How to Print int64_t Portably in C Using inttypes.h

By the end of this page, you will understand why int64_t should not be printed with hard-coded format specifiers like %lld or %I64d, and how to print it portably using the macros from inttypes.h, such as PRId64 and PRIu64. You will also see how this fits into real C codebases and how to avoid common printf format mismatches.

In C, fixed-width integer types such as int64_t and uint64_t are defined in stdint.h. They guarantee a width, such as exactly 64 bits, but they do not guarantee which base C type is used underneath.

For example, on one platform int64_t might be a long, while on another it might be a long long. Because printf requires the format string to match the actual underlying type, using a hard-coded specifier like %lld or %I64d is not fully portable.

That is why C provides inttypes.h. This header defines format macros specifically for fixed-width integer types:

• PRId64 for signed int64_t
• PRIu64 for unsigned uint64_t
• and many others for other integer widths

These macros expand to the correct format string for the current platform.

This matters because printf is a variadic function. Variadic functions do not know argument types automatically. They trust the format string. If the format string and argument type do not match, you can get:

• compiler warnings
• incorrect output
• undefined behavior

Using PRId64 and related macros makes your code portable, correct, and warning-free.

Think of int64_t as a box labeled "64-bit signed integer".

The label tells you the size of the value, but not the exact material the box is made from. On one machine, the box might really be a long; on another, a long long.

printf does not look at the label on the box. It only follows the instructions in the format string.

So if you say %lld, you are telling printf: "Expect a long long." If the value is actually represented as a long on that platform, the instruction does not match the box.

PRId64 is like asking the compiler: "Please give me the right instruction for this platform."

To print fixed-width integers portably, include both stdint.h and inttypes.h.

#include <stdio.h>
#include <stdint.h>
#include <inttypes.h>

Signed int64_t

#include <stdio.h>
#include <stdint.h>
#include <inttypes.h>

int main(void) {
    int64_t my_int = 999999999999999999;
    printf("This is my_int: %" PRId64 "\n", my_int);
    return 0;
}

Unsigned uint64_t

#include <stdio.h>
#include <stdint.h>
#include <inttypes.h>

int main(void) {
    uint64_t value = 999999999999999999ULL;
    printf("This is value: %" PRIu64 "\n", value);
    return 0;
}

Why the format looks unusual

This part:

"%" PRId64 "\n"

is just string literal concatenation in C. The compiler joins it into one format string.

If PRId64 expands to lld, the final result becomes:

"%lld\n"

If it expands to ld, then the final string becomes:

"%ld\n"

That is exactly what makes it portable.

Consider this example:

#include <stdio.h>
#include <stdint.h>
#include <inttypes.h>

int main(void) {
    int64_t count = 42;
    printf("count = %" PRId64 "\n", count);
    return 0;
}

Here is what happens step by step:

1. #include <stdint.h> defines the type int64_t.
2. #include <inttypes.h> defines the macro PRId64.
3. int64_t count = 42; stores a 64-bit signed integer.
4. The compiler sees:

   "count = %" PRId64 "\n"
   

5. It expands PRId64 to the correct platform-specific format part, such as lld or ld.
6. Adjacent string literals are merged into one string.
7. printf receives a format string that correctly matches the underlying type of count.
8. The program prints:

   count = 42
   

Because the format matches the type, the compiler warning goes away.

Portable fixed-width integer printing is common in real programs where exact integer sizes matter.

File sizes and byte counters

uint64_t bytes_processed = 1234567890ULL;
printf("Processed %" PRIu64 " bytes\n", bytes_processed);

Timestamps and IDs

int64_t event_id = 91234567890123;
printf("Event ID: %" PRId64 "\n", event_id);

Cross-platform libraries

If your code runs on Linux, macOS, and Windows, you should avoid platform-specific specifiers like %I64d.

Binary protocols and data formats

When reading or writing 64-bit values from network packets, database files, or custom binary formats, developers often use int64_t and uint64_t. Printing them correctly is important for debugging and logging.

In real C projects, developers usually follow these patterns:

Logging fixed-width values

fprintf(stderr, "offset=%" PRId64 "\n", offset);

This keeps logs portable across compilers and operating systems.

Validation and diagnostics

if (size < 0) {
    fprintf(stderr, "Invalid size: %" PRId64 "\n", size);
    return 1;
}

Configuration and parsing tools

Programs that parse configuration files or command-line arguments often store results in fixed-width types and print them back for debugging.

Wrapping format details in macros or helpers

Large codebases sometimes create helper functions for repeated logging:

void print_record_id(int64_t id) {
    printf("record id=%" PRId64 "\n", id);
}

This reduces repeated format mistakes.

Error handling

When printing values during failures, using the correct format matters even more because debugging output must be trustworthy.

Mistake 1: Using %lld for every int64_t

int64_t x = 10;
printf("%lld\n", x);  // may warn on some systems

Why it is a problem:

• int64_t is not guaranteed to be long long
• it may be long on some platforms

Use this instead:

printf("%" PRId64 "\n", x);

Mistake 2: Using Windows-specific %I64d

printf("%I64d\n", x);

Why it is a problem:

• %I64d is not standard C portable syntax
• it may fail or warn on non-Windows compilers

Use PRId64 instead.

Mistake 3: Forgetting to include inttypes.h

#include <stdio.h>
#include <stdint.h>

printf("%" PRId64 "\n", x);  // error: PRId64 not defined

Fix:

#include <inttypes.h>

Mistake 4: Mixing signed and unsigned macros

uint64_t u = 100;
printf("%" PRId64 "\n", u);  // wrong macro for unsigned type

Use:

printf("%" PRIu64 "\n", u);

Mistake 5: Adding the macro inside the string manually

Broken:

printf("%PRId64\n", x);

Correct:

printf("%" PRId64 "\n", x);

The macro is not text inside the string. It is a preprocessor token that becomes part of the string through concatenation.

int64_t vs long long

stdint.h vs inttypes.h

Use both together when you need fixed-width integers and portable printing.

#include <stdint.h>
#include <inttypes.h>

Print fixed-width integers

int64_t a;
uint64_t b;

printf("%" PRId64 "\n", a);
printf("%" PRIu64 "\n", b);

Common macros

Pattern to remember

printf("%" PRId64 "\n", value);

Rules

• Use stdint.h for the type.
• Use inttypes.h for the print macro.
• Do not assume int64_t is always long long.
• Do not use %I64d if you want portable code.
• Match signed types with PRId... and unsigned types with PRIu....

Edge case

int64_t exists only if the platform provides an exact 64-bit signed integer type. On most modern systems, it does.

What is the correct portable way to print int64_t in C?

Use PRId64 from inttypes.h:

printf("%" PRId64 "\n", value);

Why does %lld sometimes produce a warning for int64_t?

Because int64_t is not guaranteed to be long long on every platform. It may be a different underlying type.

Do I need both stdint.h and inttypes.h?

Yes. stdint.h gives you int64_t, and inttypes.h gives you PRId64.

Can I use %I64d in portable C code?

No. %I64d is compiler- or platform-specific and should not be used for portable code.

How do I print uint64_t?

Use PRIu64:

printf("%" PRIu64 "\n", value);

Why is the format written as "%" PRId64 instead of %PRId64?

Because PRId64 is a macro, not part of the string text. C concatenates adjacent string literals after macro expansion.

Does this only apply to printf?

No. The same idea applies to other formatted output functions such as fprintf and sprintf.

• printf format specifiers — directly related because mismatched format strings cause warnings and bugs.
• Variadic functions — printf relies on the format string because argument types are not automatically known.
• stdint.h fixed-width integers — int64_t comes from this header.
• inttypes.h macros — provides PRId64, PRIu64, and related formatting helpers.
• Signed vs unsigned integers — important when choosing between PRId64 and PRIu64.
• Type portability in C — helps write code that works across compilers and operating systems.