linux.git/drivers/usb/gadget/function/f_tcm.c, branch v6.7

scsi: target: Have drivers report if they support direct submissions

2023-10-13T19:53:57+00:00

In some cases, like with multiple LUN targets or where the target has to
respond to transport level requests from the receiving context it can be
better to defer cmd submission to a helper thread. If the backend driver
blocks on something like request/tag allocation it can block the entire
target submission path and other LUs and transport IO on that session.

In other cases like single LUN targets with storage that can support all
the commands that the target can queue, then it's best to submit the cmd
to the backend from the target's cmd receiving context.

Subsequent commits will allow the user to config what they prefer, but
drivers like loop can't directly submit because they can be called from a
context that can't sleep. And, drivers like vhost-scsi can support direct
submission, but need to keep their default behavior of deferring execution
to avoid possible regressions where the backend can block.

Make the drivers tell LIO core if they support direct submissions and their
current default, so we can prevent users from misconfiguring the system and
initialize devices correctly.

Signed-off-by: Mike Christie 
Link: https://lore.kernel.org/r/20230928020907.5730-2-michael.christie@oracle.com
Signed-off-by: Martin K. Petersen

Merge patch series "target: TMF and recovery fixes"

2023-03-24T21:39:15+00:00

Mike Christie  says:

The following patches apply over Martin's 6.4 branches and Linus's tree.
They fix a couple regressions in iscsit that occur when there are TMRs
executing and a connection is closed. It also includes Dimitry's fixes in
related code paths for cmd cleanup when ERL2 is used and the write pending
hang during conn cleanup.

This version of the patchset brings it back to just regressions and fixes
for bugs we have a lot of users hitting. I'm going to fix isert and get it
hooked into iscsit properly in a second patchset, because this one was
getting so large. I've also moved my cleanup type of patches for a 3rd
patchset.

Link: https://lore.kernel.org/r/20230319015620.96006-1-michael.christie@oracle.com
Signed-off-by: Martin K. Petersen

scsi: target: Pass in cmd counter to use during cmd setup

2023-03-24T21:32:23+00:00

Allow target_get_sess_cmd() users to pass in the cmd counter they want to
use. Right now we pass in the session's cmd counter but in a subsequent
commit iSCSI will switch from per session to per conn.

Signed-off-by: Mike Christie 
Link: https://lore.kernel.org/r/20230319015620.96006-4-michael.christie@oracle.com
Signed-off-by: Martin K. Petersen

scsi: usb: gadget: f_tcm: Remove default fabric ops callouts

2023-03-17T03:36:36+00:00

Remove callouts that are identical to the default implementations in TCM
Core.

Signed-off-by: Dmitry Bogdanov 
Link: https://lore.kernel.org/r/20230313181110.20566-8-d.bogdanov@yadro.com
Acked-by: Greg Kroah-Hartman 
Signed-off-by: Martin K. Petersen

usb: gadget: add _init/__exit annotations to module init/exit funcs

2022-09-22T13:52:29+00:00

Add missing _init/__exit annotations to module init/exit funcs.

Signed-off-by: Xiu Jianfeng 
Link: https://lore.kernel.org/r/20220917084413.23957-1-xiujianfeng@huawei.com
Signed-off-by: Greg Kroah-Hartman

scsi: target: usb: Replace enable attr with ops.enable

2021-10-05T03:27:39+00:00

Remove tpg/enable attribute.  Add fabric ops enable_tpg implementation
instead.

Link: https://lore.kernel.org/r/20210910084133.17956-8-d.bogdanov@yadro.com
Reviewed-by: Roman Bolshakov 
Reviewed-by: Mike Christie 
Signed-off-by: Dmitry Bogdanov 
Signed-off-by: Martin K. Petersen

usb: fix various gadgets null ptr deref on 10gbps cabling.

2021-06-09T08:37:13+00:00

This avoids a null pointer dereference in
f_{ecm,eem,hid,loopback,printer,rndis,serial,sourcesink,subset,tcm}
by simply reusing the 5gbps config for 10gbps.

Fixes: eaef50c76057 ("usb: gadget: Update usb_assign_descriptors for SuperSpeedPlus")
Cc: Christophe JAILLET 
Cc: Felipe Balbi 
Cc: Gustavo A. R. Silva 
Cc: Lorenzo Colitti 
Cc: Martin K. Petersen 
Cc: Michael R Sweet 
Cc: Mike Christie 
Cc: Pawel Laszczak 
Cc: Peter Chen 
Cc: Sudhakar Panneerselvam 
Cc: Wei Ming Chen 
Cc: Will McVicker 
Cc: Zqiang 
Reviewed-By: Lorenzo Colitti 
Cc: stable 
Signed-off-by: Maciej Żenczykowski 
Link: https://lore.kernel.org/r/20210608044141.3898496-1-zenczykowski@gmail.com
Signed-off-by: Greg Kroah-Hartman

scsi: target: usb: gadget: Convert to new submission API

2021-03-04T22:37:01+00:00

target_submit_cmd() is now only for simple drivers that do their own sync
during shutdown and do not use target_stop_session(). It will never return
a failure, so we can remove that code from the driver.

Note: Before these patches target_submit_cmd() would never return an error
for usb since it does not use target_stop_session(). If it did then we
would have hit a refcount error here:

    transport_send_check_condition_and_sense(se_cmd,
                             TCM_UNSUPPORTED_SCSI_OPCODE, 1);
    transport_generic_free_cmd(&cmd->se_cmd, 0);

transport_send_check_condition_and_sense() calls queue_status and the
driver can sometimes do transport_generic_free_cmd() from there via
uasp_status_data_cmpl(). In that case, the above
transport_generic_free_cmd() would then hit a refcount error.

So that other use of the above error path in the driver is also probably
wrong, but someone with the hardware needs to fix that.

Link: https://lore.kernel.org/r/20210227170006.5077-11-michael.christie@oracle.com
Cc: Felipe Balbi 
Reviewed-by: Christoph Hellwig 
Signed-off-by: Mike Christie 
Signed-off-by: Martin K. Petersen

scsi: target: core: Rename transport_init_se_cmd()

2021-03-04T22:37:00+00:00

Rename transport_init_se_cmd() to __target_init_cmd() to reflect that it is
more of an internal function that drivers should normally not use and
because we are going to add a new init function in the next patches.

Link: https://lore.kernel.org/r/20210227170006.5077-4-michael.christie@oracle.com
Reviewed-by: Christoph Hellwig 
Reviewed-by: Himanshu Madhani 
Signed-off-by: Mike Christie 
Signed-off-by: Martin K. Petersen

usb: gadget: config_ep_by_speed_and_alt instead of config_ep_by_speed

2020-10-02T06:57:40+00:00

This patch replace config_ep_by_speed with config_ep_by_speed_and_alt.
This change allows to select proper usb_ss_ep_comp_descriptor for each
stream capable endpoints.

f_tcm function for SS use array of headers for both BOT/UAS alternate
setting:

static struct usb_descriptor_header *uasp_ss_function_desc[] = {
        (struct usb_descriptor_header *) &bot_intf_desc,
        (struct usb_descriptor_header *) &uasp_ss_bi_desc,
        (struct usb_descriptor_header *) &bot_bi_ep_comp_desc,
        (struct usb_descriptor_header *) &uasp_ss_bo_desc,
        (struct usb_descriptor_header *) &bot_bo_ep_comp_desc,

        (struct usb_descriptor_header *) &uasp_intf_desc,
        (struct usb_descriptor_header *) &uasp_ss_bi_desc,
        (struct usb_descriptor_header *) &uasp_bi_ep_comp_desc,
        (struct usb_descriptor_header *) &uasp_bi_pipe_desc,
        (struct usb_descriptor_header *) &uasp_ss_bo_desc,
        (struct usb_descriptor_header *) &uasp_bo_ep_comp_desc,
        (struct usb_descriptor_header *) &uasp_bo_pipe_desc,
        (struct usb_descriptor_header *) &uasp_ss_status_desc,
        (struct usb_descriptor_header *) &uasp_status_in_ep_comp_desc,
        (struct usb_descriptor_header *) &uasp_status_pipe_desc,
        (struct usb_descriptor_header *) &uasp_ss_cmd_desc,
        (struct usb_descriptor_header *) &uasp_cmd_comp_desc,
        (struct usb_descriptor_header *) &uasp_cmd_pipe_desc,
        NULL,
};

The first 5 descriptors are associated with BOT alternate setting,
and others are associated with UAS.

During handling UAS alternate setting f_tcm driver invokes
config_ep_by_speed and this function sets incorrect companion endpoint
descriptor in usb_ep object.

Instead setting ep->comp_desc to uasp_bi_ep_comp_desc function in this
case set ep->comp_desc to bot_uasp_ss_bi_desc.

And in result it uses the descriptor from BOT alternate setting
instead UAS.

Finally, it causes that controller driver during enabling endpoints
detect that just enabled endpoint for bot.

Signed-off-by: Jayshri Pawar 
Signed-off-by: Pawel Laszczak 
Reviewed-by: Peter Chen 
Signed-off-by: Felipe Balbi