onsuccess: #49: 1: [TCWG CI] https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/49/

Results :
  | # reset_artifacts:
  | -10
  | # build_bmk_llvm:
  | -3
  | # benchmark -- -O3_marm:
  | 1

check_regression status : 0

8 files changed, 307 insertions, 134 deletions

diff --git a/notify/exe.improvement b/notify/exe.improvement
new file mode 100644
index 0000000..652b682
--- /dev/null
+++ b/notify/exe.improvement
@@ -0,0 +1 @@
+3,401.bzip2,bzip2_base.default,401.bzip2 reduced in size by 3%,401.bzip2 reduced in size by 3% from 77619 to 75179 bytes

diff --git a/notify/jira/comments.txt b/notify/jira/comments.txt
index 32d856b..2c45835 100644
--- a/notify/jira/comments.txt
+++ b/notify/jira/comments.txt
@@ -1,3 +1,3 @@
 [LLVM-651]
-No change
-Details: https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/artifact/artifacts/notify/mail-body.txt/*view*/
+401.bzip2 reduced in size by 3%
+Details: https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/49/artifact/artifacts/notify/mail-body.txt/*view*/
diff --git a/notify/mail-body.txt b/notify/mail-body.txt
index 3974983..31ca823 100644
--- a/notify/mail-body.txt
+++ b/notify/mail-body.txt
@@ -4,15 +4,18 @@ Please find some details below.  If you have any questions, please follow up on
 
 In CI config tcwg_bmk-code_size-spec2k6/llvm-arm-master-O3 after:
 
-  | 797 commits in llvm
-  | bd077e98e463 Released restriction that prevented implicit dynamic-to-static dimension type cast in TOSA ops.
-  | ea90e289d97d Fix typo
-  | 89e44e33edec [clang][Interp] Implement __builtin_fmax
-  | 7ef87931ddd7 [libc++][PSTL] Silence clang-tidy errors.
-  | 8698d56d996a [Transforms][LICM] Add the ability to undo unprofitable reassociation
-  | ... and 792 more commits in llvm
-
-No change
+  | 975 commits in llvm
+  | d32e68ae539d [docs] Graduate VectorizationPlan.rst from proposal.
+  | b1ada7a1d31e [DAG] Support store merging of vector constant stores (try 2)
+  | e838471bc493 [X86] Add regression test case from pr64593
+  | 0e2c1cb23329 [X86] ReplaceNodeResults - always attempt to fold TRUNCATE -> PACKSS/PACKUS nodes.
+  | 888717e85395 [mlir][transform] Enable gpu-to-nvvm via conversion patterns driven by TD
+  | ... and 970 more commits in llvm
+
+the following benchmarks reduced in size by more than 1%:
+- 401.bzip2 reduced in size by 3% from 77619 to 75179 bytes

+the following hot functions reduced in size by more than 10% (but their benchmarks reduced in size by less than 1%):
+- 401.bzip2:[.] BZ2_decompress reduced in size by 12% from 20424 to 17984 bytes

 
 Below reproducer instructions can be used to re-build both "first_bad" and "last_good" cross-toolchains used in this bisection.  Naturally, the scripts will fail when triggerring benchmarking jobs if you don\'t have access to Linaro TCWG CI.
 
@@ -29,6 +32,6 @@ This benchmarking CI is work-in-progress, and we welcome feedback and suggestion
 -----------------8<--------------------------8<--------------------------8<--------------------------
 The information below can be used to reproduce a debug environment:
 
-Current build   : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/artifact/artifacts
-Reference build : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/44/artifact/artifacts
+Current build   : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/49/artifact/artifacts
+Reference build : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/artifact/artifacts
 
diff --git a/notify/mail-subject.txt b/notify/mail-subject.txt
index 1f4d028..21f3ffa 100644
--- a/notify/mail-subject.txt
+++ b/notify/mail-subject.txt
@@ -1 +1 @@
-[Linaro-TCWG-CI] 797 commits in llvm: No change
+[Linaro-TCWG-CI] 975 commits in llvm: 401.bzip2 reduced in size by 3%
diff --git a/notify/notify-full.log b/notify/notify-full.log
index 6f88034..52d66c7 100644
--- a/notify/notify-full.log
+++ b/notify/notify-full.log
@@ -1,6 +1,6 @@
 MOUNTS: /home/tcwg-buildslave/workspace/tcwg_bmk_0:/home/tcwg-buildslave/workspace/tcwg_bmk_0 /home/tcwg-buildslave/snapshots-ref:/home/tcwg-buildslave/snapshots-ref:ro /etc/ssh/ssh_host_ecdsa_key:/etc/ssh/ssh_host_ecdsa_key:ro /etc/ssh/ssh_host_ed25519_key:/etc/ssh/ssh_host_ed25519_key:ro /etc/ssh/ssh_host_rsa_key:/etc/ssh/ssh_host_rsa_key:ro /etc/ssh/ssh_host_ed25519_key.pub:/etc/ssh/ssh_host_ed25519_key.pub:ro /etc/ssh/ssh_host_rsa_key.pub:/etc/ssh/ssh_host_rsa_key.pub:ro /etc/ssh/ssh_host_dsa_key:/etc/ssh/ssh_host_dsa_key:ro /etc/ssh/ssh_host_dsa_key.pub:/etc/ssh/ssh_host_dsa_key.pub:ro /etc/ssh/ssh_host_ecdsa_key.pub:/etc/ssh/ssh_host_ecdsa_key.pub:ro ccache-tcwg_bmk-armhf-jammy:/home/tcwg-buildslave/.ccache /home/tcwg-benchmark
-DEBUG: starting docker on tcwg-snow_bmk-02 from tcwg-jade-02, date Tue Aug  1 05:30:03 PM UTC 2023
-ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build -dtP -v /home/tcwg-buildslave/workspace/tcwg_bmk_0:/home/tcwg-buildslave/workspace/tcwg_bmk_0 -v /home/tcwg-buildslave/snapshots-ref:/home/tcwg-buildslave/snapshots-ref:ro -v /etc/ssh/ssh_host_ecdsa_key:/etc/ssh/ssh_host_ecdsa_key:ro -v /etc/ssh/ssh_host_ed25519_key:/etc/ssh/ssh_host_ed25519_key:ro -v /etc/ssh/ssh_host_rsa_key:/etc/ssh/ssh_host_rsa_key:ro -v /etc/ssh/ssh_host_ed25519_key.pub:/etc/ssh/ssh_host_ed25519_key.pub:ro -v /etc/ssh/ssh_host_rsa_key.pub:/etc/ssh/ssh_host_rsa_key.pub:ro -v /etc/ssh/ssh_host_dsa_key:/etc/ssh/ssh_host_dsa_key:ro -v /etc/ssh/ssh_host_dsa_key.pub:/etc/ssh/ssh_host_dsa_key.pub:ro -v /etc/ssh/ssh_host_ecdsa_key.pub:/etc/ssh/ssh_host_ecdsa_key.pub:ro -v ccache-tcwg_bmk-armhf-jammy:/home/tcwg-buildslave/.ccache -v /home/tcwg-benchmark --memory=160000M --pids-limit=5000 --cpu-shares=1000 --cpuset-cpus 0-79 --cap-add=SYS_PTRACE --security-opt seccomp:unconfined linaro/ci-armhf-tcwg-build-ubuntu:jammy
+DEBUG: starting docker on tcwg-snow_bmk-02 from tcwg-jade-02, date Thu Aug 10 05:01:37 PM UTC 2023
+ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 49-tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build -dtP -v /home/tcwg-buildslave/workspace/tcwg_bmk_0:/home/tcwg-buildslave/workspace/tcwg_bmk_0 -v /home/tcwg-buildslave/snapshots-ref:/home/tcwg-buildslave/snapshots-ref:ro -v /etc/ssh/ssh_host_ecdsa_key:/etc/ssh/ssh_host_ecdsa_key:ro -v /etc/ssh/ssh_host_ed25519_key:/etc/ssh/ssh_host_ed25519_key:ro -v /etc/ssh/ssh_host_rsa_key:/etc/ssh/ssh_host_rsa_key:ro -v /etc/ssh/ssh_host_ed25519_key.pub:/etc/ssh/ssh_host_ed25519_key.pub:ro -v /etc/ssh/ssh_host_rsa_key.pub:/etc/ssh/ssh_host_rsa_key.pub:ro -v /etc/ssh/ssh_host_dsa_key:/etc/ssh/ssh_host_dsa_key:ro -v /etc/ssh/ssh_host_dsa_key.pub:/etc/ssh/ssh_host_dsa_key.pub:ro -v /etc/ssh/ssh_host_ecdsa_key.pub:/etc/ssh/ssh_host_ecdsa_key.pub:ro -v ccache-tcwg_bmk-armhf-jammy:/home/tcwg-buildslave/.ccache -v /home/tcwg-benchmark --memory=160000M --pids-limit=5000 --cpu-shares=1000 --cpuset-cpus 0-79 --cap-add=SYS_PTRACE --security-opt seccomp:unconfined linaro/ci-armhf-tcwg-build-ubuntu:jammy
 
 /home/tcwg-buildslave/workspace/tcwg_bmk_0/jenkins-scripts/round-robin-notify.sh @@rr[top_artifacts] artifacts __TCWG_JIRA_TOKEN ijQW9spm0p7HwZnUtLFx7CCA __stage full __verbose true
 
@@ -28,15 +28,15 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 ++ get_baseline_manifest BUILD_URL
 ++ get_manifest base-artifacts/manifest.sh BUILD_URL false
 ++ set +x
-#   Baseline  : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/44/
-+ echo '#   Baseline  : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/44/'
+#   Baseline  : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/
++ echo '#   Baseline  : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/'
 #   Using dir : base-artifacts
 + echo '#   Using dir : base-artifacts'
 ++ get_current_manifest BUILD_URL
 ++ get_manifest artifacts/manifest.sh BUILD_URL
 ++ set +x
-#   Artifacts : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/
-+ echo '#   Artifacts : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/'
+#   Artifacts : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/49/
++ echo '#   Artifacts : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/49/'
 #   Using dir : artifacts
 
 + echo '#   Using dir : artifacts'
@@ -44,8 +44,8 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 + mkdir -p artifacts/notify
 + '[' -f /home/tcwg-buildslave/.jipdate.yml ']'
 + set +x
-+ check_source_changes
 # check_source_changes
++ check_source_changes
 + echo '# check_source_changes'
 ++ get_current_manifest '{rr[update_baseline]}'
 ++ get_manifest artifacts/manifest.sh '{rr[update_baseline]}'
@@ -60,7 +60,7 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 +++ set -euf -o pipefail
 +++ local c delim=
 +++ for c in ${rr[components]}
-+++ '[' xhttps://github.com/llvm/llvm-project.git#bd077e98e463933e72bbd7bd03c6432d529e710c '!=' xbaseline ']'
++++ '[' xhttps://github.com/llvm/llvm-project.git#origin/main '!=' xbaseline ']'
 +++ echo -ne llvm
 +++ delim=' '
 +++ echo
@@ -76,7 +76,7 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 +++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 +++ set -euf -o pipefail +x
 +++ cat base-artifacts/git/llvm_rev
-++ '[' xbd077e98e463933e72bbd7bd03c6432d529e710c '!=' x0478ef2d366c6f88678e37d54190dcdaa0ec69da ']'
+++ '[' xd32e68ae539df62935ebb74cfff3d8490d1b1e35 '!=' xbd077e98e463933e72bbd7bd03c6432d529e710c ']'
 ++ echo -ne llvm
 ++ delim=' '
 ++ echo
@@ -89,14 +89,14 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 ++ assert_with_msg 'ERROR: No llvm_rev in current git' '[' -f artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat artifacts/git/llvm_rev
-+ first_bad=bd077e98e463933e72bbd7bd03c6432d529e710c
++ first_bad=d32e68ae539df62935ebb74cfff3d8490d1b1e35
 ++ get_baseline_git llvm_rev
 ++ set -euf -o pipefail
 ++ local base_artifacts=base-artifacts
 ++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat base-artifacts/git/llvm_rev
-+ last_good=0478ef2d366c6f88678e37d54190dcdaa0ec69da
++ last_good=bd077e98e463933e72bbd7bd03c6432d529e710c
 + local res
 + res=0
 + wait 106
@@ -113,20 +113,20 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 ++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat base-artifacts/git/llvm_rev
-+ base_rev=0478ef2d366c6f88678e37d54190dcdaa0ec69da
++ base_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
 ++ get_current_git llvm_rev
 ++ set -euf -o pipefail
 ++ assert_with_msg 'ERROR: No llvm_rev in current git' '[' -f artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat artifacts/git/llvm_rev
-+ cur_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
-++ git -C llvm rev-parse 'bd077e98e463933e72bbd7bd03c6432d529e710c^@'
++ cur_rev=d32e68ae539df62935ebb74cfff3d8490d1b1e35
+++ git -C llvm rev-parse 'd32e68ae539df62935ebb74cfff3d8490d1b1e35^@'
 + for sha1 in $(git -C "$c" rev-parse "$cur_rev^@")
-+ '[' ea90e289d97dff0d9893cc4559aff6390bdb27d9 = 0478ef2d366c6f88678e37d54190dcdaa0ec69da ']'
++ '[' b1ada7a1d31ee28a7c0e57268c6c6fe24931f25d = bd077e98e463933e72bbd7bd03c6432d529e710c ']'
 + return 1
 + change_kind=single_component
-+ echo '# Debug traces :'
 # Debug traces :
++ echo '# Debug traces :'
 #   change_kind=single_component : llvm
 + echo '#   change_kind=single_component : llvm'
 + for c in "${changed_components[@]}"
@@ -136,17 +136,17 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 ++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat base-artifacts/git/llvm_rev
-+ base_rev=0478ef2d366c6f88678e37d54190dcdaa0ec69da
++ base_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
 ++ get_current_git llvm_rev
 ++ set -euf -o pipefail
 ++ assert_with_msg 'ERROR: No llvm_rev in current git' '[' -f artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat artifacts/git/llvm_rev
-+ cur_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
-++ git -C llvm rev-list --count 0478ef2d366c6f88678e37d54190dcdaa0ec69da..bd077e98e463933e72bbd7bd03c6432d529e710c
-+ c_commits=797
-#   rev for llvm : 0478ef2d366c6f88678e37d54190dcdaa0ec69da..bd077e98e463933e72bbd7bd03c6432d529e710c (797 commits)
-+ echo '#   rev for llvm : 0478ef2d366c6f88678e37d54190dcdaa0ec69da..bd077e98e463933e72bbd7bd03c6432d529e710c (797 commits)'
++ cur_rev=d32e68ae539df62935ebb74cfff3d8490d1b1e35
+++ git -C llvm rev-list --count bd077e98e463933e72bbd7bd03c6432d529e710c..d32e68ae539df62935ebb74cfff3d8490d1b1e35
++ c_commits=975
+#   rev for llvm : bd077e98e463933e72bbd7bd03c6432d529e710c..d32e68ae539df62935ebb74cfff3d8490d1b1e35 (975 commits)
++ echo '#   rev for llvm : bd077e98e463933e72bbd7bd03c6432d529e710c..d32e68ae539df62935ebb74cfff3d8490d1b1e35 (975 commits)'
 
 + echo ''
 + setup_stages_to_run
@@ -166,12 +166,12 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 + print_config_f=bmk_print_config
 + generate_extra_details
 + set -euf -o pipefail
-# generate_extra_details
 + echo '# generate_extra_details'
+# generate_extra_details
 + post_interesting_commits init
 + set -euf -o pipefail
-+ echo '# post_interesting_commits'
 # post_interesting_commits
++ echo '# post_interesting_commits'
 + local stage=init
 + '[' single_component '!=' single_commit ']'
 + return
@@ -183,13 +183,13 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 + return
 + post_interesting_commits full
 + set -euf -o pipefail
-# post_interesting_commits
 + echo '# post_interesting_commits'
+# post_interesting_commits
 + local stage=full
 + '[' single_component '!=' single_commit ']'
 + return
-# print all notification files
 + echo '# print all notification files'
+# print all notification files
 + print_mail_recipients
 + false
 + return
@@ -283,11 +283,11 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 ++ get_current_manifest BUILD_URL
 ++ get_manifest artifacts/manifest.sh BUILD_URL
 ++ set +x
-+ bad_artifacts_url=https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/artifact/artifacts
++ bad_artifacts_url=https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/49/artifact/artifacts
 ++ get_baseline_manifest BUILD_URL
 ++ get_manifest base-artifacts/manifest.sh BUILD_URL false
 ++ set +x
-+ good_artifacts_url=https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/44/artifact/artifacts
++ good_artifacts_url=https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/artifact/artifacts
 + cat
 ++ bmk_print_result --oneline
 ++ false
@@ -333,20 +333,20 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 +++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 +++ set -euf -o pipefail +x
 +++ cat base-artifacts/git/llvm_rev
-++ base_rev=0478ef2d366c6f88678e37d54190dcdaa0ec69da
+++ base_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
 +++ get_current_git llvm_rev
 +++ set -euf -o pipefail
 +++ assert_with_msg 'ERROR: No llvm_rev in current git' '[' -f artifacts/git/llvm_rev ']'
 +++ set -euf -o pipefail +x
 +++ cat artifacts/git/llvm_rev
-++ cur_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
-+++ git -C llvm rev-list --count 0478ef2d366c6f88678e37d54190dcdaa0ec69da..bd077e98e463933e72bbd7bd03c6432d529e710c
-++ c_commits=797
-++ new_commits=797
+++ cur_rev=d32e68ae539df62935ebb74cfff3d8490d1b1e35
++++ git -C llvm rev-list --count bd077e98e463933e72bbd7bd03c6432d529e710c..d32e68ae539df62935ebb74cfff3d8490d1b1e35
+++ c_commits=975
+++ new_commits=975
 +++ echo llvm
 +++ tr ' ' ,
 ++ components=llvm
-++ echo '797 commits in llvm'
+++ echo '975 commits in llvm'
 ++ print_commits --short
 ++ sed -e 's/^/  | /'
 ++ false
@@ -361,27 +361,27 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 +++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 +++ set -euf -o pipefail +x
 +++ cat base-artifacts/git/llvm_rev
-++ base_rev=0478ef2d366c6f88678e37d54190dcdaa0ec69da
+++ base_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
 +++ get_current_git llvm_rev
 +++ set -euf -o pipefail
 +++ assert_with_msg 'ERROR: No llvm_rev in current git' '[' -f artifacts/git/llvm_rev ']'
 +++ set -euf -o pipefail +x
 +++ cat artifacts/git/llvm_rev
-++ cur_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
-+++ git -C llvm rev-list --count 0478ef2d366c6f88678e37d54190dcdaa0ec69da..bd077e98e463933e72bbd7bd03c6432d529e710c
-++ c_commits=797
-++ new_commits=797
+++ cur_rev=d32e68ae539df62935ebb74cfff3d8490d1b1e35
++++ git -C llvm rev-list --count bd077e98e463933e72bbd7bd03c6432d529e710c..d32e68ae539df62935ebb74cfff3d8490d1b1e35
+++ c_commits=975
+++ new_commits=975
 ++ echo 'llvm commits:'
-+++ git -C llvm log --pretty=oneline 0478ef2d366c6f88678e37d54190dcdaa0ec69da..bd077e98e463933e72bbd7bd03c6432d529e710c
++++ git -C llvm log --pretty=oneline bd077e98e463933e72bbd7bd03c6432d529e710c..d32e68ae539df62935ebb74cfff3d8490d1b1e35
 +++ head -n5
 +++ true
-++ echo 'bd077e98e463933e72bbd7bd03c6432d529e710c Released restriction that prevented implicit dynamic-to-static dimension type cast in TOSA ops.
-ea90e289d97dff0d9893cc4559aff6390bdb27d9 Fix typo
-89e44e33edeca2fdfa1992096cd5faf98a1f7dae [clang][Interp] Implement __builtin_fmax
-7ef87931ddd714f493b9e864c9273f02895fdfef [libc++][PSTL] Silence clang-tidy errors.
-8698d56d996a72b6ea87d751ec644c9c611717e2 [Transforms][LICM] Add the ability to undo unprofitable reassociation'
-++ '[' 797 -gt 5 ']'
-++ echo '... and 792 more'
+++ echo 'd32e68ae539df62935ebb74cfff3d8490d1b1e35 [docs] Graduate VectorizationPlan.rst from proposal.
+b1ada7a1d31ee28a7c0e57268c6c6fe24931f25d [DAG] Support store merging of vector constant stores (try 2)
+e838471bc4936854ced32709f7f105af87aed798 [X86] Add regression test case from pr64593
+0e2c1cb233294035994267e00b6ba05e5e682e19 [X86] ReplaceNodeResults - always attempt to fold TRUNCATE -> PACKSS/PACKUS nodes.
+888717e85395c22dcc70a7d16ae8459e70df822c [mlir][transform] Enable gpu-to-nvvm via conversion patterns driven by TD'
+++ '[' 975 -gt 5 ']'
+++ echo '... and 970 more'
 ++ bmk_print_result --short
 ++ false
 ++ local print_arg=--short
@@ -550,46 +550,46 @@ ea90e289d97dff0d9893cc4559aff6390bdb27d9 Fix typo
 +++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 +++ set -euf -o pipefail +x
 +++ cat base-artifacts/git/llvm_rev
-++ base_rev=0478ef2d366c6f88678e37d54190dcdaa0ec69da
+++ base_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
 +++ get_current_git llvm_rev
 +++ set -euf -o pipefail
 +++ assert_with_msg 'ERROR: No llvm_rev in current git' '[' -f artifacts/git/llvm_rev ']'
 +++ set -euf -o pipefail +x
 +++ cat artifacts/git/llvm_rev
-++ cur_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
-+++ git -C llvm rev-list --count 0478ef2d366c6f88678e37d54190dcdaa0ec69da..bd077e98e463933e72bbd7bd03c6432d529e710c
-++ c_commits=797
-++ new_commits=797
+++ cur_rev=d32e68ae539df62935ebb74cfff3d8490d1b1e35
++++ git -C llvm rev-list --count bd077e98e463933e72bbd7bd03c6432d529e710c..d32e68ae539df62935ebb74cfff3d8490d1b1e35
+++ c_commits=975
+++ new_commits=975
 +++ echo llvm
 +++ tr ' ' ,
 ++ components=llvm
-++ echo '797 commits in llvm'
-+ echo '[TCWG-CI] No change after llvm commit: 797 commits in llvm'
+++ echo '975 commits in llvm'
++ echo '[TCWG-CI] No change after llvm commit: 975 commits in llvm'
 # generate dashboard
 + echo '# generate dashboard'
 + generate_dashboard_squad
 + local results_date
-# generate_dashboard_squad
 + echo '# generate_dashboard_squad'
+# generate_dashboard_squad
 ... Skipping
 + false
 + echo '... Skipping'
 + return
 + post_dashboard_squad
-# post_dashboard_squad
 + echo '# post_dashboard_squad'
-+ false
+# post_dashboard_squad
 ... Skipping
++ false
 + echo '... Skipping'
 + return
 + false
 => Not the first detection of this issue. Not sending mail.
 + echo '=> Not the first detection of this issue. Not sending mail.'
 + post_to_jira
-+ echo '# post_to_jira'
 # post_to_jira
+Full stage ran successfully.
++ echo '# post_to_jira'
 + false
 + false
-Full stage ran successfully.
 + echo 'Full stage ran successfully.'
-ec5e0e44439238373c3115a86416952ab4309515610facdc354ab95cef7b9d6b
+c8b0301854fdb128bf114589b0d3c85ebb6d12db1335f288b94dfeec1938a14a
diff --git a/notify/notify-init.log b/notify/notify-init.log
index 1558601..d53ac60 100644
--- a/notify/notify-init.log
+++ b/notify/notify-init.log
@@ -1,25 +1,18 @@
 MOUNTS: /home/tcwg-buildslave/workspace/tcwg_bmk_0:/home/tcwg-buildslave/workspace/tcwg_bmk_0 /home/tcwg-buildslave/snapshots-ref:/home/tcwg-buildslave/snapshots-ref:ro /etc/ssh/ssh_host_ecdsa_key:/etc/ssh/ssh_host_ecdsa_key:ro /etc/ssh/ssh_host_ed25519_key:/etc/ssh/ssh_host_ed25519_key:ro /etc/ssh/ssh_host_rsa_key:/etc/ssh/ssh_host_rsa_key:ro /etc/ssh/ssh_host_ed25519_key.pub:/etc/ssh/ssh_host_ed25519_key.pub:ro /etc/ssh/ssh_host_rsa_key.pub:/etc/ssh/ssh_host_rsa_key.pub:ro /etc/ssh/ssh_host_dsa_key:/etc/ssh/ssh_host_dsa_key:ro /etc/ssh/ssh_host_dsa_key.pub:/etc/ssh/ssh_host_dsa_key.pub:ro /etc/ssh/ssh_host_ecdsa_key.pub:/etc/ssh/ssh_host_ecdsa_key.pub:ro ccache-tcwg_bmk-armhf-jammy:/home/tcwg-buildslave/.ccache /home/tcwg-benchmark
-DEBUG: starting docker on tcwg-snow_bmk-02 from tcwg-jade-02, date Tue Aug  1 05:29:56 PM UTC 2023
-ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build -dtP -v /home/tcwg-buildslave/workspace/tcwg_bmk_0:/home/tcwg-buildslave/workspace/tcwg_bmk_0 -v /home/tcwg-buildslave/snapshots-ref:/home/tcwg-buildslave/snapshots-ref:ro -v /etc/ssh/ssh_host_ecdsa_key:/etc/ssh/ssh_host_ecdsa_key:ro -v /etc/ssh/ssh_host_ed25519_key:/etc/ssh/ssh_host_ed25519_key:ro -v /etc/ssh/ssh_host_rsa_key:/etc/ssh/ssh_host_rsa_key:ro -v /etc/ssh/ssh_host_ed25519_key.pub:/etc/ssh/ssh_host_ed25519_key.pub:ro -v /etc/ssh/ssh_host_rsa_key.pub:/etc/ssh/ssh_host_rsa_key.pub:ro -v /etc/ssh/ssh_host_dsa_key:/etc/ssh/ssh_host_dsa_key:ro -v /etc/ssh/ssh_host_dsa_key.pub:/etc/ssh/ssh_host_dsa_key.pub:ro -v /etc/ssh/ssh_host_ecdsa_key.pub:/etc/ssh/ssh_host_ecdsa_key.pub:ro -v ccache-tcwg_bmk-armhf-jammy:/home/tcwg-buildslave/.ccache -v /home/tcwg-benchmark --memory=160000M --pids-limit=5000 --cpu-shares=1000 --cpuset-cpus 0-79 --cap-add=SYS_PTRACE --security-opt seccomp:unconfined linaro/ci-armhf-tcwg-build-ubuntu:jammy
+DEBUG: starting docker on tcwg-snow_bmk-02 from tcwg-jade-02, date Thu Aug 10 05:01:33 PM UTC 2023
+ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 49-tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build -dtP -v /home/tcwg-buildslave/workspace/tcwg_bmk_0:/home/tcwg-buildslave/workspace/tcwg_bmk_0 -v /home/tcwg-buildslave/snapshots-ref:/home/tcwg-buildslave/snapshots-ref:ro -v /etc/ssh/ssh_host_ecdsa_key:/etc/ssh/ssh_host_ecdsa_key:ro -v /etc/ssh/ssh_host_ed25519_key:/etc/ssh/ssh_host_ed25519_key:ro -v /etc/ssh/ssh_host_rsa_key:/etc/ssh/ssh_host_rsa_key:ro -v /etc/ssh/ssh_host_ed25519_key.pub:/etc/ssh/ssh_host_ed25519_key.pub:ro -v /etc/ssh/ssh_host_rsa_key.pub:/etc/ssh/ssh_host_rsa_key.pub:ro -v /etc/ssh/ssh_host_dsa_key:/etc/ssh/ssh_host_dsa_key:ro -v /etc/ssh/ssh_host_dsa_key.pub:/etc/ssh/ssh_host_dsa_key.pub:ro -v /etc/ssh/ssh_host_ecdsa_key.pub:/etc/ssh/ssh_host_ecdsa_key.pub:ro -v ccache-tcwg_bmk-armhf-jammy:/home/tcwg-buildslave/.ccache -v /home/tcwg-benchmark --memory=160000M --pids-limit=5000 --cpu-shares=1000 --cpuset-cpus 0-79 --cap-add=SYS_PTRACE --security-opt seccomp:unconfined linaro/ci-armhf-tcwg-build-ubuntu:jammy
 
 /home/tcwg-buildslave/workspace/tcwg_bmk_0/jenkins-scripts/round-robin-notify.sh @@rr[top_artifacts] artifacts --notify ignore __stage init __verbose true
 
 + false
 + dryrun=
-# setup_notify_environment
-# Debug traces :
-#   Baseline  : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/44/
-#   Using dir : base-artifacts
-#   Artifacts : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/
-#   Using dir : artifacts
-
-# check_source_changes
 + declare top_artifacts ci_project ci_config
 + declare change_kind changed_single_component last_good first_bad
 + declare -a changed_components
 + declare print_commits_f print_result_f print_config_f print_last_icommit_f
 + dump_model_only=false
 + setup_notify_environment
+# setup_notify_environment
 + echo '# setup_notify_environment'
 + top_artifacts=artifacts
 ++ get_current_manifest '{rr[ci_project]}'
@@ -31,21 +24,28 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 ++ set +x
 + ci_config=llvm-arm-master-O3
 + echo '# Debug traces :'
+# Debug traces :
 ++ get_baseline_manifest BUILD_URL
 ++ get_manifest base-artifacts/manifest.sh BUILD_URL false
 ++ set +x
-+ echo '#   Baseline  : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/44/'
++ echo '#   Baseline  : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/'
+#   Baseline  : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/
+#   Using dir : base-artifacts
 + echo '#   Using dir : base-artifacts'
 ++ get_current_manifest BUILD_URL
 ++ get_manifest artifacts/manifest.sh BUILD_URL
 ++ set +x
-+ echo '#   Artifacts : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/46/'
+#   Artifacts : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/49/
+#   Using dir : artifacts
+
++ echo '#   Artifacts : https://ci.linaro.org/job/tcwg_bmk-code_size-spec2k6--llvm-arm-master-O3-build/49/'
 + echo '#   Using dir : artifacts'
 + echo ''
 + mkdir -p artifacts/notify
 + '[' -f /home/tcwg-buildslave/.jipdate.yml ']'
 + set +x
 + check_source_changes
+# check_source_changes
 + echo '# check_source_changes'
 ++ get_current_manifest '{rr[update_baseline]}'
 ++ get_manifest artifacts/manifest.sh '{rr[update_baseline]}'
@@ -60,7 +60,7 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 +++ set -euf -o pipefail
 +++ local c delim=
 +++ for c in ${rr[components]}
-+++ '[' xhttps://github.com/llvm/llvm-project.git#bd077e98e463933e72bbd7bd03c6432d529e710c '!=' xbaseline ']'
++++ '[' xhttps://github.com/llvm/llvm-project.git#origin/main '!=' xbaseline ']'
 +++ echo -ne llvm
 +++ delim=' '
 +++ echo
@@ -76,7 +76,7 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 +++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 +++ set -euf -o pipefail +x
 +++ cat base-artifacts/git/llvm_rev
-++ '[' xbd077e98e463933e72bbd7bd03c6432d529e710c '!=' x0478ef2d366c6f88678e37d54190dcdaa0ec69da ']'
+++ '[' xd32e68ae539df62935ebb74cfff3d8490d1b1e35 '!=' xbd077e98e463933e72bbd7bd03c6432d529e710c ']'
 ++ echo -ne llvm
 ++ delim=' '
 ++ echo
@@ -89,14 +89,14 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 ++ assert_with_msg 'ERROR: No llvm_rev in current git' '[' -f artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat artifacts/git/llvm_rev
-+ first_bad=bd077e98e463933e72bbd7bd03c6432d529e710c
++ first_bad=d32e68ae539df62935ebb74cfff3d8490d1b1e35
 ++ get_baseline_git llvm_rev
 ++ set -euf -o pipefail
 ++ local base_artifacts=base-artifacts
 ++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat base-artifacts/git/llvm_rev
-+ last_good=0478ef2d366c6f88678e37d54190dcdaa0ec69da
++ last_good=bd077e98e463933e72bbd7bd03c6432d529e710c
 + local res
 + res=0
 + wait 108
@@ -113,16 +113,16 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 ++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat base-artifacts/git/llvm_rev
-+ base_rev=0478ef2d366c6f88678e37d54190dcdaa0ec69da
++ base_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
 ++ get_current_git llvm_rev
 ++ set -euf -o pipefail
 ++ assert_with_msg 'ERROR: No llvm_rev in current git' '[' -f artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat artifacts/git/llvm_rev
-+ cur_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
-++ git -C llvm rev-parse 'bd077e98e463933e72bbd7bd03c6432d529e710c^@'
++ cur_rev=d32e68ae539df62935ebb74cfff3d8490d1b1e35
+++ git -C llvm rev-parse 'd32e68ae539df62935ebb74cfff3d8490d1b1e35^@'
 + for sha1 in $(git -C "$c" rev-parse "$cur_rev^@")
-+ '[' ea90e289d97dff0d9893cc4559aff6390bdb27d9 = 0478ef2d366c6f88678e37d54190dcdaa0ec69da ']'
++ '[' b1ada7a1d31ee28a7c0e57268c6c6fe24931f25d = bd077e98e463933e72bbd7bd03c6432d529e710c ']'
 + return 1
 + change_kind=single_component
 # Debug traces :
@@ -136,18 +136,18 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 ++ assert_with_msg 'ERROR: No llvm_rev in baseline git' '[' -f base-artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat base-artifacts/git/llvm_rev
-+ base_rev=0478ef2d366c6f88678e37d54190dcdaa0ec69da
++ base_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
 ++ get_current_git llvm_rev
 ++ set -euf -o pipefail
 ++ assert_with_msg 'ERROR: No llvm_rev in current git' '[' -f artifacts/git/llvm_rev ']'
 ++ set -euf -o pipefail +x
 ++ cat artifacts/git/llvm_rev
-+ cur_rev=bd077e98e463933e72bbd7bd03c6432d529e710c
-++ git -C llvm rev-list --count 0478ef2d366c6f88678e37d54190dcdaa0ec69da..bd077e98e463933e72bbd7bd03c6432d529e710c
-+ c_commits=797
-+ echo '#   rev for llvm : 0478ef2d366c6f88678e37d54190dcdaa0ec69da..bd077e98e463933e72bbd7bd03c6432d529e710c (797 commits)'
-#   rev for llvm : 0478ef2d366c6f88678e37d54190dcdaa0ec69da..bd077e98e463933e72bbd7bd03c6432d529e710c (797 commits)
++ cur_rev=d32e68ae539df62935ebb74cfff3d8490d1b1e35
+++ git -C llvm rev-list --count bd077e98e463933e72bbd7bd03c6432d529e710c..d32e68ae539df62935ebb74cfff3d8490d1b1e35
++ c_commits=975
+#   rev for llvm : bd077e98e463933e72bbd7bd03c6432d529e710c..d32e68ae539df62935ebb74cfff3d8490d1b1e35 (975 commits)
 
++ echo '#   rev for llvm : bd077e98e463933e72bbd7bd03c6432d529e710c..d32e68ae539df62935ebb74cfff3d8490d1b1e35 (975 commits)'
 + echo ''
 + setup_stages_to_run
 + '[' xignore == xignore ']'
@@ -166,8 +166,8 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 + print_config_f=bmk_print_config
 + generate_extra_details
 + set -euf -o pipefail
-+ echo '# generate_extra_details'
 # generate_extra_details
++ echo '# generate_extra_details'
 + post_interesting_commits init
 + set -euf -o pipefail
 # post_interesting_commits
@@ -179,4 +179,4 @@ ssh -Snone -oForwardAgent=no tcwg-snow_bmk-02 docker-wrapper run --name 46-tcwg_
 Init stage ran successfully.
 + echo 'Init stage ran successfully.'
 + exit 0
-a17ade6cabe6945bdaaadd7294d663456931ce8439ba52bfe5d4440d11427442
+8683b2bb073c5595eeb7ccff33032197fa2a2c00e786ac36502549282e92d5f3
diff --git a/notify/output-bmk-results.log b/notify/output-bmk-results.log
index f64c72a..404ee43 100644
--- a/notify/output-bmk-results.log
+++ b/notify/output-bmk-results.log
@@ -108,7 +108,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking exe.regression : 401.bzip2,bzip2_base.default : size=0% (threshold=1%)
+DEBUG: checking exe.regression : 401.bzip2,bzip2_base.default : size=3% (threshold=1%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -153,7 +153,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking exe.regression : 433.milc,milc_base.default : size=0% (threshold=1%)
+DEBUG: checking exe.regression : 433.milc,milc_base.default : size=1% (threshold=1%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -183,6 +183,21 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+DEBUG: checking exe.regression : 445.gobmk,gobmk_base.default : size=1% (threshold=1%)
+output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
+ --- modulename: output-bmk-results, funcname: is_entry_regression
+output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
+output-bmk-results.py(168):         return (result - 100 > threshold)
+output-bmk-results.py(217):             continue
+output-bmk-results.py(208):     for index, row in out_df.iterrows():
+output-bmk-results.py(210):         threshold = get_threshold(sym_type, metric, mode, row["benchmark"], row["symbol"])
+ --- modulename: output-bmk-results, funcname: get_threshold
+output-bmk-results.py(98):     if metric == "sample":
+output-bmk-results.py(108):     if metric == "num_vect_loops" or metric == "num_sve_loops":
+output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mode)]
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 DEBUG: checking exe.regression : 447.dealII,dealII_base.default : size=0% (threshold=1%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
@@ -381,12 +396,37 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking exe.improvement : 401.bzip2,bzip2_base.default : size=0% (threshold=1%)
+DEBUG: checking exe.improvement : 401.bzip2,bzip2_base.default : size=3% (threshold=1%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
 output-bmk-results.py(177):         return (100 - result > threshold)
-output-bmk-results.py(217):             continue
+output-bmk-results.py(219):         percent_change, short_diag, long_diag = get_short_long_diag(row, metric, sym_type, change_kind)
+ --- modulename: output-bmk-results, funcname: get_short_long_diag
+output-bmk-results.py(114):     bmk = row["benchmark"]
+output-bmk-results.py(115):     rel_value = row["rel_" + metric]
+output-bmk-results.py(116):     prev_value = row[metric + "_x"]
+output-bmk-results.py(117):     curr_value = row[metric + "_y"]
+output-bmk-results.py(119):     if metric == "sample":
+output-bmk-results.py(136):     suffix = ""
+output-bmk-results.py(137):     if metric == "sample":
+output-bmk-results.py(141):     elif metric == "size":
+output-bmk-results.py(142):         prefix_regression = "grew in size by"
+output-bmk-results.py(143):         prefix_improvement = "reduced in size by"
+output-bmk-results.py(144):         suffix = "bytes"
+output-bmk-results.py(151):     if sym_type=="symbol":
+output-bmk-results.py(154):       item=bmk
+output-bmk-results.py(156):     short_diag = "{0} {1} {2}%".format(item, locals()["prefix_" + change_kind], abs(rel_value - 100))
+output-bmk-results.py(157):     long_diag = "{0} from {1} to {2} {3}".format(short_diag, prev_value, curr_value, suffix)
+output-bmk-results.py(158):     return abs(rel_value - 100), short_diag, long_diag
+output-bmk-results.py(223):         if metric == "sample" \
+output-bmk-results.py(230):         print("DEBUG: *** {0},{1} : {2}".format(row["benchmark"], row["symbol"], long_diag))
+DEBUG: *** 401.bzip2,bzip2_base.default : 401.bzip2 reduced in size by 3% from 77619 to 75179 bytes
+output-bmk-results.py(232):         f_out.write_csv((percent_change, row["benchmark"], row["symbol"], short_diag, long_diag))
+ --- modulename: output-bmk-results, funcname: write_csv
+output-bmk-results.py(41):     if not self.predicate or not self.csvwriter:
+output-bmk-results.py(43):     self.csvwriter.writerow(arr)
+output-bmk-results.py(233):         if change_kind == "regression":
 output-bmk-results.py(208):     for index, row in out_df.iterrows():
 output-bmk-results.py(210):         threshold = get_threshold(sym_type, metric, mode, row["benchmark"], row["symbol"])
  --- modulename: output-bmk-results, funcname: get_threshold
@@ -426,7 +466,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking exe.improvement : 433.milc,milc_base.default : size=0% (threshold=1%)
+DEBUG: checking exe.improvement : 433.milc,milc_base.default : size=1% (threshold=1%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -456,6 +496,21 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+DEBUG: checking exe.improvement : 445.gobmk,gobmk_base.default : size=1% (threshold=1%)
+output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
+ --- modulename: output-bmk-results, funcname: is_entry_improvement
+output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
+output-bmk-results.py(177):         return (100 - result > threshold)
+output-bmk-results.py(217):             continue
+output-bmk-results.py(208):     for index, row in out_df.iterrows():
+output-bmk-results.py(210):         threshold = get_threshold(sym_type, metric, mode, row["benchmark"], row["symbol"])
+ --- modulename: output-bmk-results, funcname: get_threshold
+output-bmk-results.py(98):     if metric == "sample":
+output-bmk-results.py(108):     if metric == "num_vect_loops" or metric == "num_sve_loops":
+output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mode)]
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 DEBUG: checking exe.improvement : 447.dealII,dealII_base.default : size=0% (threshold=1%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
@@ -633,7 +688,6 @@ output-bmk-results.py(237):     f_out.close()
 output-bmk-results.py(29):     if not self.outf:
 output-bmk-results.py(31):     self.outf.close()
 output-bmk-results.py(32):     if os.stat(self.filename).st_size == 0:
-output-bmk-results.py(33):        os.remove(self.filename)
 output-bmk-results.py(257):     output_bmk_results_1(sym_df, "symbol", "regression", f_regr, f_skip, f_ebp, run_step_artifacts, metric, mode, details)
  --- modulename: output-bmk-results, funcname: output_bmk_results_1
 output-bmk-results.py(202):     f_out = Outfile("{0}/{1}.{2}".format(run_step_artifacts, sym_type, change_kind), "w", predicate=(details=="verbose"))
@@ -699,7 +753,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 401.bzip2,[.] BZ2_decompress : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 401.bzip2,[.] BZ2_decompress : size=12% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -789,7 +843,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 433.milc,[.] mult_su3_nn : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 433.milc,[.] mult_adj_su3_mat_vec : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -804,7 +858,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 433.milc,[.] mult_adj_su3_mat_vec : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 433.milc,[.] mult_su3_mat_vec : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -819,7 +873,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 433.milc,[.] mult_su3_mat_vec : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 433.milc,[.] mult_su3_nn : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -864,7 +918,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 444.namd,[.] _ZN20ComputeNonbondedUtil16calc_pair_energyEP9nonbonded : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 444.namd,[.] _ZN20ComputeNonbondedUtil32calc_pair_energy_merge_fullelectEP9nonbonded : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -879,7 +933,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 444.namd,[.] _ZN20ComputeNonbondedUtil32calc_pair_energy_merge_fullelectEP9nonbonded : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 444.namd,[.] _ZN20ComputeNonbondedUtil16calc_pair_energyEP9nonbonded : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -924,6 +978,36 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+DEBUG: checking symbol.regression : 445.gobmk,[.] do_play_move : size=0% (threshold=10%)
+output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
+ --- modulename: output-bmk-results, funcname: is_entry_regression
+output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
+output-bmk-results.py(168):         return (result - 100 > threshold)
+output-bmk-results.py(217):             continue
+output-bmk-results.py(208):     for index, row in out_df.iterrows():
+output-bmk-results.py(210):         threshold = get_threshold(sym_type, metric, mode, row["benchmark"], row["symbol"])
+ --- modulename: output-bmk-results, funcname: get_threshold
+output-bmk-results.py(98):     if metric == "sample":
+output-bmk-results.py(108):     if metric == "num_vect_loops" or metric == "num_sve_loops":
+output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mode)]
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+DEBUG: checking symbol.regression : 445.gobmk,[.] do_dfa_matchpat : size=0% (threshold=10%)
+output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
+ --- modulename: output-bmk-results, funcname: is_entry_regression
+output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
+output-bmk-results.py(168):         return (result - 100 > threshold)
+output-bmk-results.py(217):             continue
+output-bmk-results.py(208):     for index, row in out_df.iterrows():
+output-bmk-results.py(210):         threshold = get_threshold(sym_type, metric, mode, row["benchmark"], row["symbol"])
+ --- modulename: output-bmk-results, funcname: get_threshold
+output-bmk-results.py(98):     if metric == "sample":
+output-bmk-results.py(108):     if metric == "num_vect_loops" or metric == "num_sve_loops":
+output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mode)]
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 DEBUG: checking symbol.regression : 447.dealII,[.] _ZNK9MappingQ1ILi3EE12compute_fillERK12TriaIteratorILi3E15DoFCellAccessorILi3EEEjN10QProjectorILi3EE17DataSetDescriptorERNS0_12InternalDataERSt6vectorI5PointILi3EESaISE_EE : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
@@ -1014,7 +1098,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 453.povray,[.] _ZN3povL23All_Plane_IntersectionsEPNS_13Object_StructEPNS_10Ray_StructEPNS_13istack_structE : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 450.soplex,[.] _ZN6soplex9SPxFastRT8maxDeltaERdS1_RNS_12UpdateVectorERNS_6VectorES5_ii : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -1029,7 +1113,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 453.povray,[.] _ZN3povL24All_Sphere_IntersectionsEPNS_13Object_StructEPNS_10Ray_StructEPNS_13istack_structE : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 453.povray,[.] _ZN3povL23All_Plane_IntersectionsEPNS_13Object_StructEPNS_10Ray_StructEPNS_13istack_structE : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -1059,6 +1143,21 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+DEBUG: checking symbol.regression : 453.povray,[.] _ZN3povL24All_Sphere_IntersectionsEPNS_13Object_StructEPNS_10Ray_StructEPNS_13istack_structE : size=0% (threshold=10%)
+output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
+ --- modulename: output-bmk-results, funcname: is_entry_regression
+output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
+output-bmk-results.py(168):         return (result - 100 > threshold)
+output-bmk-results.py(217):             continue
+output-bmk-results.py(208):     for index, row in out_df.iterrows():
+output-bmk-results.py(210):         threshold = get_threshold(sym_type, metric, mode, row["benchmark"], row["symbol"])
+ --- modulename: output-bmk-results, funcname: get_threshold
+output-bmk-results.py(98):     if metric == "sample":
+output-bmk-results.py(108):     if metric == "num_vect_loops" or metric == "num_sve_loops":
+output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mode)]
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 DEBUG: checking symbol.regression : 453.povray,[.] _ZN3pov17Check_And_EnqueueEPNS_21Priority_Queue_StructEPNS_16BBox_Tree_StructEPNS_19Bounding_Box_StructEPNS_14Rayinfo_StructE : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
@@ -1149,7 +1248,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 462.libquantum,[.] quantum_cnot : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 462.libquantum,[.] quantum_sigma_x : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -1164,7 +1263,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 462.libquantum,[.] quantum_sigma_x : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 462.libquantum,[.] quantum_cnot : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -1344,7 +1443,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 483.xalancbmk,[.] _ZN10xalanc_1_814XalanDOMString6equalsERKS0_S2_ : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 483.xalancbmk,[.] _ZNK10xalanc_1_85XPath11executeMoreEPNS_9XalanNodeEN9__gnu_cxx17__normal_iteratorIPKiSt6vectorIiSaIiEEEERNS_21XPathExecutionContextE : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -1359,7 +1458,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.regression : 483.xalancbmk,[.] _ZNK10xalanc_1_85XPath11executeMoreEPNS_9XalanNodeEN9__gnu_cxx17__normal_iteratorIPKiSt6vectorIiSaIiEEEERNS_21XPathExecutionContextE : size=0% (threshold=10%)
+DEBUG: checking symbol.regression : 483.xalancbmk,[.] _ZN10xalanc_1_814VariablesStack9findEntryERKNS_10XalanQNameEbb : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_regression
 output-bmk-results.py(167):     if metric in metric_utils.higher_regress_metrics:
@@ -1437,12 +1536,37 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 401.bzip2,[.] BZ2_decompress : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 401.bzip2,[.] BZ2_decompress : size=12% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
 output-bmk-results.py(177):         return (100 - result > threshold)
-output-bmk-results.py(217):             continue
+output-bmk-results.py(219):         percent_change, short_diag, long_diag = get_short_long_diag(row, metric, sym_type, change_kind)
+ --- modulename: output-bmk-results, funcname: get_short_long_diag
+output-bmk-results.py(114):     bmk = row["benchmark"]
+output-bmk-results.py(115):     rel_value = row["rel_" + metric]
+output-bmk-results.py(116):     prev_value = row[metric + "_x"]
+output-bmk-results.py(117):     curr_value = row[metric + "_y"]
+output-bmk-results.py(119):     if metric == "sample":
+output-bmk-results.py(136):     suffix = ""
+output-bmk-results.py(137):     if metric == "sample":
+output-bmk-results.py(141):     elif metric == "size":
+output-bmk-results.py(142):         prefix_regression = "grew in size by"
+output-bmk-results.py(143):         prefix_improvement = "reduced in size by"
+output-bmk-results.py(144):         suffix = "bytes"
+output-bmk-results.py(151):     if sym_type=="symbol":
+output-bmk-results.py(152):       item=bmk+":"+row["symbol"]
+output-bmk-results.py(156):     short_diag = "{0} {1} {2}%".format(item, locals()["prefix_" + change_kind], abs(rel_value - 100))
+output-bmk-results.py(157):     long_diag = "{0} from {1} to {2} {3}".format(short_diag, prev_value, curr_value, suffix)
+output-bmk-results.py(158):     return abs(rel_value - 100), short_diag, long_diag
+output-bmk-results.py(223):         if metric == "sample" \
+output-bmk-results.py(230):         print("DEBUG: *** {0},{1} : {2}".format(row["benchmark"], row["symbol"], long_diag))
+DEBUG: *** 401.bzip2,[.] BZ2_decompress : 401.bzip2:[.] BZ2_decompress reduced in size by 12% from 20424 to 17984 bytes
+output-bmk-results.py(232):         f_out.write_csv((percent_change, row["benchmark"], row["symbol"], short_diag, long_diag))
+ --- modulename: output-bmk-results, funcname: write_csv
+output-bmk-results.py(41):     if not self.predicate or not self.csvwriter:
+output-bmk-results.py(43):     self.csvwriter.writerow(arr)
+output-bmk-results.py(233):         if change_kind == "regression":
 output-bmk-results.py(208):     for index, row in out_df.iterrows():
 output-bmk-results.py(210):         threshold = get_threshold(sym_type, metric, mode, row["benchmark"], row["symbol"])
  --- modulename: output-bmk-results, funcname: get_threshold
@@ -1527,7 +1651,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 433.milc,[.] mult_su3_nn : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 433.milc,[.] mult_adj_su3_mat_vec : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -1542,7 +1666,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 433.milc,[.] mult_adj_su3_mat_vec : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 433.milc,[.] mult_su3_mat_vec : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -1557,7 +1681,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 433.milc,[.] mult_su3_mat_vec : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 433.milc,[.] mult_su3_nn : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -1602,7 +1726,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 444.namd,[.] _ZN20ComputeNonbondedUtil16calc_pair_energyEP9nonbonded : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 444.namd,[.] _ZN20ComputeNonbondedUtil32calc_pair_energy_merge_fullelectEP9nonbonded : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -1617,7 +1741,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 444.namd,[.] _ZN20ComputeNonbondedUtil32calc_pair_energy_merge_fullelectEP9nonbonded : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 444.namd,[.] _ZN20ComputeNonbondedUtil16calc_pair_energyEP9nonbonded : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -1662,6 +1786,36 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+DEBUG: checking symbol.improvement : 445.gobmk,[.] do_play_move : size=0% (threshold=10%)
+output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
+ --- modulename: output-bmk-results, funcname: is_entry_improvement
+output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
+output-bmk-results.py(177):         return (100 - result > threshold)
+output-bmk-results.py(217):             continue
+output-bmk-results.py(208):     for index, row in out_df.iterrows():
+output-bmk-results.py(210):         threshold = get_threshold(sym_type, metric, mode, row["benchmark"], row["symbol"])
+ --- modulename: output-bmk-results, funcname: get_threshold
+output-bmk-results.py(98):     if metric == "sample":
+output-bmk-results.py(108):     if metric == "num_vect_loops" or metric == "num_sve_loops":
+output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mode)]
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+DEBUG: checking symbol.improvement : 445.gobmk,[.] do_dfa_matchpat : size=0% (threshold=10%)
+output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
+ --- modulename: output-bmk-results, funcname: is_entry_improvement
+output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
+output-bmk-results.py(177):         return (100 - result > threshold)
+output-bmk-results.py(217):             continue
+output-bmk-results.py(208):     for index, row in out_df.iterrows():
+output-bmk-results.py(210):         threshold = get_threshold(sym_type, metric, mode, row["benchmark"], row["symbol"])
+ --- modulename: output-bmk-results, funcname: get_threshold
+output-bmk-results.py(98):     if metric == "sample":
+output-bmk-results.py(108):     if metric == "num_vect_loops" or metric == "num_sve_loops":
+output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mode)]
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 DEBUG: checking symbol.improvement : 447.dealII,[.] _ZNK9MappingQ1ILi3EE12compute_fillERK12TriaIteratorILi3E15DoFCellAccessorILi3EEEjN10QProjectorILi3EE17DataSetDescriptorERNS0_12InternalDataERSt6vectorI5PointILi3EESaISE_EE : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
@@ -1752,7 +1906,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 453.povray,[.] _ZN3povL23All_Plane_IntersectionsEPNS_13Object_StructEPNS_10Ray_StructEPNS_13istack_structE : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 450.soplex,[.] _ZN6soplex9SPxFastRT8maxDeltaERdS1_RNS_12UpdateVectorERNS_6VectorES5_ii : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -1767,7 +1921,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 453.povray,[.] _ZN3povL24All_Sphere_IntersectionsEPNS_13Object_StructEPNS_10Ray_StructEPNS_13istack_structE : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 453.povray,[.] _ZN3povL23All_Plane_IntersectionsEPNS_13Object_StructEPNS_10Ray_StructEPNS_13istack_structE : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -1797,6 +1951,21 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+DEBUG: checking symbol.improvement : 453.povray,[.] _ZN3povL24All_Sphere_IntersectionsEPNS_13Object_StructEPNS_10Ray_StructEPNS_13istack_structE : size=0% (threshold=10%)
+output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
+ --- modulename: output-bmk-results, funcname: is_entry_improvement
+output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
+output-bmk-results.py(177):         return (100 - result > threshold)
+output-bmk-results.py(217):             continue
+output-bmk-results.py(208):     for index, row in out_df.iterrows():
+output-bmk-results.py(210):         threshold = get_threshold(sym_type, metric, mode, row["benchmark"], row["symbol"])
+ --- modulename: output-bmk-results, funcname: get_threshold
+output-bmk-results.py(98):     if metric == "sample":
+output-bmk-results.py(108):     if metric == "num_vect_loops" or metric == "num_sve_loops":
+output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mode)]
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
+output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
+output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 DEBUG: checking symbol.improvement : 453.povray,[.] _ZN3pov17Check_And_EnqueueEPNS_21Priority_Queue_StructEPNS_16BBox_Tree_StructEPNS_19Bounding_Box_StructEPNS_14Rayinfo_StructE : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
@@ -1887,7 +2056,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 462.libquantum,[.] quantum_cnot : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 462.libquantum,[.] quantum_sigma_x : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -1902,7 +2071,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 462.libquantum,[.] quantum_sigma_x : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 462.libquantum,[.] quantum_cnot : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -2082,7 +2251,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 483.xalancbmk,[.] _ZN10xalanc_1_814XalanDOMString6equalsERKS0_S2_ : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 483.xalancbmk,[.] _ZNK10xalanc_1_85XPath11executeMoreEPNS_9XalanNodeEN9__gnu_cxx17__normal_iteratorIPKiSt6vectorIiSaIiEEEERNS_21XPathExecutionContextE : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -2097,7 +2266,7 @@ output-bmk-results.py(111):     return default_threshold[(change_kind,metric,mod
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
 output-bmk-results.py(213):              .format(sym_type, change_kind, row["benchmark"], row["symbol"], metric, 100-row["rel_" + metric], threshold))
 output-bmk-results.py(212):         print("DEBUG: checking {0}.{1} : {2},{3} : {4}={5}% (threshold={6}%)"\
-DEBUG: checking symbol.improvement : 483.xalancbmk,[.] _ZNK10xalanc_1_85XPath11executeMoreEPNS_9XalanNodeEN9__gnu_cxx17__normal_iteratorIPKiSt6vectorIiSaIiEEEERNS_21XPathExecutionContextE : size=0% (threshold=10%)
+DEBUG: checking symbol.improvement : 483.xalancbmk,[.] _ZN10xalanc_1_814VariablesStack9findEntryERKNS_10XalanQNameEbb : size=0% (threshold=10%)
 output-bmk-results.py(216):         if not is_entry_xxx[change_kind](metric, row["rel_" + metric], threshold):
  --- modulename: output-bmk-results, funcname: is_entry_improvement
 output-bmk-results.py(176):     if metric in metric_utils.higher_regress_metrics:
@@ -2109,7 +2278,6 @@ output-bmk-results.py(237):     f_out.close()
 output-bmk-results.py(29):     if not self.outf:
 output-bmk-results.py(31):     self.outf.close()
 output-bmk-results.py(32):     if os.stat(self.filename).st_size == 0:
-output-bmk-results.py(33):        os.remove(self.filename)
 output-bmk-results.py(260):     f_ebp.write("\n")
  --- modulename: output-bmk-results, funcname: write
 output-bmk-results.py(36):     if not self.predicate or not self.outf:
diff --git a/notify/symbol.improvement b/notify/symbol.improvement
new file mode 100644
index 0000000..1849fa5
--- /dev/null
+++ b/notify/symbol.improvement
@@ -0,0 +1 @@
+12,401.bzip2,[.] BZ2_decompress,401.bzip2:[.] BZ2_decompress reduced in size by 12%,401.bzip2:[.] BZ2_decompress reduced in size by 12% from 20424 to 17984 bytes