Re: [PATCH v4 01/15] crypto: accumulative hashing API

[Markus Armbruster]

Daniel P. Berrangé <berrange@redhat.com> writes:

> On Wed, Aug 07, 2024 at 07:51:08PM +0000, Alejandro Zeise wrote:
>> Changes the hash API to support accumulative hashing.
>> Hash objects are created with "qcrypto_hash_new",
>> updated with data with "qcrypto_hash_update", and
>> the hash obtained with "qcrypto_hash_finalize".
>> 
>> These changes bring the hashing API more in line with the
>> hmac API.
>> 
>> Signed-off-by: Alejandro Zeise <alejandro.zeise@seagate.com>

[...]

>> diff --git a/include/crypto/hash.h b/include/crypto/hash.h
>> index 54d87aa2a1..6d7222867e 100644
>> --- a/include/crypto/hash.h
>> +++ b/include/crypto/hash.h
>> @@ -1,6 +1,7 @@
>>  /*
>>   * QEMU Crypto hash algorithms
>>   *
>> + * Copyright (c) 2024 Seagate Technology LLC and/or its Affiliates
>>   * Copyright (c) 2015 Red Hat, Inc.
>>   *
>>   * This library is free software; you can redistribute it and/or
>> @@ -25,6 +26,13 @@
>>  
>>  /* See also "QCryptoHashAlgorithm" defined in qapi/crypto.json */
>>  
>> +typedef struct QCryptoHash QCryptoHash;
>> +struct QCryptoHash {
>> +    QCryptoHashAlgorithm alg;
>> +    void *opaque;
>> +    void *driver;
>> +};
>> +
>>  /**
>>   * qcrypto_hash_supports:
>>   * @alg: the hash algorithm
>> @@ -120,6 +128,117 @@ int qcrypto_hash_digestv(QCryptoHashAlgorithm alg,
>>                           char **digest,
>>                           Error **errp);
>>  
>> +/**
>> + * qcrypto_hash_updatev:
>> + * @hash: hash object from qcrypto_hash_new
>> + * @iov: the array of memory regions to hash
>> + * @niov: the length of @iov
>> + * @errp: pointer to a NULL-initialized error object
>> + *
>> + * Updates the given hash object with all the memory regions
>> + * present in @iov.
>> + *
>> + * Returns: 0 on success, non-zero on error
>
> Minor point, this and all the other APIs should be saying
> 'or -1 on error' to follow QEMU's error reporting standards.

Specifically, qapi/error.h:

 * - Whenever practical, also return a value that indicates success /
 *   failure.  This can make the error checking more concise, and can
 *   avoid useless error object creation and destruction.  Note that
 *   we still have many functions returning void.  We recommend
 *   • bool-valued functions return true on success / false on failure,
 *   • pointer-valued functions return non-null / null pointer, and
 *   • integer-valued functions return non-negative / negative.