已合并
fix script error #138
lishengvictor创建于 3 天前
fix script error #138
已合并
lishengvictor创建于 3 天前
1 个文件变更+21-21
Mscripts/package/custom/driver/common/scripts/npu_log_collect.sh+21-21
@@ -37,7 +37,7 @@ usage() {
37 echo "$(basename $0) Collect the default content."37 echo "$(basename $0) Collect the default content."
38 echo "$(basename $0) -p /home/tmp Specify a path to store the generated content."38 echo "$(basename $0) -p /home/tmp Specify a path to store the generated content."
39 echo "$(basename $0) --off host_info Disable host os info."39 echo "$(basename $0) --off host_info Disable host os info."
40-if [[ "$CHIP_NUM_910_95" -eq 0]]; then40+if [[ "$CHIP_NUM_910_95" -eq 0 ]]; then
41 echo "$(basename $0) -m Collect the mcu logs."41 echo "$(basename $0) -m Collect the mcu logs."
42 echo "$(basename $0) -l Collect the link_down logs."42 echo "$(basename $0) -l Collect the link_down logs."
43fi43fi
@@ -59,7 +59,7 @@ function parse_args()
59{59{
60 if [ $# -eq 0 ]; then60 if [ $# -eq 0 ]; then
61 echo "No arguments provided. Performing default action..."61 echo "No arguments provided. Performing default action..."
62- main 62+ main
63 exit 063 exit 0
64 fi64 fi
65 while [ "$1" != "" ]; do65 while [ "$1" != "" ]; do
@@ -69,7 +69,7 @@ function parse_args()
69 usage69 usage
70 exit 170 exit 1
71 elif [ ! -d "$2" ] && [ ! -h "$2" ]; then71 elif [ ! -d "$2" ] && [ ! -h "$2" ]; then
72- echo "Warning: The specified path is inexistent, this tool will create the directory." 72+ echo "Warning: The specified path is inexistent, this tool will create the directory."
73 mkdir -p "$2"73 mkdir -p "$2"
74 chmod 440 "$2"74 chmod 440 "$2"
75 fi75 fi
@@ -149,7 +149,7 @@ function create_log_collect_dir()
149 if [ ! -d "$LOG_COLLECT_DIR" ]; then149 if [ ! -d "$LOG_COLLECT_DIR" ]; then
150 mkdir -p $LOG_COLLECT_DIR150 mkdir -p $LOG_COLLECT_DIR
151 if [ $? -ne 0 ]; then151 if [ $? -ne 0 ]; then
152- echo "Error: Create log collect directory failed. Please check." 152+ echo "Error: Create log collect directory failed. Please check."
153 exit 1153 exit 1
154 fi154 fi
155 fi155 fi
@@ -249,7 +249,7 @@ function collect_npu_info_log()
249 j=0249 j=0
250 for dev_id in $dev_list250 for dev_id in $dev_list
251 do251 do
252- echo "########### collect npu id $dev_id product info ###########" >> $LOG_COLLECT_DIR/npu_info_log/card_info.log 252+ echo "########### collect npu id $dev_id product info ###########" >> $LOG_COLLECT_DIR/npu_info_log/card_info.log
253 echo "npu-smi info -t product -i $dev_id -c 0 1" >> $LOG_COLLECT_DIR/npu_info_log/card_info.log253 echo "npu-smi info -t product -i $dev_id -c 0 1" >> $LOG_COLLECT_DIR/npu_info_log/card_info.log
254 timeout 10s npu-smi info -t product -i $dev_id -c 0 1>> $LOG_COLLECT_DIR/npu_info_log/card_info.log 2>> $LOG_COLLECT_RUNING_INFO254 timeout 10s npu-smi info -t product -i $dev_id -c 0 1>> $LOG_COLLECT_DIR/npu_info_log/card_info.log 2>> $LOG_COLLECT_RUNING_INFO
255 255 
@@ -300,7 +300,7 @@ function collect_mcu_log()
300 for dev_id in $dev_list300 for dev_id in $dev_list
301 do301 do
302 (302 (
303- npu-smi set -t collect-log -i $dev_id 1>> $LOG_COLLECT_RUNING_INFO 2>&1 303+ npu-smi set -t collect-log -i $dev_id 1>> $LOG_COLLECT_RUNING_INFO 2>&1
304 mkdir -p $LOG_COLLECT_DIR/mcu_log/card_${dev_id}_mcu_log304 mkdir -p $LOG_COLLECT_DIR/mcu_log/card_${dev_id}_mcu_log
305 cp -f /run/mcu_log/error_log_${dev_id}_*.log /run/mcu_log/maintaince_log_${dev_id}_*.log /run/mcu_log/operate_log_${dev_id}_*.log $LOG_COLLECT_DIR/mcu_log/card_${dev_id}_mcu_log/ 1>> $LOG_COLLECT_RUNING_INFO 2>&1305 cp -f /run/mcu_log/error_log_${dev_id}_*.log /run/mcu_log/maintaince_log_${dev_id}_*.log /run/mcu_log/operate_log_${dev_id}_*.log $LOG_COLLECT_DIR/mcu_log/card_${dev_id}_mcu_log/ 1>> $LOG_COLLECT_RUNING_INFO 2>&1
306 ) &306 ) &
@@ -395,7 +395,7 @@ function main()
395 green_echo "$(date)"395 green_echo "$(date)"
396}396}
397 397 
398-function collect_full() 398+function collect_full()
399{399{
400 show_info400 show_info
401 get_install_variable401 get_install_variable
@@ -409,16 +409,16 @@ function collect_full()
409 if [ "$CHIP_NUM_910_93" -ne 0 ]; then409 if [ "$CHIP_NUM_910_93" -ne 0 ]; then
410 collect_lingqu_log &410 collect_lingqu_log &
411 fi411 fi
412- 412+ 
413- get_version_info 413+ get_version_info
414 get_path_env &414 get_path_env &
415 get_driver_log &415 get_driver_log &
416- get_pcie_log 2>/dev/null & 416+ get_pcie_log 2>/dev/null &
417 get_os_info &417 get_os_info &
418 get_qemu_log &418 get_qemu_log &
419 419 
420- wait 420+ wait
421- get_device_log 421+ get_device_log
422 compress_file422 compress_file
423 green_echo "$(date)"423 green_echo "$(date)"
424}424}
@@ -597,7 +597,7 @@ function network_info_collect()
597 for i in $(seq 0 $npu_nums); do echo "====> $i" >> roce.log;timeout 10s hccn_tool -i $i -pfc_stat -g &>> roce.log;done597 for i in $(seq 0 $npu_nums); do echo "====> $i" >> roce.log;timeout 10s hccn_tool -i $i -pfc_stat -g &>> roce.log;done
598 echo "-----------collect pfc data packet info success---------------"598 echo "-----------collect pfc data packet info success---------------"
599 599 
600- echo "-----------collect maximum size of the buffer info success---------------" 600+ echo "-----------collect maximum size of the buffer info success---------------"
601 #查询网卡侧收发包状态,重定向输出到stat.log601 #查询网卡侧收发包状态,重定向输出到stat.log
602 echo -e "\n-----------stat_extra---------------" >> stat.log602 echo -e "\n-----------stat_extra---------------" >> stat.log
603 for i in $(seq 0 $npu_nums); do echo "====> $i" >> stat.log;timeout 10s hccn_tool -i $i -stat_extra -g &>> stat.log;done603 for i in $(seq 0 $npu_nums); do echo "====> $i" >> stat.log;timeout 10s hccn_tool -i $i -stat_extra -g &>> stat.log;done
@@ -646,7 +646,7 @@ function get_path_env() {
646 646 
647function get_version_info() {647function get_version_info() {
648 date648 date
649- 649+ 
650 echo "--------------- cat drvier version.info ---------------"650 echo "--------------- cat drvier version.info ---------------"
651 which npu-smi > /dev/null 2>&1651 which npu-smi > /dev/null 2>&1
652 if [ $? -eq 0 ];then652 if [ $? -eq 0 ];then
@@ -657,7 +657,7 @@ function get_version_info() {
657 timeout 20s ${INSTALL_PATH}/driver/tools/upgrade-tool --device_index -1 --system_version >> ${FILE_VERSION_LOG}657 timeout 20s ${INSTALL_PATH}/driver/tools/upgrade-tool --device_index -1 --system_version >> ${FILE_VERSION_LOG}
658 echo "--------------- firmware version ---------------"658 echo "--------------- firmware version ---------------"
659 timeout 20s ${INSTALL_PATH}/driver/tools/upgrade-tool --device_index -1 --component -1 --version >> ${FILE_VERSION_LOG}659 timeout 20s ${INSTALL_PATH}/driver/tools/upgrade-tool --device_index -1 --component -1 --version >> ${FILE_VERSION_LOG}
660- bdf_list=$(lspci | egrep "d100|d500|d801|d802|d803|d806" | awk '{print $1}') 660+ bdf_list=$(lspci | egrep "d100|d500|d801|d802|d803|d806" | awk '{print $1}')
661 for bdf in $bdf_list ;do lspci -vvvs $bdf -xxx | grep 4e0 >> ${FILE_VERSION_LOG};done661 for bdf in $bdf_list ;do lspci -vvvs $bdf -xxx | grep 4e0 >> ${FILE_VERSION_LOG};done
662 for bdf in $bdf_list ;do lspci -vvvs $bdf -xxxx | grep 300 >> ${FILE_VERSION_LOG}; lspci -vvvs $bdf -xxxx | grep 320 >> ${FILE_VERSION_LOG}; lspci -vvvs $bdf -xxxx | grep 450 >> ${FILE_VERSION_LOG};done662 for bdf in $bdf_list ;do lspci -vvvs $bdf -xxxx | grep 300 >> ${FILE_VERSION_LOG}; lspci -vvvs $bdf -xxxx | grep 320 >> ${FILE_VERSION_LOG}; lspci -vvvs $bdf -xxxx | grep 450 >> ${FILE_VERSION_LOG};done
663 lspci | egrep "d100|d500|d801|d802|d803|d806" | awk '{print $1}' | xargs -i lspci -xxxx -s {} | grep 300 >> ${FILE_VERSION_LOG}663 lspci | egrep "d100|d500|d801|d802|d803|d806" | awk '{print $1}' | xargs -i lspci -xxxx -s {} | grep 300 >> ${FILE_VERSION_LOG}
@@ -675,15 +675,15 @@ function get_version_info() {
675function get_pcie_log() {675function get_pcie_log() {
676 green_echo "================ Collect PCIe log ==============" | tee -a $LOG_COLLECT_RUNING_INFO676 green_echo "================ Collect PCIe log ==============" | tee -a $LOG_COLLECT_RUNING_INFO
677 date >> ${FILE_PCIE_LOG}677 date >> ${FILE_PCIE_LOG}
678- 678+ 
679 echo "--------------- lspci ---------------" >> ${FILE_PCIE_LOG}679 echo "--------------- lspci ---------------" >> ${FILE_PCIE_LOG}
680 lspci | egrep "d100|d500|d801|d802|d803|d806" >> ${FILE_PCIE_LOG}680 lspci | egrep "d100|d500|d801|d802|d803|d806" >> ${FILE_PCIE_LOG}
681 681 
682 echo "--------------- bdf_to_devid ---------------" >> ${FILE_PCIE_LOG}682 echo "--------------- bdf_to_devid ---------------" >> ${FILE_PCIE_LOG}
683 bdf_list=($(lspci | egrep "d100|d500|d801|d802|d803|d806" | awk '{print $1}'))683 bdf_list=($(lspci | egrep "d100|d500|d801|d802|d803|d806" | awk '{print $1}'))
684 cat /sys/bus/pci/devices/0000:${bdf_list[0]}/devdrv_sysfs_bdf_to_devid >> ${FILE_PCIE_LOG}684 cat /sys/bus/pci/devices/0000:${bdf_list[0]}/devdrv_sysfs_bdf_to_devid >> ${FILE_PCIE_LOG}
685- bdf_list=$(lspci | egrep "d100|d500|d801|d802|d803|d806" | awk '{print $1}') 685+ bdf_list=$(lspci | egrep "d100|d500|d801|d802|d803|d806" | awk '{print $1}')
686- 686+ 
687 echo "--------------- firmware version ---------------" >> ${FILE_PCIE_LOG}687 echo "--------------- firmware version ---------------" >> ${FILE_PCIE_LOG}
688 for bdf in $bdf_list ;do lspci -vvvs $bdf -xxx | grep 4e0;done >> ${FILE_PCIE_LOG}688 for bdf in $bdf_list ;do lspci -vvvs $bdf -xxx | grep 4e0;done >> ${FILE_PCIE_LOG}
689 for bdf in $bdf_list ;do lspci -vvvs $bdf -xxxx | grep 300 >> ${FILE_VERSION_LOG}; lspci -vvvs $bdf -xxxx | grep 320 >> ${FILE_VERSION_LOG}; lspci -vvvs $bdf -xxxx | grep 450 >> ${FILE_PCIE_LOG};done689 for bdf in $bdf_list ;do lspci -vvvs $bdf -xxxx | grep 300 >> ${FILE_VERSION_LOG}; lspci -vvvs $bdf -xxxx | grep 320 >> ${FILE_VERSION_LOG}; lspci -vvvs $bdf -xxxx | grep 450 >> ${FILE_PCIE_LOG};done
@@ -693,7 +693,7 @@ function get_pcie_log() {
693 693 
694 echo "--------------- LnkSta ---------------" >> ${FILE_PCIE_LOG}694 echo "--------------- LnkSta ---------------" >> ${FILE_PCIE_LOG}
695 date;lspci | egrep "d100|d500|d801|d802|d803|d806" | awk '{print $1}'|xargs -i lspci -vvvvs {}|grep -E 'LnkSta:' >> ${FILE_PCIE_LOG}695 date;lspci | egrep "d100|d500|d801|d802|d803|d806" | awk '{print $1}'|xargs -i lspci -vvvvs {}|grep -E 'LnkSta:' >> ${FILE_PCIE_LOG}
696- 696+ 
697 echo "--------------- NUMA ---------------" >> ${FILE_PCIE_LOG}697 echo "--------------- NUMA ---------------" >> ${FILE_PCIE_LOG}
698 for bdf in $bdf_list ;do echo ====$bdf====;lspci -vvvs $bdf | egrep "Lnk|NUMA";done >> ${FILE_PCIE_LOG}698 for bdf in $bdf_list ;do echo ====$bdf====;lspci -vvvs $bdf | egrep "Lnk|NUMA";done >> ${FILE_PCIE_LOG}
699 for bdf in $bdf_list ;do echo ====$bdf====;lspci -vvvs $bdf | egrep "NUMA";done >> ${FILE_PCIE_LOG}699 for bdf in $bdf_list ;do echo ====$bdf====;lspci -vvvs $bdf | egrep "NUMA";done >> ${FILE_PCIE_LOG}
@@ -771,7 +771,7 @@ function get_qemu_log() {
771 771 
772function get_driver_log() {772function get_driver_log() {
773 green_echo "================ Collect driver log ==============" | tee -a $LOG_COLLECT_RUNING_INFO773 green_echo "================ Collect driver log ==============" | tee -a $LOG_COLLECT_RUNING_INFO
774- cp -rf /var/log/npu "${LOG_FILE_PATH}" > /dev/null 2>&1 774+ cp -rf /var/log/npu "${LOG_FILE_PATH}" > /dev/null 2>&1
775 green_echo "================ Collect plog log ==============" | tee -a $LOG_COLLECT_RUNING_INFO775 green_echo "================ Collect plog log ==============" | tee -a $LOG_COLLECT_RUNING_INFO
776 cp -rf ~/ascend "${LOG_FILE_PATH}" > /dev/null 2>&1776 cp -rf ~/ascend "${LOG_FILE_PATH}" > /dev/null 2>&1
777 green_echo "$(date) End collect driver log" | tee -a $LOG_COLLECT_RUNING_INFO777 green_echo "$(date) End collect driver log" | tee -a $LOG_COLLECT_RUNING_INFO
@@ -782,7 +782,7 @@ function get_device_log() {
782 green_echo "================ Collect device log ==============" | tee -a $LOG_COLLECT_RUNING_INFO782 green_echo "================ Collect device log ==============" | tee -a $LOG_COLLECT_RUNING_INFO
783 msnpureport -f783 msnpureport -f
784 for d in 0 1 2 3 4 5 6 7; do784 for d in 0 1 2 3 4 5 6 7; do
785- msnpureport -r -d $d 785+ msnpureport -r -d $d
786 done786 done
787 cd ${TOP_PATH}787 cd ${TOP_PATH}
788 green_echo "$(date) End collect device log" | tee -a $LOG_COLLECT_RUNING_INFO788 green_echo "$(date) End collect device log" | tee -a $LOG_COLLECT_RUNING_INFO