[PATCH v2] Fix the reporting of leaf change statistics.
Matthias Maennich via libabigail
libabigail@sourceware.org
Wed Jan 1 00:00:00 GMT 2020
On Mon, Feb 17, 2020 at 05:54:31PM +0000, Matthias Maennich wrote:
>Hi Giuliano!
>
>On Thu, Feb 13, 2020 at 04:19:26PM +0000, Android Kernel Team wrote:
>>Leaf changes (as reported with --leaf-changes-only) to variables were
>>miscounted as changes to functions.
>>
>> * src/abg-comparison.cc
>> (apply_filters_and_compute_diff_stats): Increment the correct
>> counter for leaf variable changes.
>> * tests/data/Makefile.am: Add test case files.
>> * tests/data/test-abidiff-exit/test-leaf0-*: New test case.
>> * tests/test-abidiff-exit.cc: Run new test case.
>>
>>Signed-off-by: Giuliano Procida <gprocida@google.com>
>>---
>>src/abg-comparison.cc | 4 ++--
>>tests/data/Makefile.am | 3 +++
>>.../data/test-abidiff-exit/test-leaf0-report.txt | 13 +++++++++++++
>>tests/data/test-abidiff-exit/test-leaf0-v0.o | Bin 0 -> 2784 bytes
>>tests/data/test-abidiff-exit/test-leaf0-v1.o | Bin 0 -> 2824 bytes
>
>I assume the sources of those are rather small, so can you add the
>source snippets of the test files to the commit message?
>
>>tests/test-abidiff-exit.cc | 9 +++++++++
>>6 files changed, 27 insertions(+), 2 deletions(-)
>>create mode 100644 tests/data/test-abidiff-exit/test-leaf0-report.txt
>>create mode 100644 tests/data/test-abidiff-exit/test-leaf0-v0.o
>>create mode 100644 tests/data/test-abidiff-exit/test-leaf0-v1.o
>>
>>diff --git a/src/abg-comparison.cc b/src/abg-comparison.cc
>>index af4cd5e7..61aea47f 100644
>>--- a/src/abg-comparison.cc
>>+++ b/src/abg-comparison.cc
>>@@ -9767,8 +9767,8 @@ corpus_diff::priv::apply_filters_and_compute_diff_stats(diff_stats& stat)
>> (stat.num_leaf_var_changes_filtered_out() + 1);
>> }
>> if ((*i)->has_local_changes())
>>- stat.num_leaf_func_changes
>>- (stat.num_leaf_func_changes() + 1);
>>+ stat.num_leaf_var_changes
>>+ (stat.num_leaf_var_changes() + 1);
>> }
>>
>> stat.num_func_syms_added(added_unrefed_fn_syms_.size());
>>diff --git a/tests/data/Makefile.am b/tests/data/Makefile.am
>>index 1f574d2e..2a928ef2 100644
>>--- a/tests/data/Makefile.am
>>+++ b/tests/data/Makefile.am
>>@@ -107,6 +107,9 @@ test-abidiff-exit/test-loc-v0.bi \
>>test-abidiff-exit/test-loc-v1.bi \
>>test-abidiff-exit/test-loc-with-locs-report.txt \
>>test-abidiff-exit/test-loc-without-locs-report.txt \
>>+test-abidiff-exit/test-leaf0-v0.o \
>>+test-abidiff-exit/test-leaf0-v1.o \
>>+test-abidiff-exit/test-leaf0.report.txt \
^^^
The file is named test-leaf0-report.txt
With that, `make distcheck` runs successfully as well.
Cheers,
Matthias
>>\
>>test-diff-dwarf/test0-v0.cc \
>>test-diff-dwarf/test0-v0.o \
>>diff --git a/tests/data/test-abidiff-exit/test-leaf0-report.txt b/tests/data/test-abidiff-exit/test-leaf0-report.txt
>>new file mode 100644
>>index 00000000..46f39976
>>--- /dev/null
>>+++ b/tests/data/test-abidiff-exit/test-leaf0-report.txt
>>@@ -0,0 +1,13 @@
>>+Leaf changes summary: 2 artifacts changed
>>+Changed leaf types summary: 0 leaf type changed
>>+Removed/Changed/Added functions summary: 0 Removed, 1 Changed, 0 Added function
>>+Removed/Changed/Added variables summary: 0 Removed, 1 Changed, 0 Added variable
>>+
>>+1 function with some sub-type change:
>>+
>>+ [C]'function int changed_fun()' has some sub-type changes:
>>+ return type changed:
>>+ type name changed from 'int' to 'long int'
>>+ type size changed from 32 to 64 (in bits)
>>+
>>+
>
>Should the variable change not be reported as well? Is this a different
>issue or are variables changes are not supposed to be reported in this
>mode?
>
>>diff --git a/tests/data/test-abidiff-exit/test-leaf0-v0.o b/tests/data/test-abidiff-exit/test-leaf0-v0.o
>>new file mode 100644
>>index 0000000000000000000000000000000000000000..a79511cc3c6db039aaa1f5528db19b55f75c7a2d
>>GIT binary patch
>>literal 2784
>>zcmbtVOKTHR6h4#N*w)yPHa=P^GJ;QZ#zabO)oL_a>jM$Ng@VLMa?=ExiA*Ln>Ouq&
>>z+__LhaOa=UwSUBoKftvMJ!j^g<o0HY3s1<s=X~dR&wb3xtG8BDmIX`}T!5j(D1g0V
>>z$d@9v1bLW&duwlhekJ+gcWB`>7U?FU5HTMC@sQE@6!DgoF5@&QSRuyJEkBH-SRzd@
>>z88C4G>oZu>*5PbEBeL_^X=`RYw+|q-03v%fb1t)-xt_l`mzo6-hai)=kP#DDnF|-^
>>zSnp@A=gF-!`|<J|Mn8%Fs3Jg1jX0qOMeHh$0*>NnS--PWI7U6hWm^!+Zs=@CuTgL|
>>z3vR6zIi&M287Dw-cY5c|v~@p~Lcdzm^Vem)(&+iX{H7O-M!9zTo?Sk1qPSo$yQkfv
>>zeY#LwF5u60+F{R^E9=siL92$M7u1?7{iQ``alv-hZKvUNpZT?B(D6IHuG90Mw*0yi
>>zdesftg;Qbs2~-{wi<}$l*aWy%E}ylJU6Iw6=g0Y+rQ=58XkELb5tse^E?E7XHNR_o
>>z;vKeOKjQQ!ikf&Sn|qTgPwiMmgBVU6>EMjPuRlpjk$5s`BL<z_;wDg&L_(cXQzt+>
>>z<eE~zxF!u*0h2>P<3zk-4T)I+QzIAw0<zoSyGHOCM5eWM@f|yW_>9({!NDCljl9ux
>>z$VNAwJFpYuF;7KsC4|$ej}wSA9Z$eG-noeO82C#i(`6%~eb#r4<53*_e7Izc#PPX~
>>zb1@vn(Ytz6;cTx&xNG3AmHa@>W<9)iiW5CU$}YE!{o*#bt#DeazFu&<&)RqgaSnrc
>>zZZa3jeh6;e3q5eF-7dI+-0<AGtoGI`Ex*x0UBr9UY9KeasJ(z*ldUcrEnl(=rU19r
>>zX}2Z5vaa6=rHk*1(+$1aV--j?D~-TwOI7%PPV?wbz2$y#xRPQ9=TymLtJ6Ug2s3Xo
>>z@3U@8yy;I*G@U2Y{}ve|@%@RVQMrf;GbY)xf14fEY}%egoN|<uo~N8j9DgYR=v)!R
>>zXP@uU3SwqVit&vGq}%Z{7){$ds4+1z9#>#t;2ATfE%iC6{#kV?*q`dBSJJG%j+hw}
>>zAvmMJ8zg9C+E!4Plz&a-=RQ$>dY8=nPY|P+Tqm*-y;6xjFra46&+`k!ll=Fp%Y5V?
>>z{;!q4VFu^7JO)E8=Vj!R@;A*(3Rq%8HS_uzeveF2{~jgukG?|e&wZtO%-j^aufV1%
>>mn7#w_4x9e8C+6Kl2!1H=6bTxc>!-L${(V*d7tNPg)Bi7zRJKt7
>>
>>literal 0
>>HcmV?d00001
>>
>>diff --git a/tests/data/test-abidiff-exit/test-leaf0-v1.o b/tests/data/test-abidiff-exit/test-leaf0-v1.o
>>new file mode 100644
>>index 0000000000000000000000000000000000000000..4a433b9572abb46799391619f21538b54fb3204c
>>GIT binary patch
>>literal 2824
>>zcmbtVOOF#r5UzH!FU(r|0KsYl8k84^%&<nQ)h??AgOh~~5kg2Df`qKG-F_(ccr_kx
>>zmVlH?K;(eL0SO@?gp@y#3nzX{?#Vd^Qq|LCx7))hWx1>B>w0zdV}AYahuxeIz$D-b
>>zj5J08ZcdNnk`_y_3`=nD=I?*~N&15i8g1ivL`$<eOe$+29rBi(BUy;)GO{_z6LNNn
>>zMC~c0&PxVN96-DQu_(SQEazomwXi6bW)@BY<kmqJF6X!NTlu}^t1G!v0P+jS=dYj@
>>z#th)<3eQJ{y=8JMoqBbB1*3n1e_avGaxTp|;TVX@F8(EajiV6%7Upr(eZ*x8NK`-Z
>>zJ1T6f`3J?IR@1&y$6+o_fa>w7qqnBSkGUNB)mmZHRP{<@5CO{vVGI|O?ml#SQPXQj
>>z3Ea4S&nsWJP+a%6f=fZsyR=r^TEm~`cauS+cA6?ual3|U7}r`mhnpMz#=7S>J--q5
>>zA4auS+>3gHzCVb5Zbx-L39B8|hsyn8k#l1HgN)K|l*^aBbKk0JJB-rY{^oh72)#gB
>>zmJ!4+cR?I3h}B~=voEkEpCL|vBHhqyg@vcN^8EBJ8l-UMNQdYM{PQtuQf8i1+LU3J
>>z$J`{&WRWx{)}51J9eQh0!i=SjL<w^v#gb&YVhc${3G)*e0TS@2W{`wq6X+5mi`Kfh
>>zmuC=PvidVT#1kizH=hpN#9<hA;3&qkyTO$ZPOJV3|3r%^1IC{l{e*+#z0^b(kBIi!
>>z-nH`@1s#XiL{MPil$Yz+HaNwxyGoaoi0e6>s&LD2#*GSh9sIG;?;Gyc$vvSs`q?se
>>z3!kvx_yq15oO)pQ0fPR+E?!mSNu2T)ONly6AgG5)2tl>qhags+FsQ5Qpjl~0jULWv
>>zIjmM=bucy^#`NNhyL8%-Vi!ySL9N&ADtvf?sF$b!pC7-UgtebcAl0fg;;^es;s2AZ
>>zqCfSP`^n)-Mi`!7qm!*oXHg-{z2Usiwk^r7KRx4g&RqXHWQ=5=!7ROqpc|9y*uP7P
>>zMXoJ<eCUf;vTR87%m>sn4s@;v;(GWV?I7mHL<ru~fOI=$!_l>+I$ey6^9EcSdZvPF
>>zOMT9&zho{2`&0e&db;)35p!cA1ltDuo)n8*+X~KQ<=;2?xlfdz-VHbZK4KJ;>qJr0
>>z6O-sI1BP>c=8q81^8egiW_}+z4Er(v&G_>?^S!ahfYHfby*B=?`5gmnj$F0G>u2~S
>>z3R(SomeD`@I<Y_ZmDb_rrr6I6*fI;H?*qNhu0QRGd-o86zYTbi6pOfT3{l)H|97VT
>>KKiJT4*Z)8NIl2)5
>>
>>literal 0
>>HcmV?d00001
>>
>>diff --git a/tests/test-abidiff-exit.cc b/tests/test-abidiff-exit.cc
>>index 4cef727e..bafa7400 100644
>>--- a/tests/test-abidiff-exit.cc
>>+++ b/tests/test-abidiff-exit.cc
>>@@ -111,6 +111,15 @@ InOutSpec in_out_specs[] =
>> "data/test-abidiff-exit/test-loc-without-locs-report.txt",
>> "output/test-abidiff-exit/test-loc-without-locs-report.txt"
>> },
>>+ {
>>+ "data/test-abidiff-exit/test-leaf0-v0.o",
>>+ "data/test-abidiff-exit/test-leaf0-v1.o",
>>+ "",
>>+ "--no-show-locs --leaf-changes-only",
>
>Should we also add a test case to the default mode (not leaf changes),
>to ensure consistency?
>
>Thanks!
>
>Cheers,
>Matthias
>
>>+ abigail::tools_utils::ABIDIFF_ABI_CHANGE,
>>+ "data/test-abidiff-exit/test-leaf0-report.txt",
>>+ "output/test-abidiff-exit/test-leaf0-report.txt"
>>+ },
>> {0, 0, 0 ,0, abigail::tools_utils::ABIDIFF_OK, 0, 0}
>>};
>>
>>--
>>2.25.0.265.gbab2e86ba0-goog
>>
>>
More information about the Libabigail
mailing list