2010-05-18 Chris Johns <chrisj@rtems.org>

* libblock/src/diskdevs.c, libblock/include/rtems/blkdev.h,
        libblock/src/bdbuf.c: PR 1448/filesystem.

        * libblock/include/rtems/blkdev.h, libblock/src/bdbuf.c: PR
        1514/filesystem.
This commit is contained in:
Chris Johns
2010-05-18 02:14:05 +00:00
parent baf64cdb32
commit 8aa608df32
4 changed files with 71 additions and 25 deletions

View File

@@ -1,3 +1,11 @@
2010-05-18 Chris Johns <chrisj@rtems.org>
* libblock/src/diskdevs.c, libblock/include/rtems/blkdev.h,
libblock/src/bdbuf.c: PR 1448/filesystem.
* libblock/include/rtems/blkdev.h, libblock/src/bdbuf.c: PR
1514/filesystem.
2010-05-17 Oleg Kravtsov <Oleg.Kravtsov@oktetlabs.ru>
PR 1449/cpukit

View File

@@ -42,22 +42,17 @@ extern "C" {
/**
* Block device request type.
*
* @warning The sync request is an IO one and only used from the cache. Use the
* Block IO when operating at the device level. We need a sync request
* to avoid requests looping for ever.
*/
typedef enum rtems_blkdev_request_op {
RTEMS_BLKDEV_REQ_READ, /**< Read the requested blocks of data. */
RTEMS_BLKDEV_REQ_WRITE, /**< Write the requested blocks of data. */
RTEMS_BLKDEV_CAPABILITIES /**< Return the driver capabilities set. */
RTEMS_BLKDEV_REQ_SYNC /**< Sync any data with the media. */
} rtems_blkdev_request_op;
/**
* Only consecutive multi-sector buffer requests are supported.
*
* This option means the cache will only supply multiple buffers that are
* inorder so the ATA multi-sector command for example can be used. This is a
* hack to work around the current ATA driver.
*/
#define RTEMS_BLKDEV_CAP_MULTISECTOR_CONT (1 << 0)
/**
* @brief Block device request done callback function type.
*
@@ -147,7 +142,7 @@ typedef struct rtems_blkdev_request {
/**
* The start block in a request.
*
* Only valid if the driver has returned the @ref RTEMS_BLKDEV_CAPABILITIES of
* Only valid if the driver has returned the @ref RTEMS_BLKIO_CAPABILITIES of
* @ref RTEMS_BLKDEV_CAP_MULTISECTOR_CONT.
*/
#define RTEMS_BLKDEV_START_BLOCK(req) (req->bufs[0].block)
@@ -165,9 +160,25 @@ typedef struct rtems_blkdev_request {
#define RTEMS_BLKIO_GETSIZE _IOR('B', 5, rtems_blkdev_bnum)
#define RTEMS_BLKIO_SYNCDEV _IO('B', 6)
#define RTEMS_BLKIO_DELETED _IO('B', 7)
#define RTEMS_BLKIO_CAPABILITIES _IO('B', 8)
/** @} */
/**
* Only consecutive multi-sector buffer requests are supported.
*
* This option means the cache will only supply multiple buffers that are
* inorder so the ATA multi-sector command for example can be used. This is a
* hack to work around the current ATA driver.
*/
#define RTEMS_BLKDEV_CAP_MULTISECTOR_CONT (1 << 0)
/**
* The driver will accept a sync call. A sync call is made to a driver
* after a bdbuf cache sync has finished.
*/
#define RTEMS_BLKDEV_CAP_SYNC (1 << 1)
/**
* The device driver interface conventions suppose that a driver may contain an
* initialize, open, close, read, write and IO control entry points. These

View File

@@ -60,6 +60,7 @@ typedef struct rtems_bdbuf_swapout_transfer
{
rtems_chain_control bds; /**< The transfer list of BDs. */
dev_t dev; /**< The device the transfer is for. */
bool syncing; /**< The data is a sync'ing. */
rtems_blkdev_request* write_req; /**< The write request array. */
uint32_t bufs_per_bd; /**< Number of buffers per bd. */
} rtems_bdbuf_swapout_transfer;
@@ -74,7 +75,7 @@ typedef struct rtems_bdbuf_swapout_worker
* idle. */
rtems_id id; /**< The id of the task so we can wake
* it. */
volatile bool enabled; /**< The worked is enabled. */
volatile bool enabled; /**< The worker is enabled. */
rtems_bdbuf_swapout_transfer transfer; /**< The transfer data for this
* thread. */
} rtems_bdbuf_swapout_worker;
@@ -1319,9 +1320,9 @@ rtems_bdbuf_init (void)
if (bdbuf_cache.initialised)
{
rtems_bdbuf_restore_preemption (prev_mode);
return RTEMS_RESOURCE_IN_USE;
}
memset(&bdbuf_cache, 0, sizeof(bdbuf_cache));
bdbuf_cache.initialised = true;
rtems_bdbuf_restore_preemption (prev_mode);
@@ -2042,7 +2043,7 @@ rtems_bdbuf_read (dev_t dev,
#define bdbuf_alloc(size) __builtin_alloca (size)
req = bdbuf_alloc (sizeof (rtems_blkdev_request) +
sizeof ( rtems_blkdev_sg_buffer) *
sizeof (rtems_blkdev_sg_buffer) *
(bdbuf_config.max_read_ahead_blocks + 1));
if (rtems_bdbuf_tracer)
@@ -2260,7 +2261,7 @@ rtems_bdbuf_null_disk_ioctl (rtems_disk_device *dd, uint32_t req, void *arg)
/**
* Swapout transfer to the driver. The driver will break this I/O into groups
* of consecutive write requests is multiple consecutive buffers are required
* by the driver.
* by the driver. The cache is not locked.
*
* @param transfer The transfer transaction.
*/
@@ -2373,7 +2374,20 @@ rtems_bdbuf_swapout_write (rtems_bdbuf_swapout_transfer* transfer)
}
if (dd != &null_disk)
{
/*
* If sync'ing and the deivce is capability of handling a sync IO control
* call perform the call.
*/
if (transfer->syncing &&
(dd->phys_dev->capabilities & RTEMS_BLKDEV_CAP_SYNC))
{
/* int result = */ dd->ioctl (dd->phys_dev, RTEMS_BLKDEV_REQ_SYNC, NULL);
/* How should the error be handled ? */
}
rtems_disk_release (dd);
}
}
}
@@ -2401,20 +2415,31 @@ rtems_bdbuf_swapout_modified_processing (dev_t* dev,
if (!rtems_chain_is_empty (chain))
{
rtems_chain_node* node = rtems_chain_head (chain);
bool sync_all;
node = node->next;
/*
* A sync active with no valid dev means sync all.
*/
if (sync_active && (*dev == BDBUF_INVALID_DEV))
sync_all = true;
else
sync_all = false;
while (!rtems_chain_is_tail (chain, node))
{
rtems_bdbuf_buffer* bd = (rtems_bdbuf_buffer*) node;
/*
* Check if the buffer's hold timer has reached 0. If a sync is active
* or someone waits for a buffer force all the timers to 0.
* or someone waits for a buffer written force all the timers to 0.
*
* @note Lots of sync requests will skew this timer. It should be based
* on TOD to be accurate. Does it matter ?
*/
if (sync_active || rtems_bdbuf_has_buffer_waiters ())
if (sync_all || (sync_active && (*dev == bd->dev))
|| rtems_bdbuf_has_buffer_waiters ())
bd->hold_timer = 0;
if (bd->hold_timer)
@@ -2515,11 +2540,11 @@ rtems_bdbuf_swapout_processing (unsigned long timer_delta,
/*
* If a sync is active do not use a worker because the current code does not
* cleaning up after. We need to know the buffers have been written when
* syncing to the release sync lock and currently worker threads do not
* return to here. We do not know the worker is the last in a sequence of
* sync writes until after we have it running so we do not know to tell it to
* release the lock. The simplest solution is to get the main swap out task
* perform all sync operations.
* syncing to release sync lock and currently worker threads do not return to
* here. We do not know the worker is the last in a sequence of sync writes
* until after we have it running so we do not know to tell it to release the
* lock. The simplest solution is to get the main swap out task perform all
* sync operations.
*/
if (bdbuf_cache.sync_active)
worker = NULL;
@@ -2533,7 +2558,8 @@ rtems_bdbuf_swapout_processing (unsigned long timer_delta,
rtems_chain_initialize_empty (&transfer->bds);
transfer->dev = BDBUF_INVALID_DEV;
transfer->syncing = bdbuf_cache.sync_active;
/*
* When the sync is for a device limit the sync to that device. If the sync
* is for a buffer handle process the devices in the order on the sync
@@ -2541,7 +2567,7 @@ rtems_bdbuf_swapout_processing (unsigned long timer_delta,
*/
if (bdbuf_cache.sync_active)
transfer->dev = bdbuf_cache.sync_device;
/*
* If we have any buffers in the sync queue move them to the modified
* list. The first sync buffer will select the device we use.
@@ -2753,6 +2779,7 @@ rtems_bdbuf_swapout_task (rtems_task_argument arg)
transfer.write_req = rtems_bdbuf_swapout_writereq_alloc ();
rtems_chain_initialize_empty (&transfer.bds);
transfer.dev = BDBUF_INVALID_DEV;
transfer.syncing = false;
/*
* Localise the period.

View File

@@ -259,7 +259,7 @@ rtems_status_code rtems_disk_create_phys(
dd->ioctl = handler;
dd->driver_data = driver_data;
if ((*handler)(dd, RTEMS_BLKDEV_CAPABILITIES, &dd->capabilities) < 0) {
if ((*handler)(dd, RTEMS_BLKIO_CAPABILITIES, &dd->capabilities) < 0) {
dd->capabilities = 0;
}