Andrea Bolognani <abologna(a)redhat.com> [2017-12-11, 05:40PM +0100]:
The algorithm used to extract the CPU frequency from /proc/cpuinfo
is the same regardless of architecture, the only difference being
the label used to identify the relevant field.
Factor the parsing code out to a new helper and use it to implement
virHostCPUGetInfoParseCPUInfo(), thus removing some duplication.
Signed-off-by: Andrea Bolognani <abologna(a)redhat.com>
---
src/util/virhostcpu.c | 105 +++++++++++++++++++-------------------------------
1 file changed, 40 insertions(+), 65 deletions(-)
diff --git a/src/util/virhostcpu.c b/src/util/virhostcpu.c
index 4d5c56659..85803d527 100644
--- a/src/util/virhostcpu.c
+++ b/src/util/virhostcpu.c
@@ -509,6 +509,40 @@ virHostCPUHasValidSubcoreConfiguration(int threads_per_subcore)
}
+static int
+virHostCPUGetInfoParseCPUFrequency(const char *buf,
+ const char *label,
+ unsigned int *mhz)
+{
+ int ret = -1;
+
+ if (STRPREFIX(buf, label)) {
I'd prefer negation and early exit to get rid of one indentation level.
I also moved this conditional into the loop below. Save function calls.
+ char *p;
+ unsigned int ui;
+
+ buf += strlen(label);
+ while (*buf && c_isspace(*buf))
+ buf++;
+
+ if (*buf != ':' || !buf[1]) {
+ virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
+ _("parsing cpu MHz from cpuinfo"));
+ goto cleanup;
+ }
+
+ if (virStrToLong_ui(buf+1, &p, 10, &ui) == 0 &&
+ /* Accept trailing fractional part. */
+ (*p == '\0' || *p == '.' || c_isspace(*p)))
+ *mhz = ui;
+ }
+
+ ret = 0;
+
+ cleanup:
+ return ret;
+}
+
+
static int
virHostCPUGetInfoParseCPUInfo(FILE *cpuinfo,
virArch arch,
@@ -521,73 +555,14 @@ virHostCPUGetInfoParseCPUInfo(FILE *cpuinfo,
while (fgets(line, sizeof(line), cpuinfo) != NULL) {
if (ARCH_IS_X86(arch)) {
Probably (hopefully?) the compiler can optimize, but I get a bit iffy if
I see a loop invariant conditional inside a loop. I find this also
harder to read.
- char *buf = line;
- if (STRPREFIX(buf, "cpu MHz")) {
- char *p;
- unsigned int ui;
-
- buf += 7;
- while (*buf && c_isspace(*buf))
- buf++;
-
- if (*buf != ':' || !buf[1]) {
- virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
- _("parsing cpu MHz from cpuinfo"));
- goto cleanup;
- }
-
- if (virStrToLong_ui(buf+1, &p, 10, &ui) == 0 &&
- /* Accept trailing fractional part. */
- (*p == '\0' || *p == '.' || c_isspace(*p)))
- *mhz = ui;
- }
+ if (virHostCPUGetInfoParseCPUFrequency(line, "cpu MHz", mhz) <
0)
+ goto cleanup;
} else if (ARCH_IS_PPC(arch)) {
- char *buf = line;
- if (STRPREFIX(buf, "clock")) {
- char *p;
- unsigned int ui;
-
- buf += 5;
- while (*buf && c_isspace(*buf))
- buf++;
-
- if (*buf != ':' || !buf[1]) {
- virReportError(VIR_ERR_INTERNAL_ERROR, "%s",
- _("parsing cpu MHz from cpuinfo"));
- goto cleanup;
- }
-
- if (virStrToLong_ui(buf+1, &p, 10, &ui) == 0 &&
- /* Accept trailing fractional part. */
- (*p == '\0' || *p == '.' || c_isspace(*p)))
- *mhz = ui;
- /* No other interesting infos are available in /proc/cpuinfo.
- * However, there is a line identifying processor's version,
- * identification and machine, but we don't want it to be caught
- * and parsed in next iteration, because it is not in expected
- * format and thus lead to error. */
- }
+ if (virHostCPUGetInfoParseCPUFrequency(line, "clock", mhz) <
0)
+ goto cleanup;
} else if (ARCH_IS_ARM(arch)) {
- char *buf = line;
- if (STRPREFIX(buf, "BogoMIPS")) {
- char *p;
- unsigned int ui;
-
- buf += 8;
- while (*buf && c_isspace(*buf))
- buf++;
-
- if (*buf != ':' || !buf[1]) {
- virReportError(VIR_ERR_INTERNAL_ERROR,
- "%s", _("parsing cpu MHz from
cpuinfo"));
- goto cleanup;
- }
-
- if (virStrToLong_ui(buf+1, &p, 10, &ui) == 0
- /* Accept trailing fractional part. */
- && (*p == '\0' || *p == '.' ||
c_isspace(*p)))
- *mhz = ui;
- }
+ if (virHostCPUGetInfoParseCPUFrequency(line, "BogoMIPS", mhz) <
0)
+ goto cleanup;
} else if (ARCH_IS_S390(arch)) {
/* s390x has no realistic value for CPU speed,
* assign a value of zero to signify this */
--
2.14.3
--
libvir-list mailing list
libvir-list(a)redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
I attach my take on this refactorization for comparison. Feel free to
take stuff you like. Or don't :)
--
IBM Systems
Linux on z Systems & Virtualization Development
------------------------------------------------------------------------
IBM Deutschland
Schönaicher Str. 220
71032 Böblingen
Phone: +49 7031 16 1819
E-Mail: bwalk(a)de.ibm.com
------------------------------------------------------------------------
IBM Deutschland Research & Development GmbH
Vorsitzende des Aufsichtsrats: Martina Koederitz
Geschäftsführung: Dirk Wittkopp
Sitz der Gesellschaft: Böblingen
Registergericht: Amtsgericht Stuttgart, HRB 243294