC_ARGS $standard_kexec_args \
--command-line="$KDUMP_COMMANDLINE" \
--initrd="$TARGET_INITRD" "$KDUMP_KERNEL"
- ret=$?
- set +x
- exec 2>&12 12>&-
-
- if [[ $ret == 0 ]]; then
+ if [[ $? == 0 ]]; then
dinfo "kexec: loaded kdump kernel"
return 0
else
diff --git a/kexec-kdump-howto.txt b/kexec-kdump-howto.txt
index 6741faf..c65d45a 100644
--- a/kexec-kdump-howto.txt
+++ b/kexec-kdump-howto.txt
@@ -1016,12 +1016,7 @@ Debugging Tips
and the second kernel.
In the first kernel, you can find the historical logs with the journalctl
- command and check kdump service debugging information. In addition, the
- 'kexec -d' debugging messages are also saved to /var/log/kdump.log in the
- first kernel. For example:
-
- [root@ibm-z-109 ~]# ls -al /var/log/kdump.log
- -rw-r--r--. 1 root root 63238 Oct 28 06:40 /var/log/kdump.log
+ command and check kdump service debugging information.
If you want to get the debugging information of building kdump initramfs, you
can enable the '--debug' option for the dracut_args in the /etc/kdump.conf, and
--
2.45.2

30
0002-kdump-lib-fix-sed-expression-in-prepare_cmdline-on-a.patch
View File

@ -1,30 +0,0 @@
From d25d3d4500d1ca1357350fa4a28ac89a33f250e0 Mon Sep 17 00:00:00 2001
From: Philipp Rudo <prudo@redhat.com>
Date: Tue, 18 Jun 2024 14:57:16 +0200
Subject: [PATCH 2/7] kdump-lib: fix sed expression in prepare_cmdline on
aarch64
The sed expression misses a 's' for search and replace...
Fixes: 0f6ad91 ("kdump-lib: fix prepare_cmdline")
Signed-off-by: Philipp Rudo <prudo@redhat.com>
---
kdump-lib.sh | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kdump-lib.sh b/kdump-lib.sh
index 79714f9..35f0ac9 100755
--- a/kdump-lib.sh
+++ b/kdump-lib.sh
@@ -802,7 +802,7 @@ prepare_cmdline()
# This is a workaround on AWS platform. Always remove irqpoll since it
# may cause the hot-remove of some pci hotplug device.
- is_aws_aarch64 && out=$(echo "$out" | sed -e "/\<irqpoll\>//")
+ is_aws_aarch64 && out=$(echo "$out" | sed -e "s/\<irqpoll\>//")
# Always disable gpt-auto-generator as it hangs during boot of the
# crash kernel. Furthermore we know which disk will be used for dumping
--
2.45.2

55
0002-mkdumprd-Fix-makedumpfile-parameter-check.patch
View File

@ -1,55 +0,0 @@
From 247c7a5f39b305f9a83bad2d936d00237165b7e0 Mon Sep 17 00:00:00 2001
From: "Mamoru Nishibe (Fujitsu)" <nishibe.mamoru@fujitsu.com>
Date: Wed, 24 Apr 2024 08:11:12 +0000
Subject: [PATCH 2/7] mkdumprd: Fix makedumpfile parameter check.
If only "makedumpfile" is written in "core_collector" of /etc/kdump.conf
and try to run makedumpfile without options,
"makedumpfile --check-params" fails and terminates abnormally.
# grep ^core_collector /etc/kdump.conf
core_collector makedumpfile
# /usr/bin/kdumpctl start
:
Commandline parameter is invalid.
Try `makedumpfile --help' for more information.
kdump: makedumpfile parameter check failed.
kdump: mkdumprd: failed to make kdump initrd
kdump: Starting kdump: [FAILED]
On the other hand, "makedumpfile --check-params" works fine without any options.
# makedumpfile --check-params vmcore dumpfile
# echo $?
0
In addition, before verify_core_collector() was implemented,
initial RAM for kdump was successfully created using only "core_collector makedumpfile".
I consider it a regression.
This is due to a parameter extraction error in verify_core_collector().
Fix it to correctly extract only the options as follows.
Fixes: a1c28126 ("mkdumprd: Use makedumpfile --check-params option")
Signed-off-by: Mamoru Nishibe <nishibe.mamoru@fujitsu.com>
Reviewed-by: Coiby Xu <coxu@redhat.com>
---
mkdumprd | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mkdumprd b/mkdumprd
index 31c4b76..27eed5e 100644
--- a/mkdumprd
+++ b/mkdumprd
@@ -256,7 +256,7 @@ check_user_configured_target()
verify_core_collector()
{
local _cmd="${1%% *}"
- local _params="${1#* }"
+ local _params="${1#${_cmd}}"
if [[ $_cmd != "makedumpfile" ]]; then
if is_raw_dump_target; then
--
2.44.0

53
0003-Try-to-install-PHY-and-MDIO-bus-drivers-explicitly.patch
View File

@ -1,53 +0,0 @@
From d057153a1c3c36612a14143b29c0ff0be34e4fc2 Mon Sep 17 00:00:00 2001
From: Coiby Xu <coxu@redhat.com>
Date: Thu, 21 Sep 2023 11:50:14 +0800
Subject: [PATCH 3/7] Try to install PHY and MDIO bus drivers explicitly
Resolves: https://issues.redhat.com/browse/RHEL-7028
Currently, nfs dumping fails on some machines that has a dedicated PHY
driver (dealing with the physical layer) or MDIO bus (connecting the MAC
to PHY devices) driver. This is because kexec-tools doesn't install
dedicated PHY or MDIO driver explicitly. Usually a NIC driver shouldn't
specify the dependency on the needed PHY or MDIO driver because it
shouldn't a NIC (medium access control, MAC) driver is for dealing with
the Data link layer and a PHY driver is for physical layer. So as long
as a MAC driver can talk to the PHY layer via APIs, it shouldn't care
which PHY driver or device it's talking to. So when the
dependency on a PHY driver or MDIO driver is not found by dracut's
instmods, the PHY or MDIO driver won't be installed.
This patch passes =drivers/net/phy and =drivers/net/mdio to dracut's
instmods which will only install in-use PHY or MDIO driver(s).
Note ideally we should find out which PHY driver is used by a NIC but
unfortunately currently no universal way can be found
(/sys/class/net/NIC_NAME/phydev/driver/module can be used to find the
name of the PHY driver for some NICs but it doesn't exist for some NICs
like Qualcomm Atheros AR8031). So is it for a MDIO bus driver.
Fortunately currently no huge memory consumption is found for a PHY or
MDIO driver.
Fixes: a65dde2d ("Reduce kdump memory consumption by only installing needed NIC drivers")
Reported-by: Doreen Alongi <dalongi@redhat.com>
Signed-off-by: Coiby Xu <coxu@redhat.com>
---
dracut-module-setup.sh | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/dracut-module-setup.sh b/dracut-module-setup.sh
index acc0b28..3c9bdbe 100755
--- a/dracut-module-setup.sh
+++ b/dracut-module-setup.sh
@@ -384,7 +384,7 @@ _get_hpyerv_physical_driver() {
kdump_install_nic_driver() {
local _netif _driver _drivers
- _drivers=()
+ _drivers=('=drivers/net/phy' '=drivers/net/mdio')
for _netif in $1; do
[[ $_netif == lo ]] && continue
--
2.44.0

127
0003-kdumpctl-Simplify-fadump-handling-in-reset_crashkern.patch
View File

@ -1,127 +0,0 @@
From 8ebf2874a202a2d7116a27c69816b8621ace5224 Mon Sep 17 00:00:00 2001
From: Philipp Rudo <prudo@redhat.com>
Date: Thu, 1 Jun 2023 17:04:13 +0200
Subject: [PATCH 3/7] kdumpctl: Simplify fadump handling in reset_crashkernel
When handling fadump there are three cases we need to consider
1) When running on non-ppc64le systems
In this case _dump_mode=kdump and _new_fadump='' always. In other
words we don't need to care updating the fadump parameter on the
kernel command line. We could always remove it like the code did so
far. But should we remove it when we never set it? In particular as
that might overwrite a change explicitly made by the user (for
whatever reason)
2) When running on ppc64le and the user provided --fadump option
In this case _new_fadump and _dump_mode are set accordingly to what
the user provided. Thus we need to update both the crashkernel (in
case fadump was turned on/off) and the fadump (in case the fadump
mode changed) parameters.
3) When running on ppc64le but the user did not provide --fadump option
In this case both _new_fadump='' and _dump_mode=''. In this case we
take over the previously set fadump parameter. Which means that we
don't need to update the fadump parameter at all. We do need to check
whether the _new_dump_mode is fadump or kdump though so we use the
correct (new) default crashkernel value.
In the three cases only the second one needs to update the fadump
parameter. Reflect this discussion in code.
This also fixes a bug that always prints
kdump: Updated fadump= for kernel=$kernel. Please reboot the
system for the change to take effect.
when the crashkernel= parameter is unchanged as well as reboots the
system, if --reboot is provided. Even for non-ppc architectures.
Fixes: 140da74 ("rewrite reset_crashkernel to support fadump and to used by RPM scriptlet")
Signed-off-by: Philipp Rudo <prudo@redhat.com>
---
kdumpctl | 38 ++++++++++++++++++++++----------------
1 file changed, 22 insertions(+), 16 deletions(-)
diff --git a/kdumpctl b/kdumpctl
index 8dc56e5..8ec638b 100755
--- a/kdumpctl
+++ b/kdumpctl
@@ -1467,10 +1467,10 @@ _get_all_kernels_from_grubby()
reset_crashkernel()
{
local _opt _val _reboot _grubby_kernel_path _kernel _kernels
- local _dump_mode _new_dump_mode
+ local _dump_mode
local _has_changed _needs_reboot
local _old_ck _new_ck
- local _old_fadump _new_fadump
+ local _old_fadump _new_fadump _opt_fadump
for _opt in "$@"; do
case "$_opt" in
@@ -1479,12 +1479,11 @@ reset_crashkernel()
derror "Option --fadump only valid on PPC"
exit 1
fi
- _new_fadump=${_opt#*=}
- if ! _dump_mode=$(get_dump_mode_by_fadump_val $_new_fadump); then
+ _opt_fadump=${_opt#*=}
+ if ! _dump_mode=$(get_dump_mode_by_fadump_val $_opt_fadump); then
derror "failed to determine dump mode"
exit
fi
- [[ "$_new_fadump" == off ]] && _new_fadump=""
;;
--kernel=*)
_val=${_opt#*=}
@@ -1519,8 +1518,6 @@ reset_crashkernel()
return
fi
- [[ $(uname -m) != ppc64le ]] && _dump_mode=kdump
-
# If kernel-path not specified, either
# - use KDUMP_KERNELVER if it's defined
# - use current running kernel
@@ -1536,19 +1533,28 @@ reset_crashkernel()
for _kernel in $_kernels; do
_has_changed=""
- if [[ -z $_dump_mode ]]; then
- _new_dump_mode=$(get_dump_mode_by_kernel "$_kernel")
- _new_fadump=$(get_grub_kernel_boot_parameter "$_kernel" fadump)
+ if [[ $(uname -m) == ppc64le ]]; then
+ if [[ -n "$_opt_fadump" ]]; then
+ _new_ck=$(kdump_get_arch_recommend_crashkernel "$_dump_mode")
+ _old_fadump=$(get_grub_kernel_boot_parameter "$_kernel" fadump)
+ _new_fadump="$_opt_fadump"
+ [[ "$_new_fadump" == off ]] && _new_fadump=""
+ if _update_kernel_cmdline "$_kernel" fadump "$_old_fadump" "$_new_fadump"; then
+ if [[ -n "$_new_fadump" ]]; then
+ _has_changed="Updated fadump=$_new_fadump"
+ else
+ _has_changed="Removed fadump"
+ fi
+ fi
+ else
+ _dump_mode="$(get_dump_mode_by_kernel "$_kernel")"
+ _new_ck=$(kdump_get_arch_recommend_crashkernel "$_dump_mode")
+ fi
else
- _new_dump_mode=$_dump_mode
+ _new_ck=$(kdump_get_arch_recommend_crashkernel kdump)
fi
_old_ck=$(get_grub_kernel_boot_parameter "$_kernel" crashkernel)
- _new_ck=$(kdump_get_arch_recommend_crashkernel "$_new_dump_mode")
- _old_fadump=$(get_grub_kernel_boot_parameter "$_kernel" fadump)
- if _update_kernel_cmdline "$_kernel" fadump "$_old_fadump" "$_new_fadump"; then
- _has_changed="Updated fadump=$_new_fadump"
- fi
if _update_kernel_cmdline "$_kernel" crashkernel "$_old_ck" "$_new_ck"; then
_has_changed="Updated crashkernel=$_new_ck"
fi
--
2.45.2

50
0004-Install-the-driver-of-physical-device-for-a-SR-IOV-v.patch
View File

@ -1,50 +0,0 @@
From 7a8edc8de67dccae23b01461bc3b17c0ad42aa5f Mon Sep 17 00:00:00 2001
From: Coiby Xu <coxu@redhat.com>
Date: Wed, 27 Sep 2023 09:31:39 +0800
Subject: [PATCH 4/7] Install the driver of physical device for a SR-IOV
virtual device
Currently, network dumping failed over a NIC that is a Single Root I/O
Virtualization (SR-IOV) virtual device. Usually the driver of the
virtual device won't specify the dependency on the driver of the
physical device. So to fix this issue, the driver of the physical device
needs to be found and installed as well.
Fixes: a65dde2d ("Reduce kdump memory consumption by only installing needed NIC drivers")
Signed-off-by: Coiby Xu <coxu@redhat.com>
---
dracut-module-setup.sh | 11 +++++++++++
1 file changed, 11 insertions(+)
diff --git a/dracut-module-setup.sh b/dracut-module-setup.sh
index 3c9bdbe..89d9959 100755
--- a/dracut-module-setup.sh
+++ b/dracut-module-setup.sh
@@ -381,6 +381,14 @@ _get_hpyerv_physical_driver() {
_get_nic_driver "$_physical_nic"
}
+_get_physical_function_driver() {
+ local _physfn_dir=/sys/class/net/"$1"/device/physfn
+
+ if [[ -e "$_physfn_dir" ]]; then
+ basename "$(readlink -f "$_physfn_dir"/driver)"
+ fi
+}
+
kdump_install_nic_driver() {
local _netif _driver _drivers
@@ -408,6 +416,9 @@ kdump_install_nic_driver() {
fi
_drivers+=("$_driver")
+ # For a Single Root I/O Virtualization (SR-IOV) virtual device,
+ # the driver of physical device needs to be installed as well
+ _drivers+=("$(_get_physical_function_driver "$_netif")")
done
[[ -n ${_drivers[*]} ]] || return
--
2.44.0

32
0004-kdumpctl.8-Add-description-to-reset-crashkernel-rebo.patch
View File

@ -1,32 +0,0 @@
From de393c1bcb7d32e97d9b46bb6a73180072f03aa3 Mon Sep 17 00:00:00 2001
From: Philipp Rudo <prudo@redhat.com>
Date: Mon, 1 Jul 2024 12:52:39 +0200
Subject: [PATCH 4/7] kdumpctl.8: Add description to reset-crashkernel --reboot
There is no description for parameter --reboot for reset-crashkernel.
Thus add one.
Suggested-by: Lichen Liu <lichliu@redhat.com>
Signed-off-by: Philipp Rudo <prudo@redhat.com>
---
kdumpctl.8 | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/kdumpctl.8 b/kdumpctl.8
index 33c1115..29a6119 100644
--- a/kdumpctl.8
+++ b/kdumpctl.8
@@ -62,6 +62,10 @@ grubby's kernel-path=ALL and kernel-path=DEFAULT. ppc64le supports FADump and
supports an additional [--fadump=[on|off|nocma]] parameter to toggle FADump
on/off.
+If the optional parameter [--reboot] is provided the system will automatically
+reboot for changes to take effect. If no changes were made to the kernel
+command line the reboot is omitted.
+
Note: The memory requirements for kdump varies heavily depending on the
used hardware and system configuration. Thus the recommended
crashkernel might not work for your specific setup. Please test if
--
2.45.2

32
0005-dracut-kdump.sh-Save-kexec-dmesg.log-after-failure_a.patch
View File

@ -1,32 +0,0 @@
From afc6ddb781079dfbdf7919803114458907fc4e6a Mon Sep 17 00:00:00 2001
From: Philipp Rudo <prudo@redhat.com>
Date: Thu, 20 Jun 2024 13:19:30 +0200
Subject: [PATCH 5/7] dracut-kdump.sh: Save kexec-dmesg.log after
failure_action
The kexec-dmesg.log is only saved after the initial dump attempt but not
for the failure_action. So in case the initial dump attempt failed and
the failure_action is dump_to_rootfs the kexec-dmesg.log is missing. Fix
that by calling save_log also after executing the failure_action.
Fixes: 3d70f8b ("logger: save log after all kdump progress finished")
Signed-off-by: Philipp Rudo <prudo@redhat.com>
---
dracut-kdump.sh | 1 +
1 file changed, 1 insertion(+)
diff --git a/dracut-kdump.sh b/dracut-kdump.sh
index 7b3ad7a..2e078c7 100755
--- a/dracut-kdump.sh
+++ b/dracut-kdump.sh
@@ -579,6 +579,7 @@ fence_kdump_notify()
if [ "$1" = "--error-handler" ]; then
get_kdump_confs
do_failure_action
+ save_log
do_final_action
exit $?
--
2.45.2

47
0005-update-98-kexec-rules-for-crash-hotplug.patch
View File

@ -1,47 +0,0 @@
From 659e0aae8f00570c85e82e1317153bf89e59929c Mon Sep 17 00:00:00 2001
From: Baoquan He <bhe@redhat.com>
Date: Thu, 7 Sep 2023 10:37:20 -0500
Subject: [PATCH 5/7] update 98-kexec rules for crash hotplug
In kernel, with the support of cpu/memory hotplug on crash, kdump
reloading only needs to update the elfcorehdr.
To realize the benefits, we need prevent udev from updating kdump
kernel on hot un/plug changes when detecting that the crash_hotplug
sysfs nodes are present.
Link: https://lore.kernel.org/lkml/20230814214446.6659-1-eric.devolder@oracle.com/
Link: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=d68b4b6f307d155475cce541f2aee938032ed22e
Signed-off-by: Baoquan He <bhe@redhat.com>
---
98-kexec.rules | 4 ++++
98-kexec.rules.ppc64 | 4 ++++
2 files changed, 8 insertions(+)
diff --git a/98-kexec.rules b/98-kexec.rules
index b73b701..52b2ee8 100644
--- a/98-kexec.rules
+++ b/98-kexec.rules
@@ -1,3 +1,7 @@
+# The kernel updates the crash elfcorehdr for CPU and memory changes
+SUBSYSTEM=="cpu", ATTRS{crash_hotplug}=="1", GOTO="kdump_reload_end"
+SUBSYSTEM=="memory", ATTRS{crash_hotplug}=="1", GOTO="kdump_reload_end"
+
SUBSYSTEM=="cpu", ACTION=="add", GOTO="kdump_reload"
SUBSYSTEM=="cpu", ACTION=="remove", GOTO="kdump_reload"
SUBSYSTEM=="memory", ACTION=="online", GOTO="kdump_reload"
diff --git a/98-kexec.rules.ppc64 b/98-kexec.rules.ppc64
index e9db276..e7735b3 100644
--- a/98-kexec.rules.ppc64
+++ b/98-kexec.rules.ppc64
@@ -1,3 +1,7 @@
+# The kernel updates the crash elfcorehdr for CPU and memory changes
+SUBSYSTEM=="cpu", ATTRS{crash_hotplug}=="1", GOTO="kdump_reload_end"
+SUBSYSTEM=="memory", ATTRS{crash_hotplug}=="1", GOTO="kdump_reload_end"
+
SUBSYSTEM=="cpu", ACTION=="online", GOTO="kdump_reload_cpu"
SUBSYSTEM=="memory", ACTION=="online", GOTO="kdump_reload_mem"
SUBSYSTEM=="memory", ACTION=="offline", GOTO="kdump_reload_mem"
--
2.44.0

43
0006-Makefile-Fix-early-kdump-file-names.patch
View File

@ -1,43 +0,0 @@
From e6b134a1593036b47a2a35da3c695a0db90de73f Mon Sep 17 00:00:00 2001
From: Philipp Rudo <prudo@redhat.com>
Date: Mon, 24 Jun 2024 13:44:22 +0200
Subject: [PATCH 6/7] Makefile: Fix early-kdump file names
When creating the kdump-utils subpackage the files for earlykdump were
incorrectly renamed to kdump.sh and kdump-module-setup.sh rather than
the expected early-kdump.sh and module-setup.sh. These incorrect file
names then got transferred to the newly created Makefile with fe372af
("Upstream kdump-utils").
With those incorrect file names dracut fails when trying to add the
earlykdump module with
# dracut --add earlykdump kdump.img
dracut[E]: Module 'earlykdump' cannot be installed.
Fix the file names to allow installing the earlykdump module.
Fixes: 372b4c6 ("Add kdump-utils subpackage")
Signed-off-by: Philipp Rudo <prudo@redhat.com>
---
Makefile | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/Makefile b/Makefile
index e1a511c..538e4f4 100644
--- a/Makefile
+++ b/Makefile
@@ -26,8 +26,8 @@ dracut-modules:
install -m 644 dracut-kdump-emergency.target $(kdumpbasemoddir)/kdump-emergency.target
mkdir -p -m755 $(dracutmoddir)/99earlykdump
- install -m 755 dracut-early-kdump.sh $(dracutmoddir)/99earlykdump/kdump.sh
- install -m 755 dracut-early-kdump-module-setup.sh $(dracutmoddir)/99earlykdump/kdump-module-setup.sh
+ install -m 755 dracut-early-kdump.sh $(dracutmoddir)/99earlykdump/early-kdump.sh
+ install -m 755 dracut-early-kdump-module-setup.sh $(dracutmoddir)/99earlykdump/module-setup.sh
ifeq ($(ARCH), $(filter ppc64le ppc64,$(ARCH)))
mkdir -p -m755 $(dracutmoddir)/99zz-fadumpinit
--
2.45.2

61
0006-sysconfig-add-pcie_ports-compat-to-KDUMP_COMMANDLINE.patch
View File

@ -1,61 +0,0 @@
From ada6f5edf1ae06fc88759aa2f94d09e2a98d21ef Mon Sep 17 00:00:00 2001
From: Tao Liu <ltao@redhat.com>
Date: Wed, 1 May 2024 16:53:19 +0800
Subject: [PATCH 6/7] sysconfig: add pcie_ports compat to
KDUMP_COMMANDLINE_APPEND on x86_64
There have been some of failing cases of kdump in 2nd kernel, where
ususally only one cpu is enabled by "nr_cpus=1", but with a large
number of devices, which may easily exceed the maximum IRQ resources of
one cpu can handle. As a result, the 2nd kernel will hang and kdump
fails. This issue is often observed on machines with many cpus and many
devices.
On those systems, pcieports consume quite proportion of IRQ resources,
many following message can be seen in dmesg log:
pcieport 0000:18:01.0: PME: Signaling with IRQ 109
According to kernel doc[1], when "pcie_ports=compat" applied, it will disable
native PCIe services (PME, AER, DPC, PCIe hotplug). Those functions are
power management events, error reporting, performance, hotplug related,
which are not the must-have functions for kdump. In addition, after
testing, no side effects such as cannot writing vmcore into sdx, nvme
etc been noticed.
This patch will disable native PCIe services for 2nd kernel, to saving the
scarce IRQ resources and increase the kdump success.
Attach Prarit's comments:
This makes sense to me. The only concern anyone should have is that a PCIE
error could have been responsible for taking down the kernel in the first
place, and booting into the second kernel could then also have a fatal
problem. I'm not sure we can ever fix that type of cascade of panics :)
so it makes sense to disable these features.
[1]: https://www.kernel.org/doc/html/v6.9-rc1/admin-guide/kernel-parameters.html
Signed-off-by: Tao Liu <ltao@redhat.com>
Acked-by: Prarit Bhargava <prarit@redhat.com>
Acked-by: Dave Young <dyoung@redhat.com>
---
gen-kdump-sysconfig.sh | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gen-kdump-sysconfig.sh b/gen-kdump-sysconfig.sh
index 78b0bb7..1a2cd92 100755
--- a/gen-kdump-sysconfig.sh
+++ b/gen-kdump-sysconfig.sh
@@ -104,7 +104,7 @@ s390x)
x86_64)
update_param KEXEC_ARGS "-s"
update_param KDUMP_COMMANDLINE_APPEND \
- "irqpoll nr_cpus=1 reset_devices cgroup_disable=memory mce=off numa=off udev.children-max=2 panic=10 acpi_no_memhotplug transparent_hugepage=never nokaslr hest_disable novmcoredd cma=0 hugetlb_cma=0"
+ "irqpoll nr_cpus=1 reset_devices cgroup_disable=memory mce=off numa=off udev.children-max=2 panic=10 acpi_no_memhotplug transparent_hugepage=never nokaslr hest_disable novmcoredd cma=0 hugetlb_cma=0 pcie_ports=compat"
;;
*)
echo "Warning: Unknown architecture '$1', using default sysconfig template." >&2
--
2.44.0

41
0007-fadump-udev-do-not-re-register-fadump-if-kernel-hotp.patch
View File

@ -1,41 +0,0 @@
From b4e3d3724cf372493b404586126067ff66e550d6 Mon Sep 17 00:00:00 2001
From: Sourabh Jain <sourabhjain@linux.ibm.com>
Date: Fri, 26 Apr 2024 18:17:49 +0530
Subject: [PATCH 7/7] fadump/udev: do not re-register fadump if kernel hotplug
ready
With the introduction of kernel commit c6c5b14dac0d ("powerpc: make fadump
resilient with memory add/remove events") linux kernel now internally manages
the update of elfcorehdr during memory add/remove events. So no need to
re-register fadump if the /sys/kernel/fadump/hotplug_ready is set to 1.
No impact for kernels that do not have /sys/kernel/fadump/hotplug_ready
sysfs.
Relevant kernel commit links:
1. https://msgid.link/20240422195932.1583833-2-sourabhjain@linux.ibm.com
2. https://msgid.link/20240422195932.1583833-3-sourabhjain@linux.ibm.com
Signed-off-by: Sourabh Jain <sourabhjain@linux.ibm.com>
---
98-kexec.rules.ppc64 | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/98-kexec.rules.ppc64 b/98-kexec.rules.ppc64
index e7735b3..85fe0b1 100644
--- a/98-kexec.rules.ppc64
+++ b/98-kexec.rules.ppc64
@@ -15,7 +15,9 @@ GOTO="kdump_reload_end"
LABEL="kdump_reload_mem"
-RUN+="/bin/sh -c '/usr/bin/systemctl is-active kdump.service || exit 0; /usr/bin/systemd-run --quiet --no-block /usr/lib/udev/kdump-udev-throttler'"
+# Don't re-register fadump if /sys/kernel/fadump/hotplug_ready sysfs is set to 1.
+
+RUN+="/bin/sh -c '/usr/bin/systemctl is-active kdump.service || exit 0; ! test -f /sys/kernel/fadump/hotplug_ready || cat /sys/kernel/fadump/hotplug_ready | grep 1 || exit 0; /usr/bin/systemd-run --quiet --no-block /usr/lib/udev/kdump-udev-throttler'"
GOTO="kdump_reload_end"
--
2.44.0

36
0007-kdump-lib-Drop-file-dependency-in-is_uki.patch
View File

@ -1,36 +0,0 @@
From efc06f21a5b1b3a30ce438c15ce4a07fdfdd2440 Mon Sep 17 00:00:00 2001
From: Philipp Rudo <prudo@redhat.com>
Date: Mon, 24 Jun 2024 17:34:35 +0200
Subject: [PATCH 7/7] kdump-lib: Drop 'file' dependency in is_uki
The 'file' utility is no longer installed per default. In addition there
was an update to it so it now reports the file type as
application/vnd.microsoft.portable-executable. Thus fall back to use
objdump to avoid adding yet an other dependency for kdump-utils and deal
with different versions of 'file'.
Note: This has the small drawback that objdump is arch specific. I.e.
examining a aarch64 UKI on a x86_64 machine will return an 'file format
not recognized' error.
Signed-off-by: Philipp Rudo <prudo@redhat.com>
---
kdump-lib.sh | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kdump-lib.sh b/kdump-lib.sh
index 35f0ac9..13e16e3 100755
--- a/kdump-lib.sh
+++ b/kdump-lib.sh
@@ -18,7 +18,7 @@ is_uki()
img="$1"
[[ -f "$img" ]] || return
- [[ "$(file -b --mime-type "$img")" == application/x-dosexec ]] || return
+ [[ "$(objdump -a "$img" 2> /dev/null)" =~ pei-(x86-64|aarch64-little) ]] || return
objdump -h -j .linux "$img" &> /dev/null
}
--
2.45.2

132
kexec-tools-2.0.28-makedumfpile-0001-PATCH-ppc64-get-vmalloc-start-address-from-vmcoreinf.patch
View File

@ -1,132 +0,0 @@
From: Coiby Xu <coxu@redhat.com>
Subject: [PATCH] ppc64: get vmalloc start address from vmcoreinfo
Bugzilla: https://bugzilla.redhat.com/2269991
commit 94241fd2feed059227a243618f2acc6aabf366e8
Author: Aditya Gupta <adityag@linux.ibm.com>
Date: Sat Feb 24 00:33:42 2024 +0530
[PATCH] ppc64: get vmalloc start address from vmcoreinfo
Below error was noticed when running makedumpfile on linux-next kernel
crash (linux-next tag next-20240121):
Checking for memory holes : [100.0 %] | readpage_elf: Attempt to read non-existent page at 0xc000000000000.
[ 17.551718] kdump.sh[404]: readmem: type_addr: 0, addr:c00c000000000000, size:16384
[ 17.551793] kdump.sh[404]: __exclude_unnecessary_pages: Can't read the buffer of struct page.
[ 17.551864] kdump.sh[404]: create_2nd_bitmap: Can't exclude unnecessary pages.
[ 17.562632] kdump.sh[404]: The kernel version is not supported.
[ 17.562708] kdump.sh[404]: The makedumpfile operation may be incomplete.
[ 17.562773] kdump.sh[404]: makedumpfile Failed.
[ 17.564335] kdump[406]: saving vmcore failed, _exitcode:1
Above error was due to 'vmap_area_list' and 'vmlist' symbols missing
from the vmcore. 'vmap_area_list' was removed in the linux kernel
6.9-rc1 by the commit below:
commit 55c49fee57af99f3c663e69dedc5b85e691bbe50
mm/vmalloc: remove vmap_area_list
Subsequently the commit also introduced 'VMALLOC_START' in vmcoreinfo to
get base address of vmalloc area, instead of depending on 'vmap_area_list'
Hence if 'VMALLOC_START' symbol is there in vmcoreinfo:
1. Set vmalloc_start based on 'VMALLOC_START'
2. Don't error if vmap_area_list/vmlist are not defined
Reported-by: Sachin Sant <sachinp@linux.ibm.com>
Signed-off-by: Aditya Gupta <adityag@linux.ibm.com>
Signed-off-by: Coiby Xu <coxu@redhat.com>
diff --git a/makedumpfile-1.7.4/arch/ppc64.c b/makedumpfile-1.7.4/arch/ppc64.c
index 3b4f91981f71d035b94282f6c7e33323a4e4c1fd..a54f9a04db7f26eac2f1bd065b134a7e2fdaeb67 100644
--- a/makedumpfile-1.7.4/arch/ppc64.c
+++ b/makedumpfile-1.7.4/arch/ppc64.c
@@ -568,7 +568,9 @@ get_machdep_info_ppc64(void)
/*
* Get vmalloc_start value from either vmap_area_list or vmlist.
*/
- if ((SYMBOL(vmap_area_list) != NOT_FOUND_SYMBOL)
+ if (NUMBER(vmalloc_start) != NOT_FOUND_NUMBER) {
+ vmalloc_start = NUMBER(vmalloc_start);
+ } else if ((SYMBOL(vmap_area_list) != NOT_FOUND_SYMBOL)
&& (OFFSET(vmap_area.va_start) != NOT_FOUND_STRUCTURE)
&& (OFFSET(vmap_area.list) != NOT_FOUND_STRUCTURE)) {
if (!readmem(VADDR, SYMBOL(vmap_area_list) + OFFSET(list_head.next),
@@ -689,11 +691,16 @@ vaddr_to_paddr_ppc64(unsigned long vaddr)
if ((SYMBOL(vmap_area_list) == NOT_FOUND_SYMBOL)
|| (OFFSET(vmap_area.va_start) == NOT_FOUND_STRUCTURE)
|| (OFFSET(vmap_area.list) == NOT_FOUND_STRUCTURE)) {
- if ((SYMBOL(vmlist) == NOT_FOUND_SYMBOL)
- || (OFFSET(vm_struct.addr) == NOT_FOUND_STRUCTURE)) {
- ERRMSG("Can't get info for vmalloc translation.\n");
- return NOT_PADDR;
- }
+ /*
+ * Don't depend on vmap_area_list/vmlist if vmalloc_start is set in
+ * vmcoreinfo, in that case proceed without error
+ */
+ if (NUMBER(vmalloc_start) == NOT_FOUND_NUMBER)
+ if ((SYMBOL(vmlist) == NOT_FOUND_SYMBOL)
+ || (OFFSET(vm_struct.addr) == NOT_FOUND_STRUCTURE)) {
+ ERRMSG("Can't get info for vmalloc translation.\n");
+ return NOT_PADDR;
+ }
}
return ppc64_vtop_level4(vaddr);
diff --git a/makedumpfile-1.7.4/makedumpfile.c b/makedumpfile-1.7.4/makedumpfile.c
index 58c6639f289f19cdbf39ed3899be9893fdc317fe..d7f1dd41d2cab526d7d40e809ddccf656c586811 100644
--- a/makedumpfile-1.7.4/makedumpfile.c
+++ b/makedumpfile-1.7.4/makedumpfile.c
@@ -2978,6 +2978,8 @@ read_vmcoreinfo(void)
READ_NUMBER("PAGE_OFFLINE_MAPCOUNT_VALUE", PAGE_OFFLINE_MAPCOUNT_VALUE);
READ_NUMBER("phys_base", phys_base);
READ_NUMBER("KERNEL_IMAGE_SIZE", KERNEL_IMAGE_SIZE);
+
+ READ_NUMBER_UNSIGNED("VMALLOC_START", vmalloc_start);
#ifdef __aarch64__
READ_NUMBER("VA_BITS", VA_BITS);
READ_NUMBER("TCR_EL1_T1SZ", TCR_EL1_T1SZ);
@@ -2989,7 +2991,6 @@ read_vmcoreinfo(void)
READ_NUMBER("VA_BITS", va_bits);
READ_NUMBER_UNSIGNED("phys_ram_base", phys_ram_base);
READ_NUMBER_UNSIGNED("PAGE_OFFSET", page_offset);
- READ_NUMBER_UNSIGNED("VMALLOC_START", vmalloc_start);
READ_NUMBER_UNSIGNED("VMALLOC_END", vmalloc_end);
READ_NUMBER_UNSIGNED("VMEMMAP_START", vmemmap_start);
READ_NUMBER_UNSIGNED("VMEMMAP_END", vmemmap_end);
diff --git a/makedumpfile-1.7.4/makedumpfile.h b/makedumpfile-1.7.4/makedumpfile.h
index c04c330b69ecbe5fb232a2eabbd2d71f14b60cc0..c31f3a4371af8aae38dcba8cac4d6de1012b4cfd 100644
--- a/makedumpfile-1.7.4/makedumpfile.h
+++ b/makedumpfile-1.7.4/makedumpfile.h
@@ -541,8 +541,6 @@ do { \
* The value of dependence on machine
*/
#define PAGE_OFFSET (info->page_offset)
-#define VMALLOC_START (info->vmalloc_start)
-#define VMALLOC_END (info->vmalloc_end)
#define VMEMMAP_START (info->vmemmap_start)
#define VMEMMAP_END (info->vmemmap_end)
#define PMASK (0x7ffffffffffff000UL)
@@ -2263,6 +2261,9 @@ struct number_table {
long HUGETLB_PAGE_DTOR;
long phys_base;
long KERNEL_IMAGE_SIZE;
+
+ unsigned long vmalloc_start;
+
#ifdef __aarch64__
long VA_BITS;
long TCR_EL1_T1SZ;
@@ -2273,7 +2274,6 @@ struct number_table {
long va_bits;
unsigned long phys_ram_base;
unsigned long page_offset;
- unsigned long vmalloc_start;
unsigned long vmalloc_end;
unsigned long vmemmap_start;
unsigned long vmemmap_end;

99
kexec-tools-2.0.28-makedumpfile-0002-PATCH-Fix-failure-of-hugetlb-pages-exclusion-on-Linu.patch
View File

@ -1,99 +0,0 @@
From 985e575253f1c2de8d6876cfe685c68a24ee06e1 Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Thu, 30 May 2024 16:59:02 +0900
Subject: [PATCH 1/2] [PATCH] Fix failure of hugetlb pages exclusion on Linux
6.9 and later
* Required for kernel 6.9
Kernel commit d99e3140a4d3 ("mm: turn folio_test_hugetlb into a
PageType") moved the PG_hugetlb flag from folio._flags_1 into
page._mapcount and introduced NUMBER(PAGE_HUGETLB_MAPCOUNT_VALUE) entry
into vmcoreinfo.
Without the patch, "makedumpfile -d 8" cannot exclude hugetlb pages.
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
---
makedumpfile.c | 22 ++++++++++++++++++++--
makedumpfile.h | 3 +++
2 files changed, 23 insertions(+), 2 deletions(-)
diff --git a/makedumpfile-1.7.4/makedumpfile.c b/makedumpfile-1.7.4/makedumpfile.c
index d7f1dd4..437ad91 100644
--- a/makedumpfile-1.7.4/makedumpfile.c
+++ b/makedumpfile-1.7.4/makedumpfile.c
@@ -2975,6 +2975,7 @@ read_vmcoreinfo(void)
READ_SRCFILE("pud_t", pud_t);
READ_NUMBER("PAGE_BUDDY_MAPCOUNT_VALUE", PAGE_BUDDY_MAPCOUNT_VALUE);
+ READ_NUMBER("PAGE_HUGETLB_MAPCOUNT_VALUE", PAGE_HUGETLB_MAPCOUNT_VALUE);
READ_NUMBER("PAGE_OFFLINE_MAPCOUNT_VALUE", PAGE_OFFLINE_MAPCOUNT_VALUE);
READ_NUMBER("phys_base", phys_base);
READ_NUMBER("KERNEL_IMAGE_SIZE", KERNEL_IMAGE_SIZE);
@@ -6510,6 +6511,9 @@ __exclude_unnecessary_pages(unsigned long mem_map,
_count = UINT(pcache + OFFSET(page._refcount));
mapping = ULONG(pcache + OFFSET(page.mapping));
+ if (OFFSET(page._mapcount) != NOT_FOUND_STRUCTURE)
+ _mapcount = UINT(pcache + OFFSET(page._mapcount));
+
compound_order = 0;
compound_dtor = 0;
/*
@@ -6520,6 +6524,22 @@ __exclude_unnecessary_pages(unsigned long mem_map,
if ((index_pg < PGMM_CACHED - 1) && isCompoundHead(flags)) {
unsigned char *addr = pcache + SIZE(page);
+ /*
+ * Linux 6.9 and later kernels use _mapcount value for hugetlb pages.
+ * See kernel commit d99e3140a4d3.
+ */
+ if (NUMBER(PAGE_HUGETLB_MAPCOUNT_VALUE) != NOT_FOUND_NUMBER) {
+ unsigned long _flags_1 = ULONG(addr + OFFSET(page.flags));
+ unsigned int PG_hugetlb = ~NUMBER(PAGE_HUGETLB_MAPCOUNT_VALUE);
+
+ compound_order = _flags_1 & 0xff;
+
+ if ((_mapcount & (PAGE_TYPE_BASE | PG_hugetlb)) == PAGE_TYPE_BASE)
+ compound_dtor = IS_HUGETLB;
+
+ goto check_order;
+ }
+
/*
* Linux 6.6 and later. Kernels that have PG_hugetlb should also
* have the compound order in the low byte of folio._flags_1.
@@ -6564,8 +6584,6 @@ check_order:
if (OFFSET(page.compound_head) != NOT_FOUND_STRUCTURE)
compound_head = ULONG(pcache + OFFSET(page.compound_head));
- if (OFFSET(page._mapcount) != NOT_FOUND_STRUCTURE)
- _mapcount = UINT(pcache + OFFSET(page._mapcount));
if (OFFSET(page.private) != NOT_FOUND_STRUCTURE)
private = ULONG(pcache + OFFSET(page.private));
diff --git a/makedumpfile-1.7.4/makedumpfile.h b/makedumpfile-1.7.4/makedumpfile.h
index 75b66ce..f08c49f 100644
--- a/makedumpfile-1.7.4/makedumpfile.h
+++ b/makedumpfile-1.7.4/makedumpfile.h
@@ -165,6 +165,8 @@ test_bit(int nr, unsigned long addr)
#define isAnon(mapping, flags) (((unsigned long)mapping & PAGE_MAPPING_ANON) != 0 \
&& !isSlab(flags))
+#define PAGE_TYPE_BASE (0xf0000000)
+
#define PTOB(X) (((unsigned long long)(X)) << PAGESHIFT())
#define BTOP(X) (((unsigned long long)(X)) >> PAGESHIFT())
@@ -2255,6 +2257,7 @@ struct number_table {
long PG_hugetlb;
long PAGE_BUDDY_MAPCOUNT_VALUE;
+ long PAGE_HUGETLB_MAPCOUNT_VALUE;
long PAGE_OFFLINE_MAPCOUNT_VALUE;
long SECTION_SIZE_BITS;
long MAX_PHYSMEM_BITS;
--
2.40.1

129
kexec-tools-2.0.28-makedumpfile-0003-PATCH-Fix-wrong-exclusion-of-Slab-pages-on-Linux-6.1.patch
View File

@ -1,129 +0,0 @@
From bad2a7c4fa75d37a41578441468584963028bdda Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 7 Jun 2024 15:34:05 +0900
Subject: [PATCH 2/2] [PATCH] Fix wrong exclusion of Slab pages on Linux
6.10-rc1 and later
* Required for kernel 6.10
Kernel commit 46df8e73a4a3 ("mm: free up PG_slab") moved the PG_slab
flag from page.flags into page._mapcount (slab.__page_type), and
introduced NUMBER(PAGE_SLAB_MAPCOUNT_VALUE) entry into vmcoreinfo.
Without the patch, "makedumpfile -d 8" option wrongly excludes Slab
pages and crash cannot open the dumpfile with an error like this:
$ crash --kaslr auto vmlinux dumpfile
...
please wait... (gathering task table data)
crash: page excluded: kernel virtual address: ffff909980440270 type: "xa_node.slots[off]"
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
---
makedumpfile.c | 24 +++++++++++++++++++-----
makedumpfile.h | 6 +++---
2 files changed, 22 insertions(+), 8 deletions(-)
diff --git a/makedumpfile-1.7.4/makedumpfile.c b/makedumpfile-1.7.4/makedumpfile.c
index 437ad91..5b34712 100644
--- a/makedumpfile-1.7.4/makedumpfile.c
+++ b/makedumpfile-1.7.4/makedumpfile.c
@@ -275,13 +275,26 @@ isHugetlb(unsigned long dtor)
&& (SYMBOL(free_huge_page) == dtor));
}
+static inline int
+isSlab(unsigned long flags, unsigned int _mapcount)
+{
+ /* Linux 6.10 and later */
+ if (NUMBER(PAGE_SLAB_MAPCOUNT_VALUE) != NOT_FOUND_NUMBER) {
+ unsigned int PG_slab = ~NUMBER(PAGE_SLAB_MAPCOUNT_VALUE);
+ if ((_mapcount & (PAGE_TYPE_BASE | PG_slab)) == PAGE_TYPE_BASE)
+ return TRUE;
+ }
+
+ return flags & (1UL << NUMBER(PG_slab));
+}
+
static int
isOffline(unsigned long flags, unsigned int _mapcount)
{
if (NUMBER(PAGE_OFFLINE_MAPCOUNT_VALUE) == NOT_FOUND_NUMBER)
return FALSE;
- if (flags & (1UL << NUMBER(PG_slab)))
+ if (isSlab(flags, _mapcount))
return FALSE;
if (_mapcount == (int)NUMBER(PAGE_OFFLINE_MAPCOUNT_VALUE))
@@ -2977,6 +2990,7 @@ read_vmcoreinfo(void)
READ_NUMBER("PAGE_BUDDY_MAPCOUNT_VALUE", PAGE_BUDDY_MAPCOUNT_VALUE);
READ_NUMBER("PAGE_HUGETLB_MAPCOUNT_VALUE", PAGE_HUGETLB_MAPCOUNT_VALUE);
READ_NUMBER("PAGE_OFFLINE_MAPCOUNT_VALUE", PAGE_OFFLINE_MAPCOUNT_VALUE);
+ READ_NUMBER("PAGE_SLAB_MAPCOUNT_VALUE", PAGE_SLAB_MAPCOUNT_VALUE);
READ_NUMBER("phys_base", phys_base);
READ_NUMBER("KERNEL_IMAGE_SIZE", KERNEL_IMAGE_SIZE);
@@ -6043,7 +6057,7 @@ static int
page_is_buddy_v3(unsigned long flags, unsigned int _mapcount,
unsigned long private, unsigned int _count)
{
- if (flags & (1UL << NUMBER(PG_slab)))
+ if (isSlab(flags, _mapcount))
return FALSE;
if (_mapcount == (int)NUMBER(PAGE_BUDDY_MAPCOUNT_VALUE))
@@ -6618,7 +6632,7 @@ check_order:
*/
else if ((info->dump_level & DL_EXCLUDE_CACHE)
&& is_cache_page(flags)
- && !isPrivate(flags) && !isAnon(mapping, flags)) {
+ && !isPrivate(flags) && !isAnon(mapping, flags, _mapcount)) {
pfn_counter = &pfn_cache;
}
/*
@@ -6626,7 +6640,7 @@ check_order:
*/
else if ((info->dump_level & DL_EXCLUDE_CACHE_PRI)
&& is_cache_page(flags)
- && !isAnon(mapping, flags)) {
+ && !isAnon(mapping, flags, _mapcount)) {
if (isPrivate(flags))
pfn_counter = &pfn_cache_private;
else
@@ -6638,7 +6652,7 @@ check_order:
* - hugetlbfs pages
*/
else if ((info->dump_level & DL_EXCLUDE_USER_DATA)
- && (isAnon(mapping, flags) || isHugetlb(compound_dtor))) {
+ && (isAnon(mapping, flags, _mapcount) || isHugetlb(compound_dtor))) {
pfn_counter = &pfn_user;
}
/*
diff --git a/makedumpfile-1.7.4/makedumpfile.h b/makedumpfile-1.7.4/makedumpfile.h
index f08c49f..6b43a8b 100644
--- a/makedumpfile-1.7.4/makedumpfile.h
+++ b/makedumpfile-1.7.4/makedumpfile.h
@@ -161,9 +161,8 @@ test_bit(int nr, unsigned long addr)
#define isSwapBacked(flags) test_bit(NUMBER(PG_swapbacked), flags)
#define isHWPOISON(flags) (test_bit(NUMBER(PG_hwpoison), flags) \
&& (NUMBER(PG_hwpoison) != NOT_FOUND_NUMBER))
-#define isSlab(flags) test_bit(NUMBER(PG_slab), flags)
-#define isAnon(mapping, flags) (((unsigned long)mapping & PAGE_MAPPING_ANON) != 0 \
- && !isSlab(flags))
+#define isAnon(mapping, flags, _mapcount) \
+ (((unsigned long)mapping & PAGE_MAPPING_ANON) != 0 && !isSlab(flags, _mapcount))
#define PAGE_TYPE_BASE (0xf0000000)
@@ -2259,6 +2258,7 @@ struct number_table {
long PAGE_BUDDY_MAPCOUNT_VALUE;
long PAGE_HUGETLB_MAPCOUNT_VALUE;
long PAGE_OFFLINE_MAPCOUNT_VALUE;
+ long PAGE_SLAB_MAPCOUNT_VALUE;
long SECTION_SIZE_BITS;
long MAX_PHYSMEM_BITS;
long HUGETLB_PAGE_DTOR;
--
2.40.1

232
kexec-tools.spec
View File

@ -1,9 +1,3 @@
%global eppic_ver e8844d3793471163ae4a56d8f95897be9e5bd554
%global eppic_shortver %(c=%{eppic_ver}; echo ${c:0:7})
%global mkdf_ver 1.7.4
%global kdump_utils_ver 1.0.42
%global mkdf_shortver %(c=%{mkdf_ver}; echo ${c:0:7})
Name: kexec-tools
Version: 2.0.28
Release: 14%{?dist}
@ -13,9 +7,6 @@ Summary: The kexec/kdump userspace component
Source0: http://kernel.org/pub/linux/utils/kernel/kexec/%{name}-%{version}.tar.xz
Recommends: kdump-utils
Source1: https://github.com/rhkdump/kdump-utils/archive/v%{kdump_utils_ver}/kdump-utils-%{kdump_utils_ver}.tar.gz
Source9: https://github.com/makedumpfile/makedumpfile/archive/%{mkdf_ver}/makedumpfile-%{mkdf_shortver}.tar.gz
Source19: https://github.com/lucchouina/eppic/archive/%{eppic_ver}/eppic-%{eppic_shortver}.tar.gz
BuildRequires: automake
BuildRequires: autoconf
@ -54,25 +45,6 @@ Patch101: kexec-tools-2.0.28-Fix-building-on-x86_64-with-binutils-2.41.patch
# Author: Jiri Bohac <jbohac@suse.cz>
Patch601: kexec-tools-2.0.28-kexec-don-t-use-kexec_file_load-on-XEN.patch
Patch602: kexec-tools-2.0.28-makedumfpile-0001-PATCH-ppc64-get-vmalloc-start-address-from-vmcoreinf.patch
Patch603: 0001-dracut-module-setup-Fix-missing-systemd-system.conf-.patch
Patch604: 0002-mkdumprd-Fix-makedumpfile-parameter-check.patch
Patch605: 0003-Try-to-install-PHY-and-MDIO-bus-drivers-explicitly.patch
Patch606: 0004-Install-the-driver-of-physical-device-for-a-SR-IOV-v.patch
Patch607: 0005-update-98-kexec-rules-for-crash-hotplug.patch
Patch608: 0006-sysconfig-add-pcie_ports-compat-to-KDUMP_COMMANDLINE.patch
Patch609: 0007-fadump-udev-do-not-re-register-fadump-if-kernel-hotp.patch
Patch610: kexec-tools-2.0.28-makedumpfile-0002-PATCH-Fix-failure-of-hugetlb-pages-exclusion-on-Linu.patch
Patch611: kexec-tools-2.0.28-makedumpfile-0003-PATCH-Fix-wrong-exclusion-of-Slab-pages-on-Linux-6.1.patch
Patch612: 0001-Use-grep-q-cmd-instead-of-cmd-grep-q.patch
Patch613: 0001-kdumpctl-Drop-default-kexec-d-option.patch
Patch614: 0002-kdump-lib-fix-sed-expression-in-prepare_cmdline-on-a.patch
Patch615: 0003-kdumpctl-Simplify-fadump-handling-in-reset_crashkern.patch
Patch616: 0004-kdumpctl.8-Add-description-to-reset-crashkernel-rebo.patch
Patch617: 0005-dracut-kdump.sh-Save-kexec-dmesg.log-after-failure_a.patch
Patch618: 0006-Makefile-Fix-early-kdump-file-names.patch
Patch619: 0007-kdump-lib-Drop-file-dependency-in-is_uki.patch
%description
kexec-tools provides /sbin/kexec binary that facilitates a new
kernel to boot using the kernel's kexec feature either on a
@ -80,100 +52,13 @@ normal or a panic reboot. This package contains the /sbin/kexec
binary and ancillary utilities that together form the userspace
component of the kernel's kexec feature.
%package -n makedumpfile
Version: %{mkdf_ver}
Summary: make a small dumpfile of kdump
License: GPL-2.0-only
URL: https://github.com/makedumpfile/makedumpfile
Conflicts: kexec-tools < 2.0.28-5
BuildRequires: make
BuildRequires: gcc
BuildRequires: zlib-devel
BuildRequires: elfutils-devel
BuildRequires: glib2-devel
BuildRequires: bzip2-devel
BuildRequires: ncurses-devel
BuildRequires: bison
BuildRequires: flex
BuildRequires: lzo-devel
BuildRequires: snappy-devel
BuildRequires: libzstd-devel
BuildRequires: pkgconfig
BuildRequires: intltool
BuildRequires: gettext
%description -n makedumpfile
makedumpfile is a tool to compress and filter out unneeded data from kernel
dumps to reduce its file size. It is typically used with the kdump mechanism.
%package -n kdump-utils
Version: 1.0.42
License: GPL-2.0-only AND LGPL-2.1-or-later
URL: https://github.com/rhkdump/kdump-utils
Summary: Kernel crash dump collection utilities
%ifarch ppc64 ppc64le
Requires(post): servicelog
Recommends: keyutils
%endif
Requires(pre): coreutils
Requires(pre): sed
Requires: kexec-tools >= 2.0.28-8
Requires: makedumpfile
Requires: dracut >= 058
Requires: dracut-network >= 058
Requires: dracut-squash >= 058
Requires: ethtool
Requires: util-linux
# Needed for UKI support
Recommends: binutils
Recommends: grubby
Recommends: hostname
BuildRequires: systemd-rpm-macros
%ifnarch s390x
Requires: systemd-udev%{?_isa}
%endif
%description -n kdump-utils
kdump-utils is responsible for collecting the crash kernel dump. It builds and
loads the kdump initramfs so when a kernel crashes, the system will boot the
kdump kernel and initramfs to save the collected crash kernel dump to specified
target.
%prep
%setup -q
tar -z -x -v -f %{SOURCE1}
mkdir -p -m755 kcp
tar -z -x -v -f %{SOURCE9}
tar -z -x -v -f %{SOURCE19}
%patch 101 -p1
%patch 601 -p1
%patch 602 -p1
%patch 603 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 604 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 605 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 606 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 607 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 608 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 609 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 610 -p1
%patch 611 -p1
%patch 612 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 613 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 614 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 615 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 616 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 617 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 618 -p1 -d kdump-utils-%{kdump_utils_ver}
%patch 619 -p1 -d kdump-utils-%{kdump_utils_ver}
%ifarch ppc
%define archdef ARCH=ppc
@ -194,80 +79,10 @@ autoreconf
rm -f kexec-tools.spec.in
%make_build
# makedumpfile
make -C eppic-%{eppic_ver}/libeppic
make -C makedumpfile-%{mkdf_ver} LINKTYPE=dynamic USELZO=on USESNAPPY=on USEZSTD=on
make -C makedumpfile-%{mkdf_ver} LDFLAGS="$LDFLAGS -I../eppic-%{eppic_ver}/libeppic -L../eppic-%{eppic_ver}/libeppic" eppic_makedumpfile.so
%install
%make_install
rm -f %{buildroot}/%{_libdir}/kexec-tools/kexec_test
# kdump-utils
%define kdump_utils_dir kdump-utils-%{kdump_utils_ver}
make DESTDIR=%{buildroot} -C %kdump_utils_dir install
# makedumpfile
make DESTDIR=%{buildroot} -C makedumpfile-%{mkdf_ver} install
install -m 644 -D makedumpfile-%{mkdf_ver}/makedumpfile.conf %{buildroot}/%{_sysconfdir}/makedumpfile.conf.sample
rm %{buildroot}/%{_sbindir}/makedumpfile-R.pl
install -m 755 -D makedumpfile-%{mkdf_ver}/eppic_makedumpfile.so %{buildroot}/%{_libdir}/eppic_makedumpfile.so
%post -n kdump-utils
# don't try to systemctl preset the kdump service for old kexec-tools
#
# when the old kexec-tools gets removed, this trigger will be excuted to
# create a file. So later the posttrans scriptlet will know there is no need to
# systemctl preset the kdump service.
# This solution can be dropped in F41 when we assume no users will use old
# version of kexec-tools.
%define kexec_tools_no_preset %{_localstatedir}/lib/rpm-state/kexec-tools.no-preset
%triggerun -- kexec-tools
touch %{kexec_tools_no_preset}
touch /etc/kdump.conf
%ifarch ppc64 ppc64le
servicelog_notify --remove --command=/usr/lib/kdump/kdump-migrate-action.sh 2>/dev/null
servicelog_notify --add --command=/usr/lib/kdump/kdump-migrate-action.sh --match='refcode="#MIGRATE" and serviceable=0' --type=EVENT --method=pairs_stdin >/dev/null
%endif
:
%postun -n kdump-utils
%systemd_postun_with_restart kdump.service
%preun -n kdump-utils
%ifarch ppc64 ppc64le
servicelog_notify --remove --command=/usr/lib/kdump/kdump-migrate-action.sh >/dev/null
%endif
%systemd_preun kdump.service
%posttrans -n kdump-utils
# don't try to systemctl preset the kdump service for old kexec-tools
if [[ -f %{kexec_tools_no_preset} ]]; then
# this if branch can be removed in F41 when we assume no users will use the old kexec-tools
rm %{kexec_tools_no_preset}
else
# Initial installation
%systemd_post kdump.service
fi
# Try to reset kernel crashkernel value to new default value or set up
# crasherkernel value for new install
#
# Note
# 1. Skip ostree systems as they are not supported.
# 2. For Fedora 36 and RHEL9, "[ $1 == 1 ]" in posttrans scriptlet means both install and upgrade;
# For Fedora > 36, "[ $1 == 1 ]" only means install and "[ $1 == 2 ]" means upgrade
if [ ! -f /run/ostree-booted ] && [ $1 == 1 -o $1 == 2 ]; then
kdumpctl _reset-crashkernel-after-update
:
fi
%files
%{_sbindir}/kexec
@ -278,53 +93,6 @@ fi
%license COPYING
%doc TODO
%files -n kdump-utils
%ifarch ppc64 ppc64le
%{_sbindir}/mkfadumprd
%{_prefix}/lib/kernel/install.d/60-fadump.install
%endif
%{_sbindir}/mkdumprd
%{_bindir}/*
%{_prefix}/lib/kdump
%config(noreplace,missingok) %{_sysconfdir}/sysconfig/kdump
%config(noreplace,missingok) %verify(not mtime) %{_sysconfdir}/kdump.conf
%ifnarch s390x
%{_udevrulesdir}
%{_udevrulesdir}/../kdump-udev-throttler
%endif
%{_prefix}/lib/dracut/modules.d/*
%dir %{_localstatedir}/crash
%dir %{_sysconfdir}/kdump
%dir %{_sysconfdir}/kdump/pre.d
%dir %{_sysconfdir}/kdump/post.d
%dir %{_sharedstatedir}/kdump
%{_mandir}/man8/kdumpctl.8*
%{_mandir}/man8/mkdumprd.8*
%{_mandir}/man5/kdump.conf.5*
%{_unitdir}/kdump.service
%{_prefix}/lib/systemd/system-generators/kdump-dep-generator.sh
%{_prefix}/lib/kernel/install.d/60-kdump.install
%{_prefix}/lib/kernel/install.d/92-crashkernel.install
%license %kdump_utils_dir/COPYING
%doc %kdump_utils_dir/kexec-kdump-howto.txt
%doc %kdump_utils_dir/early-kdump-howto.txt
%doc %kdump_utils_dir/fadump-howto.txt
%doc %kdump_utils_dir/kdump-in-cluster-environment.txt
%doc %kdump_utils_dir/live-image-kdump-howto.txt
%doc %kdump_utils_dir/crashkernel-howto.txt
%doc %kdump_utils_dir/supported-kdump-targets.txt
%files -n makedumpfile
%license makedumpfile-%{mkdf_ver}/COPYING
%{_sbindir}/makedumpfile
%{_mandir}/man5/makedumpfile.conf.5.*
%{_mandir}/man8/makedumpfile.8.*
%{_sysconfdir}/makedumpfile.conf.sample
%{_libdir}/eppic_makedumpfile.so
%{_datadir}/makedumpfile/
%changelog
* Fri Jul 12 2024 Lichen Liu <lichliu@redhat.com> - 2.0.28-14
- Various fixes.