7.15.6 Using printf() with Vectors

In addition to vector operations, printf() support is provided for outputting vector data. See Section 6.12.13 of The OpenCL Specification version 1.2 for details about formatting vector data types using printf(). Note that the Texas Instruments C6000 implementation differs in the following ways from the OpenCL specification:

• Return value. The OpenCL specification indicates that printf() returns 0 or -1. To be consistent with the C99 specification for scalar printf(), the return value for vector printf() is the number of characters printed.
• Length modifier for 64-bit integers. However, the C6000 vector types do not include a longn or ulongn vector type, because both int and long are 32 bits. So, using a length modifier of %l (ell) would be inconsistent. Instead, use a length modifier of %ll (ell ell) for 64-bit long long and unsigned long long vector types.

That is, the ll length modifier specifies that a following d, i, o, u, x, or X conversion specifier applies to a longlongn or ulonglongn argument. Use the l length modifier for 64-bit doublen arguments as described in the OpenCL specification.

The following example declares, initializes, and prints a vector of four 32-bit floating values, a vector of four 8-bit unsigned char values, and a vector of two 64-bit long long values.

float4 f4 = (float4)(1.0f, 2.0f, 3.0f, 4.0f); uchar4 uc = (uchar4)(0xFA, 0xFB, 0xFC, 0xFD); longlong2 bigNums = (longlong2)(600000000000, -600000000000); printf("f4 = %2.2v4hlf\n", f); printf("uc = %#v4hhx\n", uc); printf("bigNums = %+v2lld\n", bigNums);

The example prints the vector containing float values using the %2.2v4hlf format string to output values at least 2 digits wide with a precision of 2 (2.2) a vector of length 4 (v4) with a floatn length modifier (hl) using the float type specifier (f).

The example prints the vector containing uchar values using the %#v4hhx format string to output a 0x prefix (#) followed by a vector of length 4 (v4) with a charn or ucharn length modifier (hh) using lowercase hexadecimal notation (x).

The example prints the vector containing long long values using the %+v2lld format string to output values with a + or - sign prefix (+) followed by a vector of length 2 (v2) with a longlongn or ulonglongn length modifier (ll) using decimal notation (d).

/* Output */ f4 = 1.00,2.00,3.00,4.00 uc = 0xfa,0xfb,0xfc,0xfd bigNums = +600000000000,-600000000000