Skip to content

Commit

Permalink
sa rev
Browse files Browse the repository at this point in the history
  • Loading branch information
camilo committed Feb 13, 2024
1 parent 41c4817 commit 52295d4
Show file tree
Hide file tree
Showing 4 changed files with 69 additions and 57 deletions.
92 changes: 46 additions & 46 deletions qfp16.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -36,9 +36,9 @@ static const struct _qFP16_intern_s intern = {
/*f_6_5*/ 425984,
/*one_fp16_f*/ 0.0000152587890625f,
/*one_fp16_d*/ 0.0000152587890625,
/*overflow_mask*/ 0x80000000u,
/*overflow_mask*/ 0x80000000U,
/*fraction_mask*/ 0x0000FFFFu,
/*integer_mask*/ 0xFFFF0000u
/*integer_mask*/ 0xFFFF0000U
};

const struct _qFP16_const_s qFP16 = {
Expand Down Expand Up @@ -67,7 +67,7 @@ const struct _qFP16_const_s qFP16 = {
/*f_360*/ 23592960,
};

static qFP16_Settings_t fp_default = { -2147483647, 2147483647, 1u, 0u }; //skipcq: CXX-W2009
static qFP16_Settings_t fp_default = { -2147483647, 2147483647, 1U, 0U }; //skipcq: CXX-W2009
static qFP16_Settings_t *fp = &fp_default; //skipcq: CXX-W2009

/*! @endcond */
Expand All @@ -93,7 +93,7 @@ int qFP16_SettingsSet( qFP16_Settings_t * const instance,
{
int retValue = 0;

if ( ( NULL != instance ) && ( max > min ) && ( rounding <= 1u ) && ( saturate <= 1u ) ) {
if ( ( NULL != instance ) && ( max > min ) && ( rounding <= 1U ) && ( saturate <= 1U ) ) {
instance->min = min;
instance->max = max;
instance->rounding = rounding;
Expand Down Expand Up @@ -121,7 +121,7 @@ int qFP16_FPToInt( const qFP16_t x )
{
int retValue;

if ( 1u == fp->rounding ) {
if ( 1U == fp->rounding ) {
if ( x >= 0 ) {
/*cppcheck-suppress misra-c2012-10.1 */
retValue = ( x + ( qFP16.one >> 1 ) ) / qFP16.one;
Expand Down Expand Up @@ -149,7 +149,7 @@ qFP16_t qFP16_FloatToFP( const float x )
/*cstat -CERT-FLP36-C*/
retValue = x * (float)qFP16.one;
/*cstat +CERT-FLP36-C*/
if ( 1u == fp->rounding ) {
if ( 1U == fp->rounding ) {
retValue += ( retValue >= 0.0f) ? 0.5f : -0.5f;
}

Expand All @@ -169,7 +169,7 @@ qFP16_t qFP16_DoubleToFP( const double x )
/*cstat -CERT-FLP36-C*/
retValue = x * (double)qFP16.one;
/*cstat +CERT-FLP36-C*/
if ( 1u == fp->rounding ) {
if ( 1U == fp->rounding ) {
retValue += ( retValue >= 0.0 ) ? 0.5 : -0.5;
}

Expand Down Expand Up @@ -223,7 +223,7 @@ qFP16_t qFP16_Add( const qFP16_t X,
uint32_t retValue;

retValue = x + y;
if ( ( 0u == ( ( x ^ y ) & intern.overflow_mask ) ) && ( 0u != ( ( x ^ retValue ) & intern.overflow_mask ) ) ) {
if ( ( 0U == ( ( x ^ y ) & intern.overflow_mask ) ) && ( 0U != ( ( x ^ retValue ) & intern.overflow_mask ) ) ) {
retValue = (uint32_t)qFP16.overflow;
}

Expand All @@ -237,7 +237,7 @@ qFP16_t qFP16_Sub( const qFP16_t X,
uint32_t retValue;

retValue = x - y;
if ( ( 0u != ( ( x ^ y ) & intern.overflow_mask ) ) && ( 0u != ( ( x ^ retValue ) & intern.overflow_mask ) ) ) {
if ( ( 0U != ( ( x ^ y ) & intern.overflow_mask ) ) && ( 0U != ( ( x ^ retValue ) & intern.overflow_mask ) ) ) {
retValue = (uint32_t)qFP16.overflow;
}

Expand Down Expand Up @@ -273,7 +273,7 @@ qFP16_t qFP16_Mul( const qFP16_t x,
a = ( mulH < 0 ) ? -1 : 0;
/*cppcheck-suppress misra-c2012-10.1 */
if ( a == ( mulH >> 15 ) ) {
if ( 1u == fp->rounding ) {
if ( 1U == fp->rounding ) {
uint32_t tmp2;

tmp2 = mulL;
Expand Down Expand Up @@ -301,7 +301,7 @@ qFP16_t qFP16_Div( const qFP16_t x,
qFP16_t retValue = fp->min;

if ( 0 != y ) {
uint32_t xRem, xDiv, bit = 0x10000u;
uint32_t xRem, xDiv, bit = 0x10000U;

xRem = (uint32_t)( ( x >= 0 ) ? x : -x );
xDiv = (uint32_t)( ( y >= 0 ) ? y : -y );
Expand All @@ -312,11 +312,11 @@ qFP16_t qFP16_Div( const qFP16_t x,
}
retValue = qFP16.overflow;
/*cstat -MISRAC2012-Rule-14.3_a*/
if ( 0u != bit ) { /*MISRAC2012-Rule-14.3_a false positive*/
if ( 0U != bit ) { /*MISRAC2012-Rule-14.3_a false positive*/
/*cstat +MISRAC2012-Rule-14.3_a*/
uint32_t quotient = 0u;
uint32_t quotient = 0U;

if ( 0u != ( xDiv & 0x80000000u ) ) {
if ( 0U != ( xDiv & 0x80000000U ) ) {
if ( xRem >= xDiv ) {
quotient |= bit;
xRem -= xDiv;
Expand All @@ -325,23 +325,23 @@ qFP16_t qFP16_Div( const qFP16_t x,
bit >>= 1;
}

while ( ( 0u != bit ) && ( 0u != xRem ) ) {
while ( ( 0U != bit ) && ( 0U != xRem ) ) {
if ( xRem >= xDiv ) {
quotient |= bit;
xRem -= xDiv;
}
xRem <<= 1;
bit >>= 1;
}
if ( 1u == fp->rounding ) {
if ( 1U == fp->rounding ) {
if ( xRem >= xDiv ) {
++quotient;
}
}

retValue = (qFP16_t)quotient;
/*cppcheck-suppress misra-c2012-10.8 */
if ( 0u != ( (uint32_t)( x ^ y ) & intern.overflow_mask ) ) {
if ( 0U != ( (uint32_t)( x ^ y ) & intern.overflow_mask ) ) {
if ( quotient == (uint32_t)fp->min ) {
retValue = qFP16.overflow;
}
Expand Down Expand Up @@ -377,19 +377,19 @@ qFP16_t qFP16_Sqrt( qFP16_t x )

retValue = 0;
/*cppcheck-suppress [ cert-INT31-c, misra-c2012-12.2, misra-c2012-12.1 ]*/
bit = ( 0 != ( x & (qFP16_t)4293918720 ) ) ? ( 1u << 30u ) : ( 1u << 18u );
bit = ( 0 != ( x & (qFP16_t)4293918720 ) ) ? ( 1U << 30U ) : ( 1U << 18u );
while ( bit > (uint32_t)x ) {
bit >>= 2u;
}

for ( n = 0u ; n < 2u ; ++n ) {
while ( 0u != bit ) {
for ( n = 0U ; n < 2u ; ++n ) {
while ( 0U != bit ) {
/*cppcheck-suppress misra-c2012-10.8 */
if ( x >= (qFP16_t)( (uint32_t)retValue + bit ) ) {
/*cppcheck-suppress misra-c2012-10.8 */
x -= (qFP16_t)( (uint32_t)retValue + bit );
/*cppcheck-suppress misra-c2012-10.8 */
retValue = (qFP16_t)( ( (uint32_t)retValue >> 1u ) + bit );
retValue = (qFP16_t)( ( (uint32_t)retValue >> 1U ) + bit );
}
else {
/*cppcheck-suppress misra-c2012-10.1 */
Expand All @@ -398,7 +398,7 @@ qFP16_t qFP16_Sqrt( qFP16_t x )
bit >>= 2u;
}

if ( 0u == n ) {
if ( 0U == n ) {
if ( x > 65535 ) {
x -= retValue;
/*cppcheck-suppress misra-c2012-10.1 */
Expand All @@ -411,11 +411,11 @@ qFP16_t qFP16_Sqrt( qFP16_t x )
retValue <<= 16;
}
/*cppcheck-suppress [ misra-c2012-10.6, misra-c2012-12.2 ] */
bit = 1u << 14u;
bit = 1U << 14u;
}
}
}
if ( ( 1u == fp->rounding ) && ( x > retValue ) ) {
if ( ( 1U == fp->rounding ) && ( x > retValue ) ) {
++retValue;
}

Expand Down Expand Up @@ -521,7 +521,7 @@ qFP16_t qFP16_Log2( const qFP16_t x )
retValue = qFP16_log2i( x );
}
}
if ( 1u == fp->saturate ) {
if ( 1U == fp->saturate ) {
if ( qFP16.overflow == retValue ) {
retValue = fp->min;
}
Expand Down Expand Up @@ -739,7 +739,7 @@ qFP16_t qFP16_Polyval( const qFP16_t * const p,
size_t i;
/*polynomial evaluation using Horner's method*/
fx = p[ 0 ];
for ( i = 1u ; i < n ; ++i ) {
for ( i = 1U ; i < n ; ++i ) {
qFP16_t tmp = qFP16_Mul( fx, x );

if ( qFP16.overflow == tmp ) {
Expand Down Expand Up @@ -785,7 +785,7 @@ qFP16_t qFP16_Pow( const qFP16_t x,
{
qFP16_t retValue = qFP16.overflow;

if ( ( 0u == ( (uint32_t)y & intern.fraction_mask ) ) && ( y > 0 ) ) {
if ( ( 0U == ( (uint32_t)y & intern.fraction_mask ) ) && ( y > 0 ) ) {
/*handle integer exponent explicitly*/
retValue = qFP16_IPow( x, y );
}
Expand Down Expand Up @@ -821,7 +821,7 @@ char* qFP16_FPToA( const qFP16_t num,
str[ 8 ] = '\0';
}
else {
const uint32_t iScales[ 6 ] = { 1u, 10u, 100u, 1000u, 10000u, 100000u };
const uint32_t iScales[ 6 ] = { 1U, 10U, 100U, 1000U, 10000U, 100000U };
uint32_t uValue, fPart, scale;
int32_t iPart;

Expand All @@ -845,11 +845,11 @@ char* qFP16_FPToA( const qFP16_t num,
iPart++;
fPart -= scale;
}
str = qFP16_itoa( str, 10000, (uint32_t)iPart, 1u );
str = qFP16_itoa( str, 10000, (uint32_t)iPart, 1U );

if ( 1u != scale ) {
if ( 1U != scale ) {
*str++ = '.';
str = qFP16_itoa( str, scale/10u, fPart, 0u );
str = qFP16_itoa( str, scale/10U, fPart, 0U );
}
*str = '\0';
}
Expand All @@ -860,7 +860,7 @@ char* qFP16_FPToA( const qFP16_t num,
qFP16_t qFP16_AToFP( const char *s )
{
uint8_t neg;
uint32_t iPart = 0u, fPart = 0u, scale = 1u, digit;
uint32_t iPart = 0U, fPart = 0U, scale = 1U, digit;
int32_t count = 0;
qFP16_t retValue = qFP16.overflow;
bool point_seen = false, overflow = false, badchr = false;
Expand All @@ -871,7 +871,7 @@ qFP16_t qFP16_AToFP( const char *s )
s++; /*discard whitespaces*/
}

neg = ( '-' == *s ) ? 1u : 0u;
neg = ( '-' == *s ) ? 1U : 0U;
if ( ( '+' == *s ) || ( '-' == *s ) ) {
s++; /*move to the next sign*/
}
Expand All @@ -883,17 +883,17 @@ qFP16_t qFP16_AToFP( const char *s )
else if ( 0 != isdigit( (int)c ) ) {
digit = (uint32_t)c - (uint32_t)'0';
if ( point_seen ) { /* Decode the fractional part */
scale *= 10u;
fPart *= 10u;
scale *= 10U;
fPart *= 10U;
fPart += digit;
}
else { /* Decode the decimal part */
iPart *= 10u;
iPart *= 10U;
iPart += digit;
++count;
overflow = ( ( 0 == count ) || ( count > 5 ) ||
( iPart > 32768u ) ||
( ( 0u == neg ) && ( iPart > 32767u ) ) );
( ( 0U == neg ) && ( iPart > 32767u ) ) );
}
}
else {
Expand All @@ -908,7 +908,7 @@ qFP16_t qFP16_AToFP( const char *s )
/*cppcheck-suppress misra-c2012-10.1 */
retValue = (qFP16_t)iPart << 16;
retValue += qFP16_Div( (qFP16_t)fPart, (qFP16_t)scale );
retValue = ( 1u == neg ) ? -retValue : retValue;
retValue = ( 1U == neg ) ? -retValue : retValue;
}
/*cstat +MISRAC2012-Dir-4.11_h*/
return retValue;
Expand All @@ -918,9 +918,9 @@ static qFP16_t qFP16_rs( const qFP16_t x )
{
qFP16_t retValue;

if ( 1u == fp->rounding ) {
if ( 1U == fp->rounding ) {
/*cppcheck-suppress misra-c2012-10.1 */
retValue = ( x >> 1u ) + ( x & 1 );
retValue = ( x >> 1U ) + ( x & 1 );
}
else {
/*cppcheck-suppress misra-c2012-10.1 */
Expand Down Expand Up @@ -953,7 +953,7 @@ static qFP16_t qFP16_log2i( qFP16_t x )
x = qFP16_rs( x );
}
}
if ( 1u == fp->rounding ) {
if ( 1U == fp->rounding ) {
x = qFP16_Mul( x, x );
if ( x >= intern.f_2 ) {
++retValue;
Expand All @@ -969,16 +969,16 @@ static char *qFP16_itoa( char *buf,
uint32_t value,
uint8_t skip )
{
while ( 0u != scale ) {
while ( 0U != scale ) {
uint32_t digit = ( value / scale );
if ( ( 0u == skip ) || ( 0u != digit ) || ( 1u == scale ) ) {
skip = 0u;
if ( ( 0U == skip ) || ( 0U != digit ) || ( 1U == scale ) ) {
skip = 0U;
/*cstat -MISRAC2012-Rule-10.2 -MISRAC2012-Rule-10.3*/
*buf++ = (char)'0' + (char)digit;
/*cstat +MISRAC2012-Rule-10.2 +MISRAC2012-Rule-10.3*/
value %= scale;
}
scale /= 10u;
scale /= 10U;
}

return buf;
Expand All @@ -990,7 +990,7 @@ static qFP16_t qFP16_Saturate( const qFP16_t nsInput,
{
qFP16_t retValue = nsInput;

if ( 1u == fp->saturate ) {
if ( 1U == fp->saturate ) {
if ( qFP16.overflow == nsInput ) {
retValue = ( ( x >= 0 ) == ( y >= 0 ) ) ? fp->max : fp->min;
}
Expand Down
13 changes: 9 additions & 4 deletions qssmoother.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -155,7 +155,9 @@ static int qSSmoother_Setup_LPF2( _qSSmoother_t * const f,
/*cppcheck-suppress misra-c2012-11.3 */
qSSmoother_LPF2_t * const s = (qSSmoother_LPF2_t* const)f;
/*cstat +MISRAC2012-Rule-11.3 +CERT-EXP39-C_d +CERT-EXP36-C_a*/
float aa, p1, r;
float aa;
float p1;
float r;
aa = alpha*alpha;
/*cstat -MISRAC2012-Dir-4.11_b*/
p1 = QLIB_SQRT( 2.0F*alpha ); /*arg always positive*/
Expand Down Expand Up @@ -278,9 +280,11 @@ static int qSSmoother_Setup_GMWF( _qSSmoother_t * const f,
qSSmoother_GMWF_t * const s = (qSSmoother_GMWF_t* const)f;
/*cstat +MISRAC2012-Rule-11.3 +CERT-EXP39-C_d +CERT-EXP36-C_a*/
float *kernel = &window[ ws ];
float r, sum = 0.0F;
float r;
float sum = 0.0F;
size_t i;
float l, center;
float l;
float center;
/*cstat -CERT-FLP36-C -MISRAC2012-Rule-10.8*/
/*cppcheck-suppress misra-c2012-10.8 */
l = (float)( wsize - 1U )/2.0F;
Expand Down Expand Up @@ -709,7 +713,8 @@ static float qSSmoother_Filter_ALNF( _qSSmoother_t * const f,
float *w_1 = &s->w[ s->n ];

for ( i = 0U ; i < s->n ; ++i ) {
float w = s->w[ i ], w1 = w_1[ i ];
float w = s->w[ i ];
float w1 = w_1[ i ];

s->w[ i ] += ( s->alpha*( x - xe )*s->x[ i ] ) + ( s->mu*( w - w1 ) );
w_1[ i ] = w;
Expand Down
10 changes: 7 additions & 3 deletions qtypegeneric.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -24,7 +24,8 @@ void qTypeGeneric_Swap( void * const x,
{
/*cstat -MISRAC2012-Rule-11.5 -CERT-EXP36-C_b*/
/*cppcheck-suppress misra-c2012-11.5 */
uint8_t * const a = (uint8_t * const)x, *b = (uint8_t * const)y;
uint8_t * const a = (uint8_t * const)x;
uint8_t *b = (uint8_t * const)y;
/*cstat +MISRAC2012-Rule-11.5 +CERT-EXP36-C_b*/
size_t i = 0U, j = 0U;
do {
Expand Down Expand Up @@ -177,7 +178,9 @@ void qTypeGeneric_Sort( void * const pbase,
if ( tmp_ptr != run_ptr ) {
uint8_t *tra = run_ptr + size;
while ( --tra >= run_ptr ) {
uint8_t c = *tra, *hi = tra, *lo = tra;
uint8_t c = *tra;
uint8_t *hi = tra;
uint8_t *lo = tra;
/*cppcheck-suppress misra-c2012-10.3 */
while ( (lo -= size) >= tmp_ptr ) {
*hi = *lo;
Expand All @@ -197,7 +200,8 @@ void qTypeGeneric_Reverse( void * const pbase,
const size_t end )
{
if ( ( NULL != pbase ) && ( size > 0U ) && ( end > init ) ) {
size_t s = size*init, e = size*end;
size_t s = size*init;
size_t e = size*end;
/*cstat -MISRAC2012-Rule-11.5 -CERT-EXP36-C_b*/
/*cppcheck-suppress misra-c2012-11.5 */
uint8_t *v = (uint8_t*)pbase;
Expand Down
Loading

0 comments on commit 52295d4

Please sign in to comment.