- Devel - libvirt List Archives

[Libvir] documentation on the website is not in sync
by Gildas 15 Nov '07

15 Nov '07

Hi, As reported on the irc channel, the documentation at http://libvirt.org/uri.html#URI_qemu is not in sync with what is reported/supported by libvirtd v 0.3.3 Regards, Gildas

3 2

[Libvir] The problem of the definition of tuning informations
by Daniel Veillard 15 Nov '07

15 Nov '07

I promised that mail for the beginning of the week but I still have a very hard time to try to formulate a good plan of action, I'm still stuck in a dilemna, see below. What is it? ----------- I think tuning informations are that set of parameters associated to a domain or a host, which are not stricly needed to get the domain(s) working but improve their runtime behaviour. To me this includes: - scheduling parameters the scope may be host/hypervisor/domain - vcpu affinity i.e. to which set of physical CPU each of the vcpu may be bound - and possibly others ... The problem: ------------ People would like to associate those to the XML domain informations, the goal being to be able to restore those informations when a domain (re-)starts. I have been objecting it so far because, I think those informations don't have the same lifetime and scope as the other domain informations saved in the XML. Since they are not needed to start the domain, and that once the domain is started the existing domain API can be used to change those informations, it is better to keep them separate. However I got objections from David Lutterkort [1], Jim Fehlig [2], and John Levon [3] plus of course the initial request for it from Tatsuro Enokura (and the Fujistu people in general) [4] The problem to me comes from 2 things: 1/ storing tuning informations in domains descriptions is not sufficient 2/ if we store them there we also need to always save them when exporting the XML domain file 2/ is fairly important to avoid a lot of problems as we have experienced before for example with console informations. If the input for virsh create gets different from the output of dumpxml, a lot of rather annoying things happen in practice, it certainly generate confusion. So we really need to output those tuning data if we put them in. Also I strongly believe in 1/, i.e. tuning informations are cross domain and they are vey likely to change fast as soon as the management applications will get deployed, but even in relatively small deployment the tuning is rather a per host informations, which may depend on the current workload of the machine. I don't believe in tuning being loaded at create time and never changing later. Even in my own very basic usage that doesn't match my use which lead me to load and stop domain on demand for short period of time. My opinion: ----------- We need better tools, even for simple use case to be able to save an existing tuning for a domain or a full machine, and reload it when needed. This is IMHO better done on top of the existing API which already have the entry points to implement them. My idea is to provide tuning commands in virsh [5]. If you implement tuning both at creation time and in the tool, this mean you either make them different in which case you have no coherency between what you say when you create a domain or save its config and what you do at the virsh level. If you don't make it different (for example trying to use the same kind of XML syntax), then you need code for doing this both in the tool and in the library itself, or you export as a new API the tuning load and save. Exporting as a parallel API what we have already for scheduling and VCPU affinity makes the API more complex, and less coherent. I don't want to force the decision one way or another, it is probable I missed something, but I don't think adding tuning informations to the domain configuration file to be really that convenient, I could be done better, and with less associated problems by keeping those separate. Daniel [1] https://www.redhat.com/archives/libvir-list/2007-October/msg00250.html [2] https://www.redhat.com/archives/libvir-list/2007-November/msg00003.html [3] https://www.redhat.com/archives/libvir-list/2007-October/msg00046.html [4] https://www.redhat.com/archives/libvir-list/2007-October/msg00221.html [5] https://www.redhat.com/archives/libvir-list/2007-October/msg00245.html -- Red Hat Virtualization group http://redhat.com/virtualization/ Daniel Veillard | virtualization library http://libvirt.org/ veillard(a)redhat.com | libxml GNOME XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/

5 13

[Libvir] latest NUMA/cpuset code testing
by beth kon 15 Nov '07

15 Nov '07

I tested the latest CVS libvirt on a 128-way x3950 and create, define, and start appear to work well with various cpusets specified. The only thing I noticed was that dumpxml does not grab the cpuset info. I have not looked at the code to verify, but is this expected? Maybe you discussed this Daniel, but I can't remember at the moment. -- Elizabeth Kon (Beth) IBM Linux Technology Center Open Hypervisor Team email: eak(a)us.ibm.com

3 2

[Libvir] ANNOUNCE: ocaml-libvirt 0.3.3.1 released
by Richard W.M. Jones 14 Nov '07

14 Nov '07

I'm pleased to announce the release of ocaml-libvirt 0.3.3.1, type-safe libvirt bindings for OCaml. Main page: http://et.redhat.com/~rjones/ocaml-libvirt/ Source repo: http://hg.et.redhat.com/virt/applications/virt-top--devel Change log: http://et.redhat.com/~rjones/ocaml-libvirt/ChangeLog.txt This minor update contains bindings for the new API calls virNodeGetFreeMemory and virNodeGetCellsFreeMemory. http://et.redhat.com/~rjones/ocaml-libvirt/html/Libvirt.Connect.html#VALnod… The 'mlvirsh' command now has a new 'freecell' subcommand: mlvirsh freecell - Calls virNodeGetFreeMemory and prints free memory in the whole system. mlvirsh freecell <start> - Calls virNodeGetCellsFreeMemory (conn, start, 1) and prints free memory on a single NUMA cell. mlvirsh freecell <start> <max> - Calls virNodeGetCellsFreeMemory (conn, start, max) and prints free memory on the given range of NUMA cells. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

1 1

[Libvir] [PATCH] Handle failed strdup and malloc.
by Jim Meyering 14 Nov '07

14 Nov '07

I noticed a bunch of unchecked strdup's in a row, and audited the rest of the file: Handle failed strdup and malloc. * src/remote_internal.c: Don't dereference NULL after failed strdup or malloc in doRemoteOpen. --- src/remote_internal.c | 22 ++++++++++++++++++++++ 1 files changed, 22 insertions(+), 0 deletions(-) diff --git a/src/remote_internal.c b/src/remote_internal.c index 1420a88..bd8b5a7 100644 --- a/src/remote_internal.c +++ b/src/remote_internal.c @@ -522,6 +522,10 @@ doRemoteOpen (virConnectPtr conn, struct private_data *priv, const char *uri_str sockname = strdup (LIBVIRTD_PRIV_UNIX_SOCKET_RO); else sockname = strdup (LIBVIRTD_PRIV_UNIX_SOCKET); + if (sockname == NULL) { + error (NULL, VIR_ERR_SYSTEM_ERROR, strerror (errno)); + goto failed; + } } } @@ -576,10 +580,19 @@ doRemoteOpen (virConnectPtr conn, struct private_data *priv, const char *uri_str if (no_tty) nr_args += 5; /* For -T -o BatchMode=yes -e none */ command = command ? : strdup ("ssh"); + if (command == NULL) { + error (NULL, VIR_ERR_SYSTEM_ERROR, strerror (errno)); + goto failed; + } // Generate the final command argv[] array. // ssh -p $port [-l $username] $hostname $netcat -U $sockname [NULL] cmd_argv = malloc (nr_args * sizeof (char *)); + if (cmd_argv == NULL) { + error (NULL, VIR_ERR_SYSTEM_ERROR, strerror (errno)); + goto failed; + } + j = 0; cmd_argv[j++] = strdup (command); cmd_argv[j++] = strdup ("-p"); @@ -601,6 +614,11 @@ doRemoteOpen (virConnectPtr conn, struct private_data *priv, const char *uri_str cmd_argv[j++] = strdup (sockname ? sockname : LIBVIRTD_PRIV_UNIX_SOCKET); cmd_argv[j++] = 0; assert (j == nr_args); + for (j = 0; j < nr_args; j++) + if (cmd_argv[j] == NULL) { + error (NULL, VIR_ERR_SYSTEM_ERROR, strerror (ENOMEM)); + goto failed; + } } /*FALLTHROUGH*/ @@ -633,6 +651,10 @@ doRemoteOpen (virConnectPtr conn, struct private_data *priv, const char *uri_str // Run the external process. if (!cmd_argv) { cmd_argv = malloc (2 * sizeof (char *)); + if (cmd_argv == NULL) { + error (NULL, VIR_ERR_SYSTEM_ERROR, strerror (errno)); + goto failed; + } cmd_argv[0] = command; cmd_argv[1] = 0; } -- 1.5.3.5.666.gfb5f

2 3

[Libvir] [PATCH] Enhanced stats for fullvirt domains
by Richard W.M. Jones 14 Nov '07

14 Nov '07

This patch does a couple of primary things: Firstly it allows you to use "hda", etc. as a path for getting block device stats from fullvirt domains. Secondly it separates out the stats code into a new file called 'stats_linux.c'. The reasoning behind the name is that this code can be shared between Xen & QEMU, and that the code is Linux-specific (it never worked on Solaris, but now this is explicit). I anticipate a 'stats_solaris.c' file once I can get Solaris + Xen going on a test machine. Also we try to detect the case where the block dev stats of a fullvirt domain are stuck at 0 -- caused by there being no frontend driver connected. We detect the condition by a query to xenstore. XENVBD_MAJOR is no longer hard-coded if we can get it instead from Linux header files. This patch adds bytes written/read to block devices for Xen PV domains if available. This also corrects a bug where stats from xvdb, xvdc, .. could not be read out because the device number was being miscalculated. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

3 7

[Libvir] [PATCH] Fix xen://hostname URIs with no trailing slash
by Richard W.M. Jones 14 Nov '07

14 Nov '07

It turns out that xmlParseURI gives an error if asked to parse the string "xen://". This string arises because after a URI such as "xen://hostname" has been stripped of its server field, it becomes "xen://" on the remote end of the connection, and the failure of xmlParseURI causes the whole connection to fail. I'm not sure of the best way to fix it. The attached patch is a semi-hack which at least allows xen://hostname to work. It's not a problem for qemu because these URIs should always contain a path. Nor for test for the same reason. Not sure about OpenVZ. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 3

[Libvir] [PATCH] parse URI once in src/libvirt.c, pass parsed URI to all drivers
by Richard W.M. Jones 14 Nov '07

14 Nov '07

The basic change here is that the virDrvOpen call (the internal "open" call for drivers) now takes the parsed xmlURIPtr instead of the raw name string. typedef virDrvOpenStatus (*virDrvOpen) (virConnectPtr conn, - const char *name, + xmlURIPtr uri, int flags); So we avoid the redundant URI parsing which was going on inside all the drivers, and also the ad-hoc "does-it-look-like-a-URI" string comparisons. That's straightforward enough except that all of the drivers were saving the name string in their private data so that they could implement the virConnectGetURI call. I've changed this so that the name is copied and saved in the main virConnect structure, and virConnectGetURI will return that unless the driver wants to override it. You need another patch (coming shortly) to allow URIs like xen://localhost (without the trailing slash) to work, which was IIRC the original point of this discussion. Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 3

[Libvir] [PATCH] Parse integers more carefully, cont'd.
by Jim Meyering 13 Nov '07

13 Nov '07

Here's part two in the strtol fix-up series: Parse integers more carefully, cont'd. * qemud/qemud.c: Replace uses of strtol with uses of xstrtol_i. Avoid overflow for very large --timeout=N values. * src/nodeinfo.c: In linuxNodeInfoMemPopulate and linuxNodeInfoCPUPopulate, use xstrtol_i rather than strtol. Unlike in qemud.c, here we allow trailing "isspace", and in the case of "cpuinfo cpu MHz", also allow a "." terminator, since we ignore the decimal and any following digits. * src/internal.h: Define xstrtol_iu, too. Signed-off-by: Jim Meyering <meyering(a)redhat.com> --- qemud/qemud.c | 16 ++++++---------- src/internal.h | 19 +++++++++++++++++++ src/nodeinfo.c | 24 +++++++++++++++++++----- 3 files changed, 44 insertions(+), 15 deletions(-) diff --git a/qemud/qemud.c b/qemud/qemud.c index 8e3ba54..ee6f6ce 100644 --- a/qemud/qemud.c +++ b/qemud/qemud.c @@ -1583,9 +1583,7 @@ remoteReadConfigFile (const char *filename) p = virConfGetValue (conf, "unix_sock_ro_perms"); CHECK_TYPE ("unix_sock_ro_perms", VIR_CONF_STRING); if (p && p->str) { - char *tmp = NULL; - unix_sock_ro_perms = strtol(p->str, &tmp, 8); - if (*tmp) { + if (xstrtol_i(p->str, NULL, 8, &unix_sock_ro_perms) != 0) { qemudLog (QEMUD_ERR, "Failed to parse mode '%s'", p->str); return -1; } @@ -1594,9 +1592,7 @@ remoteReadConfigFile (const char *filename) p = virConfGetValue (conf, "unix_sock_rw_perms"); CHECK_TYPE ("unix_sock_rw_perms", VIR_CONF_STRING); if (p && p->str) { - char *tmp = NULL; - unix_sock_rw_perms = strtol(p->str, &tmp, 8); - if (*tmp) { + if (xstrtol_i(p->str, NULL, 8, &unix_sock_rw_perms) != 0) { qemudLog (QEMUD_ERR, "Failed to parse mode '%s'", p->str); return -1; } @@ -1797,10 +1793,10 @@ int main(int argc, char **argv) { break; case 't': - timeout = strtol(optarg, &tmp, 10); - if (!tmp) - timeout = -1; - if (timeout <= 0) + if (xstrtol_i(optarg, &tmp, 10, &timeout) != 0 + || timeout <= 0 + /* Ensure that we can multiply by 1000 without overflowing. */ + || timeout > INT_MAX / 1000) timeout = -1; break; diff --git a/src/internal.h b/src/internal.h index 4a31ea6..f6fb1c5 100644 --- a/src/internal.h +++ b/src/internal.h @@ -264,6 +264,25 @@ xstrtol_i(char const *s, char **end_ptr, int base, int *result) return 0; } +/* Just like xstrtol_i, above, but produce an "unsigned int" value. */ +static inline int +xstrtol_ui(char const *s, char **end_ptr, int base, unsigned int *result) +{ + unsigned long int val; + char *p; + int err; + + errno = 0; + val = strtoul(s, &p, base); + err = (errno || (!end_ptr && *p) || p == s || (unsigned int) val != val); + if (end_ptr) + *end_ptr = p; + if (err) + return -1; + *result = val; + return 0; +} + #ifdef __cplusplus } #endif /* __cplusplus */ diff --git a/src/nodeinfo.c b/src/nodeinfo.c index 01d7e28..a0a26eb 100644 --- a/src/nodeinfo.c +++ b/src/nodeinfo.c @@ -63,6 +63,8 @@ int linuxNodeInfoCPUPopulate(virConnectPtr conn, FILE *cpuinfo, virNodeInfoPtr n } nodeinfo->cpus++; } else if (STREQLEN(buf, "cpu MHz", 7)) { + char *p; + unsigned int ui; buf += 9; while (*buf && isspace(*buf)) buf++; @@ -72,8 +74,12 @@ int linuxNodeInfoCPUPopulate(virConnectPtr conn, FILE *cpuinfo, virNodeInfoPtr n "parsing cpuinfo cpu MHz"); return -1; } - nodeinfo->mhz = (unsigned int)strtol(buf+1, NULL, 10); + if (xstrtol_ui(buf+1, &p, 10, &ui) == 0 + /* Accept trailing fractional part. */ + && (*p == '\0' || *p == '.' || isspace(*p))) + nodeinfo->mhz = ui; } else if (STREQLEN(buf, "cpu cores", 9)) { /* aka cores */ + char *p; unsigned int id; buf += 9; while (*buf && isspace(*buf)) @@ -84,8 +90,9 @@ int linuxNodeInfoCPUPopulate(virConnectPtr conn, FILE *cpuinfo, virNodeInfoPtr n "parsing cpuinfo cpu cores %c", *buf); return -1; } - id = (unsigned int)strtol(buf+1, NULL, 10); - if (id > nodeinfo->cores) + if (xstrtol_ui(buf+1, &p, 10, &id) == 0 + && (*p == '\0' || isspace(*p)) + && id > nodeinfo->cores) nodeinfo->cores = id; } } @@ -108,14 +115,21 @@ int linuxNodeInfoCPUPopulate(virConnectPtr conn, FILE *cpuinfo, virNodeInfoPtr n } -int linuxNodeInfoMemPopulate(virConnectPtr conn, FILE *meminfo, virNodeInfoPtr nodeinfo) { +int linuxNodeInfoMemPopulate(virConnectPtr conn, FILE *meminfo, + virNodeInfoPtr nodeinfo) { char line[1024]; nodeinfo->memory = 0; while (fgets(line, sizeof(line), meminfo) != NULL) { if (STREQLEN(line, "MemTotal:", 9)) { - nodeinfo->memory = (unsigned int)strtol(line + 10, NULL, 10); + char *p; + unsigned int ui; + if (xstrtol_ui(line + 10, &p, 10, &ui) == 0 + && (*p == '\0' || isspace(*p))) { + nodeinfo->memory = ui; + break; + } } } if (!nodeinfo->memory) { -- 1.5.3.5.643.g40e25

2 2

[Libvir] [PATCH] Arrange for tests to pass in a non-srcdir build.
by Jim Meyering 13 Nov '07

13 Nov '07

Most of the tests assumed that $(srcdir) == . This makes it so "make check" passes in a non-srcdir build, i.e., "mkdir build && cd build && ../configure && make && make check" Arrange for tests to pass in a non-srcdir build. * tests/Makefile.am: Include the contents of the *data directories in the make-dist-built tarball by adding each of that *data directories to EXTRA_DIST. Also add int-overflow (via $(test_scripts)) to EXTRA_DIST. * tests/nodeinfotest.c: Prepend "$abs_top_srcdir/tests" to each input file name. * tests/qemuxml2argvtest.c: Likewise. * tests/qemuxml2xmltest.c: Likewise. * tests/sexpr2xmltest.c: Likewise. * tests/test_conf.sh: Likewise. * tests/virshtest.c: Likewise. * tests/xencapstest.c: Likewise. * tests/xmconfigtest.c: Likewise. * tests/xml2sexprtest.c: Likewise. --- tests/Makefile.am | 18 ++++++++++++++++-- tests/nodeinfotest.c | 14 +++++++++++--- tests/qemuxml2argvtest.c | 11 +++++++++-- tests/qemuxml2xmltest.c | 8 +++++++- tests/sexpr2xmltest.c | 13 ++++++++++++- tests/test_conf.sh | 5 +++-- tests/virshtest.c | 13 +++++++++---- tests/xencapstest.c | 22 +++++++++++++++++++--- tests/xmconfigtest.c | 21 +++++++++++++++++++-- tests/xml2sexprtest.c | 13 ++++++++++++- 10 files changed, 117 insertions(+), 21 deletions(-) diff --git a/tests/Makefile.am b/tests/Makefile.am index 8a472f8..fec2f20 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -31,15 +31,29 @@ LDADDS = \ $(LIBVIRT) \ $(COVERAGE_LDFLAGS) -EXTRA_DIST = xmlrpcserver.py test_conf.sh +EXTRA_DIST = \ + xmlrpcserver.py \ + test_conf.sh \ + confdata \ + qemuxml2argvdata \ + xml2sexprdata \ + nodeinfodata \ + virshdata \ + xencapsdata \ + sexpr2xmldata \ + xmconfigdata noinst_PROGRAMS = xmlrpctest xml2sexprtest sexpr2xmltest virshtest conftest \ reconnect xmconfigtest xencapstest qemuxml2argvtest qemuxml2xmltest \ nodeinfotest +test_scripts = \ + int-overflow +EXTRA_DIST += $(test_scripts) + TESTS = xml2sexprtest sexpr2xmltest virshtest test_conf.sh xmconfigtest \ xencapstest qemuxml2argvtest qemuxml2xmltest nodeinfotest \ - int-overflow + $(test_scripts) if ENABLE_XEN_TESTS TESTS += reconnect endif diff --git a/tests/nodeinfotest.c b/tests/nodeinfotest.c index 604f5a4..8b22b71 100644 --- a/tests/nodeinfotest.c +++ b/tests/nodeinfotest.c @@ -10,6 +10,7 @@ #include "nodeinfo.h" static char *progname; +static char *abs_top_srcdir; #define MAX_FILE 4096 @@ -67,9 +68,12 @@ static int linuxTestNodeInfo(const void *data) { char cpuinfo[PATH_MAX]; char meminfo[PATH_MAX]; char output[PATH_MAX]; - snprintf(cpuinfo, PATH_MAX, "nodeinfodata/linux-%s.cpuinfo", (const char*)data); - snprintf(meminfo, PATH_MAX, "nodeinfodata/linux-%s.meminfo", (const char*)data); - snprintf(output, PATH_MAX, "nodeinfodata/linux-%s.txt", (const char*)data); + snprintf(cpuinfo, PATH_MAX, "%s/tests/nodeinfodata/linux-%s.cpuinfo", + abs_top_srcdir, (const char*)data); + snprintf(meminfo, PATH_MAX, "%s/tests/nodeinfodata/linux-%s.meminfo", + abs_top_srcdir, (const char*)data); + snprintf(output, PATH_MAX, "%s/tests/nodeinfodata/linux-%s.txt", + abs_top_srcdir, (const char*)data); return linuxTestCompareFiles(cpuinfo, meminfo, output); } #endif @@ -90,6 +94,10 @@ main(int argc, char **argv) "nodeinfo-6", }; + abs_top_srcdir = getenv("abs_top_srcdir"); + if (!abs_top_srcdir) + return EXIT_FAILURE; + progname = argv[0]; if (argc > 1) { diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index a91cd03..052f3f3 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -9,6 +9,7 @@ #include "internal.h" static char *progname; +static char *abs_top_srcdir; struct qemud_driver driver; #define MAX_FILE 4096 @@ -93,8 +94,10 @@ static int testCompareXMLToArgvFiles(const char *xml, const char *cmd) { static int testCompareXMLToArgvHelper(const void *data) { char xml[PATH_MAX]; char args[PATH_MAX]; - snprintf(xml, PATH_MAX, "qemuxml2argvdata/qemuxml2argv-%s.xml", (const char*)data); - snprintf(args, PATH_MAX, "qemuxml2argvdata/qemuxml2argv-%s.args", (const char*)data); + snprintf(xml, PATH_MAX, "%s/tests/qemuxml2argvdata/qemuxml2argv-%s.xml", + abs_top_srcdir, (const char*)data); + snprintf(args, PATH_MAX, "%s/tests/qemuxml2argvdata/qemuxml2argv-%s.args", + abs_top_srcdir, (const char*)data); return testCompareXMLToArgvFiles(xml, args); } @@ -112,6 +115,10 @@ main(int argc, char **argv) exit(EXIT_FAILURE); } + abs_top_srcdir = getenv("abs_top_srcdir"); + if (!abs_top_srcdir) + return 1; + if (virtTestRun("QEMU XML-2-ARGV minimal", 1, testCompareXMLToArgvHelper, "minimal") < 0) ret = -1; diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index 59eaf8a..11d8239 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -9,6 +9,7 @@ #include "internal.h" static char *progname; +static char *abs_top_srcdir; struct qemud_driver driver; #define MAX_FILE 4096 @@ -59,7 +60,8 @@ static int testCompareXMLToXMLFiles(const char *xml) { static int testCompareXMLToXMLHelper(const void *data) { char xml[PATH_MAX]; - snprintf(xml, PATH_MAX, "qemuxml2argvdata/qemuxml2argv-%s.xml", (const char*)data); + snprintf(xml, PATH_MAX, "%s/tests/qemuxml2argvdata/qemuxml2argv-%s.xml", + abs_top_srcdir, (const char*)data); return testCompareXMLToXMLFiles(xml); } @@ -76,6 +78,10 @@ main(int argc, char **argv) exit(EXIT_FAILURE); } + abs_top_srcdir = getenv("abs_top_srcdir"); + if (!abs_top_srcdir) + return 1; + if (virtTestRun("QEMU XML-2-ARGV minimal", 1, testCompareXMLToXMLHelper, "minimal") < 0) ret = -1; diff --git a/tests/sexpr2xmltest.c b/tests/sexpr2xmltest.c index eda1397..563941b 100644 --- a/tests/sexpr2xmltest.c +++ b/tests/sexpr2xmltest.c @@ -8,16 +8,23 @@ #include "internal.h" static char *progname; +static char *abs_top_srcdir; #define MAX_FILE 4096 -static int testCompareFiles(const char *xml, const char *sexpr, int xendConfigVersion) { +static int testCompareFiles(const char *xml_rel, const char *sexpr_rel, + int xendConfigVersion) { char xmlData[MAX_FILE]; char sexprData[MAX_FILE]; char *gotxml = NULL; char *xmlPtr = &(xmlData[0]); char *sexprPtr = &(sexprData[0]); int ret = -1; + char xml[PATH_MAX]; + char sexpr[PATH_MAX]; + + snprintf(xml, sizeof xml - 1, "%s/tests/%s", abs_top_srcdir, xml_rel); + snprintf(sexpr, sizeof sexpr - 1, "%s/tests/%s", abs_top_srcdir, sexpr_rel); if (virtTestLoadFile(xml, &xmlPtr, MAX_FILE) < 0) goto fail; @@ -175,6 +182,10 @@ main(int argc, char **argv) exit(EXIT_FAILURE); } + abs_top_srcdir = getenv("abs_top_srcdir"); + if (!abs_top_srcdir) + return 1; + if (virtTestRun("SEXPR-2-XML PV config (version 1)", 1, testComparePVversion1, NULL) != 0) ret = -1; diff --git a/tests/test_conf.sh b/tests/test_conf.sh index b5a6366..7735a66 100755 --- a/tests/test_conf.sh +++ b/tests/test_conf.sh @@ -1,9 +1,10 @@ #!/bin/bash +set -x NOK=0 -for f in confdata/*.conf +for f in $abs_top_srcdir/tests/confdata/*.conf do ./conftest $f > conftest.$$ - outfile=`echo $f | sed s+\.conf+\.out+` + outfile=`echo "$f" | sed s+\.conf$+\.out+` diff $outfile conftest.$$ > /dev/null if [ $? != 0 ] then diff --git a/tests/virshtest.c b/tests/virshtest.c index dc33c27..e26ef5c 100644 --- a/tests/virshtest.c +++ b/tests/virshtest.c @@ -9,6 +9,7 @@ #include "internal.h" static char *progname; +static char *abs_top_srcdir; #define MAX_FILE 4096 static int testFilterLine(char *buffer, @@ -27,11 +28,15 @@ static int testFilterLine(char *buffer, return 0; } -static int testCompareOutput(const char *expect, const char *filter, const char *const argv[]) { +static int testCompareOutput(const char *expect_rel, const char *filter, + const char *const argv[]) { char expectData[MAX_FILE]; char actualData[MAX_FILE]; char *expectPtr = &(expectData[0]); char *actualPtr = &(actualData[0]); + char expect[PATH_MAX]; + + snprintf(expect, sizeof expect - 1, "%s/tests/%s", abs_top_srcdir, expect_rel); if (virtTestLoadFile(expect, &expectPtr, MAX_FILE) < 0) return -1; @@ -268,13 +273,13 @@ int main(int argc, char **argv) { int ret = 0; - char cwd[PATH_MAX]; char buffer[PATH_MAX]; - if (!getcwd(cwd, PATH_MAX-1)) + abs_top_srcdir = getenv("abs_top_srcdir"); + if (!abs_top_srcdir) return 1; - snprintf(buffer, PATH_MAX-1, "test://%s/../docs/testnode.xml", cwd); + snprintf(buffer, PATH_MAX-1, "test://%s/docs/testnode.xml", abs_top_srcdir); buffer[PATH_MAX-1] = '\0'; progname = argv[0]; custom_uri = buffer; diff --git a/tests/xencapstest.c b/tests/xencapstest.c index 418bb7d..5744262 100644 --- a/tests/xencapstest.c +++ b/tests/xencapstest.c @@ -13,13 +13,14 @@ #include "xen_internal.h" static char *progname; +static char *abs_top_srcdir; #define MAX_FILE 4096 static int testCompareFiles(const char *hostmachine, - const char *xml, - const char *cpuinfo, - const char *capabilities) { + const char *xml_rel, + const char *cpuinfo_rel, + const char *capabilities_rel) { char xmlData[MAX_FILE]; char *expectxml = &(xmlData[0]); char *actualxml = NULL; @@ -27,6 +28,17 @@ static int testCompareFiles(const char *hostmachine, int ret = -1; + char xml[PATH_MAX]; + char cpuinfo[PATH_MAX]; + char capabilities[PATH_MAX]; + + snprintf(xml, sizeof xml - 1, "%s/tests/%s", + abs_top_srcdir, xml_rel); + snprintf(cpuinfo, sizeof cpuinfo - 1, "%s/tests/%s", + abs_top_srcdir, cpuinfo_rel); + snprintf(capabilities, sizeof capabilities - 1, "%s/tests/%s", + abs_top_srcdir, capabilities_rel); + if (virtTestLoadFile(xml, &expectxml, MAX_FILE) < 0) goto fail; @@ -152,6 +164,10 @@ main(int argc, char **argv) exit(EXIT_FAILURE); } + abs_top_srcdir = getenv("abs_top_srcdir"); + if (!abs_top_srcdir) + return 1; + virInitialize(); if (virtTestRun("Capabilities for i686, no PAE, no HVM", diff --git a/tests/xmconfigtest.c b/tests/xmconfigtest.c index 524ef52..10024b9 100644 --- a/tests/xmconfigtest.c +++ b/tests/xmconfigtest.c @@ -32,10 +32,12 @@ #include "conf.h" static char *progname; +static char *abs_top_srcdir; #define MAX_FILE 4096 -static int testCompareParseXML(const char *xmcfg, const char *xml, int xendConfigVersion) { +static int testCompareParseXML(const char *xmcfg_rel, const char *xml_rel, + int xendConfigVersion) { char xmlData[MAX_FILE]; char xmcfgData[MAX_FILE]; char gotxmcfgData[MAX_FILE]; @@ -48,6 +50,11 @@ static int testCompareParseXML(const char *xmcfg, const char *xml, int xendConfi int wrote = MAX_FILE; void *old_priv; struct _xenUnifiedPrivate priv; + char xmcfg[PATH_MAX]; + char xml[PATH_MAX]; + + snprintf(xmcfg, sizeof xmcfg - 1, "%s/tests/%s", abs_top_srcdir, xmcfg_rel); + snprintf(xml, sizeof xml - 1, "%s/tests/%s", abs_top_srcdir, xml_rel); conn = virConnectOpenReadOnly("test:///default"); if (!conn) goto fail; @@ -92,7 +99,8 @@ static int testCompareParseXML(const char *xmcfg, const char *xml, int xendConfi return ret; } -static int testCompareFormatXML(const char *xmcfg, const char *xml, int xendConfigVersion) { +static int testCompareFormatXML(const char *xmcfg_rel, const char *xml_rel, + int xendConfigVersion) { char xmlData[MAX_FILE]; char xmcfgData[MAX_FILE]; char *xmlPtr = &(xmlData[0]); @@ -103,6 +111,11 @@ static int testCompareFormatXML(const char *xmcfg, const char *xml, int xendConf virConnectPtr conn; void *old_priv; struct _xenUnifiedPrivate priv; + char xmcfg[PATH_MAX]; + char xml[PATH_MAX]; + + snprintf(xmcfg, sizeof xmcfg - 1, "%s/tests/%s", abs_top_srcdir, xmcfg_rel); + snprintf(xml, sizeof xml - 1, "%s/tests/%s", abs_top_srcdir, xml_rel); conn = virConnectOpenReadOnly("test:///default"); if (!conn) goto fail; @@ -259,6 +272,10 @@ main(int argc, char **argv) exit(EXIT_FAILURE); } + abs_top_srcdir = getenv("abs_top_srcdir"); + if (!abs_top_srcdir) + return 1; + /* Config -> XML */ if (virtTestRun("Paravirt old PVFB (Format)", 1, testCompareParavirtOldPVFBFormat, NULL) != 0) diff --git a/tests/xml2sexprtest.c b/tests/xml2sexprtest.c index 2fcc6b3..4b356d4 100644 --- a/tests/xml2sexprtest.c +++ b/tests/xml2sexprtest.c @@ -9,10 +9,12 @@ #include "internal.h" static char *progname; +static char *abs_top_srcdir; #define MAX_FILE 4096 -static int testCompareFiles(const char *xml, const char *sexpr, const char *name, int xendConfigVersion) { +static int testCompareFiles(const char *xml_rel, const char *sexpr_rel, + const char *name, int xendConfigVersion) { char xmlData[MAX_FILE]; char sexprData[MAX_FILE]; char *gotname = NULL; @@ -20,6 +22,11 @@ static int testCompareFiles(const char *xml, const char *sexpr, const char *name char *xmlPtr = &(xmlData[0]); char *sexprPtr = &(sexprData[0]); int ret = -1; + char xml[PATH_MAX]; + char sexpr[PATH_MAX]; + + snprintf(xml, sizeof xml - 1, "%s/tests/%s", abs_top_srcdir, xml_rel); + snprintf(sexpr, sizeof sexpr - 1, "%s/tests/%s", abs_top_srcdir, sexpr_rel); if (virtTestLoadFile(xml, &xmlPtr, MAX_FILE) < 0) goto fail; @@ -225,6 +232,10 @@ main(int argc, char **argv) progname = argv[0]; + abs_top_srcdir = getenv("abs_top_srcdir"); + if (!abs_top_srcdir) + return 1; + if (argc > 1) { fprintf(stderr, "Usage: %s\n", progname); exit(EXIT_FAILURE); -- 1.5.3.5.643.g40e25

2 1

[Libvir] bug #384443 - packages for virt-install and virt-manager
by Richard W.M. Jones 13 Nov '07

13 Nov '07

1 1

[Libvir] [PATCH] Make qemud's install-init rule work in a non-srcdir build.
by Jim Meyering 13 Nov '07

13 Nov '07

When building from other than the source directory, several things currently fail. This one fails because $(srcdir)/libvirtd.init doesn't exist. The libvirtd.init file we want to install is the one just built, in the current directory. Here's the fix: Make qemud's install-init rule work in a non-srcdir build. * qemud/Makefile.am: In the install-init, remove an unneeded $(srcdir)/ prefix. --- qemud/Makefile.am | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/qemud/Makefile.am b/qemud/Makefile.am index 506fde5..3da6265 100644 --- a/qemud/Makefile.am +++ b/qemud/Makefile.am @@ -95,7 +95,7 @@ remote_dispatch_proc_switch.h: remote_generate_stubs.pl remote_protocol.x if LIBVIRT_INIT_SCRIPTS_RED_HAT install-init: libvirtd.init mkdir -p $(DESTDIR)$(sysconfdir)/rc.d/init.d - $(INSTALL_SCRIPT) $(srcdir)/libvirtd.init $(DESTDIR)$(sysconfdir)/rc.d/init.d/libvirtd + $(INSTALL_SCRIPT) libvirtd.init $(DESTDIR)$(sysconfdir)/rc.d/init.d/libvirtd mkdir -p $(DESTDIR)$(sysconfdir)/sysconfig $(INSTALL_SCRIPT) $(srcdir)/libvirtd.sysconf $(DESTDIR)$(sysconfdir)/sysconfig/libvirtd -- 1.5.3.5.643.g40e25

2 2

[Libvir] bug #384300 - libvirt package for Debian
by Richard W.M. Jones 13 Nov '07

13 Nov '07

This is a libvirt package for Debian. It's derived from the Ubuntu libvirt package, rebuilt for Debian testing. I'm not the greatest Debian packager in the world, so any comments / feedback are welcome. The previous patch to the base Xen package is also required before this can be built. Source: http://et.redhat.com/~rjones/debian-libvirt/libvirt_0.3.3.orig.tar.gz http://et.redhat.com/~rjones/debian-libvirt/libvirt_0.3.3-1_i386.changes http://et.redhat.com/~rjones/debian-libvirt/libvirt_0.3.3-1.dsc http://et.redhat.com/~rjones/debian-libvirt/libvirt_0.3.3-1.diff.gz Binary for i386: http://et.redhat.com/~rjones/debian-libvirt/libvirt-bin_0.3.3-1_i386.deb http://et.redhat.com/~rjones/debian-libvirt/libvirt-dev_0.3.3-1_i386.deb http://et.redhat.com/~rjones/debian-libvirt/libvirt0_0.3.3-1_i386.deb http://et.redhat.com/~rjones/debian-libvirt/python-libvirt_0.3.3-1_i386.deb There is a Debian ITP for libvirt here: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=384300 Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

1 0

[Libvir] [PATCH] bug #402249 - include <xen/*.h> header files
by Richard W.M. Jones 13 Nov '07

13 Nov '07

When you build Xen from source, the "make install" rule installs extra header files such as <xen/xen.h> and <xen/dom0_ops.h> which are all required for building libvirt. However the Debian package doesn't run the install rule, but instead runs (effectively) "make -C tools install" and that only installs a partial set of header files. In particular none of the vital public Xen header files are installed, making it impossible to build libvirt, and probably other stuff as well. This patch fixes that by copying the required files by hand: http://et.redhat.com/~rjones/debian-libvirt/debian-xen-includes.patch (It is similar to what the Fedora RPM does, faced with the same situation). There is a bug which is somewhat related to this issue: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=402249 Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

1 0

[Libvir] [PATCH] conftest: fix transposed size and count arguments to fwrite.
by Jim Meyering 12 Nov '07

12 Nov '07

I noticed conftest fail and tracked it down to this: conftest: fix transposed size and count arguments to fwrite. * tests/conftest.c: Include <string.h> and <errno.h>. Also include strerror in diagnostic. --- tests/conftest.c | 6 ++++-- 1 files changed, 4 insertions(+), 2 deletions(-) diff --git a/tests/conftest.c b/tests/conftest.c index 604e410..9c8e3a6 100644 --- a/tests/conftest.c +++ b/tests/conftest.c @@ -1,6 +1,8 @@ #include <unistd.h> #include <stdlib.h> #include <stdio.h> +#include <string.h> +#include <errno.h> #include "conf.h" int main(int argc, char **argv) { @@ -25,8 +27,8 @@ int main(int argc, char **argv) { exit(3); } virConfFree(conf); - if (fwrite(buffer, len, 1, stdout) != len) { - fprintf(stderr, "Write failed\n"); + if (fwrite(buffer, 1, len, stdout) != len) { + fprintf(stderr, "Write failed: %s\n", strerror (errno)); exit(1); } exit(0); -- 1.5.3.5.643.g40e25

2 2

[Libvir] Question about specifications of virsh start
by Masayuki Sunou 12 Nov '07

12 Nov '07

Hi I have a question about specifications of virsh start. It is necessary to remove "type=ioemu" from the configuration file of Xen, when HVM domain installed Para Virtualized Driver use unmodified driver on Xen 3.0.3. But,libvirt always adds "type=ioemu" to vif of HVM domain in xm_internal.c. Therefore virsh start cannot start correctly HVM domain installed Para Virtualized Driver. Could you teach why does libvirt disregard setting of configuration file and set type=ioemu ? Thanks, Masayuki Sunou.

3 4

[Libvir] [PATCH] missing printf attributes
by Jim Meyering 08 Nov '07

08 Nov '07

I stumbled across two more missing printf attributes: 2007-11-08 Jim Meyering <meyering(a)redhat.com> * src/virsh.c (vshPrintExtra, vshDebug): Use gcc's printf attribute. Attaching differently, in an attempt to defy mailman-2.1.5:

2 1

[Libvir] Re: [collectd] collectd libvirtstats plugin
by Florian Forster 08 Nov '07

08 Nov '07

Hi Rich, as promised, I've taken a more in-depth look at the `libvirtstats' plugin (I've adopted this as a working title for now and will do the renaming on the weekend or so..) and it looks very good, thanks again :) The `_virDomainInfo' structure has the members `memory' and `maxMem'. Can one calculate `unused' or `free' (or whatever the preferred term in this particular case is) memory using `maxMem - memory'? Would that be an interesting metric? I'm not sure what the best way to store the CPU usage is. Is the `cpuTime' member of the `_virDomainInfo' structure the same as the sum over all `(struct _virVcpuInfo).cpuTime'? Does collecting the total make sense in this case? I'd much rather use the same `type' for ``normal'' CPU statistics and this case which would be much simpler if we didn't collect the total.. I've done some minor things while looking through the code. It basically a mix of coding style and personal convention. Since there's no Debian package yet I didn't install try to compile it yet. I'll install the library and fix the errors I've made as soon as I have some more spare time. The things I've changed are basically: - Rather use `sizeof (array);' than the `DATA_MAX_NAME_LEN' define. - NULL-terminate the strings unconditionally after `strncpy' or `snprintf' was used. - Rather use -- 8< -- if (status != 0) { handle_error (); } -- >8 -- than -- 8< -- if (status == -1) { handle_error (); } -- >8 -- On Mon, Nov 05, 2007 at 02:00:12PM +0000, Richard W.M. Jones wrote: > In the types.db, I have set the max for virt_cpu_total to be > 256,000,000,000. [...] Therefore I have assumed here that > max CPUs = 256. That's basically the reason why I'd rather collect each virtual CPU in a separate file and ``simply'' combine the graphs. This will work for n+1 CPUs ;) > I'm using CSV output, which works, but it seems to report the > aggregate counter values instead of the differences. Is that to be > expected? Yes, that is expected. The changes I've made for the next feature release (4.3) include an easy way to get the rates for a counter value. So I will certainly add an option to store rates instead. Regards, -octo -- Florian octo Forster Hacker in training GnuPG: 0x91523C3D http://verplant.org/

2 1

Re: [Libvir] [PATCH] * src/virsh.c (vshCommandOptInt): Correct indentation
by Jim Meyering 08 Nov '07

08 Nov '07

Well, mailman-2.1.5 mangled yet another of my messages. [ I know why, and have reported the bug. Maybe we'll get a fixed/newer version of mailman. ] But at least the patch got through. Here's the part that it removed: Here's a tiny patch to fix indentation, since I'm about to change the code in that area, I'm submitting this first, so we don't mix syntax-only changes with ones that change semantics. The last two times I tried to send an attachment, mailman removed them. I'll try one more time, with this one: Jim Meyering <meyering(a)redhat.com> wrote: > Signed-off-by: Jim Meyering <meyering(a)redhat.com> > --- > ChangeLog | 4 ++++ > src/virsh.c | 10 +++++----- > 2 files changed, 9 insertions(+), 5 deletions(-) > > diff --git a/ChangeLog b/ChangeLog > index ce68401..a6de485 100644 > --- a/ChangeLog > +++ b/ChangeLog > @@ -1,3 +1,7 @@ > +2007-11-08 Jim Meyering <meyering(a)redhat.com> > + > + * src/virsh.c (vshCommandOptInt): Correct indentation > + > Wed Nov 7 14:38:43 CET 2007 Daniel Veillard <veillard(a)redhat.com> > > * docs/libvir.html docs/site.xsl: small updates with references > diff --git a/src/virsh.c b/src/virsh.c > index 0615a7d..852dea2 100644 > --- a/src/virsh.c > +++ b/src/virsh.c > @@ -1,7 +1,7 @@ > /* > * virsh.c: a Xen shell used to exercise the libvirt API > * > - * Copyright (C) 2005 Red Hat, Inc. > + * Copyright (C) 2005, 2007 Red Hat, Inc. > * > * See COPYING.LIB for the License of this software > * > @@ -3956,10 +3956,10 @@ vshCommandOptInt(vshCmd * cmd, const char *name, int *found) > > if ((arg != NULL) && (arg->data != NULL)) { > res = strtol(arg->data, &end_p, 10); > - if ((arg->data == end_p) || (*end_p!= 0)) > - num_found = FALSE; > - else > - num_found = TRUE; > + if ((arg->data == end_p) || (*end_p!= 0)) > + num_found = FALSE; > + else > + num_found = TRUE; > } > if (found) > *found = num_found;

2 1

[Libvir] fix misleading indentation
by Jim Meyering 08 Nov '07

08 Nov '07

Here's a tiny patch to fix indentation, since I'm about to change the code in that area, I'm submitting this first, so we don't mix syntax-only changes with ones that change semantics. The last two times I tried to send an attachment, mailman removed them. I'll try one more time, with this one:

1 0

[Libvir] Strange warnings with gcc 4.1.2
by Richard W.M. Jones 07 Nov '07

07 Nov '07

This is with libvirt from CVS. They don't look like errors to me though... ------------ In function ‘open’, inlined from ‘readFile’ at virsh.c:850: /usr/include/bits/fcntl2.h:45: error: call to ‘__open_too_many_args’ declared with attribute error: open can be called either with 2 or 3 arguments, not more /usr/include/bits/fcntl2.h:51: error: call to ‘__open_missing_mode’ declared with attribute error: open with O_CREAT in second argument needs 3 arguments The code in question is: if ((fd = open(filename, O_RDONLY)) == -1) { Not a newly created file, so mode isn't needed AFAICS. ------------ In function ‘read’, inlined from ‘readFile’ at virsh.c:864: /usr/include/bits/unistd.h:43: warning: call to ‘__read_chk_warn’ declared with attribute warning: read called with bigger length than size of the destination buffer The code in question is: char buffer[1024]; char *new; int ret; if ((ret = read(fd, buffer, sizeof(buffer))) == 0) break; Also seems correct to me. ------------ Other warnings: virsh.c: In function ‘cmdSchedinfo’: virsh.c:1138: warning: ‘weight’ may be used uninitialized in this function virsh.c:1140: warning: ‘cap’ may be used uninitialized in this function For some reason gcc appears to have lots its ability to do the data flow analysis to find out that these warnings are bogus. In any case the fix is trivial: Index: src/virsh.c =================================================================== RCS file: /data/cvs/libvirt/src/virsh.c,v retrieving revision 1.105 diff -u -r1.105 virsh.c --- src/virsh.c 6 Nov 2007 09:41:18 -0000 1.105 +++ src/virsh.c 7 Nov 2007 12:37:07 -0000 @@ -1135,9 +1135,9 @@ int nr_inputparams = 0; int inputparams = 0; int weightfound = 0; - int weight; + int weight = 0; int capfound = 0; - int cap; + int cap = 0; char str_weight[] = "weight"; char str_cap[] = "cap"; int ret_val = FALSE; Rich. -- Emerging Technologies, Red Hat - http://et.redhat.com/~rjones/ Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SL4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 03798903

2 3

[Libvir] PATCH: Avoid format string abuse (also avoids gcc warnings).
by Jim Meyering 07 Nov '07

07 Nov '07

This patch was prompted by warnings like this: util.c:56: warning: format not a string literal and no format arguments and they're legitimate. Imagine a format string contains "%%..." goes through the vnsprintf call, which reduces it to "%...". If the result string is then passed to __virRaiseError as the format string, then *boom*. Instead, use "%s" as the format, with the non-literal as the matching argument. Patch below. I searched the sources for %% and *did* find one potential problem: $ git-grep -B1 %% > k po/ms.po-msgid "too many drivers registered in %s" po/ms.po:msgstr "terlalu banyak spesifikasi penukaran %% pada suffiks" -- src/xend_internal.c- case '\n': src/xend_internal.c: snprintf(ptr, 4, "%%%02x", string[i]); since "% p" does happen to be a valid format string! So if someone using Malaysian messages provoked that particular diagnostic in a code path that takes it through __virRaiseError, bad things might happen. Big "if", of course :-) I didn't try. 2007-11-06 Jim Meyering <meyering(a)redhat.com> Avoid risk of format string abuse (also avoids gcc warnings). * src/util.c (ReportError): Use a literal "%s" format string. * src/remote_internal.c (server_error): Likewise. * src/qemu_conf.c (qemudReportError): Likewise.

3 3

[Libvir] PATCH: Avoid format string abuse (also avoids gcc warnings).
by Jim Meyering 06 Nov '07

06 Nov '07

[resending with the patch in-line, not as an attachment -- mailman removed my attached patch the first time! Bad mailman. Bad. ] This patch was prompted by warnings like this: util.c:56: warning: format not a string literal and no format arguments and they're legitimate. Imagine a format string contains "%%..." goes through the vnsprintf call, which reduces it to "%...". If the result string is then passed to __virRaiseError as the format string, then *boom*. Instead, use "%s" as the format, with the non-literal as the matching argument. Patch below. I searched the sources for %% and *did* find one potential problem: $ git-grep -B1 %% > k po/ms.po-msgid "too many drivers registered in %s" po/ms.po:msgstr "terlalu banyak spesifikasi penukaran %% pada suffiks" -- src/xend_internal.c- case '\n': src/xend_internal.c: snprintf(ptr, 4, "%%%02x", string[i]); since "% p" does happen to be a valid format string! So if someone using Malaysian messages provoked that particular diagnostic in a code path that takes it through __virRaiseError, bad things might happen. Big "if", of course :-) I didn't try. 2007-11-06 Jim Meyering <meyering(a)redhat.com> Avoid risk of format string abuse (also avoids gcc warnings). * src/util.c (ReportError): Use a literal "%s" format string. * src/remote_internal.c (server_error): Likewise. * src/qemu_conf.c (qemudReportError): Likewise. ----------------------------- >From 8e0cafec021ba3cbec55a4d691f19c14ed9e587f Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyering(a)redhat.com> Date: Tue, 6 Nov 2007 20:14:21 +0100 Subject: [PATCH] Avoid risk of format string abuse (also avoids gcc warnings). * src/util.c (ReportError): Use a literal "%s" format string. * src/remote_internal.c (server_error): Likewise. * src/qemu_conf.c (qemudReportError): Likewise. Signed-off-by: Jim Meyering <meyering(a)redhat.com> --- ChangeLog | 7 +++++++ src/qemu_conf.c | 2 +- src/remote_internal.c | 2 +- src/util.c | 2 +- 4 files changed, 10 insertions(+), 3 deletions(-) diff --git a/ChangeLog b/ChangeLog index 511ed6a..535173d 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,10 @@ +2007-11-06 Jim Meyering <meyering(a)redhat.com> + + Avoid risk of format string abuse (also avoids gcc warnings). + * src/util.c (ReportError): Use a literal "%s" format string. + * src/remote_internal.c (server_error): Likewise. + * src/qemu_conf.c (qemudReportError): Likewise. + Tue Nov 6 17:24:16 CET 2007 Daniel Veillard <veillard(a)redhat.com> * src/xs_internals.c: patch from Chris Lalancette, forgot to diff --git a/src/qemu_conf.c b/src/qemu_conf.c index 78f4699..3556a9a 100644 --- a/src/qemu_conf.c +++ b/src/qemu_conf.c @@ -64,7 +64,7 @@ void qemudReportError(virConnectPtr conn, errorMessage[0] = '\0'; } __virRaiseError(conn, dom, net, VIR_FROM_QEMU, code, VIR_ERR_ERROR, - NULL, NULL, NULL, -1, -1, errorMessage); + NULL, NULL, NULL, -1, -1, "%s", errorMessage); } int qemudLoadDriverConfig(struct qemud_driver *driver, diff --git a/src/remote_internal.c b/src/remote_internal.c index 3af326f..1420a88 100644 --- a/src/remote_internal.c +++ b/src/remote_internal.c @@ -3073,7 +3073,7 @@ server_error (virConnectPtr conn, remote_error *err) err->domain, err->code, err->level, str1, str2, str3, err->int1, err->int2, - message); + "%s", message); } /* get_nonnull_domain and get_nonnull_network turn an on-wire diff --git a/src/util.c b/src/util.c index eb57859..c964a63 100644 --- a/src/util.c +++ b/src/util.c @@ -53,7 +53,7 @@ ReportError(virConnectPtr conn, errorMessage[0] = '\0'; } __virRaiseError(conn, dom, net, VIR_FROM_NONE, code, VIR_ERR_ERROR, - NULL, NULL, NULL, -1, -1, errorMessage); + NULL, NULL, NULL, -1, -1, "%s", errorMessage); } static int virSetCloseExec(int fd) { -- 1.5.3.5.561.g140d

1 0

[Libvir] [PATCH]: Fix build warning in xs_internal.c
by Chris Lalancette 06 Nov '07

06 Nov '07

All, Attached is a trivial patch to fix a build warning in xs_internal.c; basically the xenUnifiedDriver structure doesn't have a domainDumpXML method, so we had one too many NULLs. Signed-off-by: Chris Lalancette <clalance(a)redhat.com>

2 1

[Libvir] [RFC] Linux-VServer support
by Daniel Hokka Zakrisson 06 Nov '07

06 Nov '07

Hi, This is an initial stab at adding Linux-VServer support to libvirt. There are still a couple of things missing, like scheduler support in the XML parsing, and proper network support. I've got a few questions though. virsh's schedinfo hardcodes the available options, should I be adding new ones there? Would better introspection from getSchedulerType make this a non-issue? How do the network domains interact and associate with the regular domains? -- Daniel Hokka Zakrisson

4 24