waf:replace: Do not link against libpthread if not necessary
On Linux we should avoid linking everything against libpthread. Symbols
used my most application are provided by glibc and code which deals with
threads has to explicitly link against libpthread. This avoids setting
LDFLAGS=-pthread globally.
pthreadpool: Only link pthreadpool against librt if we have to
This calls clock_gettime() which is available in glibc on Linux. If the
wscript in libreplace detected that librt is needed for clock_gettime()
we have to link against it.
fdatasync() and clock_gettime() are provided by glibc on Linux, so there
is no need to link against librt. Checks have been added so if there are
platforms which require it are still functional.
Philipp Gesang [Thu, 19 Sep 2019 07:03:45 +0000 (09:03 +0200)]
s4:scripting: make samba-gpupdate heed --disable-python
Signed-off-by: Philipp Gesang <philipp.gesang@intra2net.com> Reviewed-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Noel Power<npower@samba.org>
Autobuild-User(master): Noel Power <npower@samba.org>
Autobuild-Date(master): Wed Sep 25 11:06:04 UTC 2019 on sn-devel-184
Noel Power [Thu, 8 Aug 2019 13:15:14 +0000 (13:15 +0000)]
s3/librpc/crypto: clang: Fixes Value stored to 'ret' is never read
Fixes:
source3/librpc/crypto/gse_krb5.c:63:3: warning: Value stored to 'ret' is never read <--[clang]
ret = smb_krb5_kt_free_entry(krbctx, &kt_entry);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1 warning generated.
Signed-off-by: Noel Power <noel.power@suse.com> Reviewed-by: Andreas Schneider <asn@samba.org>
Noel Power [Fri, 2 Aug 2019 10:45:48 +0000 (10:45 +0000)]
s3/librpc/crypto: clang: warning: Value stored to 'gss_maj' is never read
Fixes: Value stored to 'gss_maj' is never read
source3/librpc/crypto/gse.c:562:3: warning: Value stored to 'gss_maj' is never read <--[clang]
gss_maj = gss_release_buffer(&gss_min, &out_data);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
source3/librpc/crypto/gse.c:687:3: warning: Value stored to 'gss_maj' is never read <--[clang]
gss_maj = gss_release_buffer(&gss_min, &out_data);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
source3/librpc/crypto/gse.c:739:3: warning: Value stored to 'gss_maj' is never read <--[clang]
gss_maj = gss_release_buffer(&gss_min, &msg_min);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
source3/librpc/crypto/gse.c:742:3: warning: Value stored to 'gss_maj' is never read <--[clang]
gss_maj = gss_release_buffer(&gss_min, &msg_maj);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4 warnings generated.
Signed-off-by: Noel Power <noel.power@suse.com> Reviewed-by: Andreas Schneider <asn@samba.org>
Noel Power [Thu, 8 Aug 2019 14:10:30 +0000 (15:10 +0100)]
s4/libcli/clideltree.c: clang: Value stored to 'status' is never read
Fixes:
source4/libcli/clideltree.c:113:3: warning: Value stored to 'status' is never read <--[clang]
status = smbcli_setatr(tree, dname, FILE_ATTRIBUTE_NORMAL, 0);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
source4/libcli/clideltree.c:131:3: warning: Value stored to 'status' is never read <--[clang]
status = smbcli_setatr(dstate.tree, dname, FILE_ATTRIBUTE_NORMAL, 0);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2 warnings generated.
Signed-off-by: Noel Power <noel.power@suse.com> Reviewed-by: Andreas Schneider <asn@samba.org>
Noel Power [Mon, 29 Jul 2019 11:44:26 +0000 (11:44 +0000)]
s4/libcli/smb2: clang: Fix 'value stored to 'status' is never read'
Fixes:
source4/libcli/smb2/util.c:134:3: warning: Value stored to 'status' is never read <--[clang]
status = smb2_util_setatr(tree, dname, FILE_ATTRIBUTE_NORMAL);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
source4/libcli/smb2/util.c:187:5: warning: Value stored to 'status' is never read <--[clang]
status = smb2_util_setatr(tree, name, FILE_ATTRIBUTE_NORMAL);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
source4/libcli/smb2/util.c:209:3: warning: Value stored to 'status' is never read <--[clang]
status = smb2_util_setatr(tree, dname, FILE_ATTRIBUTE_NORMAL);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3 warnings generated.
Signed-off-by: Noel Power <noel.power@suse.com> Reviewed-by: Andreas Schneider <asn@samba.org>
Noel Power [Fri, 26 Jul 2019 16:23:45 +0000 (16:23 +0000)]
s4/libcli/raw: clang: Fix 'Value stored to 'p' is never read'
Fixes:
source4/libcli/raw/clisession.c:198:4: warning: Value stored to 'p' is never read <--[clang]
p += smbcli_req_pull_string(&req->in.bufinfo, mem_ctx, &parms->old.out.domain, p, -1, STR_TERMINATE);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
source4/libcli/raw/clisession.c:212:5: warning: Value stored to 'p' is never read <--[clang]
p += smbcli_req_pull_string(&req->in.bufinfo, mem_ctx, &parms->nt1.out.domain, p, -1, STR_TERMINATE);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
source4/libcli/raw/clisession.c:232:3: warning: Value stored to 'p' is never read <--[clang]
p += smbcli_req_pull_string(&req->in.bufinfo, mem_ctx, &parms->spnego.out.workgroup, p, -1, STR_TERMINATE);
^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Noel Power <noel.power@suse.com> Reviewed-by: Andreas Schneider <asn@samba.org>
Noel Power [Fri, 26 Jul 2019 15:06:54 +0000 (15:06 +0000)]
s4/libcli/raw: clang: Fix 'Access to field results in a deref of a null ptr
Fixes:
source4/libcli/raw/rawfileinfo.c:301:56: warning: Access to field 'transport' results in a dereference of a null pointer (loaded from variable 'session') <--[clang]
parms->standard.out.create_time = raw_pull_dos_date2(session->transport,
^
source4/libcli/raw/rawfileinfo.c:314:55: warning: Access to field 'transport' results in a dereference of a null pointer (loaded from variable 'session') <--[clang]
parms->ea_size.out.create_time = raw_pull_dos_date2(session->transport,
Signed-off-by: Noel Power <noel.power@suse.com> Reviewed-by: Andreas Schneider <asn@samba.org>
s3:libads: ads_krb5_chg_password() should always use the canonicalized principal
We should always use krb5_get_init_creds_opt_set_canonicalize()
and krb5_get_init_creds_opt_set_win2k() for heimdal
and expect the client principal to be changed.
There's no reason to have a different logic between MIT and Heimdal.
s4:auth: kinit_to_ccache() should always use the canonicalized principal
We should always use krb5_get_init_creds_opt_set_canonicalize()
and krb5_get_init_creds_opt_set_win2k() for heimdal
and expect the client principal to be changed.
There's no reason to have a different logic between MIT and Heimdal.
krb5_wrap: smb_krb5_kinit_password_ccache() should always use the canonicalized principal
We should always use krb5_get_init_creds_opt_set_canonicalize()
and krb5_get_init_creds_opt_set_win2k() for heimdal
and expect the client principal to be changed.
There's no reason to have a different logic between MIT and Heimdal.
s3:libads/kerberos: always use the canonicalized principal after kinit
We should always use krb5_get_init_creds_opt_set_canonicalize()
and krb5_get_init_creds_opt_set_win2k() for heimdal
and expect the client principal to be changed.
There's no reason to have a different logic between MIT and Heimdal.
s4:auth: use the correct client realm in gensec_gssapi_update_internal()
The function gensec_gssapi_client_creds() may call kinit and gets
a TGT for the user. The principal provided by the user may not
be canonicalized. The user may use 'given.last@example.com'
but that may be mapped to glast@AD.EXAMPLE.PRIVATE in the background.
It means we should use client_realm = AD.EXAMPLE.PRIVATE
instead of client_realm = EXAMPLE.COM
Martin Schwenke [Thu, 5 Sep 2019 03:57:35 +0000 (13:57 +1000)]
ctdb-tests: Switch simple tests to use CTDB_TEST_TMP_DIR
CTDB_TEST_TMP_DIR repaces SIMPLE_TESTS_VAR_DIR. local.bash no longer
needs to ensure that TEST_VAR_DIR is set, since it longer uses this
variable. Drop the comment because state has not been maintained
between tests for some time.
Signed-off-by: Martin Schwenke <martin@meltin.net> Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Martin Schwenke [Thu, 5 Sep 2019 05:47:13 +0000 (15:47 +1000)]
ctdb-tests: Use local_daemons.sh onnode for local daemons tests
With some upcoming changes, the setting of CTDB_BASE becomes
problematic because it will be included unconditionally whereas it is
currently being conveniently and almost accidentally not include in
some contexts.
So, instead of trying to coerce onnode into behaving as desired, have
the local daemons tests use local_daemons.sh onnode directly.
Signed-off-by: Martin Schwenke <martin@meltin.net> Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Signed-off-by: Bjoern Jacke <bjacke@samba.org> Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Sat Sep 21 20:50:17 UTC 2019 on sn-devel-184
projects / thirdparty / samba.git / log
