<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux.git/include/scsi/libiscsi.h, branch v5.19</title>
<subtitle>Linux kernel source tree</subtitle>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/'/>
<entry>
<title>scsi: iscsi: Fix harmless double shift bug</title>
<updated>2022-04-26T03:23:01+00:00</updated>
<author>
<name>Dan Carpenter</name>
<email>dan.carpenter@oracle.com</email>
</author>
<published>2022-04-21T15:03:52+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=565138ac5f8a5330669a20e5f94759764e9165ec'/>
<id>565138ac5f8a5330669a20e5f94759764e9165ec</id>
<content type='text'>
These flags are supposed to be bit numbers.  Right now they cause a double
shift bug where we use BIT(BIT(2)) instead of BIT(2).  Fortunately, the bit
numbers are small and it's done consistently so it does not cause an issue
at run time.

Link: https://lore.kernel.org/r/YmFyWHf8nrrx+SHa@kili
Fixes: 5bd856256f8c ("scsi: iscsi: Merge suspend fields")
Reviewed-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Dan Carpenter &lt;dan.carpenter@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
These flags are supposed to be bit numbers.  Right now they cause a double
shift bug where we use BIT(BIT(2)) instead of BIT(2).  Fortunately, the bit
numbers are small and it's done consistently so it does not cause an issue
at run time.

Link: https://lore.kernel.org/r/YmFyWHf8nrrx+SHa@kili
Fixes: 5bd856256f8c ("scsi: iscsi: Merge suspend fields")
Reviewed-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Dan Carpenter &lt;dan.carpenter@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>scsi: iscsi: Fix NOP handling during conn recovery</title>
<updated>2022-04-12T02:09:35+00:00</updated>
<author>
<name>Mike Christie</name>
<email>michael.christie@oracle.com</email>
</author>
<published>2022-04-08T00:13:12+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=44ac97109e42f87b1a34954704b81b6c8eca80c4'/>
<id>44ac97109e42f87b1a34954704b81b6c8eca80c4</id>
<content type='text'>
If a offload driver doesn't use the xmit workqueue, then when we are doing
ep_disconnect libiscsi can still inject PDUs to the driver. This adds a
check for if the connection is bound before trying to inject PDUs.

Link: https://lore.kernel.org/r/20220408001314.5014-9-michael.christie@oracle.com
Tested-by: Manish Rangankar &lt;mrangankar@marvell.com&gt;
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Reviewed-by: Chris Leech &lt;cleech@redhat.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
If a offload driver doesn't use the xmit workqueue, then when we are doing
ep_disconnect libiscsi can still inject PDUs to the driver. This adds a
check for if the connection is bound before trying to inject PDUs.

Link: https://lore.kernel.org/r/20220408001314.5014-9-michael.christie@oracle.com
Tested-by: Manish Rangankar &lt;mrangankar@marvell.com&gt;
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Reviewed-by: Chris Leech &lt;cleech@redhat.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>scsi: iscsi: Merge suspend fields</title>
<updated>2022-04-12T02:09:35+00:00</updated>
<author>
<name>Mike Christie</name>
<email>michael.christie@oracle.com</email>
</author>
<published>2022-04-08T00:13:11+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=5bd856256f8c03e329f8ff36d8c8efcb111fe6df'/>
<id>5bd856256f8c03e329f8ff36d8c8efcb111fe6df</id>
<content type='text'>
Move the tx and rx suspend fields into one flags field.

Link: https://lore.kernel.org/r/20220408001314.5014-8-michael.christie@oracle.com
Tested-by: Manish Rangankar &lt;mrangankar@marvell.com&gt;
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Reviewed-by: Chris Leech &lt;cleech@redhat.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Move the tx and rx suspend fields into one flags field.

Link: https://lore.kernel.org/r/20220408001314.5014-8-michael.christie@oracle.com
Tested-by: Manish Rangankar &lt;mrangankar@marvell.com&gt;
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Reviewed-by: Chris Leech &lt;cleech@redhat.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>scsi: iscsi: Drop temp workq_name</title>
<updated>2022-03-02T04:56:28+00:00</updated>
<author>
<name>Mike Christie</name>
<email>michael.christie@oracle.com</email>
</author>
<published>2022-02-26T23:04:35+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=69af1c9577aae2149f79be6f485609250fdfb0ad'/>
<id>69af1c9577aae2149f79be6f485609250fdfb0ad</id>
<content type='text'>
When the workqueue code was created it didn't allow variable args so we
have been using a temp buffer. Drop that.

Link: https://lore.kernel.org/r/20220226230435.38733-7-michael.christie@oracle.com
Reviewed-by: Chris Leech &lt;cleech@redhat.com&gt;
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
When the workqueue code was created it didn't allow variable args so we
have been using a temp buffer. Drop that.

Link: https://lore.kernel.org/r/20220226230435.38733-7-michael.christie@oracle.com
Reviewed-by: Chris Leech &lt;cleech@redhat.com&gt;
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>scsi: iscsi: Stop using the SCSI pointer</title>
<updated>2022-02-23T02:11:04+00:00</updated>
<author>
<name>Bart Van Assche</name>
<email>bvanassche@acm.org</email>
</author>
<published>2022-02-18T19:50:53+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=db22de3eb0352d2f8e7cda08f3fa65690e3fd64d'/>
<id>db22de3eb0352d2f8e7cda08f3fa65690e3fd64d</id>
<content type='text'>
Instead of storing the iSCSI task pointer and the session age in the SCSI
pointer, use command-private variables. This patch prepares for removal of
the SCSI pointer from struct scsi_cmnd.

The list of iSCSI drivers has been obtained as follows:
$ git grep -lw iscsi_host_alloc
drivers/infiniband/ulp/iser/iscsi_iser.c
drivers/scsi/be2iscsi/be_main.c
drivers/scsi/bnx2i/bnx2i_iscsi.c
drivers/scsi/cxgbi/libcxgbi.c
drivers/scsi/iscsi_tcp.c
drivers/scsi/libiscsi.c
drivers/scsi/qedi/qedi_main.c
drivers/scsi/qla4xxx/ql4_os.c
include/scsi/libiscsi.h

Note: it is not clear to me how the qla4xxx driver can work without this
patch since it uses the scsi_cmnd::SCp.ptr member for two different
purposes:
- The qla4xxx driver uses this member to store a struct srb pointer.
- libiscsi uses this member to store a struct iscsi_task pointer.

Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Reviewed-by: Himanshu Madhani &lt;himanshu.madhani@oracle.com&gt;
Cc: Chris Leech &lt;cleech@redhat.com&gt;
Cc: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Cc: Nilesh Javali &lt;njavali@marvell.com&gt;
Cc: Manish Rangankar &lt;mrangankar@marvell.com&gt;
Cc: Karen Xie &lt;kxie@chelsio.com&gt;
Cc: Ketan Mukadam &lt;ketan.mukadam@broadcom.com&gt;
Signed-off-by: Bart Van Assche &lt;bvanassche@acm.org&gt;

iscsi

Link: https://lore.kernel.org/r/20220218195117.25689-26-bvanassche@acm.org
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Instead of storing the iSCSI task pointer and the session age in the SCSI
pointer, use command-private variables. This patch prepares for removal of
the SCSI pointer from struct scsi_cmnd.

The list of iSCSI drivers has been obtained as follows:
$ git grep -lw iscsi_host_alloc
drivers/infiniband/ulp/iser/iscsi_iser.c
drivers/scsi/be2iscsi/be_main.c
drivers/scsi/bnx2i/bnx2i_iscsi.c
drivers/scsi/cxgbi/libcxgbi.c
drivers/scsi/iscsi_tcp.c
drivers/scsi/libiscsi.c
drivers/scsi/qedi/qedi_main.c
drivers/scsi/qla4xxx/ql4_os.c
include/scsi/libiscsi.h

Note: it is not clear to me how the qla4xxx driver can work without this
patch since it uses the scsi_cmnd::SCp.ptr member for two different
purposes:
- The qla4xxx driver uses this member to store a struct srb pointer.
- libiscsi uses this member to store a struct iscsi_task pointer.

Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Reviewed-by: Hannes Reinecke &lt;hare@suse.de&gt;
Reviewed-by: Himanshu Madhani &lt;himanshu.madhani@oracle.com&gt;
Cc: Chris Leech &lt;cleech@redhat.com&gt;
Cc: Sagi Grimberg &lt;sagi@grimberg.me&gt;
Cc: Nilesh Javali &lt;njavali@marvell.com&gt;
Cc: Manish Rangankar &lt;mrangankar@marvell.com&gt;
Cc: Karen Xie &lt;kxie@chelsio.com&gt;
Cc: Ketan Mukadam &lt;ketan.mukadam@broadcom.com&gt;
Signed-off-by: Bart Van Assche &lt;bvanassche@acm.org&gt;

iscsi

Link: https://lore.kernel.org/r/20220218195117.25689-26-bvanassche@acm.org
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>scsi: iscsi: Hold task ref during TMF timeout handling</title>
<updated>2021-06-02T05:28:22+00:00</updated>
<author>
<name>Mike Christie</name>
<email>michael.christie@oracle.com</email>
</author>
<published>2021-05-25T18:18:10+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=99b0603313eecb9b0b99fd2e60bae1e03e6b038b'/>
<id>99b0603313eecb9b0b99fd2e60bae1e03e6b038b</id>
<content type='text'>
For aborts, qedi needs to cleanup the FW then send the TMF from a worker
thread. While it's doing these the cmd could complete normally and the TMF
could time out. libiscsi would then complete the iscsi_task which will call
into the driver to cleanup the driver level resources while it still might
be accessing them for the cleanup/abort.

This has iscsi_eh_abort keep the iscsi_task ref if the TMF times out, so
qedi does not have to worry about if the task is being freed while in use
and does not need to get its own ref.

Link: https://lore.kernel.org/r/20210525181821.7617-18-michael.christie@oracle.com
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
For aborts, qedi needs to cleanup the FW then send the TMF from a worker
thread. While it's doing these the cmd could complete normally and the TMF
could time out. libiscsi would then complete the iscsi_task which will call
into the driver to cleanup the driver level resources while it still might
be accessing them for the cleanup/abort.

This has iscsi_eh_abort keep the iscsi_task ref if the TMF times out, so
qedi does not have to worry about if the task is being freed while in use
and does not need to get its own ref.

Link: https://lore.kernel.org/r/20210525181821.7617-18-michael.christie@oracle.com
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>scsi: iscsi: Fix conn use after free during resets</title>
<updated>2021-06-02T05:28:21+00:00</updated>
<author>
<name>Mike Christie</name>
<email>michael.christie@oracle.com</email>
</author>
<published>2021-05-25T18:18:06+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=ec29d0ac29be366450a7faffbcf8cba3a6a3b506'/>
<id>ec29d0ac29be366450a7faffbcf8cba3a6a3b506</id>
<content type='text'>
If we haven't done a unbind target call we can race where
iscsi_conn_teardown wakes up the EH thread and then frees the conn while
those threads are still accessing the conn ehwait.

We can only do one TMF per session so this just moves the TMF fields from
the conn to the session. We can then rely on the
iscsi_session_teardown-&gt;iscsi_remove_session-&gt;__iscsi_unbind_session call
to remove the target and it's devices, and know after that point there is
no device or scsi-ml callout trying to access the session.

Link: https://lore.kernel.org/r/20210525181821.7617-14-michael.christie@oracle.com
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
If we haven't done a unbind target call we can race where
iscsi_conn_teardown wakes up the EH thread and then frees the conn while
those threads are still accessing the conn ehwait.

We can only do one TMF per session so this just moves the TMF fields from
the conn to the session. We can then rely on the
iscsi_session_teardown-&gt;iscsi_remove_session-&gt;__iscsi_unbind_session call
to remove the target and it's devices, and know after that point there is
no device or scsi-ml callout trying to access the session.

Link: https://lore.kernel.org/r/20210525181821.7617-14-michael.christie@oracle.com
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>scsi: iscsi: Stop queueing during ep_disconnect</title>
<updated>2021-06-02T05:28:19+00:00</updated>
<author>
<name>Mike Christie</name>
<email>michael.christie@oracle.com</email>
</author>
<published>2021-05-25T18:17:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=891e2639deae721dc43764a44fa255890dc34313'/>
<id>891e2639deae721dc43764a44fa255890dc34313</id>
<content type='text'>
During ep_disconnect we have been doing iscsi_suspend_tx/queue to block new
I/O but every driver except cxgbi and iscsi_tcp can still get I/O from
__iscsi_conn_send_pdu() if we haven't called iscsi_conn_failure() before
ep_disconnect. This could happen if we were terminating the session, and
the logout timed out before it was even sent to libiscsi.

Fix the issue by adding a helper which reverses the bind_conn call that
allows new I/O to be queued. Drivers implementing ep_disconnect can use this
to make sure new I/O is not queued to them when handling the disconnect.

Link: https://lore.kernel.org/r/20210525181821.7617-3-michael.christie@oracle.com
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
During ep_disconnect we have been doing iscsi_suspend_tx/queue to block new
I/O but every driver except cxgbi and iscsi_tcp can still get I/O from
__iscsi_conn_send_pdu() if we haven't called iscsi_conn_failure() before
ep_disconnect. This could happen if we were terminating the session, and
the logout timed out before it was even sent to libiscsi.

Fix the issue by adding a helper which reverses the bind_conn call that
allows new I/O to be queued. Drivers implementing ep_disconnect can use this
to make sure new I/O is not queued to them when handling the disconnect.

Link: https://lore.kernel.org/r/20210525181821.7617-3-michael.christie@oracle.com
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>scsi: iscsi: Add task completion helper</title>
<updated>2021-06-02T05:28:19+00:00</updated>
<author>
<name>Mike Christie</name>
<email>michael.christie@oracle.com</email>
</author>
<published>2021-05-25T18:17:54+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=1486a4f5c2f35da5743b56037b8bbfb4eb38fa61'/>
<id>1486a4f5c2f35da5743b56037b8bbfb4eb38fa61</id>
<content type='text'>
This adds a helper to detect if a cmd has completed but is not yet freed.

Link: https://lore.kernel.org/r/20210525181821.7617-2-michael.christie@oracle.com
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This adds a helper to detect if a cmd has completed but is not yet freed.

Link: https://lore.kernel.org/r/20210525181821.7617-2-michael.christie@oracle.com
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>scsi: libiscsi: Add helper to calculate max SCSI cmds per session</title>
<updated>2021-02-09T03:39:04+00:00</updated>
<author>
<name>Mike Christie</name>
<email>michael.christie@oracle.com</email>
</author>
<published>2021-02-07T04:46:04+00:00</published>
<link rel='alternate' type='text/html' href='https://git.tavy.me/linux.git/commit/?id=b4046922b3c0740ad50a6e9c59e12f4dc43946d4'/>
<id>b4046922b3c0740ad50a6e9c59e12f4dc43946d4</id>
<content type='text'>
This patch just breaks out the code that calculates the number of SCSI cmds
that will be used for a SCSI session. It also adds a check that we don't go
over the host's can_queue value.

Link: https://lore.kernel.org/r/20210207044608.27585-6-michael.christie@oracle.com
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch just breaks out the code that calculates the number of SCSI cmds
that will be used for a SCSI session. It also adds a check that we don't go
over the host's can_queue value.

Link: https://lore.kernel.org/r/20210207044608.27585-6-michael.christie@oracle.com
Reviewed-by: Lee Duncan &lt;lduncan@suse.com&gt;
Signed-off-by: Mike Christie &lt;michael.christie@oracle.com&gt;
Signed-off-by: Martin K. Petersen &lt;martin.petersen@oracle.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
