Macro Compile Error | Microsoft Connect

Macro Compile Error by baexps_pr1

Closed

as By Design Help for as By Design

Type:	Bug
ID:	780128
Opened:	2/27/2013 9:10:12 AM
Access Restriction:	Public

Workaround(s)

view

User(s) can reproduce this bug

Unable to Compile the following code snippet :
------------------------------------------------------------------------------
#include <stdio.h>

#define FALSE 0
#define TRUE !FALSE

typedef unsigned int unsigned32_T;
typedef unsigned short int unsigned16_T;
typedef unsigned32_T u32_fixed_point_T;

typedef unsigned16_T u18pm2_T;     /* bbbbbbbbbbbbbbbb00.0 */
#define BPP_u18pm2_T -2
#define SIGNED_u18pm2_T FALSE

#define Init_Fixed_Point(value, to_type) \
( \
( (BPP_##to_type) >= 0U ) ? \
     ( ( (value) >= 0U ) ? \
         ( (SIGNED_##to_type) ? \
            ( ( (value) >= ( ( ( (u32_fixed_point_T)1U << ( 8U*sizeof(to_type)-1U ) ) - 1.0 ) / ( (u32_fixed_point_T)1U << (BPP_##to_type) ) ) ) ? \
             /* return max. signed counts on overflow for signed type */ \
             ( (to_type)( ( (to_type)1U << ( 8U*sizeof(to_type)-1U ) ) - 1 ) ): \
             ( (to_type)( ( (value) * ( (u32_fixed_point_T)1U << (BPP_##to_type) ) ) + 0.5 ) ) \
            ): \
            ( ( (value) >= ( ( (to_type)0xFFFFFFFFU + 0.0 ) / ( (u32_fixed_point_T)1U << (BPP_##to_type) ) ) ) ? \
             /* "+ 0.0" in above expression to force decimal arithmetic to get engineering limit */ \
             /* return max. unsigned counts on overflow for unsigned type */ \
             ( (to_type)0xFFFFFFFFU ): \
             ( (to_type)( ( (value) * ( (u32_fixed_point_T)1 << (BPP_##to_type) ) ) + 0.5 ) ) \
            ) \
         ): \
         ( (SIGNED_##to_type) ? \
            ( ( (value) <= -( ( ( (u32_fixed_point_T)1U << ( 8U*sizeof(to_type)-1U ) ) + 0.0 ) / ( (u32_fixed_point_T)1U << (BPP_##to_type) ) ) )? \
             /* "+ 0.0" in above expression to force decimal arithmetic to get engineering limit */ \
             /* return min. signed counts on underflow for signed type */ \
             ( (to_type)( -( (to_type)1U << ( 8U*sizeof(to_type)-1U ) ) ) ): \
             ( (to_type)( (double)(value)* ( (u32_fixed_point_T)1U << (BPP_##to_type) ) - 0.5) ) \
            ): \
            /* return min. unsigned counts on underflow for unsigned type */ \
            ((to_type)0U) \
         ) \
     ): \
     ( ( (value) >= 0U ) ? \
         ( (SIGNED_##to_type) ? \
            ( ( (value) >= ( ( ( (u32_fixed_point_T)1U << ( 8U*sizeof(to_type)-1U ) ) - 1U ) * ( (u32_fixed_point_T)1U << ( -(BPP_##to_type) ) ) ) ) ? \
             /* return max. signed counts on overflow for signed type */ \
             ( (to_type)( ( (to_type)1U << ( 8U*sizeof(to_type)-1U ) ) - 1U ) ): \
             ( (to_type)( ( ( (value) * ( 0x00010000U >> ( -(BPP_##to_type) ) ) ) + 32768U ) / 65536U ) ) \
            ): \
            ( ( (value) >= ( ( (to_type)0xFFFFFFFFU ) * ( (u32_fixed_point_T)1U << ( -(BPP_##to_type) ) ) ) ) ? \
             /* return max. unsigned counts on overflow for unsigned type */ \
             ( (to_type)0xFFFFFFFFU ): \
             ( (to_type)( /* ( ( (u32_fixed_point_T)(value) * ( 0x00010000U >> ( -(BPP_##to_type) ) ) ) + 32768 ) */ 1U /* / 65536 */ ) ) \
            ) \
         ): \
         ( (SIGNED_##to_type) ? \
            ( ( ( (signed int)( ( (value) / ( ( 1U << ( -(BPP_##to_type) ) ) + 0.0 ) ) - ( -( 1U << ( 8U*sizeof(to_type)-1U ) ) ) ) ) <= 0U ) ? \
             /* return min. signed counts on underflow for signed type */ \
             ( (to_type)( -( (to_type)1U << ( 8U*sizeof(to_type)-1U ) ) ) ): \
             ( (to_type)( ( ( (value) * ( 0x00010000U >> ( -(BPP_##to_type) ) ) ) - 32768U ) / 65536U ) ) \
            ): \
            /* return min. unsigned counts on underflow for unsigned type */ \
            ((to_type)0U) \
         ) \
     ) \
)

#define MAX_T_DELTA             Init_Fixed_Point( 262140U, u18pm2_T )

int volatile test_data = 5;

void main(void)
{

printf("Hello World\n");

if (test_data > MAX_T_DELTA)
{
     printf("Data Error\n");
}
}

--------------------------------------------------------

Details (expand)

Visual Studio 2012

Just compile the above code by copying to a file called main.c
with the following command :
"C:\Program Files\Microsoft Visual Studio 9.0\VC\bin\cl" /FD /c /TP /Zi /Fo".\ma
in.obj" /FAs /X /GS /GL /Gy /EHsc /I. /I"C:\Program Files\Microsoft SDKs\W
indows\v7.1\Include" /I"C:\Program Files\Microsoft Visual Studio 9.0\VC\include
" /MD /W3 /O1 /D "WIN32" /D "_CONSOLE" /D "_DLL" /D "_MT" /D "_HAS_TR1=0" -DDE
VICE_32_BIT_PROCESSOR=1 -DMDI .\main.c

----------------------------
The Error I get :

.\main.c(78) : error C2124: divide or mod by zero

------------------------------
Compiler Build Version :
Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 15.00.30729.01 for 80x86
Copyright (C) Microsoft Corporation. All rights reserved.
-----------------------------------

English

Windows 7 SP1

English

Compiler should not produce the above mentioned error.

I am attaching the file for easy coompiling

File Name	Submitted By	Submitted On	File Size
compile.zip	baexps_pr1	2/27/2013	1 KB

Posted by baexps_pr1 on 3/16/2013 at 7:36 PM

Hello Support,

Sorry for the wrong comment in the last mail.
Ignore my last comment.

So within C standard, during the First phase of constant folding within Compile Phase, the non-executable code is checked for NEGATIVE SHIFT even though CPU will never see that code due to CONSTANT FOLDING of the expression.

Thank you.
Regards

Posted by baexps_pr1 on 3/16/2013 at 7:11 PM

Hello Support,
Shift bit is not negative.
It is -(-2) = 2 during the [preprocessor phase].

Can you please tell me which line during the Preprocessor phase is becoming NEGATIVE SHIFT?

Thank you.
Regards

Posted by Microsoft on 3/12/2013 at 2:45 PM

Hi,
    Thanks for reporting the issue.
    It is by design. Accordong to the standard, the shift bits should be non-negative (even if the code is unreachable), otherwise the behavior is undefined. For example,

int main()
{
    (-2 > 0) ? 1 / (1 << -2) : 0;
}

Posted by Microsoft on 2/27/2013 at 10:58 PM

Thanks for your feedback.

We are rerouting this issue to the appropriate group within the Visual Studio Product Team for triage and resolution. These specialized experts will follow-up with your issue.

Posted by Microsoft on 2/27/2013 at 9:50 AM

Thank you for your feedback, we are currently reviewing the issue you have submitted. If this issue is urgent, please contact support directly(http://support.microsoft.com)