[Tarantool-patches] [PATCH v2 04/10] crc32: align memory access

[Vladislav Shpilevoy]

Thanks for the review!

On 28/05/2020 22:11, Timur Safin wrote:
> 
> 
> : -----Original Message-----
> : From: Vladislav Shpilevoy <v.shpilevoy at tarantool.org>
> : Subject: [PATCH v2 04/10] crc32: align memory access
> : 
> : 
> : diff --git a/src/cpu_feature.c b/src/cpu_feature.c
> : index 98567ccb3..9bf6223de 100644
> : --- a/src/cpu_feature.c
> : +++ b/src/cpu_feature.c
> : @@ -50,7 +51,7 @@
> : 
> : 
> :  static uint32_t
> : -crc32c_hw_byte(uint32_t crc, unsigned char const *data, unsigned int
> : length)
> : +crc32c_hw_byte(uint32_t crc, char const *data, unsigned int length)
> :  {
> :  	while (length--) {
> :  		__asm__ __volatile__(
> : @@ -68,6 +69,26 @@ crc32c_hw_byte(uint32_t crc, unsigned char const *data,
> : unsigned int length)
> :  uint32_t
> :  crc32c_hw(uint32_t crc, const char *buf, unsigned int len)
> :  {
> : +	const int align = alignof(unsigned long);
> : +	unsigned long addr = (unsigned long)buf;
> : +	unsigned int not_aligned_prefix =
> : +		((addr - 1 + align) & ~(align - 1)) - addr;
> 
> Hmm, hmm...
> 
> Isn't it simple `addr % align`? Or even `addr & (align - 1)` ?

Consider the example: addr = 14, align = 8.
Then not_aligned_prefix = 2. Need to read first 2
bytes one by one to get to 16, the closest aligned
address.

addr % align = 14 % 8 = 6 != 2
addr & (align - 1) = 14 & 7 = 1110 & 0111 = 110 = 6 != 2

But yeah, this could be done simpler: align - addr % align.
This will give how many bytes are needed to the next
aligned address. I wrote the solution above by blindly using
'aligned - not aligned' and the same schema as in
small_align().

Here is the diff:

====================
diff --git a/src/cpu_feature.c b/src/cpu_feature.c
index 9bf6223de..7b284fa98 100644
--- a/src/cpu_feature.c
+++ b/src/cpu_feature.c
@@ -69,10 +69,8 @@ crc32c_hw_byte(uint32_t crc, char const *data, unsigned int length)
 uint32_t
 crc32c_hw(uint32_t crc, const char *buf, unsigned int len)
 {
-	const int align = alignof(unsigned long);
-	unsigned long addr = (unsigned long)buf;
-	unsigned int not_aligned_prefix =
-		((addr - 1 + align) & ~(align - 1)) - addr;
+	const unsigned int align = alignof(unsigned long);
+	unsigned int not_aligned_prefix = align - (unsigned long)buf % align;
 	/*
 	 * Calculate CRC32 for the prefix byte-by-byte so as to
 	 * then use aligned words to calculate the rest. This is