ACPICA: Utilities: Fix local printf issue.

The bug can be reproduced by using a format that do not have the width.prec
specified after a format that have the width.prec specified. The second
formatted output will be wrong. The root cause is acpi_ut_vsnprintf()
doesn't reset the specifiers to the default values.

This patch fixes this issue. BZ 1094. Reported by Yizhe Wang, fixed by
Lv Zheng.

Since acpi_ut_vprintf() is only enabled for specific OSPM now, this patch
doesn't affect Linux kernel.

Link: https://bugs.acpica.org/show_bug.cgi?id=1094
Reported-and-tested-by: Yizhe Wang <yizhe.wang@intel.com>
Signed-off-by: Lv Zheng <lv.zheng@intel.com>
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

diff --git a/drivers/acpi/acpica/utprint.c b/drivers/acpi/acpica/utprint.c
index 080e22a..0ce3f5a 100644 (file)
--- a/drivers/acpi/acpica/utprint.c
+++ b/drivers/acpi/acpica/utprint.c
@@ -401,6 +401,7 @@ acpi_ut_vsnprintf(char *string,
 
                /* Process width */
 
+               width = -1;
                if (ACPI_IS_DIGIT(*format)) {
                        format = acpi_ut_scan_number(format, &number);
                        width = (s32) number;
@@ -415,6 +416,7 @@ acpi_ut_vsnprintf(char *string,
 
                /* Process precision */
 
+               precision = -1;
                if (*format == '.') {
                        ++format;
                        if (ACPI_IS_DIGIT(*format)) {
@@ -431,6 +433,7 @@ acpi_ut_vsnprintf(char *string,
 
                /* Process qualifier */
 
+               qualifier = -1;
                if (*format == 'h' || *format == 'l' || *format == 'L') {
                        qualifier = *format;
                        ++format;