freebsd-skq/sys/dev/hyperv/vmbus/hv_ring_buffer.c
Wei Hu da2f98a1cf Microsoft vmbus, storage and other related driver enhancements for HyperV.
- Vmbus multi channel support.
    - Vector interrupt support.
    - Signal optimization.
    - Storvsc driver performance improvement.
    - Scatter and gather support for storvsc driver.
    - Minor bug fix for KVP driver.
Thanks royger, jhb and delphij from FreeBSD community for the reviews
and comments. Also thanks Hovy Xu from NetApp for the contributions to
the storvsc driver.

PR:     195238
Submitted by:   whu
Reviewed by:    royger, jhb, delphij
Approved by:    royger
MFC after:      2 weeks
Relnotes:       yes
Sponsored by:   Microsoft OSTC
2015-04-29 10:12:34 +00:00

511 lines
12 KiB
C

/*-
* Copyright (c) 2009-2012 Microsoft Corp.
* Copyright (c) 2012 NetApp Inc.
* Copyright (c) 2012 Citrix Inc.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice unmodified, this list of conditions, and the following
* disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <sys/param.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include "hv_vmbus_priv.h"
/* Amount of space to write to */
#define HV_BYTES_AVAIL_TO_WRITE(r, w, z) ((w) >= (r))? \
((z) - ((w) - (r))):((r) - (w))
/**
* @brief Get number of bytes available to read and to write to
* for the specified ring buffer
*/
static inline void
get_ring_buffer_avail_bytes(
hv_vmbus_ring_buffer_info* rbi,
uint32_t* read,
uint32_t* write)
{
uint32_t read_loc, write_loc;
/*
* Capture the read/write indices before they changed
*/
read_loc = rbi->ring_buffer->read_index;
write_loc = rbi->ring_buffer->write_index;
*write = HV_BYTES_AVAIL_TO_WRITE(
read_loc, write_loc, rbi->ring_data_size);
*read = rbi->ring_data_size - *write;
}
/**
* @brief Get the next write location for the specified ring buffer
*/
static inline uint32_t
get_next_write_location(hv_vmbus_ring_buffer_info* ring_info)
{
uint32_t next = ring_info->ring_buffer->write_index;
return (next);
}
/**
* @brief Set the next write location for the specified ring buffer
*/
static inline void
set_next_write_location(
hv_vmbus_ring_buffer_info* ring_info,
uint32_t next_write_location)
{
ring_info->ring_buffer->write_index = next_write_location;
}
/**
* @brief Get the next read location for the specified ring buffer
*/
static inline uint32_t
get_next_read_location(hv_vmbus_ring_buffer_info* ring_info)
{
uint32_t next = ring_info->ring_buffer->read_index;
return (next);
}
/**
* @brief Get the next read location + offset for the specified ring buffer.
* This allows the caller to skip.
*/
static inline uint32_t
get_next_read_location_with_offset(
hv_vmbus_ring_buffer_info* ring_info,
uint32_t offset)
{
uint32_t next = ring_info->ring_buffer->read_index;
next += offset;
next %= ring_info->ring_data_size;
return (next);
}
/**
* @brief Set the next read location for the specified ring buffer
*/
static inline void
set_next_read_location(
hv_vmbus_ring_buffer_info* ring_info,
uint32_t next_read_location)
{
ring_info->ring_buffer->read_index = next_read_location;
}
/**
* @brief Get the start of the ring buffer
*/
static inline void *
get_ring_buffer(hv_vmbus_ring_buffer_info* ring_info)
{
return (void *) ring_info->ring_buffer->buffer;
}
/**
* @brief Get the size of the ring buffer.
*/
static inline uint32_t
get_ring_buffer_size(hv_vmbus_ring_buffer_info* ring_info)
{
return ring_info->ring_data_size;
}
/**
* Get the read and write indices as uint64_t of the specified ring buffer.
*/
static inline uint64_t
get_ring_buffer_indices(hv_vmbus_ring_buffer_info* ring_info)
{
return (uint64_t) ring_info->ring_buffer->write_index << 32;
}
void
hv_ring_buffer_read_begin(
hv_vmbus_ring_buffer_info* ring_info)
{
ring_info->ring_buffer->interrupt_mask = 1;
mb();
}
uint32_t
hv_ring_buffer_read_end(
hv_vmbus_ring_buffer_info* ring_info)
{
uint32_t read, write;
ring_info->ring_buffer->interrupt_mask = 0;
mb();
/*
* Now check to see if the ring buffer is still empty.
* If it is not, we raced and we need to process new
* incoming messages.
*/
get_ring_buffer_avail_bytes(ring_info, &read, &write);
return (read);
}
/*
* When we write to the ring buffer, check if the host needs to
* be signaled. Here is the details of this protocol:
*
* 1. The host guarantees that while it is draining the
* ring buffer, it will set the interrupt_mask to
* indicate it does not need to be interrupted when
* new data is placed.
*
* 2. The host guarantees that it will completely drain
* the ring buffer before exiting the read loop. Further,
* once the ring buffer is empty, it will clear the
* interrupt_mask and re-check to see if new data has
* arrived.
*/
static boolean_t
hv_ring_buffer_needsig_on_write(
uint32_t old_write_location,
hv_vmbus_ring_buffer_info* rbi)
{
mb();
if (rbi->ring_buffer->interrupt_mask)
return (FALSE);
/* Read memory barrier */
rmb();
/*
* This is the only case we need to signal when the
* ring transitions from being empty to non-empty.
*/
if (old_write_location == rbi->ring_buffer->read_index)
return (TRUE);
return (FALSE);
}
static uint32_t copy_to_ring_buffer(
hv_vmbus_ring_buffer_info* ring_info,
uint32_t start_write_offset,
char* src,
uint32_t src_len);
static uint32_t copy_from_ring_buffer(
hv_vmbus_ring_buffer_info* ring_info,
char* dest,
uint32_t dest_len,
uint32_t start_read_offset);
/**
* @brief Get the interrupt mask for the specified ring buffer.
*/
uint32_t
hv_vmbus_get_ring_buffer_interrupt_mask(hv_vmbus_ring_buffer_info *rbi)
{
return rbi->ring_buffer->interrupt_mask;
}
/**
* @brief Initialize the ring buffer.
*/
int
hv_vmbus_ring_buffer_init(
hv_vmbus_ring_buffer_info* ring_info,
void* buffer,
uint32_t buffer_len)
{
memset(ring_info, 0, sizeof(hv_vmbus_ring_buffer_info));
ring_info->ring_buffer = (hv_vmbus_ring_buffer*) buffer;
ring_info->ring_buffer->read_index =
ring_info->ring_buffer->write_index = 0;
ring_info->ring_size = buffer_len;
ring_info->ring_data_size = buffer_len - sizeof(hv_vmbus_ring_buffer);
mtx_init(&ring_info->ring_lock, "vmbus ring buffer", NULL, MTX_SPIN);
return (0);
}
/**
* @brief Cleanup the ring buffer.
*/
void hv_ring_buffer_cleanup(hv_vmbus_ring_buffer_info* ring_info)
{
mtx_destroy(&ring_info->ring_lock);
}
/**
* @brief Write to the ring buffer.
*/
int
hv_ring_buffer_write(
hv_vmbus_ring_buffer_info* out_ring_info,
hv_vmbus_sg_buffer_list sg_buffers[],
uint32_t sg_buffer_count,
boolean_t *need_sig)
{
int i = 0;
uint32_t byte_avail_to_write;
uint32_t byte_avail_to_read;
uint32_t old_write_location;
uint32_t total_bytes_to_write = 0;
volatile uint32_t next_write_location;
uint64_t prev_indices = 0;
for (i = 0; i < sg_buffer_count; i++) {
total_bytes_to_write += sg_buffers[i].length;
}
total_bytes_to_write += sizeof(uint64_t);
mtx_lock_spin(&out_ring_info->ring_lock);
get_ring_buffer_avail_bytes(out_ring_info, &byte_avail_to_read,
&byte_avail_to_write);
/*
* If there is only room for the packet, assume it is full.
* Otherwise, the next time around, we think the ring buffer
* is empty since the read index == write index
*/
if (byte_avail_to_write <= total_bytes_to_write) {
mtx_unlock_spin(&out_ring_info->ring_lock);
return (EAGAIN);
}
/*
* Write to the ring buffer
*/
next_write_location = get_next_write_location(out_ring_info);
old_write_location = next_write_location;
for (i = 0; i < sg_buffer_count; i++) {
next_write_location = copy_to_ring_buffer(out_ring_info,
next_write_location, (char *) sg_buffers[i].data,
sg_buffers[i].length);
}
/*
* Set previous packet start
*/
prev_indices = get_ring_buffer_indices(out_ring_info);
next_write_location = copy_to_ring_buffer(
out_ring_info, next_write_location,
(char *) &prev_indices, sizeof(uint64_t));
/*
* Full memory barrier before upding the write index.
*/
mb();
/*
* Now, update the write location
*/
set_next_write_location(out_ring_info, next_write_location);
mtx_unlock_spin(&out_ring_info->ring_lock);
*need_sig = hv_ring_buffer_needsig_on_write(old_write_location,
out_ring_info);
return (0);
}
/**
* @brief Read without advancing the read index.
*/
int
hv_ring_buffer_peek(
hv_vmbus_ring_buffer_info* in_ring_info,
void* buffer,
uint32_t buffer_len)
{
uint32_t bytesAvailToWrite;
uint32_t bytesAvailToRead;
uint32_t nextReadLocation = 0;
mtx_lock_spin(&in_ring_info->ring_lock);
get_ring_buffer_avail_bytes(in_ring_info, &bytesAvailToRead,
&bytesAvailToWrite);
/*
* Make sure there is something to read
*/
if (bytesAvailToRead < buffer_len) {
mtx_unlock_spin(&in_ring_info->ring_lock);
return (EAGAIN);
}
/*
* Convert to byte offset
*/
nextReadLocation = get_next_read_location(in_ring_info);
nextReadLocation = copy_from_ring_buffer(
in_ring_info, (char *)buffer, buffer_len, nextReadLocation);
mtx_unlock_spin(&in_ring_info->ring_lock);
return (0);
}
/**
* @brief Read and advance the read index.
*/
int
hv_ring_buffer_read(
hv_vmbus_ring_buffer_info* in_ring_info,
void* buffer,
uint32_t buffer_len,
uint32_t offset)
{
uint32_t bytes_avail_to_write;
uint32_t bytes_avail_to_read;
uint32_t next_read_location = 0;
uint64_t prev_indices = 0;
if (buffer_len <= 0)
return (EINVAL);
mtx_lock_spin(&in_ring_info->ring_lock);
get_ring_buffer_avail_bytes(
in_ring_info, &bytes_avail_to_read,
&bytes_avail_to_write);
/*
* Make sure there is something to read
*/
if (bytes_avail_to_read < buffer_len) {
mtx_unlock_spin(&in_ring_info->ring_lock);
return (EAGAIN);
}
next_read_location = get_next_read_location_with_offset(
in_ring_info,
offset);
next_read_location = copy_from_ring_buffer(
in_ring_info,
(char *) buffer,
buffer_len,
next_read_location);
next_read_location = copy_from_ring_buffer(
in_ring_info,
(char *) &prev_indices,
sizeof(uint64_t),
next_read_location);
/*
* Make sure all reads are done before we update the read index since
* the writer may start writing to the read area once the read index
* is updated.
*/
wmb();
/*
* Update the read index
*/
set_next_read_location(in_ring_info, next_read_location);
mtx_unlock_spin(&in_ring_info->ring_lock);
return (0);
}
/**
* @brief Helper routine to copy from source to ring buffer.
*
* Assume there is enough room. Handles wrap-around in dest case only!
*/
uint32_t
copy_to_ring_buffer(
hv_vmbus_ring_buffer_info* ring_info,
uint32_t start_write_offset,
char* src,
uint32_t src_len)
{
char *ring_buffer = get_ring_buffer(ring_info);
uint32_t ring_buffer_size = get_ring_buffer_size(ring_info);
uint32_t fragLen;
if (src_len > ring_buffer_size - start_write_offset) {
/* wrap-around detected! */
fragLen = ring_buffer_size - start_write_offset;
memcpy(ring_buffer + start_write_offset, src, fragLen);
memcpy(ring_buffer, src + fragLen, src_len - fragLen);
} else {
memcpy(ring_buffer + start_write_offset, src, src_len);
}
start_write_offset += src_len;
start_write_offset %= ring_buffer_size;
return (start_write_offset);
}
/**
* @brief Helper routine to copy to source from ring buffer.
*
* Assume there is enough room. Handles wrap-around in src case only!
*/
uint32_t
copy_from_ring_buffer(
hv_vmbus_ring_buffer_info* ring_info,
char* dest,
uint32_t dest_len,
uint32_t start_read_offset)
{
uint32_t fragLen;
char *ring_buffer = get_ring_buffer(ring_info);
uint32_t ring_buffer_size = get_ring_buffer_size(ring_info);
if (dest_len > ring_buffer_size - start_read_offset) {
/* wrap-around detected at the src */
fragLen = ring_buffer_size - start_read_offset;
memcpy(dest, ring_buffer + start_read_offset, fragLen);
memcpy(dest + fragLen, ring_buffer, dest_len - fragLen);
} else {
memcpy(dest, ring_buffer + start_read_offset, dest_len);
}
start_read_offset += dest_len;
start_read_offset %= ring_buffer_size;
return (start_read_offset);
}