BIO_S_BIO(3ossl)

📦
version 3.5.9 · 6 versions
, in fedora 43, fedora 44

okay bestie, sit down, hydrate, because BIO_S_BIO(3ossl) is one of those OpenSSL pages that looks scary and is actually just a group project 💖 It's the manual for the BIO pair — the OpenSSL thing that lets you be the transport. No gatekeeping here, we're going through every single line together. 🫶

🌸 NAME 🌸

BIO_s_bio, BIO_make_bio_pair, BIO_destroy_bio_pair, BIO_shutdown_wr, BIO_set_write_buf_size, BIO_get_write_buf_size, BIO_new_bio_pair, BIO_get_write_guarantee, BIO_ctrl_get_write_guarantee, BIO_get_read_request, BIO_ctrl_get_read_request, BIO_ctrl_reset_read_request, BIO_nread0, BIO_nread, BIO_nwrite0, BIO_nwrite — that's the whole squad, and their whole personality is: BIO pair BIO. 💅

✨ SYNOPSIS ✨

like, here's the receipts, copy-pasted exactly so you can't blame me:

#include <openssl/bio.h>

const BIO_METHOD *BIO_s_bio(void);

int BIO_make_bio_pair(BIO *b1, BIO *b2);
int BIO_destroy_bio_pair(BIO *b);
int BIO_shutdown_wr(BIO *b);

int BIO_set_write_buf_size(BIO *b, long size);
size_t BIO_get_write_buf_size(BIO *b, long size);

int BIO_new_bio_pair(BIO **bio1, size_t writebuf1, BIO **bio2, size_t writebuf2);

int BIO_get_write_guarantee(BIO *b);
size_t BIO_ctrl_get_write_guarantee(BIO *b);
int BIO_get_read_request(BIO *b);
size_t BIO_ctrl_get_read_request(BIO *b);
int BIO_ctrl_reset_read_request(BIO *b);

int BIO_nread0(BIO *bio, char **buf);
int BIO_nread(BIO *bio, char **buf, int num);
int BIO_nwrite0(BIO *bio, char **buf);
int BIO_nwrite(BIO *bio, char **buf, int num);

🦋 DESCRIPTION 🦋

okay so first things first: BIO_s_bio() returns the method for a BIO pair. And what IS a BIO pair? It's literally a pair of source/sink BIOs where anything you write into one half gets buffered, and then you can read it right back out of the other half. Cute, right? It's giving pen pals. 💌

but bestie — ⚠️ both halves must usually be handled by the same application thread, because no locking is done on the internal data structures. no cap. don't be sharing a pair across threads and acting surprised. 😭

Since BIO chains typically end in a source/sink BIO, you can make one of these the final link and have all the data processed by the chain under application control. main character energy honestly.

One typical use — and this is the big one — is putting TLS/SSL I/O under application control. You'd do this when the application wants to use a non standard transport for TLS/SSL, or when the normal socket routines are just, like, inappropriate. 🙈

  • Calls to BIO_read_ex() will read data from the buffer, or request a retry if no data is available.
  • Calls to BIO_write_ex() will place data in the buffer, or request a retry if the buffer is full.

The standard calls BIO_ctrl_pending() and BIO_ctrl_wpending() tell you how much pending data is chilling in the read buffer or the write buffer respectively. vibe check your buffers, bestie.

BIO_reset() clears any data in the write buffer. poof. gone. ✨

BIO_make_bio_pair() joins two separate BIOs into a connected pair. two become one, iconic.

BIO_destroy_pair() destroys the association between two connected BIOs. (Yes, the manpage says BIO_destroy_pair() right there in the prose while the actual function is BIO_destroy_bio_pair() — I'm not fixing it, I'm just telling you, iykyk. 👀) Freeing up any half of the pair will automatically destroy the association.

BIO_shutdown_wr() closes down a BIO b. After this call, no further writes on BIO b are allowed — they will return an error. Reads on the other half of the pair will return any pending data, or EOF once all pending data has been read. so like, it's a soft goodbye. 🥲

BIO_set_write_buf_size() sets the write buffer size of BIO b to size. If the size is not initialized, a default value is used. That default is currently 17K, which is sufficient for a maximum size TLS record. slay.

BIO_get_write_buf_size() returns the size of the write buffer. that's it, that's the tweet.

BIO_new_bio_pair() is the combo deal — it combines the calls to BIO_new(), BIO_make_bio_pair() and BIO_set_write_buf_size() to create a connected pair of BIOs bio1 and bio2 with write buffer sizes writebuf1 and writebuf2. If either size is zero then the default size is used.

🚨 IMPORTANT, bestie: BIO_new_bio_pair() does not check whether bio1 or bio2 do point to some other BIO. The values are overwritten, and BIO_free() is not called. so if you had something in there, that's a you problem and a memory leak. 😬

BIO_get_write_guarantee() and BIO_ctrl_get_write_guarantee() return the maximum length of data that can currently be written to the BIO. Writes larger than this value will return a value from BIO_write_ex() less than the amount requested, or — if the buffer is full — request a retry. One tiny distinction: BIO_ctrl_get_write_guarantee() is a function, whereas BIO_get_write_guarantee() is a macro.

BIO_get_read_request() and BIO_ctrl_get_read_request() return the amount of data requested, or the buffer size if it is less, if the last read attempt at the other half of the BIO pair failed due to an empty buffer. This tells you how much data you should write to the BIO so the next read will succeed. It's most useful in TLS/SSL applications, where the amount of data read is usually meaningful rather than just some buffer size. After a successful read this call will return zero. It also returns zero once new data has been written satisfying the read request, or part of it. And note: BIO_get_read_request() never returns an amount larger than what BIO_get_write_guarantee() returned. never ever. 💅

BIO_ctrl_reset_read_request() can also be used to reset the value returned by BIO_get_read_request() to zero. little reset button, we love her. 🎀

🧚 Non-copying Interface 🧚

okay this is the part where it gets fancy. BIO_nread0(), BIO_nread(), BIO_nwrite0(), and BIO_nwrite() give you a non-copying interface for reading from and writing to BIO pairs. These functions allow direct access to the internal buffer, avoiding the overhead of copying data. no extra copies, no wasted cycles, we're being efficient today. ✨

BIO_nread0() returns in *buf a pointer to the start of the available data in the peer's write buffer, and returns the number of bytes available. This lets you read directly from the buffer without copying. It does not consume the data — a subsequent call to BIO_nread() is needed to advance the buffer position. peek first, commit later. 👀

BIO_nread() is similar to BIO_nread0() but also advances the read position by up to num bytes. The actual number of bytes consumed is returned. The *buf pointer is set to the start of the data that was consumed. Since the data is considered consumed after this call, the pointer returned by BIO_nread() should not be used afterwards — unless the caller also controls the writing side. The typical pattern is: call BIO_nread0() first, use the data, and then call BIO_nread() to consume it. bestie please do it in that order. 😭

BIO_nwrite0() returns in *buf a pointer to the start of the available space in the write buffer, and returns the number of bytes that can be written. This lets you write directly to the buffer without copying. It does not commit the data — a subsequent call to BIO_nwrite() is needed to update the buffer length. it's giving "draft folder". 📝

BIO_nwrite() is similar to BIO_nwrite0() but also commits up to num bytes as written. The actual number of bytes committed is returned. The *buf pointer is set to the start of the region that was committed. BIO_nwrite() should only be called after the data has actually been written to the buffer obtained from BIO_nwrite0(), since committing signals data availability to the reading side. don't commit early, it's not a situationship, it's a handshake. 🫠

⚠️ heads up: due to the ring buffer implementation, if wrapping around would be required, BIO_nread0() and BIO_nwrite0() may return less than the total available space. In such cases, a second call may be needed to access the remaining data or space. so like, don't assume one call got everything, okay? 🌀

🎀 NOTES 🎀

Both halves of a BIO pair should be freed. Yes, both. Even if one half is implicitly freed due to a BIO_free_all() or SSL_free() call, the other half still needs to be freed. it's a duo, you can't just abandon one of them. 💔

When used in bidirectional applications (such as TLS/SSL), care should be taken to flush any data in the write buffer. You can do this by calling BIO_pending() on the other half of the pair and, if any data is pending, reading it and sending it to the underlying transport. This must be done before any normal processing (such as calling select()) due to a request and BIO_should_read() being true.

To see why this is important, consider a case where a request is sent using BIO_write_ex() and a response read with BIO_read_ex() — this can occur during a TLS/SSL handshake, for example. BIO_write_ex() will succeed and place data in the write buffer. BIO_read_ex() will initially fail and BIO_should_read() will be true. If the application then waits for data to be available on the underlying transport before flushing the write buffer, it will never succeed — because the request was never sent! 💀 deadlock era. absolutely not.

BIO_eof() is true if no data is in the peer BIO and the peer BIO has been shutdown. 🌙

And for the record, these are all implemented as macros: BIO_make_bio_pair(), BIO_destroy_bio_pair(), BIO_shutdown_wr(), BIO_set_write_buf_size(), BIO_get_write_buf_size(), BIO_get_write_guarantee(), and BIO_get_read_request().

💜 RETURN VALUES 💜

BIO_new_bio_pair() returns 1 on success, with the new BIOs available in bio1 and bio2, or 0 on failure, with NULL pointers stored into the locations for bio1 and bio2. Check the error stack for more information. 🔍

📌 [XXXXX: More return values need to be added here] — yes, that's literally in the manual. the OpenSSL girlies left a TODO note and I'm respecting it, we don't erase the lore. 🫡

BIO_nread0() returns the number of bytes available for reading, 0 if the peer has closed and no data remains (EOF), or -1 if no data is currently available (retry may be appropriate). If the BIO is not initialized, -2 is returned.

BIO_nwrite0() returns the number of bytes of space available for writing, or -1 if no space is currently available (retry may be appropriate) or the BIO has been closed. If the BIO is not initialized, -2 is returned.

BIO_nread() and BIO_nwrite() return the number of bytes consumed or committed respectively, or the same error values as BIO_nread0() and BIO_nwrite0().

and bestie, we're not done yet — there's more of this page coming, so don't go anywhere 💖

✨ EXAMPLES ✨

okay bestie, this is the part where it all comes together 💖 The BIO pair can be used to have full control over the network access of an application, no cap. The application can call select() on the socket as required without having to go through the SSL-interface. it's giving power.

BIO *internal_bio, *network_bio;

...
BIO_new_bio_pair(&internal_bio, 0, &network_bio, 0);
SSL_set_bio(ssl, internal_bio, internal_bio);
SSL_operations(); /* e.g. SSL_read and SSL_write */
...

and here's the whole vibe, laid out for u 🌸

application |   TLS-engine
   |        |
   +----------> SSL_operations()
            |     /\    ||
            |     ||    \/
            |   BIO-pair (internal_bio)
            |   BIO-pair (network_bio)
            |     ||     /\
            |     \/     ||
   +-----------< BIO_operations()
   |        |
   |        |
  socket
 ...
 SSL_free(ssl);                /* implicitly frees internal_bio */
 BIO_free(network_bio);
 ...

So the BIO pair will only buffer the data and never directly access the connection, which means it behaves nonblocking and will return as soon as the write buffer is full or the read buffer is drained. Then the application has to flush the write buffer and/or fill the read buffer. that's the assignment, bestie 🫶

Use the BIO_ctrl_pending() to find out whether data is buffered in the BIO and must be transferred to the network. Use BIO_ctrl_get_read_request() to find out how many bytes must be written into the buffer before the SSL_operation() can successfully be continued. iykyk 💅

A typical usage pattern for the non-copying write interface is:

int ret;
char *buf;

ret = BIO_nwrite0(bio, &buf);
if (ret > 0) {
    /* write up to 'ret' bytes directly to 'buf' */
    memcpy(buf, data, len);
    BIO_nwrite(bio, &buf, len);  /* commit the write */
}

A typical usage pattern for the non-copying read interface is:

int ret;
char *buf;

ret = BIO_nread0(bio, &buf);
if (ret > 0) {
    /* read up to 'ret' bytes directly from 'buf' */
    process_data(buf, ret);
    BIO_nread(bio, &buf, ret);  /* consume the data */
}

🚨 WARNINGS 🚨

okay this is the serious part, listen up bestie 😭

As the data is buffered, SSL_operation() may return with an ERROR_SSL_WANT_READ condition, but there is still data in the write buffer. An application must not rely on the error value of SSL_operation() but must assure that the write buffer is always flushed first. Otherwise a deadlock may occur, because the peer might be waiting for the data before being able to continue. fr fr, do not skip this one 💀

👀 SEE ALSO 👀

Copyright 2000-2026 The OpenSSL Project Authors. All Rights Reserved.

Licensed under the Apache License 2.0 (the "License"). You may not use this file except in compliance with the License. You can obtain a copy in the file LICENSE in the source distribution or at <https://www.openssl.org/source/license.html>.

and that's a wrap, gorgeous 🌈 this one's from OpenSSL version 3.5.9, dated 2026-09-29, and the page is BIO_S_BIO(3ossl). go forth and buffer responsibly, icon 👑💖

📦 where this page comes from