AP_ESC_TELEM: prevent overrun, esc_index should be < to ESC_TELEM_MAX_ESCS

This commit is contained in:
Pierre Kancir 2021-08-11 00:12:06 +02:00 committed by Randy Mackay
parent 04a4a27a14
commit 2c62d32ca8
1 changed files with 3 additions and 3 deletions

View File

@ -88,7 +88,7 @@ bool AP_ESC_Telem::get_rpm(uint8_t esc_index, float& rpm) const
{ {
const volatile AP_ESC_Telem_Backend::RpmData& rpmdata = _rpm_data[esc_index]; const volatile AP_ESC_Telem_Backend::RpmData& rpmdata = _rpm_data[esc_index];
if (esc_index > ESC_TELEM_MAX_ESCS || is_zero(rpmdata.update_rate_hz)) { if (esc_index >= ESC_TELEM_MAX_ESCS || is_zero(rpmdata.update_rate_hz)) {
return false; return false;
} }
@ -269,7 +269,7 @@ void AP_ESC_Telem::update_telem_data(const uint8_t esc_index, const AP_ESC_Telem
// can only get slightly more up-to-date information that perhaps they were expecting or might // can only get slightly more up-to-date information that perhaps they were expecting or might
// read data that has just gone stale - both of these are safe and avoid the overhead of locking // read data that has just gone stale - both of these are safe and avoid the overhead of locking
if (esc_index > ESC_TELEM_MAX_ESCS) { if (esc_index >= ESC_TELEM_MAX_ESCS) {
return; return;
} }
@ -303,7 +303,7 @@ void AP_ESC_Telem::update_telem_data(const uint8_t esc_index, const AP_ESC_Telem
// this should be called by backends when new telemetry values are available // this should be called by backends when new telemetry values are available
void AP_ESC_Telem::update_rpm(const uint8_t esc_index, const uint16_t new_rpm, const float error_rate) void AP_ESC_Telem::update_rpm(const uint8_t esc_index, const uint16_t new_rpm, const float error_rate)
{ {
if (esc_index > ESC_TELEM_MAX_ESCS) { if (esc_index >= ESC_TELEM_MAX_ESCS) {
return; return;
} }