11:20 a.m.
The first patch pulls the latest gnulibs to fetch the newest syntax-check
speedups.
Patch 2 is included since a gnulib update will trigger configure anyway.
The rest are mostly cleanups, the speed-ups they (might) provide are barely
measurable.
Ján Tomko (7):
maint: update to latest gnulib
configure.ac: drop checks for ETH flags
tests: fix the return value of test-wrap-argv
cfg.mk: use subst instead of tr
cfg.mk: drop redundant sc_prohibit_gethostby
cfg.mk: drop redundant sc_size_of_brackets
cfg.mk: join not_streq and not_strneq tests
.gnulib | 2 +-
build-aux/check-spacing.pl | 2 +-
cfg.mk | 35 +++++------------------------------
configure.ac | 5 -----
src/util/virnetdev.c | 20 ++++++++++----------
tests/test-wrap-argv.pl | 2 +-
tests/virstoragetest.c | 2 +-
7 files changed, 19 insertions(+), 49 deletions(-)
--
2.7.3
11:20 a.m.
New subject: [libvirt] [PATCH 1/7] maint: update to latest gnulib
Pick up the new syntax-check speedups.
---
.gnulib | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/.gnulib b/.gnulib
index 68b6ade..0fe8b3c 160000
--- a/.gnulib
+++ b/.gnulib
@@ -1 +1 @@
-Subproject commit 68b6adebef05670a312fb92b05e7bd089d2ed43a
+Subproject commit 0fe8b3c8116168e3b8b058df63ddedd45c3fcab0
--
2.7.3
12:18 a.m.
New subject: [libvirt] [PATCH 1/7] maint: update to latest gnulib
On 08/03/2016 12:20 PM, Ján Tomko wrote:
Pick up the new syntax-check speedups.
---
I think eblake usually pushes those without review under the "gnulib
updates are immune from review | can't be effectively reviewed" rule :-)
.gnulib | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/.gnulib b/.gnulib
index 68b6ade..0fe8b3c 160000
--- a/.gnulib
+++ b/.gnulib
@@ -1 +1 @@
-Subproject commit 68b6adebef05670a312fb92b05e7bd089d2ed43a
+Subproject commit 0fe8b3c8116168e3b8b058df63ddedd45c3fcab0
7:11 a.m.
New subject: [libvirt] [PATCH 1/7] maint: update to latest gnulib
On Thu, Aug 18, 2016 at 01:18:59AM -0400, Laine Stump wrote:
On 08/03/2016 12:20 PM, Ján Tomko wrote:
> Pick up the new syntax-check speedups.
> ---
I think eblake usually pushes those without review under the "gnulib
updates are immune from review | can't be effectively reviewed" rule :-)
My intention was to postpone pushing this until there's a change in
configure.ac or other file that requires us to re-run configure.
But since there has not been any in two weeks and patch 2/7
is hopelessly broken, I went ahead and pushed it now.
Jan
> .gnulib | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/.gnulib b/.gnulib
> index 68b6ade..0fe8b3c 160000
> --- a/.gnulib
> +++ b/.gnulib
> @@ -1 +1 @@
> -Subproject commit 68b6adebef05670a312fb92b05e7bd089d2ed43a
> +Subproject commit 0fe8b3c8116168e3b8b058df63ddedd45c3fcab0
11:20 a.m.
New subject: [libvirt] [PATCH 2/7] configure.ac: drop checks for ETH flags
No choices are made at configure time based on these checks.
Drop them and use #ifdefs in virnetdev.c.
---
configure.ac | 5 -----
src/util/virnetdev.c | 20 ++++++++++----------
2 files changed, 10 insertions(+), 15 deletions(-)
diff --git a/configure.ac b/configure.ac
index 8d7d63e..db42173 100644
--- a/configure.ac
+++ b/configure.ac
@@ -402,11 +402,6 @@ AC_CHECK_TYPE([struct sockpeercred],
[], [[#include <sys/socket.h>
]])
-AC_CHECK_DECLS([ETH_FLAG_TXVLAN, ETH_FLAG_NTUPLE, ETH_FLAG_RXHASH, ETH_FLAG_LRO,
- ETHTOOL_GGSO, ETHTOOL_GGRO, ETHTOOL_GFLAGS, ETHTOOL_GFEATURES],
- [], [], [[#include <linux/ethtool.h>
- ]])
-
dnl Our only use of libtasn1.h is in the testsuite, and can be skipped
dnl if the header is not present. Assume -ltasn1 is present if the
dnl header could be found.
diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c
index fa695d4..c3a35bb 100644
--- a/src/util/virnetdev.c
+++ b/src/util/virnetdev.c
@@ -85,7 +85,7 @@ VIR_LOG_INIT("util.netdev");
#endif
#define RESOURCE_FILE_LEN 4096
-#if HAVE_DECL_ETHTOOL_GFEATURES
+#ifdef ETHTOOL_GFEATURES
# define TX_UDP_TNL 25
# define GFEATURES_SIZE 2
# define FEATURE_WORD(blocks, index, field) ((blocks)[(index) / 32U].field)
@@ -2446,28 +2446,28 @@ virNetDevGetEthtoolFeatures(virBitmapPtr bitmap,
{ETHTOOL_GTXCSUM, VIR_NET_DEV_FEAT_GTXCSUM},
{ETHTOOL_GSG, VIR_NET_DEV_FEAT_GSG},
{ETHTOOL_GTSO, VIR_NET_DEV_FEAT_GTSO},
-# if HAVE_DECL_ETHTOOL_GGSO
+# ifdef ETHTOOL_GGSO
{ETHTOOL_GGSO, VIR_NET_DEV_FEAT_GGSO},
# endif
-# if HAVE_DECL_ETHTOOL_GGRO
+# ifdef ETHTOOL_GGRO
{ETHTOOL_GGRO, VIR_NET_DEV_FEAT_GGRO},
# endif
};
-# if HAVE_DECL_ETHTOOL_GFLAGS
+# ifdef ETHTOOL_GFLAGS
/* ethtool masks */
struct virNetDevEthtoolFeatureCmd flags[] = {
-# if HAVE_DECL_ETH_FLAG_LRO
+# ifdef ETH_FLAG_LRO
{ETH_FLAG_LRO, VIR_NET_DEV_FEAT_LRO},
# endif
-# if HAVE_DECL_ETH_FLAG_TXVLAN
+# ifdef ETH_FLAG_TXVLAN
{ETH_FLAG_RXVLAN, VIR_NET_DEV_FEAT_RXVLAN},
{ETH_FLAG_TXVLAN, VIR_NET_DEV_FEAT_TXVLAN},
# endif
-# if HAVE_DECL_ETH_FLAG_NTUBLE
+# ifdef ETH_FLAG_NTUPLE
{ETH_FLAG_NTUPLE, VIR_NET_DEV_FEAT_NTUPLE},
# endif
-# if HAVE_DECL_ETH_FLAG_RXHASH
+# ifdef ETH_FLAG_RXHASH
{ETH_FLAG_RXHASH, VIR_NET_DEV_FEAT_RXHASH},
# endif
};
@@ -2479,7 +2479,7 @@ virNetDevGetEthtoolFeatures(virBitmapPtr bitmap,
ignore_value(virBitmapSetBit(bitmap, ethtool_cmds[i].feat));
}
-# if HAVE_DECL_ETHTOOL_GFLAGS
+# ifdef ETHTOOL_GFLAGS
cmd.cmd = ETHTOOL_GFLAGS;
if (virNetDevFeatureAvailable(fd, ifr, &cmd)) {
for (i = 0; i < ARRAY_CARDINALITY(flags); i++) {
@@ -2491,7 +2491,7 @@ virNetDevGetEthtoolFeatures(virBitmapPtr bitmap,
}
-# if HAVE_DECL_ETHTOOL_GFEATURES
+# ifdef ETHTOOL_GFEATURES
/**
* virNetDevGFeatureAvailable
* This function checks for the availability of a network device gfeature
--
2.7.3
12:17 a.m.
New subject: [libvirt] [PATCH 2/7] configure.ac: drop checks for ETH flags
On 08/03/2016 12:20 PM, Ján Tomko wrote:
No choices are made at configure time based on these checks.
AC_CHECK_DECLS doesn't make sense if you're using it to check for a
#defined name just so you can use the new name that is #defined as a
result of the check. That's what's being done here with ETHTOOL_*, so
those are pointless.
However, it *can* be useful if you want to check for existence of a
function name, global data name, data type, or enum value at compile
time (since those *can't* be directly checked with #ifdef - it will
always return false). That's what's being done with ETH_FLAG_* (they are
all enum values), so I think those have to stay (if you apply this
patch, you end up trying to check for them directly with #ifdef, and the
result is that several of the items are omitted from the
virNetDevEthtoolFeatureCmd flags array).
Drop them and use #ifdefs in virnetdev.c.
---
configure.ac | 5 -----
src/util/virnetdev.c | 20 ++++++++++----------
2 files changed, 10 insertions(+), 15 deletions(-)
diff --git a/configure.ac b/configure.ac
index 8d7d63e..db42173 100644
--- a/configure.ac
+++ b/configure.ac
@@ -402,11 +402,6 @@ AC_CHECK_TYPE([struct sockpeercred],
[], [[#include <sys/socket.h>
]])
-AC_CHECK_DECLS([ETH_FLAG_TXVLAN, ETH_FLAG_NTUPLE, ETH_FLAG_RXHASH, ETH_FLAG_LRO,
- ETHTOOL_GGSO, ETHTOOL_GGRO, ETHTOOL_GFLAGS, ETHTOOL_GFEATURES],
- [], [], [[#include <linux/ethtool.h>
- ]])
-
dnl Our only use of libtasn1.h is in the testsuite, and can be skipped
dnl if the header is not present. Assume -ltasn1 is present if the
dnl header could be found.
diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c
index fa695d4..c3a35bb 100644
--- a/src/util/virnetdev.c
+++ b/src/util/virnetdev.c
@@ -85,7 +85,7 @@ VIR_LOG_INIT("util.netdev");
#endif
#define RESOURCE_FILE_LEN 4096
-#if HAVE_DECL_ETHTOOL_GFEATURES
+#ifdef ETHTOOL_GFEATURES
# define TX_UDP_TNL 25
# define GFEATURES_SIZE 2
# define FEATURE_WORD(blocks, index, field) ((blocks)[(index) / 32U].field)
@@ -2446,28 +2446,28 @@ virNetDevGetEthtoolFeatures(virBitmapPtr bitmap,
{ETHTOOL_GTXCSUM, VIR_NET_DEV_FEAT_GTXCSUM},
{ETHTOOL_GSG, VIR_NET_DEV_FEAT_GSG},
{ETHTOOL_GTSO, VIR_NET_DEV_FEAT_GTSO},
-# if HAVE_DECL_ETHTOOL_GGSO
+# ifdef ETHTOOL_GGSO
{ETHTOOL_GGSO, VIR_NET_DEV_FEAT_GGSO},
# endif
-# if HAVE_DECL_ETHTOOL_GGRO
+# ifdef ETHTOOL_GGRO
{ETHTOOL_GGRO, VIR_NET_DEV_FEAT_GGRO},
# endif
};
-# if HAVE_DECL_ETHTOOL_GFLAGS
+# ifdef ETHTOOL_GFLAGS
/* ethtool masks */
struct virNetDevEthtoolFeatureCmd flags[] = {
-# if HAVE_DECL_ETH_FLAG_LRO
+# ifdef ETH_FLAG_LRO
{ETH_FLAG_LRO, VIR_NET_DEV_FEAT_LRO},
# endif
-# if HAVE_DECL_ETH_FLAG_TXVLAN
+# ifdef ETH_FLAG_TXVLAN
{ETH_FLAG_RXVLAN, VIR_NET_DEV_FEAT_RXVLAN},
{ETH_FLAG_TXVLAN, VIR_NET_DEV_FEAT_TXVLAN},
# endif
-# if HAVE_DECL_ETH_FLAG_NTUBLE
+# ifdef ETH_FLAG_NTUPLE
{ETH_FLAG_NTUPLE, VIR_NET_DEV_FEAT_NTUPLE},
# endif
-# if HAVE_DECL_ETH_FLAG_RXHASH
+# ifdef ETH_FLAG_RXHASH
{ETH_FLAG_RXHASH, VIR_NET_DEV_FEAT_RXHASH},
# endif
};
@@ -2479,7 +2479,7 @@ virNetDevGetEthtoolFeatures(virBitmapPtr bitmap,
ignore_value(virBitmapSetBit(bitmap, ethtool_cmds[i].feat));
}
-# if HAVE_DECL_ETHTOOL_GFLAGS
+# ifdef ETHTOOL_GFLAGS
cmd.cmd = ETHTOOL_GFLAGS;
if (virNetDevFeatureAvailable(fd, ifr, &cmd)) {
for (i = 0; i < ARRAY_CARDINALITY(flags); i++) {
@@ -2491,7 +2491,7 @@ virNetDevGetEthtoolFeatures(virBitmapPtr bitmap,
}
-# if HAVE_DECL_ETHTOOL_GFEATURES
+# ifdef ETHTOOL_GFEATURES
/**
* virNetDevGFeatureAvailable
* This function checks for the availability of a network device gfeature
11:20 a.m.
New subject: [libvirt] [PATCH 3/7] tests: fix the return value of test-wrap-argv
The script was returning success unless it failed on the last file.
This went unnoticed because sc_prohibit_long_lines forbids lines
longer than 90 characters in .arg[sv] files.
---
tests/test-wrap-argv.pl | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/test-wrap-argv.pl b/tests/test-wrap-argv.pl
index 2898156..f0d3c0b 100755
--- a/tests/test-wrap-argv.pl
+++ b/tests/test-wrap-argv.pl
@@ -39,8 +39,8 @@ if (@ARGV[0] eq "--in-place") {
shift @ARGV;
}
+$ret = 0;
foreach my $file (@ARGV) {
- $ret = 0;
if (&rewrap($file) < 0) {
$ret = 1;
}
--
2.7.3
12:20 a.m.
New subject: [libvirt] [PATCH 3/7] tests: fix the return value of test-wrap-argv
On 08/03/2016 12:20 PM, Ján Tomko wrote:
The script was returning success unless it failed on the last file.
This went unnoticed because sc_prohibit_long_lines forbids lines
longer than 90 characters in .arg[sv] files.
---
tests/test-wrap-argv.pl | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/test-wrap-argv.pl b/tests/test-wrap-argv.pl
index 2898156..f0d3c0b 100755
--- a/tests/test-wrap-argv.pl
+++ b/tests/test-wrap-argv.pl
@@ -39,8 +39,8 @@ if (@ARGV[0] eq "--in-place") {
shift @ARGV;
}
+$ret = 0;
foreach my $file (@ARGV) {
- $ret = 0;
if (&rewrap($file) < 0) {
$ret = 1;
}
ACK.
11:20 a.m.
New subject: [libvirt] [PATCH 4/7] cfg.mk: use subst instead of tr
GNU make is able to replace characters, no need to call tr.
---
cfg.mk | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/cfg.mk b/cfg.mk
index cf47d4c..0604d69 100644
--- a/cfg.mk
+++ b/cfg.mk
@@ -614,8 +614,9 @@ msg_gen_function += xenapiSessionErrorHandler
# msg_gen_function += vshPrint
# msg_gen_function += vshError
-func_or := $(shell echo $(msg_gen_function)|tr -s ' ' '|')
-func_re := ($(func_or))
+space =
+space +=
+func_re= ($(subst $(space),|,$(msg_gen_function)))
# Look for diagnostics that aren't marked for translation.
# This won't find any for which error's format string is on a separate line.
--
2.7.3
11:20 a.m.
New subject: [libvirt] [PATCH 5/7] cfg.mk: drop redundant sc_prohibit_gethostby
Both gethostbyaddr and gethostbyname* are already checked
by sc_prohibit_nonreentrant.
---
cfg.mk | 9 ---------
1 file changed, 9 deletions(-)
diff --git a/cfg.mk b/cfg.mk
index 0604d69..9f236c2 100644
--- a/cfg.mk
+++ b/cfg.mk
@@ -520,13 +520,6 @@ sc_forbid_manual_xml_indent:
halt='use virBufferAdjustIndent instead of spaces when indenting xml' \
$(_sc_search_regexp)
-# Not only do they fail to deal well with ipv6, but the gethostby*
-# functions are also not thread-safe.
-sc_prohibit_gethostby:
- @prohibit='\<gethostby(addr|name2?) *\(' \
- halt='use getaddrinfo, not gethostby*' \
- $(_sc_search_regexp)
-
# dirname and basename from <libgen.h> are not required to be thread-safe
sc_prohibit_libgen:
@prohibit='( (base|dir)name *\(|include .libgen\.h)' \
@@ -1192,8 +1185,6 @@ exclude_file_name_regexp--sc_prohibit_strncpy =
^src/util/virstring\.c$$
exclude_file_name_regexp--sc_prohibit_strtol = ^examples/.*$$
-exclude_file_name_regexp--sc_prohibit_gethostby = ^docs/nss.html.in$$
-
exclude_file_name_regexp--sc_prohibit_xmlGetProp = ^src/util/virxml\.c$$
exclude_file_name_regexp--sc_prohibit_xmlURI = ^src/util/viruri\.c$$
--
2.7.3
12:05 a.m.
New subject: [libvirt] [PATCH 5/7] cfg.mk: drop redundant sc_prohibit_gethostby
On 08/03/2016 12:20 PM, Ján Tomko wrote:
Both gethostbyaddr and gethostbyname* are already checked
by sc_prohibit_nonreentrant.
The really interesting thing is that sc_prohibit_nonreentrant was
already checking for those functions *before* sc_prohibit_gethostby was
added.
ACK.
---
cfg.mk | 9 ---------
1 file changed, 9 deletions(-)
diff --git a/cfg.mk b/cfg.mk
index 0604d69..9f236c2 100644
--- a/cfg.mk
+++ b/cfg.mk
@@ -520,13 +520,6 @@ sc_forbid_manual_xml_indent:
halt='use virBufferAdjustIndent instead of spaces when indenting xml' \
$(_sc_search_regexp)
-# Not only do they fail to deal well with ipv6, but the gethostby*
-# functions are also not thread-safe.
-sc_prohibit_gethostby:
- @prohibit='\<gethostby(addr|name2?) *\(' \
- halt='use getaddrinfo, not gethostby*' \
- $(_sc_search_regexp)
-
# dirname and basename from <libgen.h> are not required to be thread-safe
sc_prohibit_libgen:
@prohibit='( (base|dir)name *\(|include .libgen\.h)' \
@@ -1192,8 +1185,6 @@ exclude_file_name_regexp--sc_prohibit_strncpy =
^src/util/virstring\.c$$
exclude_file_name_regexp--sc_prohibit_strtol = ^examples/.*$$
-exclude_file_name_regexp--sc_prohibit_gethostby = ^docs/nss.html.in$$
-
exclude_file_name_regexp--sc_prohibit_xmlGetProp = ^src/util/virxml\.c$$
exclude_file_name_regexp--sc_prohibit_xmlURI = ^src/util/viruri\.c$$
11:20 a.m.
New subject: [libvirt] [PATCH 6/7] cfg.mk: drop redundant sc_size_of_brackets
This check forbids a space after sizeof, which is already done
by check-spacing.pl.
---
build-aux/check-spacing.pl | 2 +-
cfg.mk | 6 ------
2 files changed, 1 insertion(+), 7 deletions(-)
diff --git a/build-aux/check-spacing.pl b/build-aux/check-spacing.pl
index 448acf2..afc7758 100755
--- a/build-aux/check-spacing.pl
+++ b/build-aux/check-spacing.pl
@@ -94,7 +94,7 @@ foreach my $file (@ARGV) {
if ($kw =~ /^(?:if|for|while|switch|return)$/) {
$tmpdata =~ s/(?:$kw\s\()/XXX(/;
} else {
- print "Whitespace after non-keyword:\n";
+ print "Whitespace after non-keyword or sizeof:\n";
print "$file:$.: $line";
$ret = 1;
last;
diff --git a/cfg.mk b/cfg.mk
index 9f236c2..3405c9a 100644
--- a/cfg.mk
+++ b/cfg.mk
@@ -476,12 +476,6 @@ sc_forbid_const_pointer_typedef:
halt='use correct style and type for Ptr typedefs' \
$(_sc_search_regexp)
-# Forbid sizeof foo or sizeof (foo), require sizeof(foo)
-sc_size_of_brackets:
- @prohibit='sizeof\s' \
- halt='use sizeof(foo), not sizeof (foo) or sizeof foo' \
- $(_sc_search_regexp)
-
# Ensure that no C source file, docs, or rng schema uses TABs for
# indentation. Also match *.h.in files, to get libvirt.h.in. Exclude
# files in gnulib, since they're imported.
--
2.7.3
12:39 a.m.
New subject: [libvirt] [PATCH 6/7] cfg.mk: drop redundant sc_size_of_brackets
On 08/03/2016 12:20 PM, Ján Tomko wrote:
This check forbids a space after sizeof, which is already done
by check-spacing.pl.
---
build-aux/check-spacing.pl | 2 +-
cfg.mk | 6 ------
2 files changed, 1 insertion(+), 7 deletions(-)
diff --git a/build-aux/check-spacing.pl b/build-aux/check-spacing.pl
index 448acf2..afc7758 100755
--- a/build-aux/check-spacing.pl
+++ b/build-aux/check-spacing.pl
@@ -94,7 +94,7 @@ foreach my $file (@ARGV) {
if ($kw =~ /^(?:if|for|while|switch|return)$/) {
$tmpdata =~ s/(?:$kw\s\()/XXX(/;
} else {
- print "Whitespace after non-keyword:\n";
+ print "Whitespace after non-keyword or sizeof:\n";
print "$file:$.: $line";
$ret = 1;
last;
The problem is that check-spacing.pl doesn't check for "sizeof blah",
which is one of the things prohibited by sc_sizeof_brackets.
So NACK on this one.
diff --git a/cfg.mk b/cfg.mk
index 9f236c2..3405c9a 100644
--- a/cfg.mk
+++ b/cfg.mk
@@ -476,12 +476,6 @@ sc_forbid_const_pointer_typedef:
halt='use correct style and type for Ptr typedefs' \
$(_sc_search_regexp)
-# Forbid sizeof foo or sizeof (foo), require sizeof(foo)
-sc_size_of_brackets:
- @prohibit='sizeof\s' \
- halt='use sizeof(foo), not sizeof (foo) or sizeof foo' \
- $(_sc_search_regexp)
-
# Ensure that no C source file, docs, or rng schema uses TABs for
# indentation. Also match *.h.in files, to get libvirt.h.in. Exclude
# files in gnulib, since they're imported.
11:20 a.m.
New subject: [libvirt] [PATCH 7/7] cfg.mk: join not_streq and not_strneq tests
The marginally nicer error message is not worth the extra lines in
cfg.mk.
Also drop the excludes since there was only one offender in the tests.
---
cfg.mk | 15 ++-------------
tests/virstoragetest.c | 2 +-
2 files changed, 3 insertions(+), 14 deletions(-)
diff --git a/cfg.mk b/cfg.mk
index 3405c9a..42123c3 100644
--- a/cfg.mk
+++ b/cfg.mk
@@ -980,13 +980,8 @@ sc_prohibit_pthread_create:
$(_sc_search_regexp)
sc_prohibit_not_streq:
- @prohibit='! *STREQ *\(.*\)' \
- halt='Use STRNEQ instead of !STREQ' \
- $(_sc_search_regexp)
-
-sc_prohibit_not_strneq:
- @prohibit='! *STRNEQ *\(.*\)' \
- halt='Use STREQ instead of !STRNEQ' \
+ @prohibit='! *STRN?EQ *\(.*\)' \
+ halt='Use STRNEQ instead of !STREQ and STREQ instead of !STRNEQ' \
$(_sc_search_regexp)
sc_prohibit_verbose_strcat:
@@ -1240,12 +1235,6 @@ exclude_file_name_regexp--sc_prohibit_sysconf_pagesize = \
exclude_file_name_regexp--sc_prohibit_pthread_create = \
^(cfg\.mk|src/util/virthread\.c|tests/.*)$$
-exclude_file_name_regexp--sc_prohibit_not_streq = \
- ^tests/.*\.[ch]$$
-
-exclude_file_name_regexp--sc_prohibit_not_strneq = \
- ^tests/.*\.[ch]$$
-
exclude_file_name_regexp--sc_prohibit_dt_without_code = \
^docs/(newapi\.xsl|(apps|contact)\.html\.in)$$
diff --git a/tests/virstoragetest.c b/tests/virstoragetest.c
index 3b19f59..184b936 100644
--- a/tests/virstoragetest.c
+++ b/tests/virstoragetest.c
@@ -698,7 +698,7 @@ testBackingParse(const void *args)
goto cleanup;
}
- if (!STREQ(xml, data->expect)) {
+ if (STRNEQ(xml, data->expect)) {
fprintf(stderr, "\n backing store string '%s'\n"
"expected storage source xml:\n%s\n"
"actual storage source xml:\n%s\n",
--
2.7.3
4:53 p.m.
On 08/03/2016 12:20 PM, Ján Tomko wrote:
The first patch pulls the latest gnulibs to fetch the newest
syntax-check
speedups.
Patch 2 is included since a gnulib update will trigger configure anyway.
The rest are mostly cleanups, the speed-ups they (might) provide are barely
measurable.
Ján Tomko (7):
maint: update to latest gnulib
configure.ac: drop checks for ETH flags
tests: fix the return value of test-wrap-argv
cfg.mk: use subst instead of tr
cfg.mk: drop redundant sc_prohibit_gethostby
cfg.mk: drop redundant sc_size_of_brackets
cfg.mk: join not_streq and not_strneq tests
.gnulib | 2 +-
build-aux/check-spacing.pl | 2 +-
cfg.mk | 35 +++++------------------------------
configure.ac | 5 -----
src/util/virnetdev.c | 20 ++++++++++----------
tests/test-wrap-argv.pl | 2 +-
tests/virstoragetest.c | 2 +-
7 files changed, 19 insertions(+), 49 deletions(-)
ACK series,
John
12:40 a.m.
On 08/17/2016 05:53 PM, John Ferlan wrote:
On 08/03/2016 12:20 PM, Ján Tomko wrote:
> The first patch pulls the latest gnulibs to fetch the newest syntax-check
> speedups.
>
> Patch 2 is included since a gnulib update will trigger configure anyway.
>
> The rest are mostly cleanups, the speed-ups they (might) provide are barely
> measurable.
>
> Ján Tomko (7):
> maint: update to latest gnulib
> configure.ac: drop checks for ETH flags
> tests: fix the return value of test-wrap-argv
> cfg.mk: use subst instead of tr
> cfg.mk: drop redundant sc_prohibit_gethostby
> cfg.mk: drop redundant sc_size_of_brackets
> cfg.mk: join not_streq and not_strneq tests
>
> .gnulib | 2 +-
> build-aux/check-spacing.pl | 2 +-
> cfg.mk | 35 +++++------------------------------
> configure.ac | 5 -----
> src/util/virnetdev.c | 20 ++++++++++----------
> tests/test-wrap-argv.pl | 2 +-
> tests/virstoragetest.c | 2 +-
> 7 files changed, 19 insertions(+), 49 deletions(-)
>
ACK series,
Except 2/7 and 6/7. (2/7 is especially bad, since it will silently
remove functionality).
3049
days inactive
3064
days old
15 comments
3 participants
participants (3)
-
John Ferlan -
Ján Tomko -
Laine Stump