agilent-dmm: Correctly parse negative overload.

diff --git a/src/hardware/agilent-dmm/sched.c b/src/hardware/agilent-dmm/sched.c
index 113673052cbaa729f4fe22437087714c4b45f2cc..b910f7aafd3c762b6bef5c027d732847018de7ee 100644 (file)
--- a/src/hardware/agilent-dmm/sched.c
+++ b/src/hardware/agilent-dmm/sched.c
@@ -234,6 +234,7 @@ static int recv_fetc(const struct sr_dev_inst *sdi, GMatchInfo *match)
        struct sr_datafeed_packet packet;
        struct sr_datafeed_analog analog;
        float fvalue;
+       const char *s;
        char *mstr;
 
        sr_spew("FETC reply '%s'.", g_match_info_get_string(match));
@@ -245,7 +246,8 @@ static int recv_fetc(const struct sr_dev_inst *sdi, GMatchInfo *match)
                 * get metadata soon enough. */
                return SR_OK;
 
-       if (!strcmp(g_match_info_get_string(match), "+9.90000000E+37")) {
+       s = g_match_info_get_string(match);
+       if (!strcmp(s, "-9.90000000E+37") || !strcmp(s, "+9.90000000E+37")) {
                /* An invalid measurement shows up on the display as "O.L", but
                 * comes through like this. Since comparing 38-digit floats
                 * is rather problematic, we'll cut through this here. */