Prev: Add a new structure for skb buffer from external.
Next: [PATCH 0001] Add Lenovo Thinkpad T400 to the whitelist
From: Boaz Harrosh on 7 Jun 2010 04:50
On 06/07/2010 06:50 AM, Nicholas A. Bellinger wrote:
> From: Nicholas Bellinger <nab(a)linux-iscsi.org>
>
> This patch adds initial support for the block layer implementation of the sg v4 interface
> (BSG) -> STGT with a new struct backingstore_template bsg_bst sharing code with the
> original sg_bst. It adds for new function bs_bsg_cmd_submit() for incoming write I/O
> and bs_bsg_cmd_complete() for polling read I/O using include/linux/bsg.h:struct sg_io_v4.
>
> This patch also splits up bs_sg_open() to support both BSG and SG looking for the
> path "/dev/bsg" in bs_sg_open(), and getting max_queue using SG_GET_COMMAND_Q and
> calling SG_SET_RESERVED_SIZE (following the original bs_sg code in init_sg_device)
>
> Also chk_bsg_device() currently using a hardcoded 254 major as def for BSG does not
> appear in include/linux/major.h just yet..
>
> So far this has been tested with STGT/iSCSI <-> BSG <-> TCM_Loop SPC-4 iSCSI Target
> Port emulation and is able to mkfs, fsck and mount a filesystem from a TCM/IBLOCK
> Linux LVM kernel level backstore
>
> Signed-off-by: Nicholas A. Bellinger <nab(a)linux-iscsi.org>
> ---
> usr/bs_sg.c | 178 +++++++++++++++++++++++++++++++++++++++++++++++++++---
> usr/scsi_cmnd.h | 7 ++
> 2 files changed, 175 insertions(+), 10 deletions(-)
>
> diff --git a/usr/bs_sg.c b/usr/bs_sg.c
> index 23676ef..d071714 100644
> --- a/usr/bs_sg.c
> +++ b/usr/bs_sg.c
> @@ -3,6 +3,9 @@
> *
> * Copyright (C) 2008 Alexander Nezhinsky <nezhinsky(a)gmail.com>
> *
> + * Added linux/block/bsg.c support using struct sg_io_v4.
> + * Copyright (C) 2010 Nicholas A. Bellinger <nab(a)linux-iscsi.org>
> + *
> * This program is free software; you can redistribute it and/or
> * modify it under the terms of the GNU General Public License as
> * published by the Free Software Foundation, version 2 of the
> @@ -33,6 +36,7 @@
> #include <sys/stat.h>
> #include <sys/epoll.h>
> #include <scsi/sg.h>
> +#include <linux/bsg.h>
>
> #include "list.h"
> #include "util.h"
> @@ -45,6 +49,14 @@
> #define BS_SG_TIMEOUT 2000
> #define BS_SG_SHIFT 9
>
> +//#define BSG_DEBUG_IO
> +
> +#ifdef BSG_DEBUG_IO
> +#define BSG_IO(x...) eprintf(x)
> +#else
> +#define BSG_IO(x...)
> +#endif
> +
> static int graceful_read(int fd, void *p_read, int to_read)
> {
> int err;
> @@ -111,7 +123,7 @@ static int bs_sg_rw(int host_no, struct scsi_cmd *cmd)
> return SAM_STAT_CHECK_CONDITION;
> }
>
> -static void set_cmd_failed(struct scsi_cmd *cmd)
> +static int set_cmd_failed(struct scsi_cmd *cmd)
> {
> int result = SAM_STAT_CHECK_CONDITION;
> uint16_t asc = ASC_READ_ERROR;
> @@ -119,6 +131,54 @@ static void set_cmd_failed(struct scsi_cmd *cmd)
>
> scsi_set_result(cmd, result);
> sense_data_build(cmd, key, asc);
> +
> + return result;
> +}
> +
> +static int bs_bsg_cmd_submit(struct scsi_cmd *cmd)
> +{
> + struct scsi_lu *dev = cmd->dev;
> + int fd = dev->fd;
> + struct sg_io_v4 *io_hdr = &cmd->cmd_bsg_hdr;
> + int err = 0;
> + /*
> + * Following linux/include/linux/bsg.h
> + */
> + /* [i] 'Q' to differentiate from v3 */
> + io_hdr->guard = 'Q';
> + io_hdr->protocol = BSG_PROTOCOL_SCSI;
> + io_hdr->subprotocol = BSG_SUB_PROTOCOL_SCSI_CMD;
> + io_hdr->request_len = cmd->scb_len;
> + io_hdr->request = (unsigned long )cmd->scb;
> +
> + if (scsi_get_data_dir(cmd) == DATA_WRITE) {

Why the if? both sides are fully bidi just remove the ifs
and: OUT <= OUT, IN <= IN

Boaz

> + io_hdr->dout_xfer_len = scsi_get_out_length(cmd);
> + io_hdr->dout_xferp = (unsigned long)scsi_get_out_buffer(cmd);
> + } else {
> + io_hdr->din_xfer_len = scsi_get_in_length(cmd);
> + io_hdr->din_xferp = (unsigned long)scsi_get_in_buffer(cmd);
> + }
> + io_hdr->max_response_len = sizeof(cmd->sense_buffer);
> + /* SCSI: (auto)sense data */
> + io_hdr->response = (unsigned long)cmd->sense_buffer;
> + /* Using the same 2000 millisecond timeout.. */
> + io_hdr->timeout = BS_SG_TIMEOUT;
> + /* [i->o] unused internally */
> + io_hdr->usr_ptr = (unsigned long)cmd;
> + BSG_IO("[%d] Set io_hdr->usr_ptr from cmd: %p\n", getpid(), cmd);
> + /* Bsg does Q_AT_HEAD by default */
> + io_hdr->flags |= BSG_FLAG_Q_AT_TAIL;
> +
> + BSG_IO("[%d] Calling graceful_write for CDB: 0x%02x\n", getpid(), cmd->scb[0]);
> + err = graceful_write(fd, &cmd->cmd_bsg_hdr, sizeof(struct sg_io_v4));
> + if (!err)
> + set_cmd_async(cmd);
> + else {
> + eprintf("failed to start cmd 0x%p\n", cmd);
> + return set_cmd_failed(cmd);
> + }
> +
> + return 0;
> }
>
> static int bs_sg_cmd_submit(struct scsi_cmd *cmd)
> @@ -136,11 +196,13 @@ static int bs_sg_cmd_submit(struct scsi_cmd *cmd)
> if (scsi_get_data_dir(cmd) == DATA_WRITE) {
> io_hdr.dxfer_direction = SG_DXFER_TO_DEV;
> io_hdr.dxfer_len = scsi_get_out_length(cmd);
> - io_hdr.dxferp = (void *)scsi_get_out_buffer(cmd);
> + cmd->cmd_sg_iovec.iov_base = scsi_get_out_buffer(cmd);
> + io_hdr.iovec_count = 1;
> } else {
> io_hdr.dxfer_direction = SG_DXFER_FROM_DEV;
> io_hdr.dxfer_len = scsi_get_in_length(cmd);
> - io_hdr.dxferp = (void *)scsi_get_in_buffer(cmd);
> + cmd->cmd_sg_iovec.iov_base = scsi_get_in_buffer(cmd);
> + io_hdr.iovec_count = 1;
> }
> io_hdr.mx_sb_len = sizeof(cmd->sense_buffer);
> io_hdr.sbp = cmd->sense_buffer;
> @@ -154,11 +216,47 @@ static int bs_sg_cmd_submit(struct scsi_cmd *cmd)
> set_cmd_async(cmd);
> else {
> eprintf("failed to start cmd 0x%p\n", cmd);
> - set_cmd_failed(cmd);
> + return set_cmd_failed(cmd);
> }
> return 0;
> }
>
> +static void bs_bsg_cmd_complete(int fd, int events, void *data)
> +{
> + struct sg_io_v4 io_hdr;
> + struct scsi_cmd *cmd;
> + int err;
> +
> + BSG_IO("[%d] bs_bsg_cmd_complete() called!\n", getpid());
> + memset(&io_hdr, 0, sizeof(io_hdr));
> + /* [i] 'Q' to differentiate from v3 */
> + io_hdr.guard = 'Q';
> +
> + err = graceful_read(fd, &io_hdr, sizeof(io_hdr));
> + if (err)
> + return;
> +
> + cmd = (struct scsi_cmd *)(unsigned long)io_hdr.usr_ptr;
> + BSG_IO("BSG Using cmd: %p for io_hdr.usr_ptr\n", cmd);
> + /*
> + * Check SCSI: command completion status
> + * */
> + if (!io_hdr.device_status) {
> + scsi_set_out_resid(cmd, io_hdr.dout_resid);
> + scsi_set_in_resid(cmd, io_hdr.din_resid);
> + } else {
> + /*
> + * NAB: Used by linux/block/bsg.c:bsg_ioctl(), is this
> + * right..?
> + */
> + cmd->sense_len = SCSI_SENSE_BUFFERSIZE;
> + scsi_set_out_resid_by_actual(cmd, 0);
> + scsi_set_in_resid_by_actual(cmd, 0);
> + }
> + BSG_IO("[%d] BSG() Calling cmd->scsi_cmd_done()\n", getpid());
> + cmd->scsi_cmd_done(cmd, io_hdr.device_status);
> +}
> +
> static void bs_sg_cmd_complete(int fd, int events, void *data)
> {
> struct sg_io_hdr io_hdr;
> @@ -186,6 +284,22 @@ static void bs_sg_cmd_complete(int fd, int events, void *data)
> cmd->scsi_cmd_done(cmd, io_hdr.status);
> }
>
> +static int chk_bsg_device(char *path)
> +{
> + struct stat st;
> +
> + if (stat(path, &st) < 0) {
> + eprintf("stat() failed errno: %d\n", errno);
> + return -1;
> + }
> +
> + /* This is not yet defined in include/linux/major.h.. */
> + if (S_ISCHR(st.st_mode) && major(st.st_rdev) == 254)
> + return 0;
> +
> + return -1;
> +}
> +
> static int chk_sg_device(char *path)
> {
> struct stat st;
> @@ -197,8 +311,29 @@ static int chk_sg_device(char *path)
>
> if (S_ISCHR(st.st_mode) && major(st.st_rdev) == SCSI_GENERIC_MAJOR)
> return 0;
> - else
> +
> + return -1;
> +}
> +
> +static int init_bsg_device(int fd)
> +{
> + int t, err;
> +
> + err = ioctl(fd, SG_GET_COMMAND_Q, &t);
> + if (err < 0) {
> + eprintf("SG_GET_COMMAND_Q for bsd failed: %d\n", err);
> + return -1;
> + }
> + eprintf("bsg: Using max_queue: %d\n", t);
> +
> + t = BS_SG_RESVD_SZ;
> + err = ioctl(fd, SG_SET_RESERVED_SIZE, &t);
> + if (err < 0) {
> + eprintf("SG_SET_RESERVED_SIZE errno: %d\n", errno);
> return -1;
> + }
> +
> + return 0;
> }
>
> static int init_sg_device(int fd)
> @@ -242,9 +377,14 @@ static int bs_sg_init(struct scsi_lu *lu)
>
> static int bs_sg_open(struct scsi_lu *lu, char *path, int *fd, uint64_t *size)
> {
> - int sg_fd, err;
> -
> - err = chk_sg_device(path);
> + void (*cmd_complete)(int, int, void *) = NULL;
> + int sg_fd, err, bsg = 0;
> +
> + if ((strstr(path, "/dev/bsg/") != NULL)) {
> + bsg = 1;
> + err = chk_bsg_device(path);
> + } else
> + err = chk_sg_device(path);
> if (err) {
> eprintf("Not recognized %s as an SG device\n", path);
> return -EINVAL;
> @@ -256,13 +396,19 @@ static int bs_sg_open(struct scsi_lu *lu, char *path, int *fd, uint64_t *size)
> return sg_fd;
> }
>
> - err = init_sg_device(sg_fd);
> + if (bsg) {
> + cmd_complete = &bs_bsg_cmd_complete;
> + err = init_bsg_device(sg_fd);
> + } else {
> + cmd_complete = bs_sg_cmd_complete;
> + err = init_sg_device(sg_fd);
> + }
> if (err) {
> eprintf("Failed to initialize sg device %s\n", path);
> return err;
> }
>
> - err = tgt_event_add(sg_fd, EPOLLIN, bs_sg_cmd_complete, NULL);
> + err = tgt_event_add(sg_fd, EPOLLIN, cmd_complete, NULL);
> if (err) {
> eprintf("Failed to add sg device event %s\n", path);
> return err;
> @@ -302,6 +448,17 @@ static struct backingstore_template sg_bst = {
> .bs_cmd_done = bs_sg_cmd_done,
> };
>
> +static struct backingstore_template bsg_bst = {
> + .bs_name = "bsg",
> + .bs_datasize = 0,
> + .bs_passthrough = 1,
> + .bs_init = bs_sg_init,
> + .bs_open = bs_sg_open,
> + .bs_close = bs_sg_close,
> + .bs_cmd_submit = bs_bsg_cmd_submit,
> + .bs_cmd_done = bs_sg_cmd_done,
> +};
> +
> static struct device_type_template sg_template = {
> .type = 0,
> .lu_init = bs_sg_lu_init,
> @@ -315,5 +472,6 @@ static struct device_type_template sg_template = {
> __attribute__((constructor)) static void bs_sg_constructor(void)
> {
> register_backingstore_template(&sg_bst);
> + register_backingstore_template(&bsg_bst);
> device_type_register(&sg_template);
> }
> diff --git a/usr/scsi_cmnd.h b/usr/scsi_cmnd.h
> index 011f3e6..67a46c3 100644
> --- a/usr/scsi_cmnd.h
> +++ b/usr/scsi_cmnd.h
> @@ -17,6 +17,9 @@ struct scsi_data_buffer {
> uint64_t buffer;
> };
>
> +#include <scsi/sg.h>
> +#include <linux/bsg.h>
> +
> struct scsi_cmd {
> struct target *c_target;
> /* linked it_nexus->cmd_hash_list */
> @@ -31,6 +34,10 @@ struct scsi_cmd {
> enum data_direction data_dir;
> struct scsi_data_buffer in_sdb;
> struct scsi_data_buffer out_sdb;
> + /* Used for bs_sg.c:bs_bs_cmd_submit() */
> + struct sg_iovec cmd_sg_iovec;
> + /* used for bs_sg.c:bs_bsg_cmd_submit() */
> + struct sg_io_v4 cmd_bsg_hdr;
>
> uint64_t cmd_itn_id;
> uint64_t offset;

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Nicholas A. Bellinger on 7 Jun 2010 05:10
On Mon, 2010-06-07 at 11:40 +0300, Boaz Harrosh wrote:
> On 06/07/2010 06:50 AM, Nicholas A. Bellinger wrote:
> > From: Nicholas Bellinger <nab(a)linux-iscsi.org>
> >
> > This patch adds initial support for the block layer implementation of the sg v4 interface
> > (BSG) -> STGT with a new struct backingstore_template bsg_bst sharing code with the
> > original sg_bst. It adds for new function bs_bsg_cmd_submit() for incoming write I/O
> > and bs_bsg_cmd_complete() for polling read I/O using include/linux/bsg.h:struct sg_io_v4.
> >
> > This patch also splits up bs_sg_open() to support both BSG and SG looking for the
> > path "/dev/bsg" in bs_sg_open(), and getting max_queue using SG_GET_COMMAND_Q and
> > calling SG_SET_RESERVED_SIZE (following the original bs_sg code in init_sg_device)
> >
> > Also chk_bsg_device() currently using a hardcoded 254 major as def for BSG does not
> > appear in include/linux/major.h just yet..
> >
> > So far this has been tested with STGT/iSCSI <-> BSG <-> TCM_Loop SPC-4 iSCSI Target
> > Port emulation and is able to mkfs, fsck and mount a filesystem from a TCM/IBLOCK
> > Linux LVM kernel level backstore
> >
> > Signed-off-by: Nicholas A. Bellinger <nab(a)linux-iscsi.org>
> > ---
> > usr/bs_sg.c | 178 +++++++++++++++++++++++++++++++++++++++++++++++++++---
> > usr/scsi_cmnd.h | 7 ++
> > 2 files changed, 175 insertions(+), 10 deletions(-)
> >
> > diff --git a/usr/bs_sg.c b/usr/bs_sg.c
> > index 23676ef..d071714 100644
> > --- a/usr/bs_sg.c
> > +++ b/usr/bs_sg.c

<SNIP>

> > +
> > +static int bs_bsg_cmd_submit(struct scsi_cmd *cmd)
> > +{
> > + struct scsi_lu *dev = cmd->dev;
> > + int fd = dev->fd;
> > + struct sg_io_v4 *io_hdr = &cmd->cmd_bsg_hdr;
> > + int err = 0;
> > + /*
> > + * Following linux/include/linux/bsg.h
> > + */
> > + /* [i] 'Q' to differentiate from v3 */
> > + io_hdr->guard = 'Q';
> > + io_hdr->protocol = BSG_PROTOCOL_SCSI;
> > + io_hdr->subprotocol = BSG_SUB_PROTOCOL_SCSI_CMD;
> > + io_hdr->request_len = cmd->scb_len;
> > + io_hdr->request = (unsigned long )cmd->scb;
> > +
> > + if (scsi_get_data_dir(cmd) == DATA_WRITE) {
>
> Why the if? both sides are fully bidi just remove the ifs
> and: OUT <= OUT, IN <= IN
>

Whoops, thanks for pointing this one out (again). Tested with
unidirection commands and included into my local tgt.git/pass that I
will send along to Tomo-san..

Btw, I still need to sort out BIDI support for TCM_Loop at some point in
the not too distant future. What would be a good way to generate BIDI
traffic into STGT or a local SCSI LLD..?

Best,

--nab

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Boaz Harrosh on 7 Jun 2010 05:30
On 06/07/2010 12:02 PM, Nicholas A. Bellinger wrote:
> On Mon, 2010-06-07 at 11:40 +0300, Boaz Harrosh wrote:
>> On 06/07/2010 06:50 AM, Nicholas A. Bellinger wrote:
>>> From: Nicholas Bellinger <nab(a)linux-iscsi.org>
>>>
>>> This patch adds initial support for the block layer implementation of the sg v4 interface
>>> (BSG) -> STGT with a new struct backingstore_template bsg_bst sharing code with the
>>> original sg_bst. It adds for new function bs_bsg_cmd_submit() for incoming write I/O
>>> and bs_bsg_cmd_complete() for polling read I/O using include/linux/bsg.h:struct sg_io_v4.
>>>
>>> This patch also splits up bs_sg_open() to support both BSG and SG looking for the
>>> path "/dev/bsg" in bs_sg_open(), and getting max_queue using SG_GET_COMMAND_Q and
>>> calling SG_SET_RESERVED_SIZE (following the original bs_sg code in init_sg_device)
>>>
>>> Also chk_bsg_device() currently using a hardcoded 254 major as def for BSG does not
>>> appear in include/linux/major.h just yet..
>>>
>>> So far this has been tested with STGT/iSCSI <-> BSG <-> TCM_Loop SPC-4 iSCSI Target
>>> Port emulation and is able to mkfs, fsck and mount a filesystem from a TCM/IBLOCK
>>> Linux LVM kernel level backstore
>>>
>>> Signed-off-by: Nicholas A. Bellinger <nab(a)linux-iscsi.org>
>>> ---
>>> usr/bs_sg.c | 178 +++++++++++++++++++++++++++++++++++++++++++++++++++---
>>> usr/scsi_cmnd.h | 7 ++
>>> 2 files changed, 175 insertions(+), 10 deletions(-)
>>>
>>> diff --git a/usr/bs_sg.c b/usr/bs_sg.c
>>> index 23676ef..d071714 100644
>>> --- a/usr/bs_sg.c
>>> +++ b/usr/bs_sg.c
>
> <SNIP>
>
>>> +
>>> +static int bs_bsg_cmd_submit(struct scsi_cmd *cmd)
>>> +{
>>> + struct scsi_lu *dev = cmd->dev;
>>> + int fd = dev->fd;
>>> + struct sg_io_v4 *io_hdr = &cmd->cmd_bsg_hdr;
>>> + int err = 0;
>>> + /*
>>> + * Following linux/include/linux/bsg.h
>>> + */
>>> + /* [i] 'Q' to differentiate from v3 */
>>> + io_hdr->guard = 'Q';
>>> + io_hdr->protocol = BSG_PROTOCOL_SCSI;
>>> + io_hdr->subprotocol = BSG_SUB_PROTOCOL_SCSI_CMD;
>>> + io_hdr->request_len = cmd->scb_len;
>>> + io_hdr->request = (unsigned long )cmd->scb;
>>> +
>>> + if (scsi_get_data_dir(cmd) == DATA_WRITE) {
>>
>> Why the if? both sides are fully bidi just remove the ifs
>> and: OUT <= OUT, IN <= IN
>>
>
> Whoops, thanks for pointing this one out (again). Tested with
> unidirection commands and included into my local tgt.git/pass that I
> will send along to Tomo-san..
>
> Btw, I still need to sort out BIDI support for TCM_Loop at some point in
> the not too distant future. What would be a good way to generate BIDI
> traffic into STGT or a local SCSI LLD..?
>
> Best,
>
> --nab
>

scsi_debug has some emulation for XOR commands, does that help?

Boaz
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Nicholas A. Bellinger on 7 Jun 2010 05:40
On Mon, 2010-06-07 at 12:27 +0300, Boaz Harrosh wrote:
> On 06/07/2010 12:02 PM, Nicholas A. Bellinger wrote:
> > On Mon, 2010-06-07 at 11:40 +0300, Boaz Harrosh wrote:
> >> On 06/07/2010 06:50 AM, Nicholas A. Bellinger wrote:
> >>> From: Nicholas Bellinger <nab(a)linux-iscsi.org>
> >>>
> >>> This patch adds initial support for the block layer implementation of the sg v4 interface
> >>> (BSG) -> STGT with a new struct backingstore_template bsg_bst sharing code with the
> >>> original sg_bst. It adds for new function bs_bsg_cmd_submit() for incoming write I/O
> >>> and bs_bsg_cmd_complete() for polling read I/O using include/linux/bsg.h:struct sg_io_v4.
> >>>
> >>> This patch also splits up bs_sg_open() to support both BSG and SG looking for the
> >>> path "/dev/bsg" in bs_sg_open(), and getting max_queue using SG_GET_COMMAND_Q and
> >>> calling SG_SET_RESERVED_SIZE (following the original bs_sg code in init_sg_device)
> >>>
> >>> Also chk_bsg_device() currently using a hardcoded 254 major as def for BSG does not
> >>> appear in include/linux/major.h just yet..
> >>>
> >>> So far this has been tested with STGT/iSCSI <-> BSG <-> TCM_Loop SPC-4 iSCSI Target
> >>> Port emulation and is able to mkfs, fsck and mount a filesystem from a TCM/IBLOCK
> >>> Linux LVM kernel level backstore
> >>>
> >>> Signed-off-by: Nicholas A. Bellinger <nab(a)linux-iscsi.org>
> >>> ---
> >>> usr/bs_sg.c | 178 +++++++++++++++++++++++++++++++++++++++++++++++++++---
> >>> usr/scsi_cmnd.h | 7 ++
> >>> 2 files changed, 175 insertions(+), 10 deletions(-)
> >>>
> >>> diff --git a/usr/bs_sg.c b/usr/bs_sg.c
> >>> index 23676ef..d071714 100644
> >>> --- a/usr/bs_sg.c
> >>> +++ b/usr/bs_sg.c
> >
> > <SNIP>
> >
> >>> +
> >>> +static int bs_bsg_cmd_submit(struct scsi_cmd *cmd)
> >>> +{
> >>> + struct scsi_lu *dev = cmd->dev;
> >>> + int fd = dev->fd;
> >>> + struct sg_io_v4 *io_hdr = &cmd->cmd_bsg_hdr;
> >>> + int err = 0;
> >>> + /*
> >>> + * Following linux/include/linux/bsg.h
> >>> + */
> >>> + /* [i] 'Q' to differentiate from v3 */
> >>> + io_hdr->guard = 'Q';
> >>> + io_hdr->protocol = BSG_PROTOCOL_SCSI;
> >>> + io_hdr->subprotocol = BSG_SUB_PROTOCOL_SCSI_CMD;
> >>> + io_hdr->request_len = cmd->scb_len;
> >>> + io_hdr->request = (unsigned long )cmd->scb;
> >>> +
> >>> + if (scsi_get_data_dir(cmd) == DATA_WRITE) {
> >>
> >> Why the if? both sides are fully bidi just remove the ifs
> >> and: OUT <= OUT, IN <= IN
> >>
> >
> > Whoops, thanks for pointing this one out (again). Tested with
> > unidirection commands and included into my local tgt.git/pass that I
> > will send along to Tomo-san..
> >
> > Btw, I still need to sort out BIDI support for TCM_Loop at some point in
> > the not too distant future. What would be a good way to generate BIDI
> > traffic into STGT or a local SCSI LLD..?
>
> scsi_debug has some emulation for XOR commands, does that help?
>

Great, thanks for the pointer on the LLD side. Is there any userspace
code around that will generate XDWRITEREAD_* CDBs for bulk I/O or do I
need to code this part myself..?

Best,

--nab

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Boaz Harrosh on 7 Jun 2010 06:40
On 06/07/2010 12:37 PM, Nicholas A. Bellinger wrote:
> On Mon, 2010-06-07 at 12:27 +0300, Boaz Harrosh wrote:
>> On 06/07/2010 12:02 PM, Nicholas A. Bellinger wrote:
>>> On Mon, 2010-06-07 at 11:40 +0300, Boaz Harrosh wrote:
>>>> On 06/07/2010 06:50 AM, Nicholas A. Bellinger wrote:
>>>>> From: Nicholas Bellinger <nab(a)linux-iscsi.org>
>>>>>
>>>>> This patch adds initial support for the block layer implementation of the sg v4 interface
>>>>> (BSG) -> STGT with a new struct backingstore_template bsg_bst sharing code with the
>>>>> original sg_bst. It adds for new function bs_bsg_cmd_submit() for incoming write I/O
>>>>> and bs_bsg_cmd_complete() for polling read I/O using include/linux/bsg.h:struct sg_io_v4.
>>>>>
>>>>> This patch also splits up bs_sg_open() to support both BSG and SG looking for the
>>>>> path "/dev/bsg" in bs_sg_open(), and getting max_queue using SG_GET_COMMAND_Q and
>>>>> calling SG_SET_RESERVED_SIZE (following the original bs_sg code in init_sg_device)
>>>>>
>>>>> Also chk_bsg_device() currently using a hardcoded 254 major as def for BSG does not
>>>>> appear in include/linux/major.h just yet..
>>>>>
>>>>> So far this has been tested with STGT/iSCSI <-> BSG <-> TCM_Loop SPC-4 iSCSI Target
>>>>> Port emulation and is able to mkfs, fsck and mount a filesystem from a TCM/IBLOCK
>>>>> Linux LVM kernel level backstore
>>>>>
>>>>> Signed-off-by: Nicholas A. Bellinger <nab(a)linux-iscsi.org>
>>>>> ---
>>>>> usr/bs_sg.c | 178 +++++++++++++++++++++++++++++++++++++++++++++++++++---
>>>>> usr/scsi_cmnd.h | 7 ++
>>>>> 2 files changed, 175 insertions(+), 10 deletions(-)
>>>>>
>>>>> diff --git a/usr/bs_sg.c b/usr/bs_sg.c
>>>>> index 23676ef..d071714 100644
>>>>> --- a/usr/bs_sg.c
>>>>> +++ b/usr/bs_sg.c
>>>
>>> <SNIP>
>>>
>>>>> +
>>>>> +static int bs_bsg_cmd_submit(struct scsi_cmd *cmd)
>>>>> +{
>>>>> + struct scsi_lu *dev = cmd->dev;
>>>>> + int fd = dev->fd;
>>>>> + struct sg_io_v4 *io_hdr = &cmd->cmd_bsg_hdr;
>>>>> + int err = 0;
>>>>> + /*
>>>>> + * Following linux/include/linux/bsg.h
>>>>> + */
>>>>> + /* [i] 'Q' to differentiate from v3 */
>>>>> + io_hdr->guard = 'Q';
>>>>> + io_hdr->protocol = BSG_PROTOCOL_SCSI;
>>>>> + io_hdr->subprotocol = BSG_SUB_PROTOCOL_SCSI_CMD;
>>>>> + io_hdr->request_len = cmd->scb_len;
>>>>> + io_hdr->request = (unsigned long )cmd->scb;
>>>>> +
>>>>> + if (scsi_get_data_dir(cmd) == DATA_WRITE) {
>>>>
>>>> Why the if? both sides are fully bidi just remove the ifs
>>>> and: OUT <= OUT, IN <= IN
>>>>
>>>
>>> Whoops, thanks for pointing this one out (again). Tested with
>>> unidirection commands and included into my local tgt.git/pass that I
>>> will send along to Tomo-san..
>>>
>>> Btw, I still need to sort out BIDI support for TCM_Loop at some point in
>>> the not too distant future. What would be a good way to generate BIDI
>>> traffic into STGT or a local SCSI LLD..?
>>
>> scsi_debug has some emulation for XOR commands, does that help?
>>
>
> Great, thanks for the pointer on the LLD side. Is there any userspace
> code around that will generate XDWRITEREAD_* CDBs for bulk I/O or do I
> need to code this part myself..?
>
> Best,
>
> --nab
>

You could use iscsi as LLD and generate OSD heavy traffic. (With a real osd_tgtd
on the end side). But that will need more work, right? what does the TCM_Loop
do exactly. Please forgive my laziness, I never took the time to find out.

Boaz
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
 |  Next  |  Last
Pages: 1 2
Prev: Add a new structure for skb buffer from external.
Next: [PATCH 0001] Add Lenovo Thinkpad T400 to the whitelist