]> xenbits.xensource.com Git - osstest.git/log
projects / osstest.git / log
? search:
summary | shortlog | log | commit | commitdiff | tree
first ⋅ prev ⋅ next
osstest.git
10 years agoallow files: Allow all libvirt test failures on other branches flight-35858 flight-35859 flight-35860 flight-35861 flight-35862 flight-35863 flight-35864 flight-35866 flight-35867 flight-35868 flight-35869 flight-35870 flight-35871 flight-35872 flight-35873 flight-35874 flight-35875 flight-35876 flight-35877 flight-35878 flight-35879 flight-35880 flight-35881 flight-35882 flight-35883 flight-35884 flight-35885 flight-35886 flight-35887 flight-35888 flight-35889 flight-35890 flight-35891 flight-35892 flight-35893 flight-35894 flight-35895 flight-35896 flight-35897 flight-35898 flight-35899 flight-35900 flight-35901 flight-35902 flight-35903 flight-35904 flight-35906 flight-35907 flight-35908 flight-35909 flight-35910 flight-35911 flight-35912 flight-35913 flight-35914 flight-35915 flight-35916 flight-35917 flight-35918 flight-35919 flight-35920 flight-35921 flight-35922 flight-35923 flight-35924 flight-35925 flight-35926 flight-35927 flight-35928 flight-35929 flight-35930 flight-35931 flight-35932 flight-35933 flight-35934 flight-35935 flight-35936 flight-35937 flight-35938 flight-35939 flight-35940 flight-35941 flight-35942 flight-35943 flight-35944 flight-35945 flight-35946 flight-35947 flight-35948 flight-35949 flight-35950 flight-35951 flight-35952 flight-35953 flight-35954 flight-35955 flight-35956 flight-35957 flight-35958 flight-35959 flight-35960 flight-35961
commit | commitdiff | tree
Ian Jackson [Tue, 3 Mar 2015 13:15:27 +0000 (13:15 +0000)]
allow files: Allow all libvirt test failures on other branches

We have a problem where virtd crashes, and this is causing problems
for other branches.  So ignore libvirt test failures.

Use the new ability to override allowances to cause libvirt failures
to still block libvirt pushes.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agoallow files: Support negation syntax
commit | commitdiff | tree
Ian Jackson [Tue, 3 Mar 2015 13:13:02 +0000 (13:13 +0000)]
allow files: Support negation syntax

We now support !<pattern>, which undoes the effect of a <pattern>
specified previously (perhaps in another file).

Note that the logic is a bit poor: !... only undoes the effect of an
identical (or equivalent) earlier pattern; it doesn't provide a
specific override.

Improving that would involve a more invasive change to
sg-report-flight which I don't want to do right now.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agoallow files: Pass branch-specific allow file last to sg-report-flight
commit | commitdiff | tree
Ian Jackson [Tue, 3 Mar 2015 13:12:02 +0000 (13:12 +0000)]
allow files: Pass branch-specific allow file last to sg-report-flight

We are going to make it possible for a branch-specific allow file to
override general allow information from allow.all.

No functional change yet.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agorump kernels: Send build mails to new list flight-35611 flight-35612 flight-35614 flight-35615 flight-35616 flight-35617 flight-35618 flight-35619 flight-35620 flight-35621 flight-35622 flight-35623 flight-35624 flight-35625 flight-35626 flight-35627 flight-35628 flight-35629 flight-35630 flight-35631 flight-35632 flight-35633 flight-35634 flight-35635 flight-35636 flight-35637 flight-35638 flight-35639 flight-35640 flight-35641 flight-35642 flight-35643 flight-35644 flight-35645 flight-35646 flight-35647 flight-35648 flight-35649 flight-35650 flight-35651 flight-35652 flight-35653 flight-35654 flight-35655 flight-35656 flight-35657 flight-35658 flight-35659 flight-35660 flight-35661 flight-35662 flight-35663 flight-35664 flight-35665 flight-35666 flight-35667 flight-35668 flight-35669 flight-35670 flight-35671 flight-35672 flight-35673 flight-35674 flight-35675 flight-35676 flight-35677 flight-35678 flight-35679 flight-35680 flight-35681 flight-35682 flight-35683 flight-35684 flight-35685 flight-35686 flight-35688 flight-35689 flight-35690 flight-35691 flight-35692 flight-35693 flight-35694 flight-35695 flight-35696 flight-35697 flight-35698 flight-35699 flight-35700 flight-35701 flight-35702 flight-35703 flight-35704 flight-35705 flight-35706 flight-35707 flight-35708 flight-35709 flight-35710 flight-35711 flight-35712 flight-35713 flight-35714 flight-35715 flight-35716 flight-35717 flight-35718 flight-35719 flight-35720 flight-35721 flight-35722 flight-35723 flight-35724 flight-35725 flight-35726 flight-35727 flight-35728 flight-35729 flight-35730 flight-35731 flight-35732 flight-35733 flight-35734 flight-35735 flight-35736 flight-35737 flight-35738 flight-35739 flight-35740 flight-35741 flight-35742 flight-35743 flight-35744 flight-35745 flight-35746 flight-35747 flight-35748 flight-35749 flight-35750 flight-35751 flight-35752 flight-35753 flight-35754 flight-35755 flight-35756 flight-35757 flight-35758 flight-35759 flight-35760 flight-35761 flight-35762 flight-35763 flight-35764 flight-35765 flight-35766 flight-35767 flight-35768 flight-35769 flight-35770 flight-35771 flight-35772 flight-35773 flight-35774 flight-35775 flight-35776 flight-35777 flight-35778 flight-35779 flight-35780 flight-35781 flight-35782 flight-35783 flight-35784 flight-35785 flight-35786 flight-35787 flight-35788 flight-35789 flight-35790 flight-35791 flight-35792 flight-35800 flight-35801 flight-35802 flight-35803 flight-35804 flight-35805 flight-35806 flight-35807 flight-35808 flight-35809 flight-35810 flight-35811 flight-35812 flight-35813 flight-35814 flight-35815 flight-35818 flight-35819 flight-35820 flight-35821 flight-35822 flight-35825 flight-35826 flight-35827 flight-35828 flight-35829 flight-35830 flight-35831 flight-35832 flight-35833 flight-35834 flight-35835 flight-35836 flight-35837 flight-35838 flight-35839 flight-35840 flight-35841 flight-35842 flight-35843 flight-35844 flight-35845 flight-35846 flight-35847 flight-35848 flight-35849 flight-35850 flight-35851 flight-35852 flight-35853 flight-35854 flight-35855 flight-35857
commit | commitdiff | tree
Ian Jackson [Thu, 26 Feb 2015 17:45:25 +0000 (17:45 +0000)]
rump kernels: Send build mails to new list

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Antti Kantee <pooka@rumpkernel.org>
---
v2: Bisections, too.

10 years agorump kernel tests: Fix testid of repeat xenstorels test flight-34913 flight-34914 flight-34918 flight-34919 flight-34920 flight-34921 flight-34922 flight-34923 flight-34924 flight-34925 flight-34926 flight-34927 flight-34928 flight-34930 flight-34931 flight-34932 flight-34933 flight-34934 flight-34937 flight-34938 flight-34939 flight-34940 flight-34941 flight-34942 flight-34943 flight-34944 flight-34945 flight-34946 flight-34947 flight-34948 flight-34949 flight-34950 flight-34951 flight-34952 flight-34953 flight-34954 flight-34955 flight-34956 flight-34957 flight-34958 flight-34959 flight-34960 flight-34961 flight-34962 flight-34963 flight-34964 flight-34965 flight-34966 flight-34967 flight-34968 flight-34969 flight-34970 flight-34971 flight-34972 flight-34973 flight-34974 flight-34975 flight-34976 flight-34977 flight-34978 flight-34979 flight-34980 flight-34981 flight-34982 flight-34983 flight-34984 flight-34985 flight-34986 flight-34987 flight-34988 flight-34989 flight-34990 flight-34991 flight-34992 flight-34993 flight-34994 flight-34995 flight-34996 flight-34997 flight-34998 flight-34999 flight-35000 flight-35001 flight-35002 flight-35003 flight-35004 flight-35005 flight-35006 flight-35007 flight-35008 flight-35009 flight-35010 flight-35011 flight-35012 flight-35013 flight-35014 flight-35015 flight-35016 flight-35017 flight-35018 flight-35019 flight-35020 flight-35021 flight-35022 flight-35023 flight-35024 flight-35025 flight-35026 flight-35027 flight-35028 flight-35029 flight-35030 flight-35031 flight-35032 flight-35033 flight-35034 flight-35035 flight-35036 flight-35037 flight-35038 flight-35039 flight-35040 flight-35041 flight-35042 flight-35043 flight-35044 flight-35045 flight-35046 flight-35047 flight-35048 flight-35049 flight-35050 flight-35051 flight-35052 flight-35053 flight-35054 flight-35055 flight-35056 flight-35057 flight-35058 flight-35059 flight-35060 flight-35061 flight-35062 flight-35063 flight-35064 flight-35065 flight-35066 flight-35067 flight-35068 flight-35069 flight-35070 flight-35071 flight-35072 flight-35073 flight-35074 flight-35075 flight-35076 flight-35077 flight-35078 flight-35079 flight-35080 flight-35081 flight-35082 flight-35083 flight-35084 flight-35085 flight-35086 flight-35087 flight-35088 flight-35089 flight-35090 flight-35091 flight-35092 flight-35093 flight-35094 flight-35095 flight-35096 flight-35097 flight-35098 flight-35099 flight-35100 flight-35101 flight-35102 flight-35103 flight-35104 flight-35105 flight-35106 flight-35107 flight-35108 flight-35109 flight-35110 flight-35111 flight-35112 flight-35113 flight-35114 flight-35115 flight-35116 flight-35117 flight-35118 flight-35119 flight-35120 flight-35121 flight-35122 flight-35123 flight-35124 flight-35125 flight-35126 flight-35127 flight-35128 flight-35129 flight-35130 flight-35131 flight-35132 flight-35133 flight-35134 flight-35135 flight-35136 flight-35137 flight-35138 flight-35139 flight-35140 flight-35141 flight-35142 flight-35143 flight-35144 flight-35145 flight-35146 flight-35147 flight-35148 flight-35149 flight-35150 flight-35151 flight-35152 flight-35153 flight-35154 flight-35155 flight-35156 flight-35157 flight-35158 flight-35159 flight-35160 flight-35161 flight-35162 flight-35163 flight-35164 flight-35165 flight-35166 flight-35167 flight-35168 flight-35169 flight-35170 flight-35171 flight-35172 flight-35173 flight-35174 flight-35175 flight-35176 flight-35177 flight-35178 flight-35179 flight-35180 flight-35181 flight-35182 flight-35183 flight-35184 flight-35185 flight-35186 flight-35187 flight-35188 flight-35189 flight-35190 flight-35191 flight-35192 flight-35193 flight-35194 flight-35195 flight-35196 flight-35197 flight-35198 flight-35199 flight-35200 flight-35201 flight-35202 flight-35203 flight-35204 flight-35205 flight-35206 flight-35207 flight-35208 flight-35209 flight-35210 flight-35211 flight-35212 flight-35213 flight-35214 flight-35215 flight-35216 flight-35217 flight-35218 flight-35219 flight-35220 flight-35221 flight-35222 flight-35223 flight-35224 flight-35225 flight-35226 flight-35227 flight-35228 flight-35229 flight-35230 flight-35231 flight-35232 flight-35233 flight-35234 flight-35235 flight-35236 flight-35237 flight-35238 flight-35239 flight-35240 flight-35241 flight-35242 flight-35243 flight-35244 flight-35245 flight-35246 flight-35247 flight-35248 flight-35249 flight-35250 flight-35251 flight-35252 flight-35253 flight-35254 flight-35256 flight-35257 flight-35258 flight-35259 flight-35260 flight-35261 flight-35262 flight-35263 flight-35264 flight-35265 flight-35266 flight-35267 flight-35268 flight-35269 flight-35270 flight-35271 flight-35272 flight-35275 flight-35276 flight-35277 flight-35278 flight-35279 flight-35280 flight-35281 flight-35282 flight-35283 flight-35284 flight-35285 flight-35286 flight-35287 flight-35288 flight-35289 flight-35290 flight-35291 flight-35292 flight-35293 flight-35294 flight-35295 flight-35296 flight-35297 flight-35298 flight-35299 flight-35300 flight-35301 flight-35302 flight-35303 flight-35304 flight-35305 flight-35306 flight-35307 flight-35308 flight-35309 flight-35310 flight-35311 flight-35312 flight-35313 flight-35314 flight-35315 flight-35316 flight-35317 flight-35318 flight-35319 flight-35320 flight-35321 flight-35322 flight-35323 flight-35324 flight-35325 flight-35326 flight-35327 flight-35328 flight-35329 flight-35330 flight-35331 flight-35332 flight-35333 flight-35334 flight-35335 flight-35336 flight-35337 flight-35338 flight-35339 flight-35340 flight-35341 flight-35342 flight-35343 flight-35344 flight-35345 flight-35346 flight-35348 flight-35349 flight-35350 flight-35351 flight-35352 flight-35353 flight-35354 flight-35355 flight-35356 flight-35357 flight-35358 flight-35359 flight-35360 flight-35361 flight-35362 flight-35363 flight-35364 flight-35365 flight-35366 flight-35367 flight-35368 flight-35369 flight-35370 flight-35371 flight-35372 flight-35373 flight-35374 flight-35375 flight-35376 flight-35377 flight-35378 flight-35379 flight-35380 flight-35381 flight-35382 flight-35383 flight-35384 flight-35385 flight-35386 flight-35387 flight-35388 flight-35389 flight-35390 flight-35391 flight-35392 flight-35393 flight-35394 flight-35395 flight-35396 flight-35397 flight-35398 flight-35399 flight-35400 flight-35401 flight-35402 flight-35403 flight-35404 flight-35405 flight-35406 flight-35407 flight-35408 flight-35409 flight-35410 flight-35411 flight-35412 flight-35413 flight-35414 flight-35415 flight-35416 flight-35417 flight-35418 flight-35419 flight-35420 flight-35421 flight-35422 flight-35423 flight-35424 flight-35425 flight-35426 flight-35427 flight-35428 flight-35429 flight-35430 flight-35431 flight-35432 flight-35433 flight-35434 flight-35435 flight-35436 flight-35437 flight-35438 flight-35439 flight-35440 flight-35441 flight-35442 flight-35443 flight-35444 flight-35445 flight-35446 flight-35447 flight-35449 flight-35450 flight-35451 flight-35452 flight-35453 flight-35454 flight-35455 flight-35460 flight-35461 flight-35467 flight-35470 flight-35471 flight-35472 flight-35473 flight-35474 flight-35475 flight-35476 flight-35477 flight-35478 flight-35479 flight-35480 flight-35481 flight-35482 flight-35483 flight-35484 flight-35485 flight-35486 flight-35487 flight-35488 flight-35489 flight-35490 flight-35491 flight-35492 flight-35493 flight-35494 flight-35495 flight-35496 flight-35497 flight-35498 flight-35499 flight-35500 flight-35501 flight-35502 flight-35503 flight-35504 flight-35505 flight-35506 flight-35507 flight-35508 flight-35509 flight-35510 flight-35511 flight-35512 flight-35513 flight-35514 flight-35515 flight-35516 flight-35517 flight-35518 flight-35519 flight-35520 flight-35521 flight-35522 flight-35523 flight-35524 flight-35525 flight-35526 flight-35527 flight-35528 flight-35529 flight-35530 flight-35531 flight-35532 flight-35533 flight-35534 flight-35535 flight-35536 flight-35537 flight-35538 flight-35539 flight-35541 flight-35542 flight-35543 flight-35544 flight-35545 flight-35548 flight-35549 flight-35550 flight-35551 flight-35552 flight-35553 flight-35554 flight-35555 flight-35556 flight-35557 flight-35558 flight-35559 flight-35560 flight-35561 flight-35562 flight-35563 flight-35564 flight-35565 flight-35566 flight-35567 flight-35568 flight-35569 flight-35570 flight-35571 flight-35572 flight-35573 flight-35574 flight-35575 flight-35576 flight-35577 flight-35578 flight-35579 flight-35580 flight-35581 flight-35582 flight-35583 flight-35584 flight-35585 flight-35586 flight-35587 flight-35588 flight-35589 flight-35590 flight-35591 flight-35592 flight-35593 flight-35594 flight-35595 flight-35596 flight-35597 flight-35598 flight-35599 flight-35600 flight-35601 flight-35602 flight-35603 flight-35604 flight-35605 flight-35606 flight-35607 flight-35608 flight-35609 flight-35610
commit | commitdiff | tree
Ian Jackson [Thu, 19 Feb 2015 11:57:08 +0000 (11:57 +0000)]
rump kernel tests: Fix testid of repeat xenstorels test

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
10 years agoNTP servers: Work around Debian's failure to honour preseed flight-34695 flight-34887
commit | commitdiff | tree
Ian Jackson [Tue, 17 Feb 2015 17:35:57 +0000 (17:35 +0000)]
NTP servers: Work around Debian's failure to honour preseed

Setting clock-setup/ntp-server is not sufficient: it only takes effect
in the installer (!)

I have reported this as Debian #778564.  In the meantime we should
work around it for current releases (including jessie, which is
frozen).

For later releases, the new ntp.conf editing code arranges to bomb out
if we have an NTP server configured and find it hasn't been honoured
during the install.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agoNTP servers: Use NtpServer host property
commit | commitdiff | tree
Ian Jackson [Tue, 17 Feb 2015 17:44:01 +0000 (17:44 +0000)]
NTP servers: Use NtpServer host property

Move the specification of NTP servers in the production-config from
the ad-hoc entry in DebianPreseed to a dedicated NtpServer host
property.

This results in no overall functional change, except that the order of
elements in the preseed file is slightly different.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
diff --git a/Osstest/Debian.pm b/Osstest/Debian.pm
index a044f8f..cc075a4 100644
--- a/Osstest/Debian.pm
+++ b/Osstest/Debian.pm
@@ -506,7 +506,7 @@ sub di_installcmdline_core ($$;@) {
 sub preseed_base ($$$;@) {
     my ($ho,$suite,$extra_packages,%xopts) = @_;

-    return <<"END";
+    my $preseed = <<"END";
 d-i mirror/suite string $suite

 d-i debian-installer/locale string en_GB
@@ -572,10 +572,22 @@ d-i pkgsel/include string openssh-server, ntp, ntpdate, ethtool, chiark-utils-bi

 $xopts{ExtraPreseed}

-### END OF DEBIAN PRESEED BASE
+END
+
+    my $ntpserver = get_target_property($ho,'NtpServer');
+use Data::Dumper;
+print STDERR "PRESEED NTP ", Dumper($ntpserver);
+    $preseed .= <<"END" if $ntpserver;
+d-i clock-setup/ntp-server string $ntpserver
+END

+    $preseed .= <<"END";
+
+### END OF DEBIAN PRESEED BASE
 END
-}
+
+    return $preseed;
+}

 sub preseed_create ($$;@) {
     my ($ho, $sfx, %xopts) = @_;
diff --git a/README b/README
index 6e63e97..0a0242c 100644
--- a/README
+++ b/README
@@ -334,6 +334,11 @@ HostProp_<testbox>_TftpScope
    Defines the Tftp scope (i.e. subnet) where this host resides. See
    "TftpFoo_<scope> and TftpFoo" below.

+HostProp_<testbox>_NtpServer
+   NTP server to use.  You should probably have your own local
+   NTP server for production use; the default is to use the operating
+   system's default (normally, Debian's pool.ntp.org servers).
+
 HostFlags_<testbox>
    Defines a set of flags for the host. Flags is a list separated by
    whitespace, comma or semi-colon. A flag can be unset by prepending
@@ -357,9 +362,7 @@ HostGroupFlags_<group>
    merged with the host specific flags. Only used in standalone mode.

 DebianPreseed
-   Text to add to the debian-installer preseed file.  Optional
-   but you will need to set some NTP servers here if your firewall
-   doesn't permit NTP to Debian's pool.ntp.org servers.
+   Text to add to the debian-installer preseed file.  Optional.

 ========================================

diff --git a/production-config b/production-config
index 515bd98..5fbca50 100644
--- a/production-config
+++ b/production-config
@@ -77,8 +77,9 @@ XenUseUser osstest
 #DebianMirrorHost debian.uk.xensource.com
 DebianMirrorHost 10.80.16.196

+HostProp_NtpServer ntp.uk.xensource.com
+
 DebianPreseed= <<'END'
-d-i clock-setup/ntp-server string ntp.uk.xensource.com
 END

 HostProp_Serial sympathy woking

10 years agoNTP servers: Introduce new NtpServer host property
commit | commitdiff | tree
Ian Jackson [Tue, 17 Feb 2015 17:42:42 +0000 (17:42 +0000)]
NTP servers: Introduce new NtpServer host property

Allow the specification of NTP servers in a dedicated NtpServer host
property, which we honour in preseed_create.

Change the documentation to recommend this, rather than use of
ad-hoc text in DebianPreseed.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
diff --git a/Osstest/Debian.pm b/Osstest/Debian.pm
index a044f8f..cc075a4 100644
--- a/Osstest/Debian.pm
+++ b/Osstest/Debian.pm
@@ -506,7 +506,7 @@ sub di_installcmdline_core ($$;@) {
 sub preseed_base ($$$;@) {
     my ($ho,$suite,$extra_packages,%xopts) = @_;

-    return <<"END";
+    my $preseed = <<"END";
 d-i mirror/suite string $suite

 d-i debian-installer/locale string en_GB
@@ -572,10 +572,22 @@ d-i pkgsel/include string openssh-server, ntp, ntpdate, ethtool, chiark-utils-bi

 $xopts{ExtraPreseed}

-### END OF DEBIAN PRESEED BASE
+END
+
+    my $ntpserver = get_target_property($ho,'NtpServer');
+use Data::Dumper;
+print STDERR "PRESEED NTP ", Dumper($ntpserver);
+    $preseed .= <<"END" if $ntpserver;
+d-i clock-setup/ntp-server string $ntpserver
+END

+    $preseed .= <<"END";
+
+### END OF DEBIAN PRESEED BASE
 END
-}
+
+    return $preseed;
+}

 sub preseed_create ($$;@) {
     my ($ho, $sfx, %xopts) = @_;
diff --git a/README b/README
index 6e63e97..0a0242c 100644
--- a/README
+++ b/README
@@ -334,6 +334,11 @@ HostProp_<testbox>_TftpScope
    Defines the Tftp scope (i.e. subnet) where this host resides. See
    "TftpFoo_<scope> and TftpFoo" below.

+HostProp_<testbox>_NtpServer
+   NTP server to use.  You should probably have your own local
+   NTP server for production use; the default is to use the operating
+   system's default (normally, Debian's pool.ntp.org servers).
+
 HostFlags_<testbox>
    Defines a set of flags for the host. Flags is a list separated by
    whitespace, comma or semi-colon. A flag can be unset by prepending
@@ -357,9 +362,7 @@ HostGroupFlags_<group>
    merged with the host specific flags. Only used in standalone mode.

 DebianPreseed
-   Text to add to the debian-installer preseed file.  Optional
-   but you will need to set some NTP servers here if your firewall
-   doesn't permit NTP to Debian's pool.ntp.org servers.
+   Text to add to the debian-installer preseed file.  Optional.

 ========================================

diff --git a/production-config b/production-config
index 515bd98..5fbca50 100644
--- a/production-config
+++ b/production-config
@@ -77,8 +77,9 @@ XenUseUser osstest
 #DebianMirrorHost debian.uk.xensource.com
 DebianMirrorHost 10.80.16.196

+HostProp_NtpServer ntp.uk.xensource.com
+
 DebianPreseed= <<'END'
-d-i clock-setup/ntp-server string ntp.uk.xensource.com
 END

 HostProp_Serial sympathy woking

10 years agoDebian.pm: Slightly refactor preseed_base
commit | commitdiff | tree
Ian Jackson [Tue, 17 Feb 2015 17:40:47 +0000 (17:40 +0000)]
Debian.pm: Slightly refactor preseed_base

Arrange for preseed_base to accumulate its result in a variable.  This
is going to make it easier to add entries which are determined
programatically.

No functional change.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agoDebian.pm: Pass $ho/$gho to preseed_base
commit | commitdiff | tree
Ian Jackson [Tue, 17 Feb 2015 17:20:53 +0000 (17:20 +0000)]
Debian.pm: Pass $ho/$gho to preseed_base

This is going to be useful soon.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agoTestSupport: Provide get_target_property
commit | commitdiff | tree
Ian Jackson [Tue, 17 Feb 2015 17:21:47 +0000 (17:21 +0000)]
TestSupport: Provide get_target_property

This looks in the supplied $ho, but if that's a $gho (ie it has a
$gho->{Host}) it also looks in its host.

This is going to be useful for a fair variety of host-specific or
infrastructure-determined properties.

It seems to me that whether a property ought to be looked up in the
host if not found in a particular guest depends mostly on the
property, and not on the way the guest is configured.  The easiest way
to represent that in the osstest codebase is probably to call
get_target_property instead of get_host_property in the appropriate
places.

The use of recursion will make this look through a series of nested
hosts if we have nested virtualisation going on.  Indeed, nested
virtualisation may benefit from replacement of get_host_property by
get_target_property in a number of cases.  (At the time of writing
there is no nested virt in osstest mainline, but it's on the way.)

Currently there are no callers of get_target_property.  One will
appear shortly.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Robert Hu <robert.hu@intel.com>
CC: LongtaoX Pang <longtaox.pang@intel.com>
10 years agoconfig: Tiny fixes
commit | commitdiff | tree
Ian Jackson [Tue, 17 Feb 2015 17:27:38 +0000 (17:27 +0000)]
config: Tiny fixes

* Default $c{DebianPreseed} to '' (previously, if it wasn't specified,
  there would be undefined variable warnings).

* Cope with empty <<END-notated config items in the configuration file
  parser.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agosg-report-job-history: Increase default history depth to 500
commit | commitdiff | tree
Ian Jackson [Tue, 17 Feb 2015 17:35:05 +0000 (17:35 +0000)]
sg-report-job-history: Increase default history depth to 500

We are putting through a lot more jobs and branches.  Older history is
still interesting.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agocs-adjust-flight: Add missing `use Data::Dumper'
commit | commitdiff | tree
Ian Jackson [Tue, 17 Feb 2015 17:34:25 +0000 (17:34 +0000)]
cs-adjust-flight: Add missing `use Data::Dumper'

Dumper is used in some of the error reporting so we need to `use' it.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agorump kernel tests: Repeat the xenstorels test 50 times flight-34540 flight-34590 flight-34652 flight-34653 flight-34654 flight-34655 flight-34656 flight-34657 flight-34658 flight-34659 flight-34660 flight-34661 flight-34662 flight-34663 flight-34664 flight-34665 flight-34666 flight-34667 flight-34668 flight-34669 flight-34670 flight-34671 flight-34672 flight-34673 flight-34674 flight-34675 flight-34676 flight-34677 flight-34678 flight-34679 flight-34680 flight-34682 flight-34683 flight-34684 flight-34686 flight-34687 flight-34688 flight-34689 flight-34690 flight-34691 flight-34692 flight-34693 flight-34694 flight-34696 flight-34697 flight-34698 flight-34699 flight-34700 flight-34701 flight-34702 flight-34703 flight-34704 flight-34705 flight-34706 flight-34707 flight-34708 flight-34709 flight-34710 flight-34711 flight-34712 flight-34713 flight-34714 flight-34715 flight-34716 flight-34717 flight-34718 flight-34719 flight-34720 flight-34721 flight-34722 flight-34723 flight-34724 flight-34725 flight-34726 flight-34727 flight-34728 flight-34729 flight-34730 flight-34731 flight-34732 flight-34733 flight-34734 flight-34735 flight-34736 flight-34737 flight-34738 flight-34739 flight-34740 flight-34741 flight-34742 flight-34743 flight-34744 flight-34745 flight-34746 flight-34747 flight-34748 flight-34749 flight-34750 flight-34751 flight-34752 flight-34753 flight-34754 flight-34755 flight-34756 flight-34757 flight-34758 flight-34759 flight-34760 flight-34762 flight-34763 flight-34764 flight-34765 flight-34766 flight-34767 flight-34768 flight-34769 flight-34770 flight-34771 flight-34772 flight-34773 flight-34774 flight-34775 flight-34776 flight-34777 flight-34778 flight-34779 flight-34780 flight-34781 flight-34782 flight-34783 flight-34784 flight-34785 flight-34786 flight-34787 flight-34788 flight-34789 flight-34790 flight-34791 flight-34792 flight-34793 flight-34794 flight-34795 flight-34796 flight-34797 flight-34798 flight-34799 flight-34800 flight-34801 flight-34802 flight-34803 flight-34804 flight-34805 flight-34806 flight-34807 flight-34808 flight-34809 flight-34810 flight-34811 flight-34812 flight-34813 flight-34814 flight-34815 flight-34816 flight-34817 flight-34818 flight-34819 flight-34820 flight-34821 flight-34822 flight-34823 flight-34824 flight-34825 flight-34826 flight-34827 flight-34828 flight-34829 flight-34830 flight-34831 flight-34832 flight-34833 flight-34834 flight-34835 flight-34836 flight-34837 flight-34838 flight-34839 flight-34840 flight-34841 flight-34842 flight-34843 flight-34844 flight-34845 flight-34846 flight-34848 flight-34849 flight-34850 flight-34851 flight-34852 flight-34853 flight-34854 flight-34857 flight-34858 flight-34859 flight-34860 flight-34862 flight-34863 flight-34864 flight-34865 flight-34866 flight-34867 flight-34868 flight-34869 flight-34870 flight-34871 flight-34872 flight-34873 flight-34874 flight-34875 flight-34876 flight-34877 flight-34878 flight-34879 flight-34880 flight-34881 flight-34882 flight-34883 flight-34884 flight-34885 flight-34886 flight-34888 flight-34889 flight-34890 flight-34891 flight-34892 flight-34893 flight-34894 flight-34895 flight-34896 flight-34897 flight-34898 flight-34899 flight-34900 flight-34901 flight-34902 flight-34903 flight-34904 flight-34905 flight-34906 flight-34907 flight-34908 flight-34909 flight-34910 flight-34911
commit | commitdiff | tree
Ian Jackson [Fri, 6 Feb 2015 17:09:40 +0000 (17:09 +0000)]
rump kernel tests: Repeat the xenstorels test 50 times

Add a new step which uses repeat-ts to run
ts-rumpuserxen-demo-xenstorels many times.

We have to run ts-guest-destroy-hard after each time, to destroy the
guest which the demo script leaves lying about.

Strategically placed `+'s in the repeat-ts command line arrange that
the testid ends up being
   rumpuserxen-demo-xenstorels/xenstorels.repeat

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
---
v2: Run the test after, rather than before, the explicit
     ts-guest-destroy-hard.  That will avoid blocking the single
     destroy test if the repeat fails.

    No longer specify to tolerate failures of the post-run-demo
     destroy, as if the test passes so must the destroy.  Now by-hand
     testing may need a different ts-repeat-test rune, but in practice
     by-hand testing will probably involve a shell loop or something
     anyway.

10 years agorump kernel tests: Cancel unneeded edits of guest config
commit | commitdiff | tree
Ian Jackson [Fri, 6 Feb 2015 17:08:31 +0000 (17:08 +0000)]
rump kernel tests: Cancel unneeded edits of guest config

If the guest config is already set up to preserve, cancel the edit.
We are going to repeat this test, and this avoids creating many
identical copies of the same file in the log output.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbelL@citrix.com>
10 years agosg-run-job, etc.: Infrastructure for test script repetition
commit | commitdiff | tree
Ian Jackson [Fri, 6 Feb 2015 17:07:16 +0000 (17:07 +0000)]
sg-run-job, etc.: Infrastructure for test script repetition

Provide:
 * ts-repeat-test, a script to run multiple other test scripts in a loop
 * repeat-ts, a proc in sg-run-job which invokes it

No callers yet.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agosg-run-job: testid generation: Process ts more like rest of args
commit | commitdiff | tree
Ian Jackson [Fri, 6 Feb 2015 15:36:34 +0000 (15:36 +0000)]
sg-run-job: testid generation: Process ts more like rest of args

Remove ts as a separate parameter to spawn-ts.  The test script now
becomes the first entry in args.

We process it through the arg loop as before.  Currently there are no
calls where the first arg is `+' so the test script name ends up in
both real_args and testid_args.

We split it out of real_args into the ts variable with lshift.
We split it out of testid_args into the deftestid with lshift.

So afterwards in spawn-ts, all the variables (including real_args, ts,
deftestid and testid_args and hence host_testid_suffix) have the
values they would have had before.

Therefore there is no functional change for any existing calls.

However, because the first argument is not treated specially for the
`+' procesing loop, it is now possible to specify `+' as the first
entry in args to spawn-ts (ie where ts used to be) to arrange that the
deftestid (and hence, probably, the testid) is computed using later
arguments.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agosg-run-job: testid generation: Introduce $testid_args
commit | commitdiff | tree
Ian Jackson [Fri, 6 Feb 2015 15:30:38 +0000 (15:30 +0000)]
sg-run-job: testid generation: Introduce $testid_args

Rather than accumulating into host_testid_suffix directly, accumulate
into a list testid_args first.

No functional change: all we do is defer the construction of
host_testid_suffix, which is not used until later in this function.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agosg-run-job: testid generation: Move spawn-step-begin
commit | commitdiff | tree
Ian Jackson [Fri, 6 Feb 2015 15:24:31 +0000 (15:24 +0000)]
sg-run-job: testid generation: Move spawn-step-begin

Run spawn-step-begin until after the loop over $args, and after the
computation of the basic deftestid.

No functional change: nothing in that loop looks at stepno.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agotcl/osstestlib.tcl: Provide lshift
commit | commitdiff | tree
Ian Jackson [Fri, 6 Feb 2015 15:36:13 +0000 (15:36 +0000)]
tcl/osstestlib.tcl: Provide lshift

No callers yet.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoTestSupport: Make next_unique_name count in decimal, not unary
commit | commitdiff | tree
Ian Jackson [Fri, 6 Feb 2015 16:59:26 +0000 (16:59 +0000)]
TestSupport: Make next_unique_name count in decimal, not unary

Ie, we add `+<counter>' rather than an ever-longer series of `+'s.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoTestSupport: Introduce next_unique_name
commit | commitdiff | tree
Ian Jackson [Fri, 6 Feb 2015 16:55:52 +0000 (16:55 +0000)]
TestSupport: Introduce next_unique_name

No functional change.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoTestSupport: Provide target_editfile_cancel
commit | commitdiff | tree
Ian Jackson [Fri, 6 Feb 2015 13:57:31 +0000 (13:57 +0000)]
TestSupport: Provide target_editfile_cancel

This allows a caller of target_editfile to cancel the edit from within
their supplied code block.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agosg-report-job-history: Show which host test ran on
commit | commitdiff | tree
Ian Jackson [Mon, 9 Feb 2015 17:55:05 +0000 (17:55 +0000)]
sg-report-job-history: Show which host test ran on

Add a column listing the host(s) used.  We first find the relevant set
of host runvars (with a SELECT DISTINCT) and then look up each var for
each actual flight.

We do the pattern-matching on runvar names in perl to avoid giving the
postgresql optimiser a chance to turn this query into a full table
scan of the runvars table.  (A previous iteration of this patch
searched, in the per-flight loop, for all runvars whose name was
appropriately LIKE, with appalling performance.)

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agosg-report-job-history: Separate out $cond in queries
commit | commitdiff | tree
Ian Jackson [Mon, 9 Feb 2015 17:54:19 +0000 (17:54 +0000)]
sg-report-job-history: Separate out $cond in queries

We are going to want to reuse this, so separate out the computation of
$cond and @params.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoallow.all: Do not regard libvirt guest start failures as regressions flight-34395 flight-34472 flight-34473 flight-34475 flight-34477 flight-34478 flight-34480 flight-34481 flight-34482 flight-34483 flight-34484 flight-34485 flight-34486 flight-34487 flight-34488 flight-34489 flight-34490 flight-34491 flight-34492 flight-34493 flight-34494 flight-34495 flight-34496 flight-34497 flight-34498 flight-34499 flight-34500 flight-34501 flight-34502 flight-34503 flight-34504 flight-34505 flight-34506 flight-34507 flight-34508 flight-34509 flight-34510 flight-34511 flight-34512 flight-34513 flight-34514 flight-34515 flight-34516 flight-34517 flight-34519 flight-34520 flight-34521 flight-34522 flight-34523 flight-34524 flight-34525 flight-34526 flight-34527 flight-34528 flight-34529 flight-34530 flight-34531 flight-34532 flight-34533 flight-34534 flight-34535 flight-34536 flight-34537 flight-34538 flight-34539 flight-34541 flight-34542 flight-34543 flight-34544 flight-34545 flight-34546 flight-34547 flight-34548 flight-34549 flight-34550 flight-34551 flight-34552 flight-34553 flight-34554 flight-34555 flight-34556 flight-34557 flight-34558 flight-34559 flight-34560 flight-34561 flight-34562 flight-34563 flight-34564 flight-34565 flight-34566 flight-34567 flight-34568 flight-34569 flight-34570 flight-34571 flight-34572 flight-34573 flight-34574 flight-34575 flight-34576 flight-34577 flight-34578 flight-34579 flight-34580 flight-34581 flight-34582 flight-34583 flight-34584 flight-34585 flight-34586 flight-34587 flight-34588 flight-34589 flight-34591 flight-34592 flight-34593 flight-34594 flight-34595 flight-34596 flight-34597 flight-34598 flight-34599 flight-34600 flight-34601 flight-34602 flight-34603 flight-34604 flight-34605 flight-34606 flight-34607 flight-34608 flight-34609 flight-34610 flight-34611 flight-34612 flight-34613 flight-34614 flight-34615 flight-34616 flight-34617 flight-34618 flight-34619 flight-34620 flight-34621 flight-34622 flight-34623 flight-34624 flight-34625 flight-34626 flight-34627 flight-34628 flight-34629 flight-34630 flight-34631 flight-34632 flight-34633 flight-34634 flight-34635 flight-34636 flight-34637 flight-34638 flight-34639 flight-34640 flight-34641 flight-34642 flight-34643 flight-34644 flight-34645 flight-34646 flight-34649 flight-34650 flight-34651
commit | commitdiff | tree
Ian Jackson [Mon, 9 Feb 2015 14:22:50 +0000 (14:22 +0000)]
allow.all: Do not regard libvirt guest start failures as regressions

We have been experiencing some regressions in test-amd64-*-libvirt
guest-start.

These appear to be host-specific: for example, test-amd64-i386-libvirt
succeeded in 34278 on itch-mite but failed in 34234 on bush-cricket.
The only difference in osstest's records of the versions is a change
to linuxfirmware.

I booked out field-cricket and have repro'd some problems there.  It
appears that libvirtd can crash.  The symptoms are not the same as in
the push gate failures in osstest (34316, 34234) and linux-linus
(34299) but similar to those in linux-3.0 (34291) and linux-3.16
(34285).

When running libvirtd in the foreground we can get this:
  libvirtd: libxl_event.c:1791: libxl__ao_complete_check_progress_reports: Assertion `ao->in_initiator' failed.

So, it appears to be some kind of race.  The libvirt tests are new so
this is not a real regression in any of the trees gated by osstest.
So, permit this test for now.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agomg-debian-installer-update: workaround caching proxies flight-34174 flight-34234 flight-34316 flight-34387
commit | commitdiff | tree
Ian Campbell [Sat, 10 Jan 2015 16:14:15 +0000 (16:14 +0000)]
mg-debian-installer-update: workaround caching proxies

Which otherwise interfere with updating d-i shortly after the upstream
update.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoREADME.dev: Updates to commissioning process
commit | commitdiff | tree
Ian Campbell [Fri, 30 Jan 2015 10:55:03 +0000 (10:55 +0000)]
README.dev: Updates to commissioning process

- mkpxedir requires the resource to be allocated
- mention the need to bless the hosts for comissioning run, and to add
  proper blessings afterwards.
- run mg-execute-flight in screen

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-logs-capture: ensure captured log files are readable.
commit | commitdiff | tree
Ian Campbell [Wed, 4 Feb 2015 09:49:44 +0000 (09:49 +0000)]
ts-logs-capture: ensure captured log files are readable.

At least /var/log/libvirt/libvirtd.log is 0600 on the target machine
and this gets preserved in the captured logfile, eventually meaning
that apache on the log server cannot read the file to serve it ("You
don't have permission to access...").

Set all logfiles to 0666, modified by umask.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-rumpuserxen-build: Cope with rumprun-xen rename
commit | commitdiff | tree
Ian Jackson [Wed, 4 Feb 2015 14:39:12 +0000 (14:39 +0000)]
ts-rumpuserxen-build: Cope with rumprun-xen rename

rumpxen-app-* has been renamed to rumprun-xen-*.  We need to cope with
either, at least for the transition.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Martin Lucina <martin@lucina.net>
10 years agots-xen-build-prep: install nasm flight-34086 flight-34114 flight-34167 flight-34168 flight-34169 flight-34170 flight-34171 flight-34172 flight-34176 flight-34177 flight-34178 flight-34179 flight-34180 flight-34181 flight-34182 flight-34183 flight-34184 flight-34185 flight-34186 flight-34187 flight-34188 flight-34189 flight-34190 flight-34191 flight-34192 flight-34193 flight-34194 flight-34195 flight-34196 flight-34197 flight-34199 flight-34200 flight-34201 flight-34202 flight-34203 flight-34204 flight-34205 flight-34206 flight-34207 flight-34208 flight-34209 flight-34210 flight-34211 flight-34212 flight-34213 flight-34214 flight-34215 flight-34216 flight-34217 flight-34218 flight-34221 flight-34223 flight-34224 flight-34225 flight-34226 flight-34227 flight-34228 flight-34229 flight-34230 flight-34231 flight-34232 flight-34233 flight-34235 flight-34236 flight-34237 flight-34238 flight-34239 flight-34240 flight-34241 flight-34242 flight-34243 flight-34244 flight-34245 flight-34246 flight-34247 flight-34248 flight-34249 flight-34250 flight-34251 flight-34252 flight-34253 flight-34254 flight-34255 flight-34256 flight-34257 flight-34258 flight-34259 flight-34260 flight-34261 flight-34262 flight-34264 flight-34265 flight-34266 flight-34267 flight-34268 flight-34270 flight-34271 flight-34274 flight-34275 flight-34276 flight-34277 flight-34278 flight-34280 flight-34281 flight-34282 flight-34285 flight-34290 flight-34291 flight-34292 flight-34293 flight-34296 flight-34297 flight-34298 flight-34299 flight-34300 flight-34315 flight-34317 flight-34318 flight-34319 flight-34320 flight-34321 flight-34322 flight-34323 flight-34329 flight-34330 flight-34335 flight-34336 flight-34337 flight-34338 flight-34339 flight-34340 flight-34341 flight-34342 flight-34343 flight-34344 flight-34345 flight-34346 flight-34347 flight-34348 flight-34349 flight-34350 flight-34351 flight-34352 flight-34353 flight-34354 flight-34355 flight-34356 flight-34357 flight-34358 flight-34359 flight-34360 flight-34361 flight-34362 flight-34363 flight-34364 flight-34365 flight-34367 flight-34368 flight-34369 flight-34370 flight-34371 flight-34372 flight-34373 flight-34381 flight-34382 flight-34383 flight-34384 flight-34385 flight-34386 flight-34388 flight-34389 flight-34390 flight-34392 flight-34393 flight-34394 flight-34396 flight-34397 flight-34398 flight-34399 flight-34400 flight-34401 flight-34402 flight-34403 flight-34404 flight-34405 flight-34406 flight-34407 flight-34408 flight-34409 flight-34410 flight-34411 flight-34412 flight-34420 flight-34421 flight-34422 flight-34423 flight-34424 flight-34425 flight-34426 flight-34427 flight-34428 flight-34429 flight-34430 flight-34431 flight-34432 flight-34433 flight-34434 flight-34435 flight-34436 flight-34437 flight-34438 flight-34439 flight-34440 flight-34441 flight-34442 flight-34443 flight-34444 flight-34445 flight-34446 flight-34447 flight-34448 flight-34449 flight-34450 flight-34451 flight-34452 flight-34453 flight-34454 flight-34455 flight-34456 flight-34457 flight-34458 flight-34459 flight-34460 flight-34461 flight-34462 flight-34463 flight-34464 flight-34465 flight-34466 flight-34467 flight-34468 flight-34469 flight-34470 flight-34471
commit | commitdiff | tree
Wei Liu [Sat, 31 Jan 2015 22:51:50 +0000 (22:51 +0000)]
ts-xen-build-prep: install nasm

OVMF requires nasm to build.

Signed-off-by: Wei Liu <wei.liu2@citrix.com>
Cc: Ian Campbell <ian.campbell@citrix.com>
Cc: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoToolstack: Pass $gho to create method
commit | commitdiff | tree
Ian Campbell [Wed, 21 Jan 2015 16:12:21 +0000 (16:12 +0000)]
Toolstack: Pass $gho to create method

The cfg file can be obtained from $gho->{CfgPath}. This is more
consistent with other methods.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-guest-start: Use guest_create
commit | commitdiff | tree
Ian Campbell [Tue, 2 Dec 2014 12:53:26 +0000 (12:53 +0000)]
ts-guest-start: Use guest_create

This allows us to abolish CfgPathVar which was inconsistently used,
appears redundant with $gho->{CfgPath} and in any case never set to
anything other than 'cfgpath'.

This was intended to deal with toolstacks with a different cfg format
(such as very early versions of xl) but is no longer relevant.

I think if this need arises again in a future toolstack this
functionality could be reintroduced pretty trivially via the toolstack
abstraction which is added by this series.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoToolstack: Remove Command field for all toolstacks.
commit | commitdiff | tree
Ian Campbell [Mon, 1 Dec 2014 16:36:01 +0000 (16:36 +0000)]
Toolstack: Remove Command field for all toolstacks.

Nothing in generic code uses this now, so remove.

xl+xend retain as _Command for internal use only.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agolibvirt: Implement shutdown_wait
commit | commitdiff | tree
Ian Campbell [Tue, 2 Dec 2014 12:11:47 +0000 (12:11 +0000)]
libvirt: Implement shutdown_wait

virsh does not include a --wait option to shutdown as xl and xm do, so
we implement it by hand.

Needs new guest_await_destroy helper. Note the guest_await_shutdown
requires on_shutdown='preserve'

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agolibvirt: Implement initscript restart which has some hope of working.
commit | commitdiff | tree
Ian Campbell [Mon, 1 Dec 2014 14:36:33 +0000 (14:36 +0000)]
libvirt: Implement initscript restart which has some hope of working.

... by refactoring start/stop actions into the functions which are
expected by restart.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoToolstack: Refactor save/restore support
commit | commitdiff | tree
Ian Campbell [Mon, 1 Dec 2014 16:34:39 +0000 (16:34 +0000)]
Toolstack: Refactor save/restore support

Use $gho->{CfgPath} rather than constructing it again.

Still stubbed out for libvirt.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoToolstack: Refactor migration support.
commit | commitdiff | tree
Ian Campbell [Fri, 28 Nov 2014 17:35:09 +0000 (17:35 +0000)]
Toolstack: Refactor migration support.

Note that since the previous patch arranges for
ts-migration-support-check to continue to fail for libvirt the libvirt
code is not actually called yet (and will die if it is). This patch is
mainly included to reduce the number of users of
toolstack()->{Command} closer to zero.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoToolstack: Refactor migration support check.
commit | commitdiff | tree
Ian Campbell [Mon, 16 Jun 2014 16:41:38 +0000 (17:41 +0100)]
Toolstack: Refactor migration support check.

Not implemented for libvirt (the check itself that is, the hook is
present).

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoToolstack: Refactor shutdown support
commit | commitdiff | tree
Ian Campbell [Mon, 16 Jun 2014 16:48:57 +0000 (17:48 +0100)]
Toolstack: Refactor shutdown support

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoToolstack: Refactor consolecmd handling
commit | commitdiff | tree
Ian Campbell [Mon, 16 Jun 2014 16:46:32 +0000 (17:46 +0100)]
Toolstack: Refactor consolecmd handling

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoToolstack: Refactor guest lifecycle.
commit | commitdiff | tree
Ian Campbell [Mon, 16 Jun 2014 15:46:40 +0000 (16:46 +0100)]
Toolstack: Refactor guest lifecycle.

Implement destory/create as per toolstack methods, including
implementing the libvirt version which previously didn't work. To do
this we use the virsh capability to convert an xl/xm style config file
into the correct XML.

xend basically calls into the xl helper since they are compatible.

xl/xm, uses ->{Command} which will eventually become private.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoTestSupport: guest_create and guest_destroy take only a $gho.
commit | commitdiff | tree
Ian Campbell [Mon, 16 Jun 2014 16:30:48 +0000 (17:30 +0100)]
TestSupport: guest_create and guest_destroy take only a $gho.

The host can be looked up from $gho->{Host} and the toolstack can be
looked up from the host.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoTestSupport: always use xl for generic operations.
commit | commitdiff | tree
Ian Campbell [Mon, 16 Jun 2014 12:29:13 +0000 (13:29 +0100)]
TestSupport: always use xl for generic operations.

Unless the toolstack is xend (for compatibility with pre-xl Xen
versions), when we use xm.

For several operations in TestSupport.pm the actual toolstack isn't
really relevant, since we want info straight from Xen. For simplicity
just use xl (or xm) in these cases, to avoid needing to implement the
following specially for each toolstack:
  - host_get_free_memory
  - guest_get_state
  - guest_find_domid
  - listing assignable pci devices

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoToolstack: use get_host_method_object() to manage toolstack selection
commit | commitdiff | tree
Ian Campbell [Mon, 16 Jun 2014 08:18:35 +0000 (09:18 +0100)]
Toolstack: use get_host_method_object() to manage toolstack selection

This will allow us to more easily have per-toolstack methods etc.

The previous hash of toolstack parameters is now a blessed object. For
now the callers don't need to change but over the following patches we
will refactor things to use method calls. In particular we will be
aiming to remove Command from the hash and use method calls for
everything.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-rumpuserxen-demo-xenstorels: Use standard functions for things
commit | commitdiff | tree
Ian Campbell [Mon, 1 Dec 2014 16:26:34 +0000 (16:26 +0000)]
ts-rumpuserxen-demo-xenstorels: Use standard functions for things

Specifically guest_create and guest_find_domid.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoPass host to toolstack()
commit | commitdiff | tree
Ian Campbell [Mon, 16 Jun 2014 12:00:14 +0000 (13:00 +0100)]
Pass host to toolstack()

This will be needed in a future patch.

Everywhere already has a $ho in hand. Also cache the answer as
$ho->{Toolstack}.

I scanned the source with:
    find -name \*.pm -exec perl -c {} \;
    for i in ts-* ; do perl -c $i; done
which reported "Not enough arguments for Osstest::TestSupport::toolstack"
for each callsite which needed changing.

Also don't pass the toolstack command name directly to
host_get_free_memory().  Look it up instead.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-logs-capture: Collect some libvirt logs and capabilities
commit | commitdiff | tree
Ian Campbell [Mon, 16 Jun 2014 16:15:43 +0000 (17:15 +0100)]
ts-logs-capture: Collect some libvirt logs and capabilities

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoapt: lock osstest's usages of apt-get against each other
commit | commitdiff | tree
Ian Campbell [Tue, 15 Apr 2014 10:39:32 +0000 (11:39 +0100)]
apt: lock osstest's usages of apt-get against each other

Currently we rely on all apt-get invocations being in a single
ts-xen-build-prep job which can't run on a shared host.

That is a bit inflexible so instead use our own lock. We wait
indefinitely and rely on osstest's existing command timeout
infrastructure to catch problems.

target_install_packages*() previous estimated the time taken to
install the packages based on the number of packages. This no longer
applies because the install might get stuck behind some other large
install. Use a 3000s (nearly an hour) timeout instead (I expect
failures here to be unusual so erred on the big side)

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoTestSupport: Add helper to wait for a guest to shutdown
commit | commitdiff | tree
Ian Campbell [Fri, 7 Feb 2014 12:41:28 +0000 (12:41 +0000)]
TestSupport: Add helper to wait for a guest to shutdown

Refactor the guts of guest_await_reboot into a helper and use for both
shutdown and reboot handling.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-hosts-allocate-Executive: Fix an undefined warning
commit | commitdiff | tree
Ian Jackson [Thu, 29 Jan 2015 12:49:48 +0000 (12:49 +0000)]
ts-hosts-allocate-Executive: Fix an undefined warning

$esrow is autovivified as { } by
            $candrow->{EquivMostRecentStatus} = $esrow->{status}
so use %$esrow which in scalar context is true iff the
hash is nonempty.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agolinux-3.16: New branch
commit | commitdiff | tree
Ian Jackson [Mon, 26 Jan 2015 18:05:11 +0000 (18:05 +0000)]
linux-3.16: New branch

Test this too.  We fetch this from the Canonical Kernel Team (ckt)
repo, but do not encode `ckt' in the branch name, in case this becomes
a proper stable tree.

(I have set up this branch on the osstest with the mg-branch-setup
script in this version of this tree.)

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agomg-branch-setup: New script
commit | commitdiff | tree
Ian Jackson [Mon, 26 Jan 2015 18:05:10 +0000 (18:05 +0000)]
mg-branch-setup: New script

Also, update the README.dev accordingly

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoap-common: Have ap-push for linux trees use refs/heads/
commit | commitdiff | tree
Ian Jackson [Mon, 26 Jan 2015 18:05:09 +0000 (18:05 +0000)]
ap-common: Have ap-push for linux trees use refs/heads/

As per the docs in README.dev.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoDebian: Handle systems where u-boot includes 0x on $filesize flight-33915 flight-33957 flight-33958 flight-33959 flight-33960 flight-33962 flight-33969 flight-33974 flight-33982 flight-33986 flight-33987 flight-33988 flight-33991 flight-33992 flight-33993 flight-33994 flight-33995 flight-34001 flight-34002 flight-34003 flight-34004 flight-34005 flight-34006 flight-34007 flight-34008 flight-34009 flight-34011 flight-34018 flight-34020 flight-34021 flight-34022 flight-34023 flight-34024 flight-34025 flight-34026 flight-34027 flight-34028 flight-34029 flight-34030 flight-34031 flight-34032 flight-34033 flight-34034 flight-34035 flight-34036 flight-34037 flight-34038 flight-34039 flight-34040 flight-34041 flight-34042 flight-34043 flight-34044 flight-34045 flight-34046 flight-34047 flight-34048 flight-34049 flight-34050 flight-34051 flight-34052 flight-34053 flight-34054 flight-34056 flight-34057 flight-34058 flight-34060 flight-34061 flight-34063 flight-34064 flight-34066 flight-34068 flight-34069 flight-34070 flight-34071 flight-34072 flight-34073 flight-34074 flight-34075 flight-34076 flight-34077 flight-34082 flight-34083 flight-34085 flight-34087 flight-34088 flight-34089 flight-34090 flight-34091 flight-34092 flight-34093 flight-34094 flight-34095 flight-34096 flight-34097 flight-34098 flight-34099 flight-34100 flight-34101 flight-34102 flight-34103 flight-34104 flight-34105 flight-34106 flight-34107 flight-34108 flight-34109 flight-34110 flight-34111 flight-34112 flight-34113 flight-34115 flight-34116 flight-34117 flight-34118 flight-34119 flight-34120 flight-34121 flight-34122 flight-34123 flight-34124 flight-34125 flight-34126 flight-34127 flight-34128 flight-34129 flight-34130 flight-34131 flight-34132 flight-34133 flight-34134 flight-34135 flight-34136 flight-34137 flight-34138 flight-34139 flight-34140 flight-34141 flight-34142 flight-34143 flight-34146 flight-34147 flight-34148 flight-34150 flight-34151 flight-34153 flight-34154 flight-34156 flight-34157 flight-34158 flight-34159 flight-34160 flight-34161 flight-34162 flight-34163 flight-34164 flight-34165 flight-34166
commit | commitdiff | tree
Ian Campbell [Thu, 29 Jan 2015 11:37:39 +0000 (11:37 +0000)]
Debian: Handle systems where u-boot includes 0x on $filesize

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoREADME.dev: Document the usage of stop files flight-33818 flight-33841 flight-33887 flight-33888
commit | commitdiff | tree
Ian Campbell [Fri, 23 Jan 2015 15:34:51 +0000 (15:34 +0000)]
README.dev: Document the usage of stop files

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoREADME.dev: Document how to do a force push.
commit | commitdiff | tree
Ian Campbell [Tue, 6 Jan 2015 13:51:17 +0000 (13:51 +0000)]
README.dev: Document how to do a force push.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoAdd a flight to test OVMF master branch
commit | commitdiff | tree
Wei Liu [Mon, 12 Jan 2015 17:15:32 +0000 (17:15 +0000)]
Add a flight to test OVMF master branch

Do the usual stuffs for adding a new branch (ap-* cr-* etc).

Modify ts-xen-build so that it builds Xen with the specified ovmf tree
and revision.

Only build and test on x86 by modifying make-flight and mfi-common.

New branch is added to cr-daily-branch. It exports the tree and
changeset used for the test if set. Otherwise the values in Config.mk
are used.

Signed-off-by: Wei Liu <wei.liu2@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agomake-flight: reorganize scheduling related test jobs
commit | commitdiff | tree
Dario Faggioli [Thu, 8 Jan 2015 17:47:02 +0000 (18:47 +0100)]
make-flight: reorganize scheduling related test jobs

Scheduling related tests are ok to run on ARM, so do
not cut them off. They also do not depend on a
particular Dom0 architecture.

The net effect is that the following tests are removed:
 test-amd64-i386-xl-credit2
 test-amd64-i386-xl-multivcpu

while the following new ones are created:
 test-amd64-amd64-xl-credit2
 test-amd64-amd64-xl-multivcpu
 test-armhf-armhf-xl-credit2
 test-armhf-armhf-xl-multivcpu
 test-armhf-armhf-xl-sedf
 test-armhf-armhf-xl-sedf-pin

Signed-off-by: Dario Faggioli <dario.faggioli@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoSwitch ARM testing back to linux-arm-xen branch.
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 15:53:35 +0000 (15:53 +0000)]
Switch ARM testing back to linux-arm-xen branch.

The cubietruck and arndale boards need a newer kernel than our current
baseline of 3.14. Therefore switch all ARM testing to the
linux-arm-kernel branch which has been moved up to test v3.16.

This effectively reverts commit 38b2fe2020883 "Switch armhf to use
common default Linux."

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoDebian: create boot.scr.nat when installing Xen on u-boot.
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:05:07 +0000 (11:05 +0000)]
Debian: create boot.scr.nat when installing Xen on u-boot.

This boots the kernel we've build natively, where boot.xen boots Xen
and boot.deb boots the original Debian kernel.

Sometimes convenient when debugging issues.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoTestSupport: replace power_cycle_timeout with power_cycle_sleep
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:05:06 +0000 (11:05 +0000)]
TestSupport: replace power_cycle_timeout with power_cycle_sleep

Which does the actual sleep and logs that it is doing so.

Otherwise on a machine which has a longish PowerCycleTime I keep
wondering why osstest has hung...

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoDebian: Create boot.scr with a suffix and copy to boot.scr
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:05:05 +0000 (11:05 +0000)]
Debian: Create boot.scr with a suffix and copy to boot.scr

This ensures that we always have a boot script to boot at least the
native Debian kernel and Xen available, even after multiple iterations
of installation, which is handy when debugging a system.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-kernel-build: Adjust kernel .config to work on the arndale boards.
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:05:04 +0000 (11:05 +0000)]
ts-kernel-build: Adjust kernel .config to work on the arndale boards.

Various drivers are missing from multi_v7_defconfig in v3.16, also
some drivers which don't play nice are enabled by default, so remove
them.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoOsstest/Debian: Add 0x prefix to $filesize
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:05:03 +0000 (11:05 +0000)]
Osstest/Debian: Add 0x prefix to $filesize

$filesize is an unprefixed hex number, but fdt set requires the 0x to
interpret it properly. See
http://lists.denx.de/pipermail/u-boot/2014-October/193622.html,
http://lists.denx.de/pipermail/u-boot/2014-November/194150.html and
http://lists.denx.de/pipermail/u-boot/2014-November/194150.html.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoOsstest/Debian: Add "clk_ignore_unused" to default command line
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:05:02 +0000 (11:05 +0000)]
Osstest/Debian: Add "clk_ignore_unused" to default command line

dom0 is not aware that some clocks are actually in use (e.g. by the
hypervisor), so this stops the kernel from messing with (specifically,
disabling) those clocks. It's harmless even when not needed.

Really there ought to be some interface to communicate this from Xen
to dom0, or some other mechanism to gate things. See
http://bugs.xenproject.org/xen/bug/45

This bug is present in at least Wheezy and Jessie so apply this
workaround to those.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoOsstest/Debian: Remove hardcoded addresses from boot.scr
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:05:01 +0000 (11:05 +0000)]
Osstest/Debian: Remove hardcoded addresses from boot.scr

Those are specific to the midway platform.

This depends on a change to the hostdb to add "0x01000000" as the
value of a new UBootSetXenAddrR property of the midway machines.

Most platforms will need something similar. For both cubietruck and
arndale 0x41000000.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoOsstest/Debian: support adding a rootdelay property to bootargs
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:05:00 +0000 (11:05 +0000)]
Osstest/Debian: support adding a rootdelay property to bootargs

arndale appears to be quite slow (and asynchronous) at finding it's
scsi controller. rootdelay=3 seems to do the trick.

For the dom0 case this involved refactoring the existing provision of
the commandline to be less open-coded in the boot script here doc.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoOsstest/Debian.pm: Refactor common kernel cmdline setup.
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:04:59 +0000 (11:04 +0000)]
Osstest/Debian.pm: Refactor common kernel cmdline setup.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoOsstest/Debian: Refactor setting of dom0 kernel command line.
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:04:58 +0000 (11:04 +0000)]
Osstest/Debian: Refactor setting of dom0 kernel command line.

This will make it easier to make components optional in the future.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoOsstest/Debian: Support for loading an FDT from u-boot script
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:04:57 +0000 (11:04 +0000)]
Osstest/Debian: Support for loading an FDT from u-boot script

The currently supported platform provides an FDT preloaded at 0x1000.
Replace this with ${fdt_addr} (which the current platform exposes) and
for platforms which do not provide an fdt arrange to load the relevant
file as named in the ${fdtfile} (which is conventionally provided by
u-boot for platforms which need this).

Drop some random memory clearing rune, I've no idea what the intended
purpose was, 0x800000 doesn't correspond to any $foo_addr_r on the
midway systems for example.

Also get rid of the scsi scan which must necessarily have already
happened (since the boot.scr itselfs lives on a scsi drive).

Lastly, add some echos to show progress through the script, as an aid
to debugging.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoOsstest/Debian: Install dtbs into target filesystem in /boot/dtbs
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:04:56 +0000 (11:04 +0000)]
Osstest/Debian: Install dtbs into target filesystem in /boot/dtbs

This is done whenever dtbs.tar.gz exists. mg-debian-installer-update
produces the necessary inputs on the relevant platform (armhf).

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoOsstest/PDU: Add eth008.pm method to control the ARM rack PDU
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:04:55 +0000 (11:04 +0000)]
Osstest/PDU: Add eth008.pm method to control the ARM rack PDU

This controls the eth008 relay board:
http://www.robot-electronics.co.uk/htm/eth008tech.htm

Due to the use of the CGI interface this requires firmware version 4+.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agomake-flight: Run a basic test on each arm platform
commit | commitdiff | tree
Ian Campbell [Thu, 22 Jan 2015 11:04:54 +0000 (11:04 +0000)]
make-flight: Run a basic test on each arm platform

Unlike x86 there is enough variation in the ARM platforms that it is
worth having a basic test on each as part of a standard run. This
relies on each host having an appropriate platform-$platform host
flag.

The existing test-ARCH-ARCH-xl test is retained as a floating test,
while a new variant is added for each distinct platform present in the
hostdb which is tied to that platform type. The intention is that only
arm platforms will have platforms at first, although perhaps
platform-intel and platform-amd could be added in the future too.

There are currently no platform-* flags in the hostdb, so tested with
s/platform-/equiv-/ and:

  ( set -ex ;
    source ./cri-getplatforms ;
    blessing=real ;
    export OSSTEST_CONFIG=production-config ;
    for p in '' `getplatforms "armhf"` ; do
      set +x ;
      echo PLATFORM: $p ;
    done
  )
which prints:
  PLATFORM:
  PLATFORM: marilith
and with s/armhf/amd64/:
  PLATFORM:
  PLATFORM: rackservers-s40670
  PLATFORM: r310-moth
  PLATFORM: rackservers-s40680
  PLATFORM: dell-r310
  PLATFORM: rackservers-s40679
  PLATFORM: rackservers-s40663
  PLATFORM: rackservers-q21011

Also tested in standalone mode with a ~/.xen-osstest/config
containing:

  PlatformsArmhf midway cubietruck arndale

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoTftpDiVersion 2015-01-10 flight-33362 flight-33400 flight-33417 flight-33418 flight-33419 flight-33420 flight-33421 flight-33422 flight-33423 flight-33424 flight-33425 flight-33426 flight-33427 flight-33428 flight-33429 flight-33430 flight-33431 flight-33432 flight-33433 flight-33434 flight-33435 flight-33436 flight-33437 flight-33438 flight-33439 flight-33440 flight-33441 flight-33442 flight-33443 flight-33444 flight-33445 flight-33446 flight-33447 flight-33448 flight-33449 flight-33450 flight-33451 flight-33452 flight-33453 flight-33454 flight-33455 flight-33456 flight-33457 flight-33458 flight-33459 flight-33460 flight-33461 flight-33462 flight-33463 flight-33464 flight-33465 flight-33466 flight-33467 flight-33468 flight-33469 flight-33470 flight-33471 flight-33472 flight-33473 flight-33474 flight-33475 flight-33476 flight-33477 flight-33478 flight-33479 flight-33480 flight-33481 flight-33482 flight-33483 flight-33484 flight-33485 flight-33486 flight-33487 flight-33488 flight-33489 flight-33490 flight-33491 flight-33492 flight-33493 flight-33494 flight-33495 flight-33496 flight-33497 flight-33498 flight-33499 flight-33500 flight-33501 flight-33502 flight-33503 flight-33504 flight-33505 flight-33506 flight-33507 flight-33508 flight-33509 flight-33510 flight-33511 flight-33512 flight-33513 flight-33514 flight-33515 flight-33516 flight-33517 flight-33518 flight-33519 flight-33520 flight-33521 flight-33522 flight-33523 flight-33524 flight-33525 flight-33526 flight-33527 flight-33528 flight-33529 flight-33530 flight-33531 flight-33532 flight-33533 flight-33534 flight-33535 flight-33536 flight-33537 flight-33538 flight-33539 flight-33540 flight-33541 flight-33542 flight-33543 flight-33544 flight-33545 flight-33546 flight-33547 flight-33548 flight-33549 flight-33550 flight-33551 flight-33552 flight-33553 flight-33554 flight-33555 flight-33556 flight-33557 flight-33558 flight-33559 flight-33560 flight-33561 flight-33562 flight-33563 flight-33564 flight-33565 flight-33566 flight-33567 flight-33568 flight-33569 flight-33570 flight-33571 flight-33572 flight-33573 flight-33574 flight-33575 flight-33576 flight-33577 flight-33578 flight-33579 flight-33580 flight-33581 flight-33582 flight-33583 flight-33584 flight-33585 flight-33586 flight-33587 flight-33588 flight-33589 flight-33590 flight-33591 flight-33592 flight-33593 flight-33594 flight-33595 flight-33596 flight-33597 flight-33598 flight-33599 flight-33600 flight-33601 flight-33602 flight-33603 flight-33604 flight-33605 flight-33606 flight-33607 flight-33608 flight-33609 flight-33610 flight-33611 flight-33612 flight-33613 flight-33614 flight-33615 flight-33616 flight-33617 flight-33618 flight-33619 flight-33620 flight-33621 flight-33622 flight-33623 flight-33624 flight-33625 flight-33626 flight-33627 flight-33628 flight-33629 flight-33630 flight-33631 flight-33632 flight-33633 flight-33634 flight-33635 flight-33636 flight-33637 flight-33638 flight-33639 flight-33640 flight-33641 flight-33642 flight-33643 flight-33644 flight-33645 flight-33646 flight-33647 flight-33648 flight-33649 flight-33650 flight-33651 flight-33652 flight-33653 flight-33654 flight-33655 flight-33656 flight-33657 flight-33658 flight-33659 flight-33660 flight-33661 flight-33662 flight-33663 flight-33664 flight-33665 flight-33666 flight-33667 flight-33668 flight-33669 flight-33670 flight-33671 flight-33672 flight-33673 flight-33674 flight-33675 flight-33676 flight-33677 flight-33678 flight-33679 flight-33680 flight-33681 flight-33682 flight-33685 flight-33687 flight-33688 flight-33689 flight-33690 flight-33691 flight-33692 flight-33693 flight-33694 flight-33695 flight-33696 flight-33697 flight-33698 flight-33699 flight-33700 flight-33701 flight-33702 flight-33703 flight-33704 flight-33705 flight-33706 flight-33707 flight-33708 flight-33709 flight-33710 flight-33711 flight-33712 flight-33713 flight-33714 flight-33715 flight-33716 flight-33717 flight-33718 flight-33719 flight-33720 flight-33721 flight-33722 flight-33723 flight-33724 flight-33725 flight-33726 flight-33727 flight-33728 flight-33729 flight-33730 flight-33731 flight-33732 flight-33733 flight-33734 flight-33735 flight-33736 flight-33737 flight-33738 flight-33739 flight-33740 flight-33741 flight-33742 flight-33743 flight-33744 flight-33745 flight-33746 flight-33747 flight-33748 flight-33749 flight-33750 flight-33751 flight-33752 flight-33753 flight-33754 flight-33755 flight-33756 flight-33757 flight-33758 flight-33759 flight-33760 flight-33761 flight-33762 flight-33763 flight-33764 flight-33765 flight-33766 flight-33767 flight-33768 flight-33769 flight-33770 flight-33771 flight-33772 flight-33773 flight-33774 flight-33775 flight-33776 flight-33777 flight-33778 flight-33779 flight-33780 flight-33781 flight-33782 flight-33783 flight-33784 flight-33785 flight-33786 flight-33787 flight-33788 flight-33789 flight-33790 flight-33791 flight-33792 flight-33793 flight-33794 flight-33795 flight-33796 flight-33797 flight-33798 flight-33799 flight-33800 flight-33801 flight-33802 flight-33803 flight-33804 flight-33805 flight-33806 flight-33807 flight-33808 flight-33809 flight-33810 flight-33811 flight-33812 flight-33813 flight-33814 flight-33815 flight-33816 flight-33817 flight-33819 flight-33820 flight-33821 flight-33822 flight-33823 flight-33824 flight-33825 flight-33826 flight-33827 flight-33828 flight-33829 flight-33830 flight-33831 flight-33832 flight-33833 flight-33834 flight-33835 flight-33836 flight-33837 flight-33838 flight-33839 flight-33840 flight-33842 flight-33843 flight-33844 flight-33845 flight-33846 flight-33847 flight-33848 flight-33849 flight-33850 flight-33851 flight-33852 flight-33853 flight-33854 flight-33855 flight-33856 flight-33857 flight-33858 flight-33859 flight-33860 flight-33861 flight-33862 flight-33863 flight-33864 flight-33865 flight-33866 flight-33867 flight-33868 flight-33869 flight-33870 flight-33871 flight-33872 flight-33873 flight-33874 flight-33875 flight-33876 flight-33877 flight-33878 flight-33879 flight-33880 flight-33881 flight-33882 flight-33883 flight-33884 flight-33885 flight-33886 flight-33889 flight-33890 flight-33891 flight-33892 flight-33893 flight-33894 flight-33895 flight-33896 flight-33897 flight-33898 flight-33899 flight-33900 flight-33901 flight-33902 flight-33903 flight-33904 flight-33905 flight-33906 flight-33907 flight-33908 flight-33909 flight-33910 flight-33911 flight-33912 flight-33913 flight-33914 flight-33916 flight-33917 flight-33918 flight-33919 flight-33920 flight-33921 flight-33922 flight-33923 flight-33924 flight-33925 flight-33926 flight-33927 flight-33928 flight-33929 flight-33930 flight-33931 flight-33932 flight-33934 flight-33935 flight-33939 flight-33940 flight-33941 flight-33943 flight-33944 flight-33946 flight-33947 flight-33948 flight-33949 flight-33950 flight-33952 flight-33954 flight-33955 flight-33956
commit | commitdiff | tree
Ian Campbell [Sat, 10 Jan 2015 16:01:00 +0000 (16:01 +0000)]
TftpDiVersion 2015-01-10

Picks up updated kernel + installer from Debian 7.8 point release.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years agomg-debian-installer-update: produce deterministic output
commit | commitdiff | tree
Ian Campbell [Sat, 10 Jan 2015 14:58:02 +0000 (14:58 +0000)]
mg-debian-installer-update: produce deterministic output

Currently rerunning mg-debian-install-update when the external files
have changed still produces differences in the local files produced
during post-processing.

Avoid these differences by:

  - Using gzip -n, which avoids storing a timestamp in the gzip
    header (as well as the name, which we don't need).
  - Using pax -M norm, which normalises all timestamps (among other
    things, such as the owner, which we don't care about)
  - Using tar --mtime, with a reference within the dpkg-deb created
    hierarchy (which has timestamps from the package and is therefore
    dependent only on the downloaded package revision)

With this the results of two invocations of
mg-debian-installer-update(-all) are identical (assuming no changes to
the downloaded files) as demonstrated by runnign this quick hack:

#!/bin/bash

set -ex

TMP=$HOME/tmp/mg-di
rm -rf $TMP
mkdir -p $TMP

cat >$TMP/config <<EOF
TftpPath $TMP
TftpDiBase debian-installer
EOF
cat $TMP/config

mkdir -p $TMP/debian-installer

export OSSTEST_CONFIG=production-config:$TMP/config

# ./mg-debian-installer-update wheezy armhf firmware-bnx2
./mg-debian-installer-update-all

find $TMP/debian-installer -type f -print0 | xargs -0x md5sum > $TMP/SUMS.BASE

#rm -rf $TMP/debian-installer/*
mv $TMP/debian-installer $TMP/debian-installer.org
mkdir -p $TMP/debian-installer

# ./mg-debian-installer-update wheezy armhf firmware-bnx2
./mg-debian-installer-update-all
md5sum --quiet -c $TMP/SUMS.BASE

exit 0

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoEmpty commit to force push gate run flight-33275 flight-33320 flight-33363 flight-33364 flight-33365 flight-33366 flight-33367 flight-33368 flight-33369 flight-33370 flight-33371 flight-33372 flight-33373 flight-33374 flight-33375 flight-33376 flight-33377 flight-33378 flight-33379 flight-33380 flight-33381 flight-33382 flight-33383 flight-33384 flight-33385 flight-33386 flight-33387 flight-33388 flight-33389 flight-33390 flight-33391 flight-33392 flight-33393 flight-33394 flight-33395 flight-33396 flight-33397 flight-33398 flight-33399 flight-33401 flight-33402 flight-33403 flight-33404 flight-33405 flight-33406 flight-33407 flight-33408 flight-33409 flight-33410 flight-33411 flight-33412 flight-33413 flight-33414 flight-33415 flight-33416
commit | commitdiff | tree
Ian Jackson [Tue, 6 Jan 2015 17:59:31 +0000 (17:59 +0000)]
Empty commit to force push gate run

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agocr-for-branches: Branches for Xen 4.5 flight-33194 flight-33195 flight-33196 flight-33197 flight-33198 flight-33199 flight-33200 flight-33201 flight-33202 flight-33203 flight-33204 flight-33205 flight-33206 flight-33207 flight-33208 flight-33209 flight-33210 flight-33211 flight-33212 flight-33213 flight-33214 flight-33215 flight-33216 flight-33217 flight-33218 flight-33219 flight-33220 flight-33221 flight-33222 flight-33223 flight-33224 flight-33225 flight-33226 flight-33227 flight-33228 flight-33229 flight-33230 flight-33231 flight-33232 flight-33233 flight-33234 flight-33235 flight-33236 flight-33237 flight-33238 flight-33239 flight-33240 flight-33241 flight-33242 flight-33243 flight-33244 flight-33245 flight-33246 flight-33247 flight-33248 flight-33250 flight-33251 flight-33252 flight-33253 flight-33254 flight-33255 flight-33256 flight-33257 flight-33258 flight-33259 flight-33260 flight-33261 flight-33262 flight-33263 flight-33264 flight-33266 flight-33267 flight-33268 flight-33269 flight-33270 flight-33271 flight-33272 flight-33273 flight-33274 flight-33276 flight-33277 flight-33278 flight-33279 flight-33280 flight-33281 flight-33282 flight-33283 flight-33284 flight-33285 flight-33286 flight-33287 flight-33288 flight-33289 flight-33290 flight-33291 flight-33292 flight-33293 flight-33294 flight-33295 flight-33296 flight-33297 flight-33298 flight-33299 flight-33300 flight-33301 flight-33302 flight-33303 flight-33304 flight-33305 flight-33306 flight-33307 flight-33308 flight-33309 flight-33310 flight-33311 flight-33312 flight-33313 flight-33314 flight-33315 flight-33316 flight-33317 flight-33318 flight-33319 flight-33321 flight-33322 flight-33323 flight-33324 flight-33325 flight-33326 flight-33327 flight-33328 flight-33329 flight-33330 flight-33331 flight-33332 flight-33333 flight-33334 flight-33335 flight-33336 flight-33337 flight-33338 flight-33339 flight-33340 flight-33341 flight-33342 flight-33343 flight-33344 flight-33345 flight-33346 flight-33347 flight-33348 flight-33349 flight-33350 flight-33351 flight-33352 flight-33353 flight-33354 flight-33355 flight-33356 flight-33357 flight-33358 flight-33359 flight-33360 flight-33361
commit | commitdiff | tree
Ian Jackson [Tue, 6 Jan 2015 15:46:32 +0000 (15:46 +0000)]
cr-for-branches: Branches for Xen 4.5

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agots-libvirt-build: use Osstest::BuildSupport::submodulefixup flight-33126 flight-33182 flight-33183 flight-33184 flight-33185 flight-33186 flight-33187 flight-33188 flight-33189 flight-33190 flight-33191 flight-33192 flight-33193
commit | commitdiff | tree
Ian Campbell [Mon, 5 Jan 2015 12:57:08 +0000 (12:57 +0000)]
ts-libvirt-build: use Osstest::BuildSupport::submodulefixup

Instead of cloning gnulib manually which can break if upstream gnulib
gets ahead of libvirt.git (which applies patches on the fly etc). By
using submodulefixup we automatically DTRT and use the version of
gnulib specified by the libvirt.git submodule metadata, but with a
runvar override if necessary.

This also removes a whole bunch of faffing in ap-*, cr-daily-branch
and mfi-common to get the version of gnulib to use, which was always a
bit of a wart (ungated for one thing...).

We continue to use --no-git and GNULIB_SRCDIR because otherwise
autogen.sh (via bootstrap) will force its own version, overwriting
what submodulefixup has done. For this we need a way to get the hash
representing the module, so introduce submodule_find (and rework
submodule_have in terms of it).

Tested in standalone mode with build-amd64-libvirt and
build-amd64-rumpuserxen (because I touched submodule_have, AFAICT the
bodges were not run). The libvirt build was tested both with the
automatic revisions and with:
    revision_libvirt=2360fe5d24175835d3f5fd1c7e8e6e13addab629
    revision_libvirt_gnulib=16518d9ed8f25d3e53931dd1aa343072933e4604
(used in successful libvirt flight 32648), in both cases confirming
that the build used the desired versions.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@citrix.com>
---
v2: Honour revision_libvirt_gnulib.
v3: Fix submodule_have, defined(&sub) is always true because defined
    is special wrt &sub.

10 years agoTftpDiVersion 2014-12-14 flight-32427 flight-32454 flight-32514 flight-32515 flight-32516 flight-32517 flight-32518 flight-32519 flight-32520 flight-32521 flight-32522 flight-32523 flight-32524 flight-32525 flight-32526 flight-32527 flight-32528 flight-32529 flight-32530 flight-32531 flight-32533 flight-32534 flight-32535 flight-32536 flight-32537 flight-32538 flight-32539 flight-32540 flight-32541 flight-32542 flight-32543 flight-32544 flight-32545 flight-32546 flight-32547 flight-32548 flight-32549 flight-32550 flight-32551 flight-32552 flight-32553 flight-32554 flight-32555 flight-32556 flight-32557 flight-32558 flight-32560 flight-32561 flight-32562 flight-32563 flight-32564 flight-32565 flight-32566 flight-32567 flight-32568 flight-32569 flight-32570 flight-32571 flight-32572 flight-32573 flight-32574 flight-32575 flight-32576 flight-32577 flight-32578 flight-32579 flight-32580 flight-32581 flight-32582 flight-32583 flight-32584 flight-32585 flight-32586 flight-32587 flight-32588 flight-32589 flight-32590 flight-32591 flight-32592 flight-32593 flight-32594 flight-32595 flight-32596 flight-32597 flight-32598 flight-32599 flight-32600 flight-32601 flight-32602 flight-32603 flight-32604 flight-32605 flight-32606 flight-32607 flight-32608 flight-32609 flight-32610 flight-32611 flight-32612 flight-32613 flight-32614 flight-32615 flight-32616 flight-32617 flight-32618 flight-32619 flight-32620 flight-32621 flight-32622 flight-32623 flight-32624 flight-32625 flight-32626 flight-32627 flight-32628 flight-32629 flight-32630 flight-32631 flight-32632 flight-32633 flight-32634 flight-32635 flight-32636 flight-32637 flight-32638 flight-32639 flight-32640 flight-32641 flight-32642 flight-32643 flight-32644 flight-32645 flight-32646 flight-32647 flight-32648 flight-32649 flight-32650 flight-32651 flight-32652 flight-32653 flight-32654 flight-32655 flight-32656 flight-32657 flight-32658 flight-32659 flight-32660 flight-32661 flight-32662 flight-32663 flight-32664 flight-32665 flight-32666 flight-32667 flight-32668 flight-32669 flight-32670 flight-32671 flight-32672 flight-32673 flight-32674 flight-32675 flight-32676 flight-32677 flight-32678 flight-32679 flight-32680 flight-32681 flight-32682 flight-32683 flight-32684 flight-32685 flight-32686 flight-32687 flight-32688 flight-32689 flight-32690 flight-32691 flight-32692 flight-32693 flight-32694 flight-32695 flight-32696 flight-32697 flight-32698 flight-32699 flight-32700 flight-32701 flight-32702 flight-32703 flight-32704 flight-32705 flight-32706 flight-32707 flight-32708 flight-32709 flight-32710 flight-32711 flight-32712 flight-32713 flight-32714 flight-32715 flight-32716 flight-32717 flight-32718 flight-32719 flight-32720 flight-32721 flight-32722 flight-32723 flight-32724 flight-32725 flight-32726 flight-32727 flight-32728 flight-32733 flight-32746 flight-32747 flight-32748 flight-32749 flight-32750 flight-32751 flight-32752 flight-32753 flight-32754 flight-32755 flight-32756 flight-32757 flight-32758 flight-32759 flight-32760 flight-32761 flight-32762 flight-32763 flight-32764 flight-32765 flight-32766 flight-32767 flight-32768 flight-32769 flight-32770 flight-32771 flight-32772 flight-32773 flight-32774 flight-32775 flight-32776 flight-32777 flight-32778 flight-32779 flight-32780 flight-32781 flight-32782 flight-32783 flight-32784 flight-32785 flight-32786 flight-32787 flight-32788 flight-32789 flight-32790 flight-32791 flight-32792 flight-32793 flight-32794 flight-32795 flight-32796 flight-32797 flight-32798 flight-32799 flight-32800 flight-32801 flight-32802 flight-32803 flight-32804 flight-32805 flight-32806 flight-32807 flight-32808 flight-32809 flight-32810 flight-32811 flight-32812 flight-32813 flight-32814 flight-32815 flight-32816 flight-32817 flight-32818 flight-32819 flight-32820 flight-32821 flight-32822 flight-32823 flight-32824 flight-32825 flight-32826 flight-32827 flight-32828 flight-32829 flight-32830 flight-32831 flight-32832 flight-32833 flight-32834 flight-32835 flight-32836 flight-32837 flight-32838 flight-32839 flight-32840 flight-32841 flight-32842 flight-32843 flight-32844 flight-32845 flight-32846 flight-32847 flight-32848 flight-32849 flight-32850 flight-32851 flight-32852 flight-32853 flight-32854 flight-32855 flight-32856 flight-32857 flight-32858 flight-32859 flight-32860 flight-32861 flight-32862 flight-32864 flight-32865 flight-32866 flight-32867 flight-32868 flight-32869 flight-32870 flight-32871 flight-32872 flight-32873 flight-32874 flight-32875 flight-32876 flight-32877 flight-32878 flight-32879 flight-32880 flight-32881 flight-32882 flight-32883 flight-32884 flight-32885 flight-32886 flight-32887 flight-32888 flight-32889 flight-32890 flight-32891 flight-32892 flight-32893 flight-32894 flight-32895 flight-32896 flight-32897 flight-32898 flight-32899 flight-32900 flight-32901 flight-32902 flight-32903 flight-32904 flight-32905 flight-32906 flight-32907 flight-32908 flight-32909 flight-32910 flight-32911 flight-32912 flight-32913 flight-32914 flight-32915 flight-32916 flight-32917 flight-32918 flight-32919 flight-32920 flight-32921 flight-32922 flight-32923 flight-32924 flight-32925 flight-32926 flight-32927 flight-32928 flight-32929 flight-32930 flight-32931 flight-32932 flight-32933 flight-32934 flight-32935 flight-32936 flight-32937 flight-32938 flight-32939 flight-32940 flight-32941 flight-32942 flight-32943 flight-32944 flight-32945 flight-32946 flight-32947 flight-32948 flight-32949 flight-32950 flight-32951 flight-32952 flight-32953 flight-32954 flight-32955 flight-32956 flight-32957 flight-32958 flight-32959 flight-32960 flight-32961 flight-32962 flight-32963 flight-32964 flight-32965 flight-32966 flight-32967 flight-32968 flight-32969 flight-32970 flight-32971 flight-32972 flight-32973 flight-32974 flight-32975 flight-32976 flight-32977 flight-32978 flight-32979 flight-32980 flight-32981 flight-32982 flight-32983 flight-32984 flight-32985 flight-32986 flight-32987 flight-32988 flight-32989 flight-32990 flight-32991 flight-32992 flight-32993 flight-32994 flight-32995 flight-32996 flight-32997 flight-32998 flight-32999 flight-33000 flight-33001 flight-33002 flight-33003 flight-33004 flight-33005 flight-33006 flight-33007 flight-33008 flight-33009 flight-33010 flight-33011 flight-33012 flight-33013 flight-33014 flight-33015 flight-33016 flight-33017 flight-33018 flight-33019 flight-33020 flight-33021 flight-33022 flight-33023 flight-33024 flight-33025 flight-33026 flight-33027 flight-33028 flight-33029 flight-33030 flight-33031 flight-33032 flight-33033 flight-33034 flight-33035 flight-33036 flight-33037 flight-33038 flight-33039 flight-33040 flight-33041 flight-33042 flight-33043 flight-33044 flight-33045 flight-33046 flight-33047 flight-33048 flight-33049 flight-33050 flight-33051 flight-33052 flight-33053 flight-33054 flight-33055 flight-33056 flight-33057 flight-33058 flight-33059 flight-33060 flight-33061 flight-33062 flight-33063 flight-33064 flight-33065 flight-33066 flight-33067 flight-33068 flight-33069 flight-33070 flight-33071 flight-33072 flight-33073 flight-33074 flight-33075 flight-33076 flight-33077 flight-33078 flight-33079 flight-33080 flight-33081 flight-33082 flight-33083 flight-33084 flight-33085 flight-33086 flight-33087 flight-33088 flight-33089 flight-33090 flight-33091 flight-33092 flight-33093 flight-33094 flight-33095 flight-33096 flight-33097 flight-33098 flight-33099 flight-33100 flight-33101 flight-33102 flight-33103 flight-33104 flight-33105 flight-33106 flight-33107 flight-33108 flight-33109 flight-33110 flight-33111 flight-33112 flight-33113 flight-33114 flight-33115 flight-33116 flight-33117 flight-33118 flight-33119 flight-33120 flight-33121 flight-33122 flight-33123 flight-33124 flight-33125 flight-33127 flight-33128 flight-33129 flight-33130 flight-33131 flight-33132 flight-33133 flight-33134 flight-33135 flight-33136 flight-33137 flight-33138 flight-33139 flight-33140 flight-33141 flight-33142 flight-33143 flight-33144 flight-33145 flight-33146 flight-33147 flight-33148 flight-33149 flight-33150 flight-33151 flight-33152 flight-33153 flight-33154 flight-33155 flight-33156 flight-33157 flight-33158 flight-33159 flight-33160 flight-33161 flight-33162 flight-33163 flight-33164 flight-33165 flight-33166 flight-33167 flight-33168 flight-33169 flight-33170 flight-33171 flight-33172 flight-33173 flight-33174 flight-33175 flight-33176 flight-33177 flight-33178 flight-33179 flight-33180 flight-33181
commit | commitdiff | tree
Ian Campbell [Sun, 14 Dec 2014 19:18:38 +0000 (19:18 +0000)]
TftpDiVersion 2014-12-14

Picks up changes on armhf to include dtbs and more modules in the initrd
overlay.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoAdd basic PVH flights. flight-32331 flight-32346 flight-32367 flight-32388 flight-32416 flight-32421 flight-32422 flight-32423 flight-32424 flight-32425 flight-32426 flight-32428 flight-32429 flight-32430 flight-32431 flight-32432 flight-32433 flight-32434 flight-32435 flight-32436 flight-32437 flight-32438 flight-32439 flight-32440 flight-32441 flight-32442 flight-32443 flight-32444 flight-32445 flight-32446 flight-32447 flight-32448 flight-32449 flight-32450 flight-32451 flight-32452 flight-32453 flight-32455 flight-32456 flight-32457 flight-32458 flight-32459 flight-32460 flight-32461 flight-32462 flight-32463 flight-32464 flight-32465 flight-32466 flight-32467 flight-32468 flight-32469 flight-32470 flight-32471 flight-32472 flight-32473 flight-32474 flight-32475 flight-32476 flight-32477 flight-32478 flight-32479 flight-32480 flight-32481 flight-32482 flight-32483 flight-32484 flight-32485 flight-32486 flight-32487 flight-32488 flight-32489 flight-32490 flight-32491 flight-32492 flight-32493 flight-32494 flight-32495 flight-32496 flight-32497 flight-32498 flight-32499 flight-32500 flight-32501 flight-32502 flight-32503 flight-32504 flight-32506 flight-32507 flight-32508 flight-32509 flight-32510 flight-32512 flight-32513
commit | commitdiff | tree
Ian Campbell [Thu, 4 Dec 2014 09:58:34 +0000 (09:58 +0000)]
Add basic PVH flights.

These are the usual PV debian flights with pvh=1 added to the
configuration file.

A job is created for each of Intel and AMD, although obviously AMD is
expected to fail at the moment.

In my testing I got:
    (XEN) Attempt to create a PVH guest on a system without necessary hardware support
because my test box happens to be AMD.

I have confirmed that the pvh=1 option is correctly present in the
guest cfg for the new pvh job, and that no pvh= is present at all in
the existing test-amd64-amd64-xl job (which is expected and desired if
no pvh runvar is present).

Beyond that I've not tested this at all I fully expect even Intel to
fail in the first instance, due to issues such as lack of necessary
kernel options etc. I suggest to take this now and iterate on any
further changes.

For a xen-unstable flight this results in these runvars:
$ ./mg-show-flight-runvars pvh| grep -- -pvh | sort
test-amd64-amd64-xl-pvh-amd               all_hostflags               arch-amd64,arch-xen-amd64,suite-wheezy,purpose-test,hvm-amd
test-amd64-amd64-xl-pvh-amd               arch                        amd64
test-amd64-amd64-xl-pvh-amd               buildjob                    build-amd64
test-amd64-amd64-xl-pvh-amd               debian_arch                 amd64
test-amd64-amd64-xl-pvh-amd               debian_kernkind             pvops
test-amd64-amd64-xl-pvh-amd               debian_pvh                  1
test-amd64-amd64-xl-pvh-amd               kernbuildjob                build-amd64-pvops
test-amd64-amd64-xl-pvh-amd               kernkind                    pvops
test-amd64-amd64-xl-pvh-amd               toolstack                   xl
test-amd64-amd64-xl-pvh-amd               xenbuildjob                 build-amd64
test-amd64-amd64-xl-pvh-intel             all_hostflags               arch-amd64,arch-xen-amd64,suite-wheezy,purpose-test,hvm-intel
test-amd64-amd64-xl-pvh-intel             arch                        amd64
test-amd64-amd64-xl-pvh-intel             buildjob                    build-amd64
test-amd64-amd64-xl-pvh-intel             debian_arch                 amd64
test-amd64-amd64-xl-pvh-intel             debian_kernkind             pvops
test-amd64-amd64-xl-pvh-intel             debian_pvh                  1
test-amd64-amd64-xl-pvh-intel             kernbuildjob                build-amd64-pvops
test-amd64-amd64-xl-pvh-intel             kernkind                    pvops
test-amd64-amd64-xl-pvh-intel             toolstack                   xl
test-amd64-amd64-xl-pvh-intel             xenbuildjob                 build-amd64

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
---
v2: pvh\b.*.

10 years agoOsstest/Debian: Add support for "ExtraInitramfsModules" host property
commit | commitdiff | tree
Ian Campbell [Fri, 21 Nov 2014 13:16:58 +0000 (13:16 +0000)]
Osstest/Debian: Add support for "ExtraInitramfsModules" host property

The arndale platform needs a bunch of clk, phy and regulator stuff in order to
access its root filesystem. However mkinitramfs is not (currently?) able to
figure this out and therefore doesn't include them in the initrd. See
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=762042

Add a new host prop which can list these required additional module and
arranges for a suitable /etc/initramfs-tools/modules to be created on install.

Using the new HostGroupProp syntax the required modules are:

HostGroupProp_arndale_ExtraInitramfsModules clk-s2mps11 s5m8767 i2c-s3c2410 phy-exynos5250-sata

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoOsstest/Debian: Refactor code to set bootargs in u-boot script
commit | commitdiff | tree
Ian Campbell [Fri, 21 Nov 2014 13:16:55 +0000 (13:16 +0000)]
Osstest/Debian: Refactor code to set bootargs in u-boot script

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-debian-install: rename cfg_xend to cfg
commit | commitdiff | tree
Wei Liu [Tue, 14 Oct 2014 21:50:08 +0000 (22:50 +0100)]
ts-debian-install: rename cfg_xend to cfg

... as this config file is just a config file in general, not strictly a
Xend format config file.

Signed-off-by: Wei Liu <wei.liu2@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agogitignore: ignore images directory
commit | commitdiff | tree
Wei Liu [Tue, 14 Oct 2014 21:50:07 +0000 (22:50 +0100)]
gitignore: ignore images directory

Signed-off-by: Wei Liu <wei.liu2@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoREADME: list chiark-utils-bin as requirement
commit | commitdiff | tree
Wei Liu [Tue, 14 Oct 2014 21:50:06 +0000 (22:50 +0100)]
README: list chiark-utils-bin as requirement

Signed-off-by: Wei Liu <wei.liu2@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoTestSupport: allow overriding of on_* in prepareguest_part_xencfg
commit | commitdiff | tree
Ian Campbell [Mon, 1 Dec 2014 12:57:20 +0000 (12:57 +0000)]
TestSupport: allow overriding of on_* in prepareguest_part_xencfg

Currently only on_reboot can be overridden

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoTestSupport: use timeout(1) flight-32167 flight-32235 flight-32307 flight-32332 flight-32333 flight-32334 flight-32335 flight-32336 flight-32337 flight-32338 flight-32339 flight-32340 flight-32341 flight-32342 flight-32343 flight-32344 flight-32345 flight-32347 flight-32348 flight-32349 flight-32350 flight-32351 flight-32352 flight-32353 flight-32354 flight-32355 flight-32356 flight-32357 flight-32358 flight-32359 flight-32360 flight-32361 flight-32362 flight-32363 flight-32364 flight-32365 flight-32366 flight-32368 flight-32369 flight-32370 flight-32371 flight-32372 flight-32373 flight-32374 flight-32375 flight-32376 flight-32377 flight-32378 flight-32379 flight-32380 flight-32381 flight-32383 flight-32384 flight-32385 flight-32386 flight-32387 flight-32389 flight-32390 flight-32391 flight-32392 flight-32393 flight-32394 flight-32395 flight-32396 flight-32397 flight-32398 flight-32399 flight-32400 flight-32401 flight-32402 flight-32403 flight-32404 flight-32405 flight-32406 flight-32407 flight-32408 flight-32409 flight-32410 flight-32411 flight-32412 flight-32413 flight-32414 flight-32415 flight-32417 flight-32418 flight-32419 flight-32420
commit | commitdiff | tree
Ian Jackson [Mon, 8 Dec 2014 12:10:24 +0000 (12:10 +0000)]
TestSupport: use timeout(1)

If a command we run times out, the machinery in cmd() will arrange
for the ts-* script to spot the timeout, and stop waiting for it.

However it is also necessary for the command we ran to die.  It has a
copy of the owner daemon fd, so if it doesn't, our resources won't get
freed.  In sufficiently exciting bugs, our allocation might continue
indefinitely, while a subprocess of ours hangs on after we are long
gone.

timeout(1) does not print a message when the process times out (!)  So
we can't do away with the logic in cmd().  We set the timeout(1)
timeout to 30s more than our own timeout, so that cmd() will time
out first and print a message.

We could use alarm(1) as we do in Osstest/Serial/sympathy.pm but that
program isn't packaged and its unsophisticated approach is not really
appropriate for arbitrary nonconsenting programs.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agoAdd simple helper to update DI for all architectures. flight-32111 flight-32136 flight-32137 flight-32138 flight-32139 flight-32140 flight-32141 flight-32142 flight-32143 flight-32144 flight-32145 flight-32146 flight-32147 flight-32148 flight-32149 flight-32150 flight-32151 flight-32152 flight-32153 flight-32154 flight-32155 flight-32156 flight-32157 flight-32161 flight-32162 flight-32163 flight-32164 flight-32165 flight-32166 flight-32168 flight-32169 flight-32170 flight-32171 flight-32172 flight-32173 flight-32175 flight-32176 flight-32177 flight-32178 flight-32179 flight-32180 flight-32181 flight-32182 flight-32183 flight-32184 flight-32185 flight-32186 flight-32187 flight-32188 flight-32189 flight-32190 flight-32191 flight-32192 flight-32193 flight-32194 flight-32195 flight-32196 flight-32197 flight-32198 flight-32199 flight-32200 flight-32201 flight-32202 flight-32203 flight-32204 flight-32205 flight-32206 flight-32207 flight-32208 flight-32209 flight-32210 flight-32211 flight-32212 flight-32213 flight-32214 flight-32215 flight-32216 flight-32217 flight-32218 flight-32219 flight-32220 flight-32221 flight-32222 flight-32223 flight-32224 flight-32225 flight-32226 flight-32227 flight-32228 flight-32229 flight-32230 flight-32231 flight-32232 flight-32233 flight-32234 flight-32236 flight-32237 flight-32238 flight-32239 flight-32240 flight-32241 flight-32242 flight-32243 flight-32244 flight-32245 flight-32246 flight-32247 flight-32248 flight-32249 flight-32250 flight-32251 flight-32252 flight-32253 flight-32254 flight-32255 flight-32256 flight-32257 flight-32258 flight-32259 flight-32260 flight-32261 flight-32262 flight-32263 flight-32264 flight-32265 flight-32266 flight-32267 flight-32268 flight-32270 flight-32271 flight-32272 flight-32273 flight-32274 flight-32275 flight-32276 flight-32277 flight-32278 flight-32279 flight-32280 flight-32281 flight-32282 flight-32283 flight-32284 flight-32285 flight-32286 flight-32287 flight-32288 flight-32289 flight-32290 flight-32291 flight-32292 flight-32293 flight-32294 flight-32295 flight-32296 flight-32297 flight-32298 flight-32299 flight-32300 flight-32301 flight-32302 flight-32303 flight-32304 flight-32305 flight-32306 flight-32308 flight-32309 flight-32310 flight-32311 flight-32312 flight-32313 flight-32314 flight-32315 flight-32316 flight-32317 flight-32318 flight-32319 flight-32320 flight-32321 flight-32322 flight-32323 flight-32324 flight-32325 flight-32326 flight-32327 flight-32328 flight-32329 flight-32330
commit | commitdiff | tree
Ian Campbell [Fri, 21 Nov 2014 13:16:52 +0000 (13:16 +0000)]
Add simple helper to update DI for all architectures.

Uses DebianNonfreeFirmware, even (especially) for production, so move
the README stanza out of standalone only section. The current default
matches what is in the current production versions of DI.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-kernel-build: enable CONFIG_IKCONFIG{_PROC}
commit | commitdiff | tree
Ian Campbell [Fri, 21 Nov 2014 13:16:51 +0000 (13:16 +0000)]
ts-kernel-build: enable CONFIG_IKCONFIG{_PROC}

This makes the kernel's .config available in /proc/config.gz and
embeds a copy which can be extracted with
linux/scripts/extract-ikconfig (which I've not tried, but have no
reason to doubt).

Having this around can be handy with an older osstest installed test
box and to confirm you've booted the kernel you think you have when
you are messing with .config options.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agostandalone: Introduce "HostGroups" for use in OSSTEST_CONFIG
commit | commitdiff | tree
Ian Campbell [Fri, 21 Nov 2014 13:16:50 +0000 (13:16 +0000)]
standalone: Introduce "HostGroups" for use in OSSTEST_CONFIG

This saves repeating identical HostProp and HostFlags for sets of
identical machines. e.g.

    HostGroupProp_cubietruck_LinuxSerialConsole ttyS0
    HostGroupProp_cubietruck_Build_Make_Flags -j12
    HostGroupProp_cubietruck_XenSerialConsole dtuart
    HostGroupProp_cubietruck_XenDTUARTPath /soc@01c00000/serial@01c28000
    HostGroupFlags_cubietruck suite-wheezy,equiv-cubietruck,need-kernel-deb-armmp,no-di-kernel,need-uboot-bootscr

    HostGroup_braque cubietruck
    HostProp_braque_Fqdn braque.uk.xensource.com

    HostGroup_picaso cubietruck
    HostProp_picaso_Fqdn picaso.uk.xensource.com

    HostGroup_metzinger cubietruck
    HostProp_metzinger metzinger.uk.xensource.com

    HostGroup_gleizes cubietruck
    HostProp_gleizes_Fqdn gleizes.uk.xensource.com

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-xen-build-prep: Install libxml-xpath-perl on build machines
commit | commitdiff | tree
Ian Campbell [Fri, 5 Dec 2014 14:54:08 +0000 (14:54 +0000)]
ts-xen-build-prep: Install libxml-xpath-perl on build machines

Required by latest libvirt, to build docs.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@citrix.com>
10 years agolinux-next tests: Use correct branch for baseline
commit | commitdiff | tree
Ian Jackson [Thu, 20 Nov 2014 13:48:56 +0000 (13:48 +0000)]
linux-next tests: Use correct branch for baseline

Make cr-daily-branch honour an environment or setting variable
EXTRA_SGR_ARGS.  In branch-settings.linux-next set it appropriately to
arrange that the linux-next test reports consider linux-linus tests as
interesting as well as just linux-next ones.

(We already use a flight from linux-linus for selecting the baseline
linux version.)

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agocs-adjust-flight: runvar-perlop: Do not report non-changes flight-31654 flight-31678 flight-31756 flight-31803 flight-31841 flight-31844 flight-31848 flight-31849 flight-31850 flight-31851 flight-31852 flight-31853 flight-31854 flight-31855 flight-31856 flight-31857 flight-31858 flight-31859 flight-31860 flight-31861 flight-31862 flight-31863 flight-31864 flight-31865 flight-31866 flight-31867 flight-31868 flight-31869 flight-31870 flight-31871 flight-31872 flight-31873 flight-31874 flight-31875 flight-31876 flight-31877 flight-31878 flight-31879 flight-31880 flight-31881 flight-31882 flight-31883 flight-31884 flight-31885 flight-31886 flight-31887 flight-31888 flight-31889 flight-31890 flight-31891 flight-31892 flight-31893 flight-31894 flight-31895 flight-31896 flight-31897 flight-31898 flight-31899 flight-31900 flight-31901 flight-31902 flight-31903 flight-31904 flight-31905 flight-31906 flight-31907 flight-31908 flight-31909 flight-31910 flight-31911 flight-31914 flight-31916 flight-31917 flight-31918 flight-31919 flight-31920 flight-31922 flight-31923 flight-31924 flight-31925 flight-31926 flight-31927 flight-31928 flight-31929 flight-31930 flight-31932 flight-31933 flight-31934 flight-31935 flight-31936 flight-31937 flight-31938 flight-31939 flight-31940 flight-31941 flight-31942 flight-31943 flight-31944 flight-31945 flight-31946 flight-31947 flight-31948 flight-31949 flight-31950 flight-31951 flight-31952 flight-31953 flight-31954 flight-31955 flight-31956 flight-31957 flight-31959 flight-31960 flight-31962 flight-31963 flight-31965 flight-31966 flight-31968 flight-31969 flight-31970 flight-31971 flight-31972 flight-31973 flight-31974 flight-31975 flight-31976 flight-31977 flight-31978 flight-31979 flight-31980 flight-31981 flight-31983 flight-31984 flight-31985 flight-31986 flight-31987 flight-31988 flight-31989 flight-31990 flight-31991 flight-31992 flight-31993 flight-31994 flight-31995 flight-31996 flight-31997 flight-31998 flight-31999 flight-32000 flight-32001 flight-32002 flight-32003 flight-32004 flight-32005 flight-32006 flight-32007 flight-32008 flight-32009 flight-32010 flight-32011 flight-32012 flight-32013 flight-32014 flight-32015 flight-32016 flight-32017 flight-32019 flight-32020 flight-32021 flight-32022 flight-32023 flight-32024 flight-32025 flight-32026 flight-32027 flight-32028 flight-32029 flight-32030 flight-32031 flight-32032 flight-32033 flight-32034 flight-32035 flight-32036 flight-32037 flight-32038 flight-32039 flight-32040 flight-32041 flight-32042 flight-32043 flight-32044 flight-32045 flight-32046 flight-32047 flight-32048 flight-32049 flight-32050 flight-32051 flight-32052 flight-32053 flight-32054 flight-32055 flight-32056 flight-32057 flight-32058 flight-32060 flight-32061 flight-32062 flight-32063 flight-32064 flight-32065 flight-32066 flight-32067 flight-32068 flight-32069 flight-32070 flight-32071 flight-32072 flight-32073 flight-32074 flight-32075 flight-32076 flight-32077 flight-32078 flight-32079 flight-32080 flight-32081 flight-32082 flight-32083 flight-32084 flight-32085 flight-32086 flight-32087 flight-32088 flight-32089 flight-32090 flight-32092 flight-32093 flight-32094 flight-32095 flight-32096 flight-32097 flight-32098 flight-32099 flight-32100 flight-32101 flight-32102 flight-32103 flight-32104 flight-32105 flight-32106 flight-32107 flight-32108 flight-32109 flight-32110 flight-32112 flight-32113 flight-32114 flight-32115 flight-32116 flight-32117 flight-32118 flight-32119 flight-32120 flight-32121 flight-32122 flight-32123 flight-32124 flight-32125 flight-32126 flight-32127 flight-32128 flight-32129 flight-32130 flight-32131 flight-32132 flight-32133 flight-32134 flight-32135
commit | commitdiff | tree
Ian Jackson [Thu, 13 Nov 2014 16:05:15 +0000 (16:05 +0000)]
cs-adjust-flight: runvar-perlop: Do not report non-changes

With -v, runvar-perlop would unconditionally print a message about the
changed variable.  Instead, only call runvar_set if the value is to
change.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agotarget_editfile: Improve doc comment
commit | commitdiff | tree
Ian Jackson [Thu, 13 Nov 2014 16:02:54 +0000 (16:02 +0000)]
target_editfile: Improve doc comment

More clearly state which arguments are optional.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agots-hosts-allocate-Executive: Radically reduce the previously_failed bonus
commit | commitdiff | tree
Ian Jackson [Tue, 11 Nov 2014 19:38:03 +0000 (19:38 +0000)]
ts-hosts-allocate-Executive: Radically reduce the previously_failed bonus

Make osstest less obsessive about sticking to failing hosts if they
are persistently unavailable.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-hosts-allocate-Executive: Redo variation_bonus scoring
commit | commitdiff | tree
Ian Jackson [Tue, 11 Nov 2014 19:34:07 +0000 (19:34 +0000)]
ts-hosts-allocate-Executive: Redo variation_bonus scoring

Use a logarithmic scale.  Cap the bonus at 12h rather than 5d/30 = 4h.
When we have previously failed, make sure we apply a reverse bonus,
rather than a penalty.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
---
v2: Fix cap to really be 12h not 12d.

10 years agots-hosts-allocate-Executive: Score for equivalent previous failures
commit | commitdiff | tree
Ian Jackson [Tue, 11 Nov 2014 19:30:38 +0000 (19:30 +0000)]
ts-hosts-allocate-Executive: Score for equivalent previous failures

Look to see whether the last run on any hosts which are equivalent to
the ones we're looking at, failed.  This means that when host X is
failing and we are considering host Y which is equivalent to X, we
give Y a selection bonus.

This means that osstest will be less obsessive about sticking to the
very same failing host.

Also, no longer multiply by the number of previously-failing hosts;
instead, just consider whether _all_ the hosts are previously-failing
in their corresponding slots (*_host runvars).  It doesn't make sense
to make the bonus for using failing hosts depend on the number of
hosts used by the test.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
---
v2: Improve commit message.

10 years agots-hosts-allocate-Executive: Clarify an expression with //
commit | commitdiff | tree
Ian Jackson [Tue, 11 Nov 2014 19:28:24 +0000 (19:28 +0000)]
ts-hosts-allocate-Executive: Clarify an expression with //

No functional change.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
next
Xen Test Framework and Suite, used for Open Source Xen CI push gate