lkml.org 
[lkml]   [2009]   [Feb]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Subject[PATCH 5/7] [Target_Core_Mod/FILEIO]: Add 4k sector/block size emulation support
From
Date
>From 2baeebdcbf070af9052545870c589ca2c8be2378 Mon Sep 17 00:00:00 2001
From: Nicholas Bellinger <nab@linux-iscsi.org>
Date: Thu, 19 Feb 2009 19:39:59 -0800
Subject: [PATCH 5/7] [Target_Core_Mod/FILEIO]: Add 4k sector/block size emulation support

This patch adds support for 4k (and all non 512 byte) sector sizes for the
Target_Core_Mod/ConfigFS interaction with Linux/VFS struct file objects.

This patch updates fd_emulate_read_cap(), fd_emulate_read_cap16() for control path
CDB emulation, and fd_seek(), fd_do_readv(), fd_do_writev(), fd_do_sendfile()
fd_do_aio_write(). The last two functions have been updated, but are still currently
stubbed out and will be enabled in a future patch to Target_Core_Mod/FILEIO.

Signed-off-by: Nicholas A. Bellinger <nab@linux-iscsi.org>
---
drivers/lio-core/target_core_file.c | 58 ++++++++++++++++++----------------
1 files changed, 31 insertions(+), 27 deletions(-)

diff --git a/drivers/lio-core/target_core_file.c b/drivers/lio-core/target_core_file.c
index 12faa15..0dd7cc0 100644
--- a/drivers/lio-core/target_core_file.c
+++ b/drivers/lio-core/target_core_file.c
@@ -435,20 +435,22 @@ extern int fd_emulate_inquiry (se_task_t *task)
static int fd_emulate_read_cap (se_task_t *task)
{
fd_dev_t *fd_dev = (fd_dev_t *) task->se_dev->dev_ptr;
- u32 blocks = (fd_dev->fd_dev_size / FD_BLOCKSIZE);
+ u32 blocks = ((fd_dev->fd_dev_size / DEV_ATTRIB(task->se_dev)->block_size) - 1);

- if ((fd_dev->fd_dev_size / FD_BLOCKSIZE) >= 0x00000000ffffffff)
+ if (((fd_dev->fd_dev_size / DEV_ATTRIB(task->se_dev)->block_size) - 1) >=
+ 0x00000000ffffffff)
blocks = 0xffffffff;

- return(transport_generic_emulate_readcapacity(TASK_CMD(task), blocks, FD_BLOCKSIZE));
+ return(transport_generic_emulate_readcapacity(TASK_CMD(task), blocks));
}

static int fd_emulate_read_cap16 (se_task_t *task)
{
fd_dev_t *fd_dev = (fd_dev_t *) task->se_dev->dev_ptr;
- unsigned long long blocks_long = (fd_dev->fd_dev_size / FD_BLOCKSIZE);
+ unsigned long long blocks_long = ((fd_dev->fd_dev_size /
+ DEV_ATTRIB(task->se_dev)->block_size) - 1);

- return(transport_generic_emulate_readcapacity_16(TASK_CMD(task), blocks_long, FD_BLOCKSIZE));
+ return(transport_generic_emulate_readcapacity_16(TASK_CMD(task), blocks_long));
}

/* fd_emulate_scsi_cdb():
@@ -526,7 +528,7 @@ static inline int fd_iovec_alloc (fd_request_t *req)
return(0);
}

-static inline int fd_seek (struct file *fd, unsigned long long lba)
+static inline int fd_seek (struct file *fd, unsigned long long lba, u32 block_size)
{
mm_segment_t old_fs;
unsigned long long offset;
@@ -534,18 +536,18 @@ static inline int fd_seek (struct file *fd, unsigned long long lba)
old_fs = get_fs();
set_fs(get_ds());
if (fd->f_op->llseek)
- offset = fd->f_op->llseek(fd, lba * FD_BLOCKSIZE, 0);
+ offset = fd->f_op->llseek(fd, lba * block_size, 0);
else
- offset = default_llseek(fd, lba * FD_BLOCKSIZE, 0);
+ offset = default_llseek(fd, lba * block_size, 0);
set_fs(old_fs);
#if 0
- PYXPRINT("lba: %llu : FD_BLOCKSIZE: %d\n", lba, FD_BLOCKSIZE);
+ PYXPRINT("lba: %llu : block_size: %d\n", lba, block_size);
PYXPRINT("offset from llseek: %llu\n", offset);
- PYXPRINT("(lba * FD_BLOCKSIZE): %llu\n", (lba * FD_BLOCKSIZE));
+ PYXPRINT("(lba * block_size): %llu\n", (lba * block_size));
#endif
- if (offset != (lba * FD_BLOCKSIZE)) {
+ if (offset != (lba * block_size)) {
TRACE_ERROR("offset: %llu not equal to LBA: %llu\n",
- offset, (lba * FD_BLOCKSIZE));
+ offset, (lba * block_size));
return(-1);
}

@@ -563,7 +565,7 @@ static int fd_do_readv (fd_request_t *req, se_task_t *task)

memset(iov, 0, sizeof(struct iovec) + req->fd_sg_count);

- if (fd_seek(fd, req->fd_lba) < 0)
+ if (fd_seek(fd, req->fd_lba, DEV_ATTRIB(task->se_dev)->block_size) < 0)
return(-1);

for (i = 0; i < req->fd_sg_count; i++) {
@@ -607,6 +609,7 @@ static int fd_do_aio_read (fd_request_t *req, se_task_t *task)
{
int ret = 0;
u32 i, length = 0;
+ u32 block_size = DEV_ATTRIB(task->se_dev)->block_size;
unsigned long long offset, lba = req->fd_lba;;
mm_segment_t old_fs;
struct file *fd = req->fd_dev->fd_file;
@@ -620,18 +623,18 @@ static int fd_do_aio_read (fd_request_t *req, se_task_t *task)
old_fs = get_fs();
set_fs(get_ds());
if (fd->f_op->llseek)
- offset = fd->f_op->llseek(fd, lba * FD_BLOCKSIZE, 0);
+ offset = fd->f_op->llseek(fd, lba * block_size, 0);
else
- offset = default_llseek(fd, lba * FD_BLOCKSIZE, 0);
+ offset = default_llseek(fd, lba * block_size, 0);
set_fs(old_fs);

- PYXPRINT("lba: %llu : FD_BLOCKSIZE: %d\n", lba, FD_BLOCKSIZE);
+ PYXPRINT("lba: %llu : block_size: %d\n", lba, block_size);
PYXPRINT("offset from llseek: %llu\n", offset);
- PYXPRINT("(lba * FD_BLOCKSIZE): %llu\n", (lba * FD_BLOCKSIZE));
+ PYXPRINT("(lba * block_size): %llu\n", (lba * block_size));

- if (offset != (lba * FD_BLOCKSIZE)) {
+ if (offset != (lba * block_size)) {
TRACE_ERROR("offset: %llu not equal to LBA: %llu\n",
- offset, (lba * FD_BLOCKSIZE));
+ offset, (lba * block_size));
return(-1);
}

@@ -717,7 +720,7 @@ static int fd_do_sendfile (fd_request_t *req, se_task_t *task)
int ret = 0;
struct file *fd = req->fd_dev->fd_file;

- if (fd_seek(fd, req->fd_lba) < 0)
+ if (fd_seek(fd, req->fd_lba, DEV_ATTRIB(task->se_dev)->block_size) < 0)
return(-1);

TASK_CMD(task)->transport_free_DMA = &fd_sendfile_free_DMA;
@@ -745,7 +748,7 @@ static int fd_do_writev (fd_request_t *req, se_task_t *task)

memset(iov, 0, sizeof(struct iovec) + req->fd_sg_count);

- if (fd_seek(fd, req->fd_lba) < 0)
+ if (fd_seek(fd, req->fd_lba, DEV_ATTRIB(task->se_dev)->block_size) < 0)
return(-1);

for (i = 0; i < req->fd_sg_count; i++) {
@@ -772,6 +775,7 @@ static int fd_do_aio_write (fd_request_t *req, se_task_t *task)
{
int ret = 0;
u32 i, length = 0;
+ u32 block_size = DEV_ATTRIB(task->se_dev)->block_size;
unsigned long long offset, lba = req->fd_lba;
mm_segment_t old_fs;
struct file *fd = req->fd_dev->fd_file;
@@ -785,18 +789,18 @@ static int fd_do_aio_write (fd_request_t *req, se_task_t *task)
old_fs = get_fs();
set_fs(get_ds());
if (fd->f_op->llseek)
- offset = fd->f_op->llseek(fd, lba * FD_BLOCKSIZE, 0);
+ offset = fd->f_op->llseek(fd, lba * block_size, 0);
else
- offset = default_llseek(fd, lba * FD_BLOCKSIZE, 0);
+ offset = default_llseek(fd, lba * block_size, 0);
set_fs(old_fs);

- PYXPRINT("lba: %llu : FD_BLOCKSIZE: %d\n", lba, FD_BLOCKSIZE);
+ PYXPRINT("lba: %llu : block_size: %d\n", lba, block_size);
PYXPRINT("offset from llseek: %llu\n", offset);
- PYXPRINT("(lba * FD_BLOCKSIZE): %llu\n", (lba * FD_BLOCKSIZE));
+ PYXPRINT("(lba * block_size): %llu\n", (lba * block_size));

- if (offset != (lba * FD_BLOCKSIZE)) {
+ if (offset != (lba * block_size)) {
TRACE_ERROR("offset: %llu not equal to LBA: %llu\n",
- offset, (lba * FD_BLOCKSIZE));
+ offset, (lba * block_size));
return(-1);
}

--
1.5.4.1





\
 
 \ /
  Last update: 2009-02-20 09:07    [W:0.026 / U:0.444 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site